点击显示 收起
【关键词】 kidney
The Water and Salt Research Center and Institute of Anatomy, University of Aarhus, Aarhus C
Department of Physiology, College of Medicine, University of Arizona, Tucson, Arizona
Department of Biochemistry and Cell Biology, School of Medicine, Kyungpook National University, Taegu, Korea
Institute of Clinical Medicine, Aarhus University Hospital, Aarhus N, Denmark
ABSTRACT
Lithium treatment is associated with development of nephrogenic diabetes insipidus, caused in part by downregulation of collecting duct aquaporin-2 (AQP2) and AQP3 expression. In the present study, we carried out cDNA microarray screening of gene expression in the inner medulla (IM) of lithium-treated and control rats, and selected genes were then investigated at the protein level by immunoblotting and/or immunohistochemistry. The following genes exhibited significantly altered transcription and mRNA expression levels, and these were compatible with the changes in protein expression. 11-Hydroxysteroid dehydrogenase type 2 protein expression in the IM was markedly increased (198 ± 25% of controls, n = 6), and immunocytochemistry demonstrated an increased labeling of IM collecting duct (IMCD) principal cells. This indicated altered renal mineralocorticoid/glucocorticoid responses in lithium-treated rats. The inhibitor of cyclin-dependent kinases p27 (KIP) protein expression was significantly decreased or undetectable in the IMCD cells, pointing to increased cellular proliferation and remodeling. Heat shock protein 27 protein expression was decreased in the IM (64 ± 6% of controls, n = 6), likely to be associated with the decreased medullary osmolality in lithium-treated rats. Consistent with this, lens aldose reductase protein expression was markedly decreased in the IM (16 ± 2% of controls, n = 6), and immunocytochemistry revealed decreased expression in the thin limb cells in the middle and terminal parts of the IM. Ezrin protein expression was upregulated in the IM (158 ± 16% of controls, n = 6), where it was predominantly expressed in the apical and cytoplasmic domain of the IMCD cells. Increased ezrin expression indicated remodeling of the actin cytoskeleton and/or altered regulation of IMCD transporters. In conclusion, the present study demonstrates changes in gene expression not only in the collecting duct but also in the thin limb of the loop of Henle in the IM, and several of these genes are linked to altered sodium and water reabsorption, cell cycling, and changes in interstitial osmolality.
nephrogenic diabetes insipidus; 11-hydroxysteroid dehydrogenase type 2; aldose reductase; inhibitor of cyclin-dependent kinases p27; heat shock protein 27; urine concentration
LITHIUM HAS BEEN WIDELY USED for treating bipolar affective disorders in human patients. However, lithium treatment is associated with a variety of renal side effects, including nephrogenic diabetes insipidus (NDI; i.e., a pronounced vasopressin-resistant polyuria and inability to concentrate urine) (16, 19), increased urinary sodium excretion (23), and distal renal tubular acidosis (14). We have recently examined the underlying cellular and molecular mechanisms for these side effects of long-term lithium treatment (14, 16, 19, 23). In particular, lithium-induced NDI is associated with dysregulation of collecting duct water channels [aquaporin-2 (AQP2) and AQP3] and epithelial sodium channel (ENaC) subunits, which play a major role in the water and sodium reabsorption in the collecting ducts (16, 19). Moreover, decreased expression of inner medullary urea transporters UT-A1 (localized at the inner medullary collecting duct cells) and UT-B (descending vasa recta) has been demonstrated to be associated with the reduced urine concentrating ability in lithium-treated rats (15). Thus downregulation of AQP2, AQP3, and UT-A1 in the collecting ducts and dysregulation of ENaC subunits in the collecting ducts appear to play a major role in the development of lithium-induced polyuria, decreased urinary concentration, and increased urinary sodium excretion. In contrast, no major changes were observed in the expression of renal sodium transporters in the thick ascending limbs [e.g., the Na-K-2Cl cotransporter BSC-1, type 3 Na/H exchanger (NHE3), and Na-K-ATPase], where active NaCl reabsorption occurs as a consequence of countercurrent multiplication (16, 23). Moreover, the expression of proximal nephron water channel AQP1 was not altered in lithium-induced NDI, further supporting the notion that lithium-induced NDI is likely due to altered collecting duct function. Additionally, we have recently observed that the cellular composition of collecting ducts undergoes substantial modifications in response to chronic lithium treatment, in addition to the changes in protein expression at the transcriptional level (6). Thus changes in the expression of genes and proteins involved in the cell proliferation, apoptosis, dedifferentiation, or cell hypertrophy are likely to be associated with chronic lithium treatment.
Recently, cDNA array experiments are carried out to identify possible new direct or indirect gene targets in the kidney that could be involved in the physiology and pathophysiology of urinary concentration. In particular, this approach was employed to reveal genes regulated in kidney inner medulla in response to vasopressin infusion in Brattleboro rats, which lack endogenous vasopressin (4). The expression of several transcripts including 11-hydroxysteroid dehydrogenase type 2 (11-HSD2), neurofibromin, casein kinase II, the -subunit of the epithelial sodium channel (-ENaC), AQP3, and c-Fos were increased in the inner medulla. This indicates that the changes in the expression of these proteins could be involved in the regulation of water and sodium excretion and reabsorption in response to high plasma vasopressin level. In particular, 11-HSD2 is believed to play a role in the regulation of ion transport in the renal collecting duct through its ability to break down glucocorticoids (cortisol in humans and corticosterone in rodents) to inactive forms. The upregulation of inner medullary 11-HSD2 gene expression in response to long-term vasopressin treatment therefore suggests possible implications for the regulation of transporter proteins in the collecting duct, which are recognized targets for glucocorticoids, mineralocorticoids, and vasopressin. This includes Na-K-ATPase (10), ENaC (28), AQP3 (17), and UT-A1 (25), in addition to AQP2 (19).
In the present study, we used the same approach to further identify the changes in the expression of genes in the inner medulla of lithium-treated rats, which are potentially involved in the development of NDI or in a compensatory process. Selected genes were investigated at the protein level by semiquantitative immunoblotting and immunohistochemistry, including 11-HSD2, p27/KIP, heat shock protein 27 (Hsp27), lens aldose reductase (AR), and ezrin. 11-HSD2 is believed to play a role in the regulation of ion transport in the renal collecting duct. P27/KIP is an inhibitor of cyclin-dependent kinase, regulating progression through the cell cycle. Hsp 27 may act as a chaperone assisting protein folding and is involved in the process of cell adaptation to changes in tonicity. AR is an NADPH-dependent enzyme converting D-glucose to sorbitol and is also thought to protect cells against damage due to high osmolality of the inner medullary interstitium. Ezrin is a multifunctional protein that may be involved in the regulation of the actions of transporters and receptors. Thus changes in these genes in the inner medulla of rats with lithium-induced NDI could be associated with altered sodium and water reabsorption, cell proliferation, and the changes in interstitial osmolality.
MATERIALS AND METHODS
Animal Protocols
The animal protocols have been approved by an institutional animal care and use committee and by the boards of the Institute of Anatomy and Institute of Clinical Medicine, University of Aarhus, according to the licenses for care and use of experimental animals issued by the Danish Ministry of Justice.
Chronic lithium treatment. Two different rat strains were used for the experiments. For microarray experiments, male inbred Wistar-Kyoto rats (Taconic M&B, Ry, Denmark) were used to minimize the animal-to-animal variability in gene expression due to different genetic backgrounds of the individual animals. For the semiquantitative immunoblotting and immunocytochemistry, male Wistar rats (Taconic M&B), which were lithium treated according to the same protocol as the Wistar-Kyoto rats, were examined. The functional data were described previously in detail (23). The lithium-treated rats received a standard diet (Altromin, Lage, Germany) supplemented with 40 mmol of lithium chloride/kg of dry food for 4 wk. The control rats received a standard diet. The rats in both groups had free access to food and water. This protocol was previously shown to result in plasma lithium levels of 0.61 ± 0.02 mM (16). The Wistar-Kyoto rats receiving the lithium-containing diet (n = 5) were matched with sibling controls on a normal diet (n = 5). During the final 7 days of lithium treatment, the rats were transferred to metabolic cages and daily water intake and urine output were monitored. At the end of lithium treatment, the rats were anesthetized under halothane inhalation. A blood sample was collected from the portal vein for measurement of plasma sodium, osmolality, and creatinine. Plasma and urine sodium concentrations were measured by standard flame photometry and creatinine and osmolality by standard tests. For RNA or protein extraction, the kidneys were rapidly removed, dissected into different zones (cortex, inner stripe of outer medulla, and inner medulla), and homogenized in the appropriate buffer (see below). For immunocytochemistry, the kidneys were perfusion fixed as described below.
RNA Isolation
Total RNA was isolated from the kidney inner medullas using a Qiagen RNeasy Midi kit (Qiagen, Valencia, CA) according to the manufacturer's instructions. The RNA was treated with DNase on the column. The quality of the RNA was controlled by electrophoresis, and concentration was measured by spectrophotometry.
cDNA Array Analysis
5k-Rat (MWG-Biotech) glass microarrays were used for the expression analysis using procedures described previously (34) with minor modifications. Briefly, 50 μg of total RNA pooled from two control rats or two lithium-treated rats, respectively, were used for each labeling reaction. Aminoallyl-dUTP (Sigma) was incorporated into cDNA in the reverse-transcription reaction using a 2:3 ratio of aminoallyl-dUTP:dTTP, SuperScript II reverse transcriptase (Life Technologies), and oligo-dT-primer overnight at 42°C. Following reverse transcription, the RNA was degraded by adding NaOH to a final concentration 250 mM and heating the reaction at 65°C for 30 min, and afterwards it was neutralized by adding an equal amount of HCl. cDNA was purified using a Qiaquick PCR purification kit (Qiagen), except the buffers in the kit were substituted for phosphate-based wash buffer (5 mM phosphate buffer, pH 8, 80% ethanol), and phosphate-based elution buffer (4 mM phosphate buffer, pH 8). The purified cDNA was lyophilized and resuspended in 4.5 μl of 0.1 M sodium bicarbonate buffer, pH 9. The fluorescent dyes Cy-3 and Cy-5 esters (PA23001 and PA25001, Amersham Pharmacia) were resuspended in 72 μl DMSO. The dye-coupling reactions were conducted for 1 h at room temperature, followed by neutralization with sodium acetate. The labeled cDNA was purified using a Qiaquick PCR purification kit (Qiagen). The efficiency of reverse transcription and dye incorporation was controlled by measuring absorption at 260, 280, 550, and 650 nm. Before hybridization, the arrays were blocked for 1 h at 42°C in blocking buffer (50% deionized formamid, 5x SSC, 0.1% SDS, 0.5% acetylated BSA) followed by five 5-min washes in H2O. The labeled cDNA from matched control and lithium-treated animals was pooled, lyophilized, resuspended in 30 μl hybridization buffer (MWG-Biotech), heated for 2 min at 95°C, cooled to 42°C, and applied to the previously blocked glass array. The hybridization proceeded overnight at 42°C in a hybridization chamber. The following day, the array was washed in decreasing stringency buffers A (2x SSC, 0.1% SDS), B (1x SSC, 0.1% SDS), and C [0.5x SSC at 30°C for 5 min, followed by a 15-s wash in buffer D (0.2 x SSC at RT)]. The arrays were scanned using a GMS 418 Array Scanner and analyzed with Imagene software. The local background was subtracted from the spot intensity for each spot. The normalized spot intensity for each dye was calculated by adjusting for the overall intensity of all spots that passed the quality control. The extraction of the genes with a sufficient expression level and the computation of the average lithium/control signal ratio for three independent experiments, were calculated using Microsoft Excel software.
Protein Sample Preparation
The dissected inner medulla was minced finely and homogenized in dissection buffer (0.3 M sucrose, 25 mM imidazole, 1 mM EDTA, pH 7.2, containing 8.5 μM leupeptin, 1 mM phenylmethylsulfonyl fluoride) using an ultra-turrax T8 homogenizer (IKA Labortechnik, Staufen, Germany). For semiquantitative immunoblotting using antibodies against ezrin, AR, 11-HSD2, and Hsp27 (see below), the homogenate was centrifuged at 4,000 g for 15 min at 4°C to remove whole cells, nuclei, and mitochondria. The supernatant was then used for preparing gel samples by adding one-third volume of Laemmli sample buffer containing 2% SDS. For semiquantitative immunoblotting of p27, the kidney inner medullary homogenate was used for preparing gel samples.
SDS-PAGE and Immunoblotting
Samples of rat kidney membranes were loaded on 9 or 12% polyacrylamide minigels and run for 1.5 h at 130 V. After transfer by electroelution (100 V, 1 h) to nitrocellulose membranes, the membranes were blocked with 5% milk in PBS-T (80 mM Na2HPO4, 20 mM NaH2PO4, 100 mM NaCl, and 0.1% Tween 20, pH 7.5) for 1 h and incubated overnight at 4°C with primary antibodies, followed by incubation with secondary anti-rabbit (P448 diluted 1:3,000, DAKO, Glostrup, Denmark) or donkey anti-sheep IgG (1713035-147, diluted to 1:5,000, Jackson Laboratories) horseradish peroxidase-conjugated antibodies. The labeling was visualized by the enhanced chemiluminescence (ECL or ECL+plus) system and exposure to photographic film (Hyperfilm ECL, RPN3103K, Amersham Pharmacia Biotech, Little Chalfont, UK). The densitometry values were obtained by scanning the films, and band densities were quantitated with Scion Image software, using a "rolling ball" background subtraction filter.
Immunohistochemistry
The kidneys were fixed by retrograde perfusion via the aorta with 3% paraformaldehyde in 0.1 M cacodylate buffer, pH 7.4, and postfixed for 1 h in the same fixative. Kidney slices containing all kidney zones were dehydrated and embedded in paraffin. The paraffin-embedded tissues were cut at 2 μm on a rotary microtome (Leica, Heidelberg, Germany). For immunolabeling, the sections were dewaxed with xylene and rehydrated with graded ethanol. Endogenous peroxidase activity was blocked with 0.5% H2O2 in absolute methanol for 10 min. In a microwave oven, the sections were boiled in a target retrieval solution (1 mM Tris, pH 9.0, with 0.5 mM EGTA) for 10 min. After cooling of the sections, nonspecific binding was blocked with 50 mM NH4Cl in PBS for 30 min followed by 3 x 10 min with PBS blocking-buffer containing 1% BSA, 0.05% saponin, and 0.2% gelatin. The sections were incubated with primary antibody (diluted in PBS with 0.1% BSA and 0.3% Triton X-100) overnight at 4°C. The sections were washed for 3 x 10 min with PBS wash buffer containing 0.1% BSA, 0.05% saponin, and 0.2% gelatin and incubated with appropriate horseradish peroxidase-conjugated secondary antibody for 1 h at room temperature. After 3 x 10-min rinses with PBS wash buffer, the sites of antibody-antigen reaction were visualized with a brown chromogen produced within 10 min by incubation with 0.05% 3,3'-diaminobenzidine tetrachloride (Kem-en-Tek, Copenhagen, Denmark) dissolved in distilled water with 0.1% H2O2. Mayer's hematoxylin was used for counterstaining, and after dehydration coverslips were mounted with hydrophobic medium (Eukitt, Kindler, Freiburg, Germany). Light microscopy was carried out with Leica DMRE (Leica Microsystems, Herlev, Denmark).
Antibodies
In this study, the following polyclonal antibodies were used: sheep antibody against rat 11-HSD2 was purchased from Chemicon (AB1296). It has previously been characterized for use in the kidney for immunoblotting and immunohistochemistry (4). In the current study, the antibody was used at a 1:1,500 dilution for both immunohistochemistry and immunoblotting.
Affinity-purified goat polyclonal antibody against rat lens AR was a kind gift from Dr. Peter Kador. The antibody was previously characterized in detail (including in Ref. 30). It was used for immunoblotting at a dilution of 1:400,000 and for immunohistochemistry at a dilution of 1:500,000.
The affinity-purified goat polyclonal antibody against Hsp27 was purchased from Santa Cruz Biotechnology (sc-1049; citations regarding previous characterizations are listed at http//www.scbt.com). It was used for immunohistochemistry at 1:80 dilution, and for immunoblotting at a 1:100 dilution. The rabbit polyclonal antibody against P27 was purchased from Abcam (ab7961). It was used at a 1:1,000 dilution for both immunohistochemistry and immunoblotting. The affinity-purified goat polyclonal antibody against human ezrin was purchased from Santa Cruz Biotechnology (sc-6409; citations regarding previous characterizations are listed at http://www.scbt.com). It was used at a 1:100 dilution for immunohistochemistry and immunoblotting.
RESULTS
Lithium Treatment-Induced Severe NDI
The Wistar-Kyoto rats that received standard rat chow containing 40 mmol lithium/kg food for 4 wk showed a marked polyuria and decreased urine osmolality (Table 1) compared with control rats not receiving lithium, consistent with the previously demonstrated lithium-induced NDI (16, 23).
View this table:
Expression Microarray Analyses Showed Altered Gene Expression in Inner Medulla
To assess the changes of gene expression in kidney inner medulla in response to chronic lithium treatment, we utilized a microarray containing 50-nucleotide-long sequences of 5,000 known rat genes (5k-rat, MWG-Biotech). Figure 1A shows the average signal intensity for each gene (corresponding to gene expression level) vs. the ratio of normalized mean signal intensities of the red (lithium) and green (control) fluorescence, showing general linearity of the measured signal ratios over six orders of magnitude, with a weak dye bias for the most strongly expressed genes (red signal appeared generally stronger than the green signal for those genes). The expression levels of the majority of the genes were predominantly unchanged between lithium-treated and control rats, or changed only marginally (Fig. 1A). The arrows in Fig. 1A indicate two examples of genes exhibiting significant changes in gene expression: 11-HSD2 showed a threefold increase in response to chronic lithium treatment and the AR gene showed a twofold decrease in expression. Figure 1, B and C, illustrates the appearance of the hybridization signals corresponding to the 11-HSD2 and AR genes.
The average changes in the expression level for each gene were determined from three independent experiments. Table 2 lists some of the genes for which the expression levels were changed; however, we omitted listing genes where dye bias was suspected. The full data set can be seen at the GEO site http://www.ncbi.nlm.nih.gov/geo/ (accession nos. GSE1341, GSM21824-GSM21826).
View this table:
Corresponding Changes Between mRNA and Protein Expression for Five Selected Genes
Changes in mRNA expression are often considered to be associated with similar changes in protein abundance and, thereby, physiological changes. Altered gene transcription and mRNA expression are, however, not always associated with a corresponding change in protein abundance (11). We therefore selected five genes (see below) from the cDNA microarray analysis that were significantly changed in response to chronic lithium treatment and carried out an analysis of protein expression using semiquantitative immunoblotting and immunohistochemistry. The selected genes are each potentially involved in the pathology of lithium-induced NDI. The proposed mechanisms and the consequences of regulation are presented in detail in DISCUSSION.
11-HSD2. 11-HSD2 is involved in regulation of sodium excretion by breaking down the active glucocorticoids. The marked increase in 11-HSD2 mRNA expression in the inner medulla of the lithium-treated rats (Fig. 1, A and B, and Table 2) was associated with significantly increased 11-HSD2 protein abundance (198 ± 25% of the control level 100 ± 26%, P < 0.05, Fig. 2, A and B). Consistently, immunohistochemistry showed labeling of 11-HSD2 in the IMCD cells throughout the initial and the middle parts of the inner medulla in the lithium-treated rats (Fig. 2, D and F). In contrast, in the control rats the 11-HSD2 labeling was significantly higher in the IMCD cells in the initial part of the inner medulla (initial part of IM1 or initial IMCD), whereas the IMCD cells in the IM2 and M3 regions were weakly or unlabeled, (Fig. 2, C and E).
In addition to the changes in the inner medulla, the expression of 11-HSD2 was analyzed in the kidney cortex and outer medulla. In contrast to the increased abundance in the inner medulla, 11-HSD2 protein abundance was dramatically decreased in both the kidney cortex (34 ± 4% of controls, P < 0.05) and in the inner stripe of the outer medulla (36 ± 12% of the control levels, P < 0.05) of lithium-treated rats (Fig. 3, AD). Immunoperoxidase labeling was consistent with immunoblotting results. In the kidney cortex of control rats, strong 11-HSD2 labeling was observed in the connecting tubule (CNT; not shown) and cortical collecting duct ( in Fig. 3E),with exclusive labeling of CNT and principal cells, respectively. Less abundant labeling was also noted in distal convoluted tubule cells compared with CNT in the same rat (not shown), consistent with previous observations (3, 8, 27). In lithium-treated animals, there was distinctly less 11-HSD2 labeling in the cortical collecting duct ( in Fig. 3F) but not in CNT and distal collecting duct (not shown). In the outer medullary collecting duct, 11-HSD2 labeling of principal cells was also reduced compared with control rats (Fig. 3, G and H).
AR. AR is an enzyme producing the organic osmolyte sorbitol that protects the cells against damage by high electrolyte concentrations in the inner medullary interstitium. Microarray analysis indicated a decreased level of lens AR transcript in the inner medulla of lithium-treated rats. We confirmed this finding at the protein level. Semiquantitative immunoblotting revealed a decrease in protein abundance of AR in the inner medulla down to 16 ± 2% of controls (P < 0.05, Fig. 4A).
Immunohistochemistry of control rat kidneys revealed AR staining of the thin limb cells in the inner medulla (arrows in Fig. 5, A and C), consistent with a previous study (13). Additionally, in the terminal one-third of the inner medulla cytoplasmatic staining of the IMCD cells and interstitial cells was observed, with labeling intensity increasing toward the papillary tip (not shown). In the kidneys from lithium-treated rats, the intensity of staining of the thin limb segment was markedly decreased in the middle (arrows in Fig. 5D) and terminal parts (not shown) of the inner medulla but remained unchanged in the initial part of the inner medulla (arrows in Fig. 5B). In contrast, the immunolabeling of IMCD and interstitial cells remained unchanged in the lithium-treated rats (not shown).
Hsp27. Hsp27 has a potential function in cell adaptation to changes of tonicity in the inner medulla. Microarray analysis showed a decrease in the mRNA level of Hsp27 in the inner medulla in response to prolonged lithium treatment compared with controls. Immunoblotting revealed a significant decrease in Hsp27 protein abundance in kidney inner medulla of lithium-treated rats (64 ± 6% of control levels, P < 0.05, Fig. 4B).
In control rats, immunohistochemistry showed that in the cortex and outer medulla Hsp27 was only present in the vascular endothelial cells (not shown). In contrast, in the inner medulla Hsp27 was detected in the cytoplasm in different types of cells: in the IMCD cells with the highest concentration near the apical membrane, in thin limb epithelial cells, in vascular endothelial cells, and in interstitial cells (not shown). Hsp27 showed a characteristic pattern of expression, with increasing labeling toward the papillary tip (i.e., following the increase in interstitial osmolality). In the lithium-treated animals, the labeling of Hsp27 in the IMCD cells was decreased (not shown). In contrast, the level of Hsp27 in the vascular endothelial cells, thin limbs epithelial cells, and interstitial cells appeared unchanged.
p27/KIP. p27/Kip is a regulatory protein governing the progression through the cell cycle, and its level determines resulting cell proliferation, cell hypertrophy, or apoptosis. Microarray analysis demonstrated a decrease in the p27 mRNA level in the kidney inner medulla after lithium treatment. Consistently, immunoblotting showed a decrease in p27 protein abundance down to 69 ± 8% of the control level (P < 0.05, Fig. 4C).
In the control rats, immunohistochemistry revealed that essentially all cell nuclei in the IMCD cells, thin limb cells, vascular endothelial cells, and the interstitial cells in the middle (Fig. 6, A and C) and terminal part of the inner medulla (papillary tip, not shown) were p27 positive. Very sporadically, p27-negative cells appeared in these cells (arrowheads in Fig. 6, A and C). In the inner medullas from the lithium-treated rats, there appeared to be an increased number of IMCD cells per tubular length, as previously demonstrated (12). Moreover, we observed a marked decrease in nuclear p27 labeling intensity in the IMCD cells and more frequently occurring p27-negative IMCD cells per tubular length (Fig. 6, B and D). Decreased p27 labeling intensity and an increased number of p27-negative IMCD cells were most pronounced in IM1 and the IM2 regions of the inner medulla (Fig. 6, B and D).
To determine the apparent increase in the number of cells per IMCD tubule and p27-negative IMCD cells more accurately, we conducted a semiquantitative cell-counting analysis of the total IMCD nuclei and p27-negative cells in the IM2 region of six lithium-treated and six control animals. This was done by counting nuclei in the area from 10 microscope images of each rat kidney taken across the inner medulla at similar distances from the IM1 region with identical magnification and microscope settings. The analysis revealed 2.5 times more IMCD nuclei in identical areas of inner medulla in lithium-treated rats compared with control rats (1,176 ± 48 vs. 458 ± 14 nuclei, P < 0.05), thus confirming the qualitative observations. Moreover, lithium-treated rats showed a fivefold increase in the fraction of p27-negative cells (0.4 ± 0.1% in the control to 1.9 ± 0.2% in the lithium-treated animals, P < 0.05). These results suggests a potential role of p27 in the proliferation of IMCD cells, as previously observed (6).
Ezrin. Ezrin is a multifunctional protein potentially modifying the actions of transporters and receptors by forming a "signaling complex" anchored in the actin cytoskeleton. Microarray analysis indicated an increase of ezrin mRNA level in the inner medulla of lithium-treated rats compared with control rats. This was also demonstrated by immunoblotting and immunohistochemistry. Immunoblotting revealed an increased ezrin protein abundance in the inner medulla (158 ± 16% of the controls, P < 0.05, Fig. 4D). Immunohistochemistry of the control kidneys showed apical staining of the IMCD cells (arrows in Fig. 7A) and, to a lesser extent, of the thin limbs in the inner medulla (not shown). In lithium-treated rats, the labeling intensity of the IMCD and the thin limbs was increased and this was most pronounced in the IM1 (not shown) and IM2 (Fig. 7B) regions of the inner medulla. Also, intracellular labeling of ezrin was observed in the IMCD cells in lithium-treated rats (arrows in Fig. 7B).
DISCUSSION
Microarray analysis of the gene expression profile in the kidneys from lithium-treated animals offers a unique possibility for finding novel genes that are potentially involved in the pathogenesis of lithium-induced NDI. In the present study, we carried out cDNA microarray screening of gene expression in the inner medulla of lithium-treated and control rats, and selected genes were investigated at the protein level by immunoblotting and immunohistochemistry. These genes are potentially involved in the development of NDI or are involved in compensatory processes that are activated to reduce excessive loss of water and electrolytes into urine.
It should be underscored that changes in mRNA expression may or may not necessarily be accompanied by similar changes in protein expression. Thus, analogous to a previous study (4), we have evaluated the changes in expression of selected genes by use of immunoblotting and immunohistochemistry. Each of these selected genes are discussed below.
11-HSD2
11-HSD2 is an enzyme expressed in mineralocorticoid-sensitive epithelia. It converts the glucocorticoid cortisol in humans or corticosterone in rodents to inactive keto-derivatives. 11-HSD2 deficiency, therefore, results in the syndrome of "apparent mineralocorticoid excess." In those patients, the free cortisol occupies and activates the mineralocorticoid receptor (MR) in the kidney and the colon, which mimics the condition with high plasma aldosterone levels. It has been demonstrated to be associated with increased expression of the thiazide-sensitive Na-Cl cotransporter (NCC), the -ENaC and the 70-kDa form of -ENaC in the kidney (20), and various other genes involved in sodium reabsorption in the colon, which results in abnormal high sodium reabsorption and hypertension.
In the lithium-treated rats, we observed an increase in the expression of 11-HSD2 in the inner medulla. An increase in 11-HSD2 transcript and protein level was previously observed in the inner medulla of the AVP-deficient Brattleboro rats (central NDI rat model) treated with vasopressin (4). Another study performed in isolated rat renal collecting ducts showed that AVP or DDAVP caused a rapid increase in 11-HSD2 catalytic activity, further potentiated by the treatment with aldosterone (1). Although it is known that plasma AVP levels in lithium-treated rats are increased potentially due to relative dehydration (9), it is unclear whether vasopressin contributes directly to the changes in 11-HSD2 expression in the inner medulla of lithium-treated rats. This would appear to be unlikely based on earlier studies demonstrating the inhibitory action of lithium on the signaling transduction pathways downstream of the vasopressin V2 receptor (i.e., inhibition of vasopressin-induced adenylyl cyclase activity and cAMP levels in medullary collecting duct, thereby decreasing vasopressin-regulated AQP2 expression) (7). Moreover, of particular interest there was a marked decrease in the expression of 11-HSD2 in the outer medulla and cortex, contrasting the opposite increased expression in the inner medulla. This is, to our knowledge, the first report of such a reciprocal regulation of this enzyme in different segments of the collecting duct system. The mechanisms behind this reciprocal regulation of 11-HSD2 in kidney collecting ducts remain unknown.
Similarly, we have recently demonstrated the selective and segment-specific downregulation of -ENaC and -ENaC in lithium-induced NDI (23). Rats with lithium-induced NDI had markedly reduced -ENaC and -ENaC protein abundances in the kidney cortex and outer medulla, compatible with the impaired ENaC regulation by aldosterone and vasopressin (23). In contrast, ENaC regulation in the CNT and IMCD was consistent with normal actions mediated by increased plasma aldosterone and vasopressin concentrations, i.e., increase in -ENaC and the 70-kDa form of -ENaC as well as an increase in the total -ENaC abundance (23). Thus future studies will be needed to define the underlying mechanism that governs the selective and segment-specific alteration of 11-HSD2 and ENaC subunits in lithium-induced NDI.
Lens AR
AR is an NADPH-dependent enzyme converting D-glucose to sorbitol. The intracellular accumulation of sorbitol is thought to protect the cells against damage due to high osmolality of the inner medullary interstitium during antidiuresis. Intracellular accumulation of sorbitol enables the cells to lower the intracellular concentration of electrolytes. AR labeling was associated with the ascending thin limb of the loop of Henle in the whole inner medulla and in the IMCD in the terminal part of the renal papilla, consistent with a previous study (13).
AR transcription is controlled by the binding of tonicity-responsive enhancer binding protein (TonEBP) to the tonicity-responsive enhancer (TonE) in the AR promoter. TonEBP also regulates other tonicity-responsive genes, e.g., transporters of organic osmolytes, including the sodium-myo-inositol cotransporter (26), taurine transporter (2), and sodium-chloride-betaine cotransporter (21); and certain cytokines (18), a molecular chaperone HSP702 (32), cyclooxygenase-2 (35), and, as recently shown, AQP2 (29) and UT-A (22).
Cell culture studies showed that TonEBP activity is regulated by several mechanisms, e.g., transcription rate, nuclear translocation, and phosphorylation of the trans-activating domain. However, in vivo under physiological conditions of diuresis or antidiuresis, TonEBP activity was shown to be regulated mainly by changing its subcellular distribution, nuclear or cytoplasmic, with no change in overall TonEBP abundance (5). Interestingly, in dehydrated rats nuclear translocation was most evident in the thin limbs, especially in the IM1 regions, whereas no change in distribution was observed in the IMCDs. TonEBP activity (measured as sodium-myo-inositol cotransporter mRNA level) also changed dramatically in response to dehydration or water loading in the thin limb in the IM1 region but remained unaffected in the IMCD.
In the present study, AR expression was decreased in the ascending thin limb in the IM2 and IM3 regions of the inner medulla in response to lithium treatment, whereas the ascending thin limb in the IM1 region maintained high AR expression level. This indicates that the mechanism of regulation of AR under normal physiological dehydration is different from the AR regulation in lithium-induced NDI, where interstitial osmolality is decreased (13).
p27
In the normal kidney, there is very little cell proliferation: the majority of the cells are in the quiescent phase G0. Due to various stimuli such as hyperglycemia, ischemia, urinary obstruction, or the presence of mitogens, cells can leave the quiescent state. The outcome could be cell proliferation, apoptosis, dedifferentiation, or cell hypertrophy, depending on the level and/or the phosphorylation state of various regulatory proteins controlling the progression through the cell cycle, e.g., cyclins, cyclin-dependent kinases (Cdk), and inhibitors of cyclin-dependent kinases.
In short, activation of Cdk4 by binding of cyclin D governs the transition from the G0 to the G1 phase. Cdk4 phosphorylates and thereby inactivates Rb protein, which results in activation of the E2F transcription factor and subsequently activation of genes necessary for cell-cycle progression. The cells increase in size and the amount of protein and RNA increases in anticipation of the DNA synthesis phase. The rise in cyclin D levels in the early G1 phase also serves to titrate the p27/Kip inhibitor protein away from cyclinE/Cdk2 complexes. The increase in the levels of active Cdk2/cyclinE governs the progression through the late G1 phase, whereas Cdk2/cyclinA activity is necessary for the entry into the S phase. However, if the level of p27 inhibitor in the cell is high due to some extracellular signals, it binds to and inactivates the Cdk2/cyclin complexes. This leads to cell-cycle arrest in the early G1 phase and cell hypertrophy without cell division, as observed in mesangial cells in the glomerulus with diabetes mellitus (31). p27 Is also involved in apoptosis, and in p27 knockout mice cell proliferation and apoptosis were increased following ureteral obstruction (24).
In the lithium-treated rats, we observed an increase in the number of cells composing the IMCD and a decrease in the level of p27. We often observed a cytoplasmic staining of p27, which may be associated with p27 ubiquitinization and degradation in the cytoplasm. Consistent with this, we have recently demonstrated evidence of increased proliferation of IMCD cells, with an increase in the fraction of intercalated cells in the collecting duct (6).
Hsp27
Hsp27 is widely expressed in many cell types in the presence or absence of stress. However, when the cells are stressed, the abundance of Hsp27 increases. Hsp27 functions as a chaperone, assisting protein folding. It is also involved in the remodeling of the actin cytoskeleton. This process is regulated by Hsp27 phosphorylation, resulting in the dissociation of Hsp27 from the barbed ends of actin microfilaments, allowing addition or removal of actin monomers. In the kidney inner medulla, Hsp27 has been suggested to participate in the process of cell adaptation to changes in tonicity, accompanied by cell swelling and shrinkage. Hsp27 distribution therefore follows the corticopapillary osmotic gradient; i.e., it is low in the cortex and high in the papilla.
The results of the present study demonstrated a decrease in Hsp27 abundance in IMCD cells in response to lithium treatment. It is therefore possible that the decreased expression of Hsp27 is secondary to the prolonged state of relatively low interstitial osmolality in the IM. It is noteworthy that the Hsp27 level in interstitial cells and thin limb cells remains at the control levels, which may indicate tonicity-independent pathways regulating Hsp27 level in those cells.
Ezrin
Ezrin is a multifunctional protein with the highest expression levels found in the proximal tubule, at the base of the microvilli, where it stabilizes microvillar structure by binding to actin filaments. Another proposed function of ezrin is tethering of transmembrane cell adhesion molecules to the actin cytoskeleton, e.g., ICAM-1, -2, -3, CD44, MBS, and CD43. Ezrin is also involved in the acute regulation of the actions of transporters and receptors. Due to the presence of multiple protein binding domains, it has been proposed that ezrin can bring different regulatory molecules, effectors, and linker molecules to form together a so-called "signaling complex" anchored in the actin cytoskeleton. The most studied signaling complex system in this context is NHE3 regulation by PKA through interactions with sodium/hydrogen exchanger-regulatory factor 1 and ezrin.
In the lithium-treated rats, we observed an increase in the expression level of ezrin in the inner medulla. In some cells, in the middle part of the inner medulla we observed cytoplasmic labeling of ezrin. It is possible that the change in ezrin localization in the cells reflects the changes in actin microfilament architecture secondary to the IMCD proliferation and morphological changes in the inner medulla. Similarly, ezrin upregulation has been detected in uterine or prostate tissues undergoing massive proliferation due to sex hormone treatment, and ezrin is downregulated in association with tissue reduction by apoptosis on hormone withdrawal (castration) (33). Moreover, ezrin upregulation has also been associated with a highly metastatic phenotype of cancer associated with increased cell motility. Alternatively, the change in both ezrin expression and intracellular distribution in lithium-treated rats may be an indicator of an altered function of some transporters or receptors in the IMCD.
Conclusions
The expression profiling of diseased vs. healthy tissue is a method of identifying genes potentially involved in the pathogenic processes. However, great care has to be taken when gene expression data are interpreted. One component might be the heterogeneous cellular composition of the studied organs or tissues that could be subject to hypertrophy/hyperplasia due to the pathogenic process. We demonstrated that chronic lithium treatment in rats caused major changes in the morphology of the inner medullary structures, consistent with our previous studies (6, 14). This also includes a major change in the fraction of principal/intercalated cells in the IMCD in response to lithium treatment (6). This can cause overrepresentation of certain transcripts specific for certain cell types, e.g., the intercalated cells. Such histological changes in tissue composition might make the identification of genes potentially involved in the etiology of the disease difficult. We therefore took the approach of using cDNA array analysis to provide a list of genes that are regulated, and selected genes were then investigated at the protein level by semiquantitative immunoblotting and immunohistochemistry.
In this study, we described the regulation of several genes in lithium-induced NDI and confirmed the change in the expression level by immunohistochemistry and immunoblotting. Further experiments are required to clarify the function of those and other genes in lithium-induced NDI and in other models of NDI.
GRANTS
The Water and Salt Research Center at the University of Aarhus is established and supported by the Danish National Research Foundation (Danmarks Grundforskningsfond). Support for this study was provided by The Karen Elise Jensen Foundation, The Commission of the European Union (QRLT-200000987 and QLRT-200000778), The Human Frontier Science Program, The Novo Nordisk Foundation The Danish Medical Research Council, The University of Aarhus Research Foundation, The Skovgaard Foundation, The Danish Research Academy, The University of Aarhus, and The Advanced Medical Technology Cluster for Diagnosis and Prediction at Kyungpook National University from MOCIE (T.-H. Kwon).
ACKNOWLEDGMENTS
The authors thank Lotte Vallentin Holbech, Ida Maria Jalk, Gitte Kall, Inger Merete Paulsen, Helle Hyer, Zhila Nikrozi, Mette Vistisen, and Dorte Wulff for expert technical assistance.
FOOTNOTES
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.
REFERENCES
Alfaidy N, Blot-Chabaud M, Bonvalet JP, and Farman N. Vasopressin potentiates mineralocorticoid selectivity by stimulating 11 beta hydroxysteroid dehydrogenase in rat collecting duct. J Clin Invest 100: 24372442, 1997.
Bitoun M, Levillain O, and Tappaz M. Gene expression of the taurine transporter and taurine biosynthetic enzymes in rat kidney after antidiuresis and salt loading. Pflügers Arch 442: 8795, 2001.
Bostanjoglo M, Reeves WB, Reilly RF, Velazquez H, Robertson N, Litwack G, Morsing P, Dorup J, Bachmann S, Ellison DH, and Bostonjoglo M. 11-Hydroxysteroid dehydrogenase, mineralocorticoid receptor, and thiazide-sensitive Na-Cl cotransporter expression by distal tubules. J Am Soc Nephrol 9: 13471358, 1998.
Brooks HL, Ageloff S, Kwon TH, Brandt W, Terris JM, Seth A, Michea L, Nielsen S, Fenton R, and Knepper MA. cDNA array identification of genes regulated in rat renal medulla in response to vasopressin infusion. Am J Physiol Renal Physiol 284: F218F228, 2003.
Cha JH, Woo SK, Han KH, Kim YH, Handler JS, Kim J, and Kwon HM. Hydration status affects nuclear distribution of transcription factor tonicity responsive enhancer binding protein in rat kidney. J Am Soc Nephrol 12: 22212230, 2001.
Christensen BM, Marples D, Kim YH, Wang W, Frkir J, and Nielsen S. Changes in cellular composition of kidney collecting duct cells in rats with lithium-induced NDI. Am J Physiol Cell Physiol 286: C952C964, 2004.
Christensen S, Kusano E, Yusufi AN, Murayama N, and Dousa TP. Pathogenesis of nephrogenic diabetes insipidus due to chronic administration of lithium in rats. J Clin Invest 75: 18691879, 1985.
Cole TJ. Cloning of the mouse 11-hydroxysteroid dehydrogenase type 2 gene: tissue specific expression and localization in distal convoluted tubules and collecting ducts of the kidney. Endocrinology 136: 46934696, 1995.
Davies J, Waller S, Zeng Q, Wells S, and Murphy D. Further delineation of the sequences required for the expression and physiological regulation of the vasopressin gene in transgenic rat hypothalamic magnocellular neurones. J Neuroendocrinol 15: 4250, 2003.
Feraille E and Doucet A. Sodium-potassium-adenosinetriphosphatase-dependent sodium transport in the kidney: hormonal control. Physiol Rev 81: 345418, 2001.
Ideker T, Thorsson V, Ranish JA, Christmas R, Buhler J, Eng JK, Bumgarner R, Goodlett DR, Aebersold R, and Hood L. Integrated genomic and proteomic analyses of a systematically perturbed metabolic network. Science 292: 929934, 2001.
Jacobsen NO, Olesen OV, Thomsen K, Ottosen PD, and Olsen S. Early changes in renal distal convoluted tubules and collecting ducts of lithium-treated rats: light microscopy, enzyme histochemistry, and 3H-thymidine autoradiography. Lab Invest 46: 298305, 1982.
Jung JY, Kim YH, Cha JH, Han KH, Kim MK, Madsen KM, and Kim J. Expression of aldose reductase in developing rat kidney. Am J Physiol Renal Physiol 283: F481F491, 2002.
Kim YH, Kwon TH, Christensen BM, Nielsen J, Wall SM, Madsen KM, Frkir J, and Nielsen S. Altered expression of renal acid-base transporters in rats with lithium-induced NDI. Am J Physiol Renal Physiol 285: F1244F1257, 2003.
Klein JD, Gunn RB, Roberts BR, and Sands JM. Down-regulation of urea transporters in the renal inner medulla of lithium-fed rats. Kidney Int 61: 9951002, 2002.
Kwon TH, Laursen UH, Marples D, Maunsbach AB, Knepper MA, Frkir J, and Nielsen S. Altered expression of renal AQPs and Na+ transporters in rats with lithium-induced NDI. Am J Physiol Renal Physiol 279: F552F564, 2000.
Kwon TH, Nielsen J, Masilamani S, Hager H, Knepper MA, Frkir J, and Nielsen S. Regulation of collecting duct AQP3 expression: response to mineralocorticoid. Am J Physiol Renal Physiol 283: F1403F1421, 2002.
Lopez-Rodriguez C, Aramburu J, Jin L, Rakeman AS, Michino M, and Rao A. Bridging the NFAT and NF-B families: NFAT5 dimerization regulates cytokine gene transcription in response to osmotic stress. Immunity 15: 4758, 2001.
Marples D, Christensen S, Christensen EI, Ottosen PD, and Nielsen S. Lithium-induced downregulation of aquaporin-2 water channel expression in rat kidney medulla. J Clin Invest 95: 18381845, 1995.
Masilamani S, Wang X, Kim GH, Brooks H, Nielsen J, Nielsen S, Nakamura K, Stokes JB, and Knepper MA. Time course of renal Na-K-ATPase, NHE3, NKCC2, NCC, and ENaC abundance changes with dietary NaCl restriction. Am J Physiol Renal Physiol 283: F648F657, 2002.
Miyakawa H, Woo SK, Chen CP, Dahl SC, Handler JS, and Kwon HM. Cis- and trans-acting factors regulating transcription of the BGT1 gene in response to hypertonicity. Am J Physiol Renal Physiol 274: F753F761, 1998.
Nakayama Y, Peng T, Sands JM, and Bagnasco SM. The TonE/TonEBP pathway mediates tonicity-responsive regulation of UT-A urea transporter expression. J Biol Chem 275: 3827538280, 2000.
Nielsen J, Kwon TH, Praetorius J, Kim YH, Frkir J, Knepper MA, and Nielsen S. Segment-specific ENaC downregulation in kidney of rats with lithium-induced NDI. Am J Physiol Renal Physiol 285: F1198F1209, 2003.
Ophascharoensuk V, Fero ML, Hughes J, Roberts JM, and Shankland SJ. The cyclin-dependent kinase inhibitor p27Kip1 safeguards against inflammatory injury. Nat Med 4: 575580, 1998.
Peng T, Sands JM, and Bagnasco SM. Glucocorticoids inhibit transcription and expression of the UT-A urea transporter gene. Am J Physiol Renal Physiol 282: F853F858, 2002.
Rim JS, Atta MG, Dahl SC, Berry GT, Handler JS, and Kwon HM. Transcription of the sodium/myo-inositol cotransporter gene is regulated by multiple tonicity-responsive enhancers spread over 50 kilobase pairs in the 5'-flanking region. J Biol Chem 273: 2061520621, 1998.
Stewart PM, Whorwood CB, Barber P, Gregory J, Monder C, Franklyn JA, and Sheppard MC. Localization of renal 11-dehydrogenase by in situ hybridization: autocrine not paracrine protector of the mineralocorticoid receptor. Endocrinology 128: 21292135, 1991.
Stokes JB and Sigmund RD. Regulation of rENaC mRNA by dietary NaCl and steroids: organ, tissue, and steroid heterogeneity. Am J Physiol Cell Physiol 274: C1699C1707, 1998.
Storm R, Klussmann E, Geelhaar A, Rosenthal W, and Maric K. Osmolality and solute composition are strong regulators of AQP2 expression in renal principal cells. Am J Physiol Renal Physiol 284: F189F198, 2003.
Terubayashi H, Sato S, Nishimura C, Kador PF, and Kinoshita JH. Localization of aldose and aldehyde reductase in the kidney. Kidney Int 36: 843851, 1989.
Wolf G, Reinking R, Zahner G, Stahl RA, and Shankland SJ. Erk 1,2 phosphorylates p27(Kip1): functional evidence for a role in high glucose-induced hypertrophy of mesangial cells. Diabetologia 46: 10901099, 2003.
Woo SK, Lee SD, Na KY, Park WK, and Kwon HM. TonEBP/NFAT5 stimulates transcription of HSP70 in response to hypertonicity. Mol Cell Biol 22: 57535760, 2002.
Wu X, Pang ST, Sahlin L, Blanck A, Norstedt G, and Flores-Morales A. Gene expression profiling of the effects of castration and estrogen treatment in the rat uterus. Biol Reprod 69: 13081317, 2003.
Yang IV, Chen E, Hasseman JP, Liang W, Frank BC, Wang S, Sharov V, Saeed AI, White J, Li J, Lee NH, Yeatman TJ, and Quackenbush J. Within the fold: assessing differential expression measures and reproducibility in microarray assays. Genome Biol 3: research 0062, 2002.
Zhao H, Tian W, Tai C, and Cohen DM. Hypertonic induction of COX-2 expression in renal medullary epithelial cells requires transactivation of the EGFR. Am J Physiol Renal Physiol 285: F281F288, 2003.