点击显示 收起
【摘要】 Vimentin, an intermediate filament protein mainly expressed in mesenchyma-derived cells, is reexpressed in renal tubular epithelial cells under many pathological conditions, characterized by intense cell proliferation. Whether vimentin reexpression is only a marker of cell dedifferentiation or is instrumental in the maintenance of cell structure and/or function is still unknown. Here, we used vimentin knockout mice ( Vim -/- ) and an experimental model of acute renal injury (30-min bilateral renal ischemia) to explore the role of vimentin. Bilateral renal ischemia induced an initial phase of acute tubular necrosis that did not require vimentin and was similar, in terms of morphological and functional changes, in Vim +/+ and Vim -/- mice. However, vimentin was essential to favor Na-glucose cotransporter 1 localization to brush-border membranes and to restore Na-glucose cotransport activity in regenerating tubular cells. We show that the effect of vimentin inactivation is specific and results in persistent glucosuria. We propose that vimentin is part of a structural network that favors carrier localization to plasma membranes to restore transport activity in injured kidneys.
【关键词】 recovery sodiumglucose cotransporter glucosuria
THE FUNCTIONAL ROLE of vimentin reexpression in epithelial cells represents a fascinating enigma in biology. Vimentin is a class III intermediate filament component of the cytoskeleton that is expressed differentially under physiological and pathological conditions as follows: in mesenchyma-derived cells in normal adult tissues and also in many epithelial proliferating cells during diseases, such as cancer or postischemic and toxic tissue regeneration ( 31, 39 ). In the mature kidney, vimentin is detected in glomeruli, vessels, and interstitial cells but not in tubular epithelial cells ( 2, 18 ). However, vimentin is expressed in proximal tubular cells during the recovery phase that follows ischemia or nephrotoxic tubular necrosis ( 15, 28, 41, 49, 51 ), in renal cell carcinoma ( 11, 19, 47 ), and in proximal tubular cells in culture ( 13, 17 ). In all these conditions, transient vimentin expression is considered as a marker of epithelial dedifferentiation ( 4 ), but the question still remained whether vimentin is directly involved in recovery of cell functions.
The paradigm of vimentin-null mice, which develop and reproduce without an obvious phenotype ( 9 ), shows that vimentin is not required for the survival of individuals and apparently argues against a major role of this protein, at least under normal physiological conditions. However, various cell culture models, including those derived from vimentin-null mice, suggest that vimentin could play a role in: 1 ) organization of other cytoplasmic structures, such as microtubules and microfilaments, 2 ) maintenance of cellular and nuclear shape and cell migration, 3 ) metabolism of lipoprotein-derived cholesterol and recycling of sphingosine and glycosphingolipids, 4 ) cell cycle (vimentin is considered as an immediate-early gene), and 5 ) interaction with several nonstructural proteins, such as protein kinase C or stress response proteins ( 12 ). Many of these functions are severely affected by an ischemic injury ( 5 ).
In the kidney, ischemia results in disruption of actin microfilament and microtubule networks ( 1, 22 ). This initiates a cascade of structural and functional alterations of proximal tubular cells, including disappearance of brush-border microvilli, opening of tight junctions, and loss of cell-matrix and cell-cell adhesions ( 25, 45, 48 ). Together, these changes induce the loss of lipid (i.e., cholesterol, sphingomyelin, and sphingolipids) and protein (i.e., Na-K-ATPase and adhesion molecules) polarity ( 27 ) that ultimately leads to decreased reabsorption of water and solutes by proximal tubular cells ( 24 ). Moreover, the abundance of Na-K-ATPase, in terms of mRNA and protein levels, is severely reduced after renal ischemia, and this contributes to the impairment of tubular reabsorption ( 23, 43, 50 ). Parallel, the synthesis of stress proteins ( 33, 44 ) and that of cytokines, chemoattractant chemokines, adhesion molecules, and nitric oxide ( 8, 10 ), increases in postischemic cells. Altogether, these molecules potentiate ischemia-induced renal injury, by favoring leukocyte infiltration. Next, surviving cells proliferate and migrate to restore the damaged tubules ( 40 ).
Here we have used vimentin knockout mice to examine the role of vimentin in functional recovery of proximal tubules during the regenerating phase that follows an acute tubular necrosis. We show that the morphological recovery of postischemic kidneys did not require vimentin. However, vimentin was essential to restore Na-glucose cotransport activity and to prevent glucosuria in postischemic mice. Moreover, we demonstrate that the effect of vimentin inactivation involved reduction of Na-glucose cotransporter 1 (SGLT1) localization to regenerating brush-border membranes (BBM).
MATERIALS AND METHODS
Animals
Mice with a mutated Vim allele have been previously described ( 9 ). Animals were fed ad libitum and housed in a room with constant ambient temperature and a 12:12-h light-dark cycle. All animal procedures were conducted in accordance with French government policies (Services Vétérinaires de la Santé et de la Production Animale, Ministère de l'Agriculture).
Experimental Protocol
Adult (2-3 mo) Vim +/+ and Vim -/- mice originating from the same litter were studied. Bilateral renal ischemia was performed as previously described, with minor modifications ( 38 ). Briefly, right and left flank incisions were made, and the renal pedicles were clamped with microaneurysm clamps (Moria, Paris, France) for 30 min. Reperfusion was assessed by visual examination of the kidneys, which recovered the usual color within 20-30 s. Surgery was performed under xylazine (Rompun 2%; Bayer Pharma, Puteaux, France) and ketamine (Clorketam 1000; Vétoquinol, Lure, France) anesthesia. During the duration of anesthesia, mice were maintained on a heating table (37°C) to avoid the fall in temperature.
Mice were killed at 2, 10, and 28 days after surgery. At each time point, 16-22 Vim +/+ and Vim -/- mice were examined. At the same time points, 16-22 nonoperated Vim +/+ and Vim -/- mice from the same gender and litter were analyzed and used as controls.
At the time of death, kidneys were removed. On recovery day 2 and 10, urine samples were collected using metabolic cages over the 24 h before death, and blood samples were obtained by cava puncture in 10 animals of each experimental group, for determination of Na, creatinine, and glucose concentrations.
The experimental protocol was chosen according to the results obtained in a pilot study that aimed at defining a model of transient bilateral postischemic tubular necrosis in mice. In this study, the two renal pedicles were clamped for 15, 30, and 50 min, and mice were killed at different times after surgery from day 1 to day 28. We observed that 50 min of clamping resulted in 100% mortality of mice within 3 days after ischemia, whereas 15 min of clamping induced only scarce and moderate renal changes. In contrast, 30 min of clamping resulted in a reversible (recovery on day 10 ) model of acute renal failure.
Renal Function
Serum and urine determinations were performed using a Monarch multiparametric autoanalyzer (Instrumentation Laboratory, Paris, France).
Renal Morphology
Kidneys from six mice for each experimental group and time point were frozen in 2-methylbutane and kept at -80°C. Section (4 µm thick) were fixed in 4% formaldehyde at room temperature for 10 min and stained with Periodic acid-Schiff.
Actin staining. Sections (4 µm thick) were fixed in 4% formaldehyde at room temperature for 10 min and permeabilized with 0.1% Triton X-100 for 5 min. Next, sections were incubated at 37°C for 20 min with phalloidin conjugated to TRITC (Sigma Aldrich Chimie, Saint Quentin Fallavier, France), diluted 1:1,000.
BBM Vesicle Preparation
Brush-border membrane vesicles (BBMV) were isolated as described by Booth and Kenny ( 6 ) according to the modifications of Biber et al. ( 3 ). Briefly, kidneys were removed, decapsulated, and rinsed in ice-cold Hanks' solution. The cortex from the two kidneys of four mice was separated from the medulla, and cortical slices were pooled and homogenized with a Ultra-turax T25 homogenizer (IKA; Werke, Janke and Kunkel, Staufen, Germany) in a buffer consisting of 300 mM mannitol, 5 mM EGTA, 12 mM Tris·HCl (pH 7.4), and a cocktail of anti-proteases (Complete; Roche, Meylan, France). BBMV were then prepared by the MgCl 2 precipitation and differential centrifugation procedures, and the final pellet was resuspended in a buffer containing 300 mM mannitol, 16 mM HEPES, and 10 mM Tris (ph 7.5). The enrichment of BBMV content of the precipitation was assessed by measuring enzymatic activities in the homogenates and in the final membrane preparations. A nine- and fourfold enrichment factor was routinely obtained for -glutamyltranspeptidase and Na-K-ATPase activity, respectively, in both Vim +/+ and Vim -/- membrane preparations. Because BBMV were prepared four times by pooling the two kidneys of four animals, 16 control and postischemic mice were studied for each genotype and experimental time point.
Enzymatic Activities
-Glutamyltranspeptidase (EC 2.3.2.2 ) activity was determined using an adaptation of the technique of Orlowski and Meister ( 29 ). Na-K-ATPase (EC 3.6.1.3 ) activity was determined according to the technique of Post et al. ( 30 ). Specificity was assessed using ouabain (8 mM).
BBMV Transport Studies
BBMV were equilibrated for 60 min at room temperature in a buffer containing 300 mM mannitol, 16 mM HEPES, and 10 mM Tris·HCl (pH 7.4). Total uptake was measured in an incubation media [150 mM NaCl, 16 mM HEPES, and 10 mM Tris·HCl (pH 7.4)] containing variable concentrations of the compound under study: - D -[ methyl -U- 14 C]glucopyranoside (293 mCi/mmol; Amersham Pharmacia Biotech, Orsay, France), D -[U- 14 C]galactose (200 mCi/mmol; Amersham Pharmacia Biotech), phosphorus-32 (10 mCi/ml; Amersham Pharmacia Biotech), and L -[2,3- 3 H]alanine (52 Ci/mmol; Amersham Pharmacia Biotech). After timed incubation at room temperature, ice-cold stop solution [300 mM mannitol, 10 mM Tris·HCl (pH 7.5), and 2 mM CaCl 2 ] was added, and samples were transferred to 0.65-µm filters (Gelman Laboratory, Pall France, Saint Germain en Laye, France). Filters were washed with ice-cold stop solution, and the radioactivity was counted in a LKB Wallac counter (model 1209; Rackbeta, Turku, Finland). Na-independent uptakes were determined in a medium where Na was substituted to N -methylglucamine (137 mM). Na-dependent uptakes were calculated as the difference between total and Na-independent uptakes.
Western Blot Analysis
Immunoblotting was performed as previously described ( 32 ). The primary antibodies used were: 1 ) a rabbit polyclonal anti-SGLT1 antibody (Chemicon, Temecula, CA), diluted 1:2,000; 2 ) a rabbit polyclonal anti-Npt2a antibody (kindly provided by H. Murer and J. Biber, Institut of Physiology, Zurich, Switzerland), diluted 1:6,000; 3 ) a mouse monoclonal anti-P-glycoprotein antibody (Calbiochem, San Diego, CA), diluted 1/100; 4 ) a mouse monoclonal anti- -actin antibody (Sigma Aldrich Chimie), diluted 1:5,000; and 5 ) a mouse monoclonal anti-villin antibody (kindly provided by S. Robine, Institut Curie, Paris, France), diluted 1:3,000. The secondary antibodies used were as follows: 1 ) a sheep anti-mouse horseradish peroxidase-linked Ig antibody (Amersham Pharmacia Biotech), diluted 1:5,000 for P-glycoprotein, 1:8,000 for actin, and 1:5,000 for villin; 2 ) a donkey anti-rabbit horseradish peroxidase-linked Ig antibody (Amersham Pharmacia Biotech), diluted 1:2,000 for SGLT1 and 1:6,000 for Npt2a.
Expression of Data and Statistical Analysis
Data are expressed as means ± SE. Differences between the experimental groups were evaluated using one-way ANOVA, followed, when significant, by the Bonferroni test.
RESULTS
Recovery of Na-glucose Cotransport Activity is Impaired in the Absence of Vimentin
We performed bilateral renal ischemia in Vim +/+ and Vim -/- mice and studied Na-glucose cotransport activity at 2, 10, and 28 days after surgery. As expected, uptake of -methylglucopyranoside (MGP), a nonmetabolized analog of D -glucose, decreased markedly in BBMV prepared from postischemic kidneys 2 days after renal ischemia ( Fig. 1 A ). The decrement was of the same magnitude in Vim +/+ (50 ± 5%) and Vim -/- (68 ± 8%) mice. In Vim +/+ postischemic kidneys, Na-glucose cotransport activity increased back to baseline levels on day 10 after ischemia. In contrast, MGP uptake remained significantly lower (64 ± 7%) in Vim -/- postischemic kidneys compared with controls ( Fig. 1 A ). In mutant mice, glucose uptake was still reduced (50 ± 4%) on recovery day 28. No difference in MGP uptake was detected in control nonoperated kidneys from Vim +/+ and Vim -/- mice ( Fig. 1 A ).
Fig. 1. Na-dependent cotransport activities in brush-border membrane vesicles (BBMV) from control (C) and postischemic kidneys of Vim +/+ (open bars) and Vim -/- (filled bars) mice. A : Na-glucose ( left ) and Na-galactose ( right ) cotransport activities. The uptake of -methylglucopyranoside (MGP) was evaluated at 2, 10, and 28 days after surgery, whereas the uptake of galactose was measured at day 10 after surgery. Because glucose uptake of control nonoperated kidneys was unchanged throughout the course of experiment, only one experimental time point is shown. The absence of vimentin impaired the recovery of Na-glucose cotransport activity in postischemic kidneys. B : Na-alanine ( left ) and Na-phosphate ( right ) cotransport activities on recovery day 10. Alanine and phosphate uptakes were affected neither by ischemia nor by the mouse genotype. Data are means ± SE of 4 separate experiments. Each experiment was performed in triplicate by pooling the two kidneys of 4 mice. Sixteen mice were analyzed for each experimental group and time point. Statistical analysis: ANOVA, postischemic vs. control kidneys: * P < 0.05 and ** P < 0.01; Vim -/- vs. Vim +/+ kidneys: P < 0.05 and P < 0.01.
Two specific cotransporters ensure mainly the uptake of glucose across BBM [SGLT1 and SGLT2 ( 52 )]. Although glucose is transported by both proteins, galactose is selectively taken up by SGLT1, but not by SGLT2 ( 21 ). Hence, we measured Na-galactose transport activity in Vim +/+ and Vim -/- postischemic kidneys. As shown in Fig. 1 A, at day 10 after ischemia, Na-galactose cotransport activity was reduced significantly in BBMV from Vim -/- mice compared with Vim +/+ mice.
Next, we evaluated whether the absence of vimentin affects other Na-dependent cotransport activities in postischemic kidneys. For this purpose, we measured Na-dependent neutral amino acid transport and Na-dependent phosphate transport activities in BBMV from Vim +/+ and Vim -/- mice. As shown in Fig. 1 B, on recovery day 10, Na-alanine and Na-phosphate cotransport activities were similar in postischemic and control nonoperated kidneys, and values did not differ in Vim +/+ and Vim -/- mice. The fact that both phosphate and alanine uptakes were not modified in Vim -/- postischemic kidneys made a primary effect of vimentin gene inactivation on Na-K-ATPase activity unlikely. Indeed, as shown in Table 1, ouabain-sensitive rubidium uptake, an ion that competes with K at Na-K-ATPase, was affected neither by ischemia nor by the mouse genotype on recovery day 10.
Table 1. Na-K-ATPase and -glutamyltranspeptidase activities after ischemia
Normal BBM Morphological Recovery in the Absence of Vimentin
The recovery of transport activities in postischemic kidneys is paralleled by the morphological renewal of BBM ( 35 ). Hence, we analyzed the changes of BBM throughout the recovery phase in Vim +/+ and Vim -/- mice. Phalloidin staining showed that actin microfilaments were concentrated in the terminal web and microvilli regions and, to a lesser extent, below the apical surface and along the basal membranes of proximal tubular cells in control nonoperated kidneys. Actin disappeared in BBM of many proximal tubules 2 days after injury ( Fig. 2 A ). After day 10 of ischemia, the actin network was completely reestablished, and the appearance of postischemic BBM could not be distinguished from that of control nonoperated kidneys. The morphology of BBM did not change thereafter. The same pattern was observed in Vim +/+ and Vim -/- postischemic tubules ( Fig. 2 A ). Western blots showed that the expression of villin and actin, two major constituents of BBM, paralleled the morphological changes ( Fig. 2 B ). Indeed, the abundance of both proteins was significantly decreased 2 days after ischemia, and the decrement was comparable in Vim +/+ and Vim -/- mice. On recovery day 10, BBM villin and actin content reached the values of control nonoperated animals and did not change thereafter, regardless of the mouse genotype.
Fig. 2. Morphology and composition of brush-border membranes (BBM) throughout the course of experiment in Vim +/+ and Vim -/- mice. A : phalloidin staining in control nonoperated ( a and e ) and postischemic ( b-d and f-h ) kidneys of Vim +/+ ( a - d ) and Vim -/- ( e-h ) mice 2, 10, and 28 days after ischemia. Because BBM appearance of control nonoperated kidneys was unchanged throughout the course of experiment, only one experimental time point is shown. Six mice were analyzed for each experimental group and time point. The absence of vimentin did not affect brush-border appearance at each experimental time point. Magnification, x 400 B : Western blot analysis of villin ( top ) and -actin ( bottom ) protein expression in BBM from control and postischemic kidneys of Vim +/+ (open bars) and Vim -/- (filled bars) mice. Membranes were incubated with a mouse monoclonal anti-villin antibody or a mouse monoclonal anti- -actin antibody. The absence of vimentin affected neither villin nor actin expression at each experimental time point. Blots are representative samples from 4 separate experiments, each performed by pooling the two kidneys of 4 mice. Sixteen mice were analyzed for each experimental group and time point. Data are means ± SE. Statistical analysis: ANOVA, postischemic vs. control kidneys: * P < 0.05 and ** P < 0.01; Vim -/- vs. Vim +/+ kidneys: not significant; *** P < 0.001.
The activity of -glutamyltranspeptidase, an enzyme expressed in differentiated BBM, was similar in postischemic and control nonoperated kidneys in both Vim +/+ and Vim -/- mice 10 days after ischemia ( Table 1 ), suggesting that regeneration of BBM was not grossly impaired in mutant mice.
Decreased SGLT1 Protein Expression in the Absence of Vimentin
The activity of Na-glucose cotransport in proximal tubules depends on carrier insertion in BBM. We previously showed that, in proximal tubular cells in primary culture, the absence of vimentin affected the localization of SGLT1 to specialized cholesterol- and sphingolipid-enriched domains of BBM, the rafts ( 32 ). Hence, we monitored the expression of SGLT1 in BBM from control nonoperated and postischemic kidneys. The abundance of SGLT1 was reduced significantly in postischemic BBM on recovery day 2 ( Fig. 3 A ). The decrement was of the same magnitude in Vim +/+ (54 ± 9%) and Vim -/- (64 ± 1%) postischemic kidneys compared with control kidneys. Because actin levels were also significantly reduced (51 ± 3 and 61 ± 3% in Vim +/+ and Vim -/- postischemic kidneys, respectively), the SGLT1-to-actin ratio was comparable at day 2 after ischemia in postischemic and control nonoperated mice regardless of the genotype. To evaluate whether differences in loading could account for the reduction of SGLT1 in postischemic BBM, we stained each membrane with Ponceau red. Quantification of the major bands confirmed that equal amounts of proteins were loaded on each line (data not shown). In Vim +/+ postischemic kidneys, SGLT1 protein levels returned back to baseline levels at day 10 and did not change thereafter ( Fig. 3 A ). In Vim -/- postischemic animals, the abundance of SGLT1 protein remained below the control levels at both days 10 and 28 after ischemia. In contrast, actin protein levels increased up to control values in both Vim +/+ and Vim -/- postischemic kidneys from day 10 after ischemia ( Figs. 2 B and 3 A ). Neither ischemia nor vimentin gene inactivation affected the abundance of SGLT1 in whole kidneys (data not shown).
Fig. 3. Western blot analysis of Na-dependent cotransporter expression in BBM from control and postischemic kidneys of Vim +/+ (open bars) and Vim -/- (filled bars) mice. A : SGLT1 expression at 2, 10, and 28 days after surgery. Membranes were incubated with a rabbit polyclonal anti-SGLT1 antibody and a mouse monoclonal anti- -actin antibody. The absence of vimentin resulted in a persistent reduction of SGLT1 protein in regenerating BBM of postischemic kidneys. Note that, on recovery day 2, the disorganization of BBM resulted in decreased actin expression. Because this decrement paralleled the reduction of SGLT1, the SGLT1-to-actin ratio was similar in control nonoperated and postischemic kidneys, regardless of the genotype. B : Npt2a expression at 10 days after surgery. Membranes were incubated with a rabbit polyclonal anti-Npt2a antibody and a mouse monoclonal anti- -actin antibody. Neither ischemia nor vimentin gene inactivation affected Npt2a expression. Blots are representative samples from 4 separate experiments, each performed by pooling the 2 kidneys of 4 mice. Sixteen mice were analyzed for each experimental group and time point. Data are means ± SE. Statistical analysis: ANOVA, postischemic vs. control kidneys: * P < 0.05 and ** P < 0.01; Vim -/- vs. Vim +/+ kidneys: P < 0.01.
We then analyzed whether the absence of vimentin affected the expression of another Na-dependent cotransporter located to proximal tubules, including the S 3 segment, Npt2a. As shown in Fig. 3 B, on recovery day 10, the abundance of Npt2a did not change in BBM of postischemic kidneys compared with that of control nonoperated kidneys. Furthermore, Npt2a levels were identical in Vim +/+ and Vim -/- kidneys, confirming that phosphate transport, unlike glucose transport activity, is unaffected in mutant mice. Similarly, the expression of another apical-located transporter, the P-glycoprotein, was unaffected by vimentin gene inactivation. In fact, Western blot analysis failed to detect any differences in P-glycoprotein levels between Vim +/+ and Vim -/- postischemic kidneys on recovery day 10. The P-glycoprotein-to-actin protein ratio was 1.234 ± 0.45 and 1.889 ± 0.19 arbitrary units in Vim +/+ and Vim -/- mice, respectively ( P = not significant).
Persistent Glucosuria in the Absence of Vimentin
We next investigated whether the impairment of glucose transport activity affects urinary glucose excretion in Vim -/- postischemic mice. As expected, the glucose-to-creatinine ratio markedly increased 2 days after ischemia in both Vim +/+ and Vim -/- mice ( Fig. 4 A ). However, whereas glucose excretion decreased back to the baseline levels on recovery day 10 in wild-type mice, glucosuria remained significantly higher (3-fold) in Vim -/- animals. This effect seems specific. Indeed, in agreement with the transport data, the absence of vimentin did not affect urinary phosphate excretion throughout the course of experiment. The phosphate-to-creatinine ratio was similar in postischemic and control nonoperated mice, irrespective of the time point or genotype ( Fig. 4 B ). Similarly, the pattern of urinary Na excretion was indistinguishable in Vim +/+ and Vim -/- mice throughout the study: it strikingly increased on recovery day 2 and decreased back to baseline levels from day 10 after ischemia ( Fig. 4 C ).
Fig. 4. Urinary solute excretion in control and postischemic Vim +/+ (open bars) and Vim -/- (filled bars) mice at 2 and 10 days after surgery. Because values of control nonoperated mice were unchanged throughout the course of the experiment, only one experimental time point is shown. A : glucose-to-creatinine ratio. B : phosphate-to-creatinine ratio. C : Na-to-creatinine ratio. The absence of vimentin resulted in persistent glucosuria. Data are means ± SE. Ten mice were analyzed for each experimental group and time point. Statistical analysis: ANOVA, postischemic vs. control mice: * P < 0.05 and ** P < 0.01; Vim -/- vs. Vim +/+ mice: P < 0.01.
To assess whether an increase in filtered glucose load may account for persistent glucosuria in Vim -/- mice, we measured plasma creatinine and glucose concentrations throughout the course of experiment. As shown in Table 2, plasma creatinine levels markedly increased 2 days after ischemia in both Vim +/+ and Vim -/- mice. However, from recovery day 10, plasma creatinine returned back to baseline levels, and the values were identical in Vim +/+ and Vim -/- mice. Plasma glucose concentration did not change irrespective of the time point or genotype ( Table 2 ).
Table 2. Biological data at death
DISCUSSION
Genetic manipulation in the mouse offers a powerful approach to define specific roles for individual factors in the complex pathogenesis of human diseases. Vimentin, an intermediate filament of class III, is one of the most intriguing proteins of the cytoskeleton. Mainly expressed in mesenchyma-derived cells in healthy mature kidneys, it is detected in epithelial cells under pathological conditions. The functional role of this expression is not yet known. We combined a genetic approach ( Vim -/- mice) and an experimental model of acute renal injury (bilateral ischemia) to show that vimentin reexpression favors the recovery of tubular Na-glucose cotransport activity after an initial injury. The beneficial effect of vimentin expression in tubular cells is associated with the reexpression of SGLT1 transporters in regenerating BBM. We propose that vimentin acts as an essential element in the restoration of specific transport function, possibly via the maintenance of membrane physical state.
SGLT1 is Reduced in BBM of Postischemic Kidneys
Previous studies have shown that renal ischemia results in reduced proximal tubule glucose transport ( 20, 26, 35 ). The kinetic parameters of MGP uptake (reduced V max with unchanged K d values) and the reduction of phlorizin-binding sites suggest that the transport defect results from a decreased number of accessible Na-glucose carriers in the apical membrane ( 26 ). In the present study, we provide the first evidence in favor of this hypothesis by showing that the amount of SGLT1 protein is reduced in BBM of postischemic kidneys. Two specific cotransporters ensure mainly the uptake of glucose across renal BBM: SGLT1 and SGLT2 ( 52 ). Although the lack of available antibodies limited the analysis of SGLT2 expression, several data of the present study are in favor of a major role of SGLT1. First, the changes of SGLT1 protein paralleled the modification of Na-glucose activity throughout the course of the experiment. Second, the uptake of galactose, a hexose transported by SGLT1 exclusively ( 21 ), was reduced after ischemia in Vim -/- mice. Third, the abundance of SGLT2 mRNA changed neither after ischemia nor in the absence of vimentin (data not shown). Moreover, the previous observations that SGLT1 is the major transporter of the S 3 segment ( 42, 53 ) and that this segment is susceptible to ischemic injury ( 45 ) strongly support this view. Whether ischemia also affects SGLT2 localization to BBM is unknown and merits further investigation.
The Absence of Vimentin Affects Apical SGLT1 Expression in Regenerating Kidneys
We show that the absence of vimentin impaired the reestablishment of Na-glucose cotransport activity during the regeneration phase that follows ischemia. This effect is specific, since the activity of other cotransporters of BBM, such as the Na-phosphate cotransporter Npt2a or Na-neutral amino acid cotransporter, was not modified in mutant mice. In contrast, the early (first 2 days) morphological and functional changes induced by ischemia were not affected by vimentin gene inactivation. It is noteworthy that the pattern of vimentin expression is consistent with a prominent role of this protein in the regenerating phase rather than in the ischemic phase; the peak of vimentin expression was detected between days 5 and 10 after ischemia, depending on the model used ( 28, 38, 41, 49, 51 ). The factor(s) that trigger the transcription of vimentin gene in postischemic tubular cells remain(s) to be elucidated.
The molecular and cellular mechanisms underlying the recovery of glucose transport activity after acute ischemic or toxic tubular necrosis are not completely elucidated. It has been suggested that the reestablishment of apical and basolateral membrane polarity is mandatory for the recovery of tubular functions ( 24 ). In the present study, we showed that the absence of vimentin did not affect the recovery of differentiated BBM, as judged by morphological, biochemical, and enzymatic analysis. Similarly, in a previous study, we showed that the absence of vimentin affects neither cell proliferation nor morphological repair of postischemic tubules ( 38 ). Nevertheless, the abundance of SGLT1 protein was selectively reduced in Vim -/- regenerating BBM, whereas protein concentrations did not change in whole kidney extracts. This phenotype paralleled the activity of glucose transport. In contrast, but in agreement with the functional data, the abundance of Npt2a, another Na-dependent cotransporter of BBM, was not affected by the absence of vimentin. Similarly, the expression of P-glycoprotein, another apically located transporter, did not change in mutant kidneys. Taken together, these results suggest that vimentin participates in the recovery of glucose transport function by selectively favoring the expression of transporters in BBM.
With regard to the mechanism(s) by which vimentin affects carrier localization to apical membranes, it can be proposed that, in postischemic kidneys, vimentin is necessary to locate SGLT1 to rafts, particular cholesterol- and sphingolipid-enriched microdomains of plasma membranes ( 34 ). Various data are in agreement with this hypothesis. First, our previous in vitro observations showed that, in proximal tubular cells in primary culture: 1 ) activity of SGLT1 depends on its localization to rafts, which could favor its dimerization; 2 ) vimentin is located on rafts and forms a physical complex with caveolin; and 3 ) disruption of these microdomains by methyl- -cyclodextrin results in reduced glucose transport activity, a change that is paralleled by a decrease in SGLT1 and vimentin content in rafts ( 32 ). Second, it was recently reported that SGLT1 sorting could be carried out by direct interaction with rafts ( 36 ). Third, it has been observed that acute proximal tubular injury results in damage of raft structure, as suggested by cholesterol missorting and caveolin release ( 54 ). Finally, it has to be pointed out that vimentin expression in postischemic BBM ( 28, 38, 41, 49, 51 ) parallels the changes of lipid composition ( 35 ). Interestingly, vimentin is reexpressed when fluidity of the apical membrane increases, a condition that favors the raft's disruption and impairs Na-glucose cotransport activity ( 14, 46 ). Whether vimentin could maintain the structure of rafts during the recovery phase after ischemia, when the fluidity of apical membranes is increased, is an attractive hypothesis that deserves further investigations.
The Absence of Vimentin Resulted in Persistent Glucosuria
The impairment of transport activities in postischemic kidneys results in natriuresis and glucosuria. Urine analysis confirmed that, in our experimental conditions, both Na and glucose excretion increased in the early phase after ischemia and that the recovery of Na-glucose transport activity was accompanied by urine normalization in Vim +/+ mice. In contrast, the glucose transport defect resulted in persistent glucosuria in Vim -/- animals, whereas phosphate excretion was normal. In a previous study, we showed that the absence of vimentin impairs hemodynamic adaptations to nephron reduction ( 37 ). The fact that, on recovery day 10, glomerular filtration rate, as reflected by plasma creatinine concentration, was comparable in Vim +/+ and Vim -/- mice and did not differ from values of control nonoperated littermates argues against the possibility that glucosuria resulted from increased filtered glucose and supports a tubular reabsorption defect. The observation that in Vim -/- mice glucosuria decreased dramatically from day 2 to day 10 after ischemia, whereas the magnitude of SGLT1 downregulation in BBM did not change, is intriguing. However, the global impairment of proximal reabsorption, including that of Na ( 23 ), may account for the higher glucosuria on recovery day 2. This is supported by the finding that Na excretion, markedly increased at day 2 after ischemia, dropped back to control values by day 10. Alternatively, SGLT2 activity may be transiently impaired in the early phase after ischemia and recovers faster than that of SGLT1.
In conclusion, combining a genetic approach ( Vim -/- mice) and an experimental model of renal injury (bilateral ischemia), we provide the first evidence, to our knowledge, for a role of vimentin reexpression in epithelial cells and demonstrate that vimentin may prevent glucosuria associated with pathological conditions. Whether vimentin might be involved in human renal regeneration and account for different evolutions of postischemic tubular necrosis in patients ( 7, 16 ) is an attractive hypothesis that deserves further investigations.
GRANTS
This work was supported by Institut National de la Santé et de la Recherche Médicale, Université René Descartes, Laboratoires de Recherches Physiologiques, Association pour la Recherche contre le Cancer (no. 9896), and Centre de Recherche Industrielle et Technique.
ACKNOWLEDGMENTS
We are grateful to Rohia Alili and Martine Muffat-Joly for technical assistance at the Centre d'Explorations Fonctionnelles Integrées, Institut Fédératif de Recherche 2. We thank S. Robin for anti-villin antibodies and H. Murer and J. Biber for anti-Npt2a antibodies. We are deeply grateful to G. Trugnan for critical advice.
【参考文献】
Abbate M, Bonventre JV, and Brown D. The microtubule network of renal epithelial cells is disrupted by ischemia and reperfusion. Am J Physiol Renal Fluid Electrolyte Physiol 267: F971-F978, 1994.
Bachmann S, Kriz W, Kuhn C, and Franke WW. Differentiation of cell types in the mammalian kidney in immunofluorescence microscopy using antibodies to intermediate filament proteins and desmoplakins. Histochemistry 77: 365-394, 1983.
Biber J, Stieger B, Haase W, and Murer H. A high yield preparation for rat kidney brush border membranes. Different behaviour of lysosomal markers. Biochim Biophys Acta 647: 169-176, 1981.
Bonventre JV. Dedifferentiation and proliferation of surviving epithelial cells in acute renal failure. J Am Soc Nephrol 14: S55-S61, 2003.
Bonventre JV and Weinberg JM. Recent advances in the pathophysiology of ischemic acute renal failure. J Am Soc Nephrol 14: 2199-2210, 2003.
Booth AG and Kenny AJ. A rapid method for the preparation of microvilli from rabbit kidney. Biochem J 142: 575-581, 1974.
Brady HR, Brenner BM, Clarkson MR, and Leieberthal W. Acute renal failure. In: The Kidney, edited by Brenner BM. Philadelphia, PA: Saunders, 2004, p. 1215-1292.
Chiao H, Kohda Y, McLeroy P, Craig L, Housini I, and Star RA. Alpha-melanocyte-stimulating hormone protects against renal injury after ischemia in mice and rats. J Clin Invest 99: 1165-1172, 1997.
Colucci-Guyon E, Portier MM, Dunia I, Paulin D, Pournin S, and Babinet C. Mice lacking vimentin develop and reproduce without an obvious phenotype. Cell 79: 679-694, 1994.
Deng J, Kohda Y, Chiao H, Wang Y, Hu X, Hewitt SM, Miyaji T, McLeroy P, Nibhanupudy B, Li S, and Star RA. Interleukin-10 inhibits ischemic and cisplatin-induced acute renal injury. Kidney Int 60: 2118-2128, 2001.
Dierick AM, Praet M, Verbeeck P, Robyns C, and Oosterlinck W. Vimentin expression of renal cell carcinoma in relation of DNA content and histological grading: a combined light microscopy, immunocytochemical, and cytophotometrical analysis. Histopathology 18: 315-322, 1991.
Evans RM. Vimentin: the conundrum of the intermediate filament gene family. Bioessays 20: 79-86, 1998. <a href="/cgi/external_ref?access_num=10.1002/(SICI)1521-1878(199801)20:1
Franke WW, Schmid E, Winter S, Osborn M, and Weber K. Widespread occurence of intermediate-sized filaments of vimentin-type in cultured cells from diverse vertebrates. Exp Cell Res 123: 25-46, 1979.
Friedlander G, Shahedi M, Le Grimellec C, and Amiel C. Increase in membrane fluidity and opening of tight junctions have similar effects on sodium-coupled uptakes in renal epithelial cells. J Biol Chem 263: 11183-11188, 1988.
Gossrau R, Günther T, and Graf R. Enhancement of gentamicin-induced nephrotoxicity by Mg deficiency in non-pregnant rats. Histochemistry 90: 489-496, 1989.
Green J, Abassi Z, Winaver J, and Skorecki KL. Acute renal failure: clinical and pathophysiolocic aspects. In: The Kidney, edited by Seldin DW and Giebisch G. Philadelphia, PA: Lippincott Williams & Wilkins, 2000, p. 2329-2374.
Hatzinger PB, Chen Q, Dong L, and Stevens JL. Alterations in intermediate filament proteins in rat kidney proximal tubule epithelial cells. Biochem Biophys Res Commun 157: 1316-1322, 1988.
Holthöfer H, Miettinen A, Lehto VP, Lehtonen E, and Virtanen I. Expression of vimentin and cytokeratin types of intermediate filament proteins in developing and adult human kidneys. Lab Invest 50: 552-559, 1984.
Holthöfer H, Miettinen A, Paasivuo R, Lehto VP, Linder E, Alfthan O, and Virtanen I. Cellular origin and differentiation of renal carcinomas. Lab Invest 49: 317-326, 1983.
Johnston PA, Rennke H, and Levinsky NG. Recovery of proximal tubular function from ischemic injury. Am J Physiol Renal Fluid Electrolyte Physiol 246: F159-F166, 1984.
Kanai Y, Lee WS, You G, Brown D, and Hediger MA. The human kidney low affinity Na+/glucose cotransporter SGLT2. Delineation of the major renal reabsorptive mechanism for D -glucose. J Clin Invest 93: 397-404, 1994.
Kellerman PS, Clark RA, Hoilien CA, Linas SL, and Molitoris BA. Role of microfilaments in maintenance of proximal tubule structural and functional integrity. Am J Physiol Renal Fluid Electrolyte Physiol 259: F279-F285, 1990.
Kwon TH, Frokiaer J, Han JS, Knepper MA, and Nielsen S. Decreased abundance of major Na + transporters in kidneys of rats with ischemia-induced acute renal failure. Am J Physiol Renal Physiol 278: F925-F939, 2000.
Molitoris BA. Ischemia-induced loss of epithelial polarity: potential role of the actin cytoskeleton. Am J Physiol Renal Fluid Electrolyte Physiol 260: F769-F778, 1991.
Molitoris BA, Falk SA, and Dahl RH. Ischemia-induced loss of epithelial polarity: role of the tight junction. J Clin Invest 84: 1334-1339, 1989.
Molitoris BA and Kinne R. Ischemia induces surface membrane dysfunction. Mechanism of altered Na + -dependent glucose transport. J Clin Invest 80: 647-654, 1987.
Molitoris BA, Wilson PD, Schrier RW, and Simon FR. Ischemia induces partial loss of surface membrane polarity and accumulation of putative calcium ionophores. J Clin Invest 76: 2097-2105, 1985.
Nouwen EJ, Verstrepen WA, Buyssens N, Zhu MQ, and De Broe ME. Hyperplasia, hypertrophy, and phenotypic alterations in the distal nephron after acute proximal tubular injury in the rat. Lab Invest 70: 479-493, 1994.
Orlowski M and Meister A. Isolation of highly purified gamma-glutamylcysteine synthetase from rat kidney. Biochemistry 10: 372-380, 1971.
Post RL, Albright CD, and Dayani K. Resolution of pump and leak components of sodium and potassium ion transport in human erythrocytes. J Gen Physiol 50: 1201-1220, 1967.
Prasad S, Soldatenkov VA, Srinivasarao G, and Dritschilo A. Intermediate filament proteins during carcinogenesis and apoptosis. Int J Oncol 14: 563-570, 1999.
Runembert I, Queffeulou G, Federici P, Vrtovsnik F, Colucci-Guyon E, Babinet C, Briand P, Trugnan G, Friedlander G, and Terzi F. Vimentin affects localization and activity of sodium-glucose cotransporter SGLT1 in membrane rafts. J Cell Sci 115: 713-724, 2002.
Semenza GL. Series introduction: tissue ischemia: pathophysiology and therapeutics. J Clin Invest 106: 613-614, 2000.
Smart EJ, Graf GA, McNiven MA, Sessa WC, Engelman JA, Scherer PE, Okamoto T, and Lisanti MP. Caveolins, liquid-ordered domains, and signal transduction. Mol Cell Biol 19: 7289-7304, 1999.
Spiegel DM, Wilson PD, and Molitoris BA. Epithelial polarity following ischemia: a requirement for normal cell function. Am J Physiol Renal Fluid Electrolyte Physiol 256: F430-F436, 1989.
Suzuki T, Fujikura K, Koyama H, Matsuzaki T, Takahashi Y, and Takata K. The apical localization of SGLT1 glucose transporter is determined by the short amino acid sequence in its N-terminal domain. Eur J Cell Biol 80: 765-774, 2001.
Terzi F, Henrion D, Colucci-Guyon E, Federici P, Babinet, Levy BI, Briand P, and Friedlander G. Reduction of renal mass is lethal in mice lacking vimentin. Role of endothelin-nitric oxide imbalance. J Clin Invest 100: 1520-1528, 1997.
Terzi F, Maunoury R, Colucci-Guyon E, Babinet C, Federici P, Briand P, and Friedlander G. Normal tubular regeneration and differentiation of the post-ischemic kidney in mice lacking vimentin. Am J Pathol 150: 1361-1371, 1997.
Thornell LE, Holmbom B, Eriksson A, Reiz S, Marklund S, and Naslund U. Enzyme and immunohistochemical assessment of myocardial damage after ischaemia and reperfusion in a closed-chest pig model. Histochemistry 98: 341-353, 1992.
Toback FG. Regeneration after acute tubular necrosis. Kidney Int 41: 226-246, 1992.
Toubeau G, Nonclercq D, Zanen J, Laurent G, Schaudies PR, and Heuson-Stiennon JA. Renal tissue expression of EGF and EGF receptor after ischemic tubular injury: an immunohistological study. Exp Nephrol 2: 229-239, 1994.
Turner RJ and Moran A. Heterogeneity of sodium-dependent D -glucose transport sites along the proximal tubule: evidence from vesicle studies. Am J Physiol Renal Fluid Electrolyte Physiol 242: F406-F414, 1982.
Van Why SK, Mann AS, Ardito T, Siegel NJ, and Kashgarian M. Expression and molecular regulation of Na + -K + -ATPase after renal ischemia. Am J Physiol Renal Fluid Electrolyte Physiol 267: F75-F85, 1994.
Van Why SK and Siegel NJ. Heat shock proteins in renal injury and recovery. Curr Opin Nephrol Hypertens 7: 407-412, 1998.
Venkatachalam MA, Bernard DB, Donohoe JF, and Levinsky NG. Ischemic damage and repair in the rat proximal tubule: differences among the S1, S2, and S3 segments. Kidney Int 14: 31-49, 1978.
Vrtovsnik F, el Yandouzi EH, Le Grimellec C, and Friedlander G. Sphingomyelin and cholesterol modulate sodium coupled uptakes in proximal tubular cells. Kidney Int 41: 983-991, 1992.
Waldherr R and Schwechheimer K. Co-expression of cytokeratin and vimentin intermediate-sized filaments in renal cell carcinomas. Virchows Arch 408: 15-27, 1985.
Walker PD. Alterations in renal tubular extracellular matrix components after ischemia-reperfusion injury to the kidney. Lab Invest 70: 339-345, 1994.
Wallin A, Zhang G, Jones TW, Jaken S, and Stevens JL. Mechanism of nephrogenic repair response: studies on proliferation and vimentin expression after 35 S-1,2-di-chlorovinyl- L -cysteine nephrotoxicity in vivo and in cultured proximal tubule epithelial cells. Lab Invest 66: 474-484, 1992.
Wang Z, Rabb H, Haq M, Shull GE, and Soleimani M. A possible molecular basis of natriuresis during ischemic-reperfusion injury in the kidney. J Am Soc Nephrol 9: 605-613, 1998.
Witzgall R, Brown D, Schwarz C, and Bonventre JV. Localization of proliferating cell nuclear antigen, vimentin, c-Fos, and clusterin in the postischemic kidney. Evidence for a heterogenous genetic response among nephron segments, and a large pool of mitotically active and dedifferentiated cells. J Clin Invest 93: 2175-2188, 1994.
Wright EM. Renal Na + -glucose cotransporters. Am J Physiol Renal Physiol 280: F10-F18, 2001.
You G, Lee WS, Barros EJ, Kanai Y, Huo TL, Khawaja S, Wells RG, Nigam, SK, and Hediger MA. Molecular characteristics of Na + -coupled glucose transporters in adult and embryonic rat kidney. J Biol Chem 270: 29365-29371, 1995.
Zager RA, Johnson A, Hanson S, and dela Rosa V. Altered cholesterol localization and caveolin expression during the evolution of acute renal failure. Kidney Int 61: 1674-1683, 2002.
作者单位：1 Institut National de la Santé et de la Recherche Médicale Unitè 426 Department of Physiologie, Faculté de Médecine Xavier Bichat, 75870 Paris, Cedex 18; 2 Institut Cochin de Génétique Moléculaire, 75014 Paris; and 3 Unité de Recherche