点击显示 收起
【摘要】 Glomerular podocytes are critical regulators of glomerular permeability via the slit diaphragm and may play a role in cleaning the glomerular filter. Whether podocytes are able to endocytose proteins is uncertain. We studied protein endocytosis in conditionally immortalized mouse and human podocytes using FITC-albumin by direct quantitative assay and by fluorescence microscopy and electron microscopy in mouse podocytes. Furthermore, in vivo uptake was studied in human, rat, and mouse podocytes. Both mouse and human podocytes displayed specific one-site binding for FITC-albumin with K d of 0.91 or 0.44 mg/ml and B max of 3.15 or 0.81 µg/mg cell protein, respectively. In addition, they showed avid endocytosis of FITC-albumin with K m of 9.48 or 4.5 mg/ml and V max of 474.3 or 97.4 µg·mg cell protein -1 ·h -1, respectively. Immunoglobulin and transferrin were inefficient competitors of this process, indicating some specificity for albumin. Accumulation of endocytosed albumin could be demonstrated in intracellular vesicles by fluorescence confocal microscopy and electron microscopy. Endocytosis was sensitive to pretreatment with simvastatin. In vivo accumulation of albumin was found in all three species but was most pronounced in the rat. We conclude that podocytes are able to endocytose protein in a statin-sensitive manner. This function is likely to be highly significant in health and disease. In addition, protein endocytosis by podocytes may represent a useful, measurable phenotypic characteristic against which potentially injurious or beneficial interventions can be assessed.
【关键词】 proteinuria renal nephropathy permeability slit diaphragm
HEALTHY INDIVIDUALS EXCRETE only tiny quantities of protein in their urine, and the exclusion of circulating macromolecules from very large volumes of glomerular filtrate is a crucial aspect of normal renal function. Based on the results of many studies using inert tracers of different molecular size and charge, the most widely held view is that the glomerular filtration barrier restricts macromolecular filtration on the basis of increasing size and negative charge ( 16 ).
The precise location of the permselectivity barrier in the glomerulus has been the subject of considerable research and controversy. The glomerular filtration barrier is composed of three layers: the fenestrated endothelium of the glomerular capillaries, the glomerular basement membrane (GBM), and the foot processes of the glomerular visceral epithelial cells or podocytes ( 16 ). Mature differentiated podocytes have a complex cellular architecture, with interdigitating foot processes extending out from the cell body and embedded into the GBM. Slit diaphragms bridge the spaces between the foot processes or filtration slits. Together, the GBM, podocyte foot processes, and slit diaphragms form a dynamic and interdependent filter system. However, it is currently believed that while the endothelium and GBM together provide a coarse filter with some charge selectivity, the major permselective obstacle to macromolecular filtration within the glomerulus resides in the slit diaphragm ( 23 ).
Mechanisms of glomerular protein handling have received much attention in the study of nephrotic syndrome and following the recognition of proteinuria as an independent risk factor for both renal failure and cardiovascular disease. Considerable research effort has been directed toward understanding the barrier to macromolecular filtration provided by the components of the slit diaphragm ( 23 ), but little attention has been paid to the potential for podocytes to handle plasma proteins beyond the specialization of the slit diaphragm. A number of problems with the currently held models of glomerular permeability, together with intriguing clinical observations, could be explained by podocyte endocytosis of plasma proteins. Given the high-volume vectorial flux of solvent and solute from the glomerular capillaries toward the urinary space, some have hypothesized that the filter should clog as a result of protein hold-up at the slit diaphragm ( 17, 26 ). While this theoretical phenomenon has been avidly proposed ( 26 ), there is little evidence to support its existence, and it may be prevented by uptake and removal of trapped proteins by podocytes. Evidence that podocyte endocytosis of proteins does occur derives from the clinical evaluation of renal biopsies from heavily proteinuric patients, where extensive protein vacuolation of podocytes is commonly observed ( 28 ). Additionally, tracer experiments using ferritin have identified this large protein in podocyte lysosomes, also suggesting that podocytes have an endocytic recovery function for ferritin that passes the GBM ( 13, 15 ). Whether protein uptake by podocytes only occurs in proteinuric states, or whether a healthy cell function adapts to a pathological environment, or whether such an adaptive process may become maladaptive in disease is unknown, and the underlying mechanisms are obscure.
We have had a long-standing interest in protein handling by kidney proximal tubular epithelial cells (PTEC) and have much experience in the study of albumin endocytosis by these cells, recently describing inhibition of PTEC albumin endocytosis by statins ( 4, 6, 25 ). Statins have also been shown to abrogate oxidized LDL-induced injury of podocytes and to preserve their nephrin expression ( 7 ). Therefore, we hypothesized that podocytes may possess the ability to endocytose albumin and that this process may be regulated by HMG-CoA reductase activity.
MATERIALS AND METHODS
Reagents and materials. Conditionally immortalized mouse and human podocytes carrying a temperature-sensitive variant of the simian virus (SV40) large tumor antigen were obtained from Peter Mundel (Albert Einstein College of Medicine, New York, NY) and Moin Saleem (University of Bristol, Bristol, UK), respectively ( 22, 24 ). Free FITC human serum albumin, mevalonic acid, and geranylgeranyl pyrophosphate (GGPP) were purchased from Sigma. Sodium salt of simvastatin was obtained as a gift from AstraZeneca. Colloidal gold-labeled albumin (gold-alb) was from BB International and supplied by Agar Scientific.
Antibodies. For rat kidneys, affinity-purified sheep anti-rat albumin (dilution 1:4,000, Biogenesis, Poole, UK) was used, and for human and mouse kidneys rabbit anti-rat albumin (dilutions 1:200 or 1:400, Nordic Immunological, Tilbury, The Netherlands) were used.
The following secondary antibodies were used conjugated to 10-nm gold particles: donkey anti-sheep and goat anti-rabbit (British BioCell International, Cardiff, UK).
Cell culture. All podocytes were cultured as previously described by ( 22, 24 ). Podocytes were grown in flasks coated with 0.1 mg/ml bovine calfskin collagen type I (Sigma) under growth-permissive conditions at 33°C maintained in RPMI 1640 medium with glutamine and supplemented with 10% fetal bovine serum, 100 U/ml penicillin plus 100 µg/ml streptomycin and 10 U/ml mouse recombinant interferon- (for mouse cells only). Cells were incubated in a humidified atmosphere of 5% CO 2 -95% air, media was renewed every second day, and cells were split at confluence approximately once a week using a trypsin-EDTA solution. Podocyte differentiation was induced under nonpermissive conditions by thermo shifting the cells to 37°C and removing interferon- from the culture medium for 14 days. All cells used were between passages 15 and 25. Before all experiments, cells were maintained in serum-free media for 23 h.
Albumin binding and uptake. Cells were plated in 24-well plates at a density of 25,000 cells/well and grown to 95% confluence over 3 days. For all experiments, human serum albumin was labeled with FITC. FITC-albumin was separated from free FITC by gel filtration using PD-10 desalting columns containing Sephadex G-25. To determine whether FITC-albumin solutions were contaminated by free FITC, they were precipitated with trichloroacetic acid, centrifuged, and fluorescence in the supernatant was measured. In addition, FITC-albumin solutions were separated by PAGE, and the location of fluorescence in the gel was determined under UV light before the dye front had migrated off the gel.
FITC-albumin was added to podocyte cell monolayers in Ringer solution (in mM: 122.5 NaCl, 5.4 KCl, 1.2 CaCl 2, 0.8 MgCl 2, 0.8 Na 2 HPO 4, 0.2 NaH 2 PO 4, 5.5 glucose, and 10 HEPES), pH 7.4, at various concentrations and incubated at either 37 or 4°C for various times. Incubations were terminated by placing plates on ice, removing the FITC-albumin/Ringer solution, and washing each well six times with 1-ml aliquots of ice-cold Ringer solution, pH 7.4. After the final wash, cells were lysed by the addition of 0.1% Triton X-100 (vol/vol) in 20 mM MOPS, and the cell-associated fluorescence was measured using a spectrofluorometer. Total protein content was determined by the method of Lowry. In some experiments, 100 x excess unlabeled albumin was added and incubated with the cells along with FITC-albumin. In others, podocytes were incubated with 1.5 mg/ml FITC-albumin together with an equimolar concentration of either IgG or transferrin. The effect of statin on FITC-albumin uptake was assessed after a 23-h preincubation with the sodium salt of simvastatin. The mechanisms of simvastatin's effects on FITC-albumin uptake were further assessed by treating cells with either mevalonate or GGPP during the 23-h preincubation.
Cell viability. Cell viability was analyzed using methylthiazoletetrazolium (MTT assay) as previously described ( 8 ).
Confocal microscopy. Differentiated mouse podocyte cells were grown to 50% confluence on collagen type I-coated chamber slides. Twenty-three hours before the start of the experiment, the serum was removed from the media. Cells were incubated with Ringer solution, pH 7.4, containing 3 mg/ml FITC-albumin at 37°C for 1 h. Control cells were incubated in Ringer solution only. Unbound FITC-albumin was removed by rinsing six times with ice-cold Ringer solution. Cells were fixed by submerging them in 2% paraformaldehyde for 2 min and then rinsed for 15 min with PBS (10 mM phosphate buffer, 2.7 mM KCl, and 137 mM NaCl, pH 7.4). Coverslips were viewed on an Olympus Fluoview laser-scanning confocal microscope on an inverted stage. The FITC-albumin was excited at a wavelength of 488 nm, and the emitted light was captured using a 510-nm barrier filter and displayed as an image on a computer running Olympus Fluoview software.
For the real-time experiments, differentiated podocyte cells were grown to 50% confluence on collagen type I-coated glass coverslips and placed in serum-free media 24 h before the start of the experiment. Coverslips were then placed in a perfusion chamber on the stage of the inverted microscope and perfused with Ringer solution, pH 7.4, via a peristaltic pump until the temperature was stable at 37°C. Ringer solution containing 3 mg/ml FITC-albumin at 37°C was then perfused onto the cells for 15 min. All FITC-albumin was then removed by continuous perfusion of the cells with Ringer solution. The first image of the cells was captured 25 min after the start of the FITC-albumin incubation, using a computer running Olympus Fluoview software. Further images were captured every 10 min up to a final time of 75 min. The images obtained were then compiled into movies enabling the trafficking of endocytosed FITC-albumin within vesicles to be visualized further.
Electron microscopy: in vitro experiments. BSA conjugated to 10-nm colloidal gold was dialyzed overnight against an excess of Ringer solution containing 10 mg/ml BSA to remove sodium azide preservative and to concentrate the gold-alb. Differentiated mouse podocyte cells were grown to confluence on collagen type I-covered glass coverslips and growth arrested in serum-free media 23 h before the start of the experiment. The media was removed and replaced with Ringer solution, pH 7.4, containing the concentrated gold-alb and incubated for 1 h at 37°C. Uptake was terminated by washing cell monolayers six times with ice-cold Ringer solution, pH 7.4. Cells were then fixed in 2.5% glutaraldehyde in 0.1 M Sorensen's buffer, pH 7.2, for 3 h. This primary fixation was followed by three washes (10-min periods) in 0.1 M phosphate buffer, pH 7.2. Secondary fixation of the cells was carried out using 0.5% osmium tetroxide in 0.1 M phosphate buffer, pH 7.2, for 1.5 h. Cells were then washed for 45 min in doubly distilled H 2 O, dehydrated in graded ethanol solutions, washed for 20 min in propylene oxide, and then left overnight in 50:50 propylene oxide/Spurr's resin, which was then replaced with fresh resin for 5 h. Embedded samples were polymerized at 60°C for 16 h before being sectioned using a diamond knife on a Reichert Ultracut S onto copper grids. Sections were then stained with 2% aqueous uranyl acetate for 20 min followed by lead citrate for 2 min and viewed on a JEOL 1220 transmission electron microscope.
Preparation of renal tissue. Normal, uninvolved human renal tissue obtained from resected renal carcinoma kidneys was fixed in 4% formaldehyde in 0.1 M sodium cacodylate buffer, pH 7.2. The tissue was trimmed into small blocks, further fixed by immersion for 1 h in 1% formaldehyde, infiltrated with 2.3 M sucrose containing 2% formaldehyde for 30 min, and frozen in liquid nitrogen.
Male mouse (C57/BL76J, 25 g) and male rat (Wistar, 250 g) kidneys were fixed by perfusion retrograde through the abdominal aorta with 2% paraformaldehyde in 0.1 M sodium cacodylate buffer, pH 7.2. The tissue was trimmed into small blocks and processed as above. All animal experiments were approved and carried out in accordance with provisions for the animal care license provided by the Danish National Animal Experiments Inspectorate. Informed consent was obtained from each human patient.
Immunocytochemistry. For electron microscopy, 70- to 90-nm cryosections were obtained at -100°C with an FCS Reichert Ultracut S cryoultramicrotome as previously described ( 11 ). For electron microscopic immunolabeling, the sections were incubated with primary antibodies at 4°C overnight followed by incubation for 1 h at room temperature with 10-nm gold particles coupled to relevant IgGs (BioCell). The cryosections were embedded in methylcellulose containing 0.3% uranyl acetate and studied with a Philips CM100 electron microscope. For the controls, sections were incubated with secondary antibodies alone or with nonspecific IgG.
Statistical analyses. Curve fitting of data was performed with GraphPad Prism software, and statistical analysis was performed using GraphPad Instat by ANOVA followed by the Tukey-Kramer post hoc test for multiple comparisons.
RESULTS
Negligible fluorescence was found in supernatants of centrifuged acid-precipitated FITC-albumin (data not shown). Significant fluorescence was only associated with the FITC-albumin band (data not shown) when subjected to PAGE. We therefore judged that FITC-albumin preparations were not significantly contaminated by free FITC.
Binding and uptake of albumin by differentiated podocytes. The concentration dependence of FITC-albumin uptake alone, or in the presence of a 100-fold excess of unlabeled albumin, is shown in Fig. 1, A and B, after 1 h of incubation. When incubated at 37°C, both mouse and human podocytes take up FITC-albumin. This process is specific and demonstrates saturation kinetics consistent with a receptor-mediated process. Analysis of these data by nonlinear curve fitting indicates the presence of a single uptake transporting system for FITC-albumin in both cell types. The concentrations of albumin leading to half-maximal activation of uptake ( K m ) were 9.48 and 4.5 mg/ml, and the maximal rates of albumin uptake ( V max ) were 474.3 and 97.4 µg·mg cell protein -1 ·h -1 for mouse and human cells, respectively. Uptake was completely inhibited by a 100-fold excess of unlabeled albumin, demonstrating specificity and also indicating that increasing cellular fluorescence was not simply the result of accumulating free-FITC. Figure 1, C and D, shows the time course of uptake of 3.0 mg/ml FITC-albumin by podocytes.
Fig. 1. Concentration dependence and time course of FITC-albumin uptake into differentiated mouse and human podocytes at 37°C after 1-h incubation. Mouse podocytes ( A ) and human podocytes ( B ) were incubated with varying concentrations of FITC-albumin alone (solid line) or in the presence of 100-fold excess of unlabeled albumin (dashed line) at 37°C. Mouse podocytes ( C ) and human podocytes ( D ) were incubated with 3 mg/ml FITC-albumin for varying time periods at 37 (solid line) and at 4°C (dashed line). Cells were then washed, and cell-associated fluorescence was determined as described. Values are means ± SE of 3 experiments.
To assess binding of FITC-albumin to podocytes, rather than uptake, experiments were performed at 4°C. Values representing FITC-albumin binding in the presence of excess unlabeled albumin were subtracted from those obtained for identical conditions in the presence of FITC-albumin alone, and the data are presented in Fig. 2. Binding to both mouse and human podocytes was saturable. The data fit best to a one site-binding model with equilibrium dissociation constants ( K d ) of 0.91 and 0.44 mg/ml and a maximum number of binding sites (B max ) of 3.15 and 0.81 µg/mg cell protein for mouse and human cells, respectively. Uptake of FITC-albumin by podocytes was significantly reduced by coincubation with an equimolar amount of either IgG or transferrin ( Fig. 3 ). However, only 20% of FITC-albumin uptake was inhibited under these conditions, indicating some specificity of the uptake process for albumin.
Fig. 2. Albumin binding to podocytes. Podocytes were incubated with varying concentrations of FITC-albumin for 1 h at 4°C, washed, and cell-associated fluorescence was measured. Values corresponding to nonspecific binding for each condition measured in the presence of 100-fold excess unlabeled albumin have been subtracted. Solid line, mouse podocytes; broken line, human podocytes. Values are means ± SE of 3 experiments.
Fig. 3. Specificity of uptake for albumin. Mouse podocytes were incubated at 37°C with 1.5 mg/ml FITC-albumin alone or in the presence of an equimolar amount of either IgG or transferrin. Cells were then washed, and cell-associated fluorescence was determined as described. Values are means ± SE of 3 experiments. *** P < 0.001 compared with control.
Microscopic evaluation of podocyte albumin endocytosis in vitro. To confirm and visualize the endocytic uptake of albumin by mouse podocyte cells, two methods were used. First, the uptake of FITC-albumin was examined by confocal microscopy ( Fig. 4 ). After 1-h incubation with 3 mg/ml FITC-albumin in Ringer solution at 37°C, FITC-albumin could be seen in abundant large vesicles within the podocyte cell bodies. The majority of FITC-albumin-containing vesicles assembled in the perinuclear region; however, it was also possible to see FITC-albumin in vesicles within the podocyte processes. Using real-time confocal microscopy of living cells, endocytosed FITC-albumin could be seen in vesicles moving from processes and the cell periphery toward the perinuclear area (see supplementary data available in the online version of the journal). Small stringlike structures appeared to aggregate into larger spherical structures.
3 experiments.
The ultrastructure of podocytes binding and endocytosing albumin was then further visualized by transmission electron microscopy using gold-alb. Figure 5 depicts four representative sections. The sections show that podocytes incubated with gold-alb demonstrate a number of gold particles bound to the cell surface and within intracellular vesicular structures of varying electron density, presumably corresponding to different components of the endosomal/lysosomal trafficking apparatus. All intracellular gold-alb appears to be confined to membrane-bound vesicular structures as opposed to free in the cytoplasm.
Fig. 5. Electron microscopy of 10-nm gold-albumin (alb)-incubated podocytes. Shown are representative photomicrographs of mouse podocytes following exposure to Ringer solution containing 10-nm gold-labeled albumin for 1 h at 37°C. Gold particles can be seen bound to the cell surface membrane (arrowheads) and in endocytic vesicles (arrows). Magnification is as indicated by scale bars.
Effect of simvastatin on podocyte albumin endocytosis. The effect of 24-h incubation with simvastatin on podocyte cell viability was determined by MTT assay. Podocytes were found to be fully viable when exposed to concentrations of simvastatin up to 1 µM. Greater concentrations led to a significant loss in cell viability, as shown in Fig. 6 A. Exposure to increasing, but nontoxic, concentrations of simvastatin significantly inhibited podocyte FITC-albumin endocytosis. This inhibition was statistically significant at 0.75 µM simvastatin, when albumin endocytosis compared with the control was reduced by 28% ( Fig. 6 B ). The effect of simvastatin was reversed by the addition of either mevalonate or GGPP ( Fig. 6 C ).
Fig. 6. Effect of simvastatin on podocyte viability and albumin endocytosis. A : mouse podocytes were incubated with varying concentrations of simvastatin for 24 h, and viability was determined by MTT assay as described. Values are means ± SE expressed as a percentage of the control (no statin) of 2 different experiments performed in duplicate. * P < 0.05, ** P < 0.01 compared with controls. B : mouse podocytes were incubated with various concentrations of simvastatin for 23 h followed by 3 mg/ml FITC-albumin for 1 h additionally at 37°C. C : mouse podocytes were incubated with 1 µM simvastatin and either mevalonate or GGPP for 23 h followed by 3 mg/ml FITC-albumin for 1 h additionally at 37°C. Values are means ± SE of 3 different experiments performed in duplicate. *** P < 0.001 compared with controls.
Podocyte accumulation of albumin in vivo. To test for in vivo uptake of endogenous albumin, we studied human, rat, and mouse glomeruli by immunogold labeling on ultrathin cryosections. Accumulation of albumin in podocytes of rat glomeruli under physiological conditions was relatively extensive. Gold particles were often found in large lysosome-like structures, as seen in Fig. 7 A. In human podocytes, accumulation was seen in similar structures ( Fig. 7 B ); however, cells accumulating albumin in human glomeruli were much more sparse compared with those from rats. In mice, albumin-containing vesicles like the one demonstrated in Fig. 7 C were seen only very rarely.
Fig. 7. Electron microscope immunohistochemistry of albumin accumulation in podocytes in vivo. A : lysosome-like cytoplasmic body exhibiting intensive labeling for albumin in foot process of rat podocyte (arrow). B : similar labeling in human podocyte. C : small labeled vesicle in the perinuclear area of mouse podocyte. N, nucleus; BM, basement membrane. Magnification is as indicated by scale bars.
DISCUSSION
Glomerular podocytes have attracted intense study in recent years, largely due to their role in the maintenance of normal glomerular permeability barrier function and as key players in progressive renal diseases ( 23 ). Much is now known about podocyte proteins that contribute to slit diaphragm function in a structural sense and as signaling intermediates and also about podocyte cell proteins supporting functional interactions with the GBM ( 23 ).
Anatomically, the podocyte is well placed to clean the glomerular filter by removing macromolecules that may otherwise cause clogging, but this presumptive function has never been definitively delineated. Indeed, although cultured differentiated podocytes express a characteristic proteome ( 23 ), no precise physiological functions are known or can be measured. Only one previous report has briefly described protein endocytosis by cultured podocytes ( 20 ). Londono and Bendayan ( 19 ) described rat glomerular handling of native and modified albumins in vivo. While podocyte endocytosis was not the focus of their study and not specifically commented on in their paper, some of the published electron micrographs appear to show localization of albumin to the podocyte. Now, for the first time we have quantified binding and kinetic parameters of protein endocytosis in podocytes from two species using albumin as a prototype. However, these transformed cultured cells may not accurately represent true mature podocyte structure and function in vivo, and therefore, in addition, we have demonstrated albumin accumulation in podocytes from three different species under physiological conditions. Our results unequivocally confirm the ability of podocytes to reabsorb protein by endocytosis and are in keeping with morphological observations that podocytes have many coated membrane pits and a large number of multivesicular bodies and other endosomal/lysosomal structures within their cell body, suggestive of high endocytic activity and vesicular trafficking ( 21 ).
Receptor-mediated endocytosis of proteins is an important differentiated phenotype of kidney PTEC and has been well studied in these cells by us and other workers ( 4, 6, 9, 14, 25 ). Using opossum kidney cells as a model, two binding sites for albumin in PTEC have been described ( 5 ): a high-affinity, low-capacity binding site with a K d of 0.15 mg/ml and a B max of 0.2 µg/mg cell protein and a low-affinity, high-capacity binding site with a K d of 8.3 of mg/ml and a B max of 4.6 µg/mg cell protein. The kinetics of albumin reabsorption has also been studied in PTEC ( 4 ). Corresponding to the two binding sites for albumin in opposum kidney cells, two transport processes for uptake were identified: the higher affinity one with a K m of 0.024 mg/ml and a V max of 1.2 µg·mg -1 ·h -1 and the lower affinity one with a K m of 15.9 mg/ml and a V max of 32.0 µg·mg -1 ·h -1.
Compared with the higher affinity PTEC albumin binding site, the binding affinity of the receptor for albumin in both mouse and human cultured podocytes is 3- to 9-fold lower but the number of these sites is on the order of 4- to 10-fold greater. Uptake rates of albumin uptake into podocytes are correspondingly greater than those into cultured PTEC. These results may seem to indicate that compared with PTEC the single podocyte uptake systems described in the current experiments represent a greater capacity for albumin endocytosis by podocytes than by PTEC. While these observations are in keeping with the belief that podocytes will encounter higher albumin concentrations than PTEC, at least in the vicinity of their basal membrane on the endothelial side of the slit diaphragm, care is required in their interpretation. Comparison of a phenomenon quantified in vitro in different cell types and then extrapolating it to the in vivo situation may not be straightforward. Nonetheless, if podocytes do perform a glomerular filter-cleaning function as postulated, a higher capacity uptake system for albumin and other proteins than that found in PTEC would be required due to the higher protein concentrations encountered.
It would be expected that if albumin and podocytes were coincubated with an equimolar amount of a protein with equal affinity for the uptake system, then albumin uptake should be reduced by 50%. Equimolar amounts of transferrin and IgG were only able to block albumin uptake by 20%. Thus the uptake system has some specificity for albumin. Furthermore, given that FITC-albumin preparations were not contaminated by free FITC, and that podocyte uptake of FITC-albumin was completely blocked by excess unlabeled albumin, the possibility that the endocytosis assay-measured uptake of free FITC is excluded.
In PTEC, megalin and cubilin are a mutually dependent receptor complex acting as a scavenger for many filtered macromolecules including albumin ( 2, 10, 12, 29 ). This complex is believed to represent the higher affinity binding site in these cells. While megalin expression is described in podocytes ( 18, 27 ), cubilin has not been identified in these cells. Given this, and the different binding affinities for albumin between these two cell types, it is possible that the podocyte albumin receptor is distinct from that in PTEC. Receptor-mediated endocytosis by PTEC occurs across the apical cell membrane, and expression of megalin/cubilin is restricted to this membrane domain. The polarity of podocytes in culture, however, is unclear, and we are unsure whether we are measuring a polarized transport process in these cells. Studies to address this question are underway.
Nonetheless, there are similarities between the albumin endocytosis in PTEC and podocytes. Albumin endocytosis in both cell types is sensitive to the statin inhibitors of HMG-CoA reductase. When we recently described statin-sensitive inhibition of albumin uptake in PTEC, we showed that this was due to depletion of nonsterol products of the mevalonate pathway and provided proof of principle that defective posttranslational modification of low-molecular-weight GTP-binding proteins underpinned the failure of endocytosis in the presence of statins ( 25 ). In mouse podocytes, statins at similar concentrations exhibit a similar inhibitory effect on albumin transport. It is likely that the mechanism is the same. We observed that the statin effect on albumin endocytosis by podocytes was prevented by repletion of mevalonate and the isoprenoid precursor GGPP. This suggests that, as a result of inhibition of HMG-CoA reductase, the depletion of GGPP or one of its products is critical to the effect of statins on protein reabsorption, rather than nonspecific toxicity.
The relevance of this observation in podocytes to statin treatment of patients is unclear. Individuals on statins do not seem to develop features suggestive of impaired cleaning of the glomerular barrier, whatever they may be. It is not known whether protein uptake by podocytes in proteinuric disease may become maladaptive as it is in PTEC ( 3 ). Toxicity in podocytes may be induced by oxidized LDL and is manifest as redistribution of nephrin but via uncertain mechanisms ( 7 ). This effect is abrogated by statin treatment. Delivery of albumin-bound lipids into the interior of PTEC is toxic in proteinuria, and similar intracellular delivery of injurious lipid into podocytes in proteinuria may have a similar toxic effect ( 1, 3 ). Thus statin inhibition of protein endocytosis by podocytes may be favorable and protective.
In summary, we have described and quantified an albumin endocytic function in podocytes that is, partially at least, inhibitable by statins. The capacity in vitro of podocytes to endocytose albumin, and presumably other proteins, is several-fold higher than that in the extensively studied cultured PTEC. It should be emphasized that these findings do not reflect in vivo physiological conditions in which proximal tubular accumulation of endocytosed protein is many-fold that of podocytes. Clearly, however, these observations not only have significant implications for our understanding of podocyte and glomerular biology, but they also provide for the first time an easily quantifiable podocyte function against which toxic or protective interventions can be measured.
GRANTS
J. Eyre was supportd by a departmental PhD studentship from the Dept. of Infection, Immunity, and Inflammation at the Univ. of Leicester.
【参考文献】
Arici M, Chana R, Lewington A, Brown J, Brunskill NJ. Stimulation of proximal tubular cell apoptosis by albumin-bound fatty acids mediated by peroxisome proliferator activated receptor-gamma. J Am Soc Nephrol 14: 17-27, 2003.
Birn H, Fyfe JC, Jacobsen C, Mounier F, Verroust PJ, Ørskov H, Willnow TE, Moestrup SK, Christensen EI. Cubilin is an albumin binding protein important for renal tubular albumin reabsorption. J Clin Invest 105: 1353-1361, 2000.
Brunskill NJ. Albumin signals the coming of age of proteinuric nephropathy. J Am Soc Nephrol 15: 504-505, 2004.
Brunskill NJ, Cockcroft N, Nahorski S, Walls J. Albumin endocytosis is regulated by heterotrimeric GTP-binding protein G i-3 in opossum kidney cells. Am J Physiol Renal Fluid Electrolyte Physiol 271: F356-F364, 1996.
Brunskill NJ, Nahorski S, Walls J. Characteristics of albumin binding to opossum kidney cells, and identification of potential receptors. Pflügers Arch 433: 497-504, 1997.
Brunskill NJ, Stuart J, Tobin AB, Walls J, Nahorski S. Receptor-mediated endocytosis of albumin by kidney proximal tubule cells is regulated by phosphatidylinositide 3-kinase. J Clin Invest 101: 2140-2150, 1998.
Bussolati B, Deregibus MC, Fonsato V, Doublier S, Spatola T, Procida S, Di Carlo F, Camussi G. Statins prevent oxidized LDL-induced injury of glomerular podocytes by activating the phosphatidylinositol 3-kinase/AKT-signaling pathway. J Am Soc Nephrol 16: 1936-1947, 2005.
Chana RS, Lewington AJ, Brunskill NJ. Differential effects of peroxisome proliferator activated receptor-gamma (PPAR gamma) ligands in proximal tubular cells: thiazolidinediones are partial PPAR gamma agonists. Kidney Int 65: 2081-2090, 2004.
Christensen EI, Birn H. Megalin and cubilin: multifunctional endocytic receptors. Nat Rev Mol Cell Biol 3: 256-266, 2002.
Christensen EI, Birn H. Megalin, and cubilin: synergistic endocytic receptors in renal proximal tubule. Am J Physiol Renal Physiol 280: F562-F573, 2001.
Christensen EI, Nielsen S, Moestrup SK, Borre C, Maunsbach AB, de Heer E, Ronco P, Hammond TG, Verroust P. Segmental distribution of the endocytosis receptor gp330 in renal proximal tubules. Eur J Cell Biol 66: 349-364, 1995.
Cui S, Verroust P, Moestrup SK, Christensen EI. Megalin/gp330 mediates uptake of albumin in renal proximal tubule. Am J Physiol Renal Fluid Electrolyte Physiol 271: F900-F907, 1996.
Farquhar MG, Wissig SL, Palade GE. Glomerular permeability. I. Ferritin transfer across the normal capillary wall. J Exp Med 113: 47-66, 1961.
Gekle M. Renal tubule albumin transport. Annu Rev Physiol 67: 573-594, 2005.
Graham RC, Karnovsky MJ. Glomerular permeability. Ultra-structural cytochemical studies using peroxidases as protein tracers. J Exp Med 124: 1123-1134, 1966.
Haraldsson B, Sörensson J. Why do we not all have proteinuria? An update of our current understanding of the glomerular barrier. News Physiol Sci 19: 7-10, 2004.
Kanwar YS, Rosenzweig LJ. Clogging of the glomerular basement membrane. J Cell Biol 93: 489-494, 1982.
Kerjaschki D, Farquhar MG. Immunocytochemical localization of the Heymann nephritis antigen (GP330) in glomerular epithelial cells of normal Lewis rats. J Exp Med 157: 667-686, 1983.
Londono I, Bendayan M. Glomerular handling of native albumin in the presence of circulating modified albumins by the normal rat kidney. Am J Physiol Renal Physiol 289: F1201-F1209, 2005.
Morigi M, Buelli S, Angioletti S, Zanchi C, Longaretti L, Zoja C, Galbusera M, Gastoldi S, Mundel P, Remuzzi G, Benigni A. In response to protein load podocytes reorganize cytoskeleton, and modulate endothelin-1 gene: implication for permselective dysfunction of chronic nephropathies. Am J Pathol 166: 1309-1320, 2005.
Mundel P, Kriz W. Structure, and function of podocytes: an update. Anat Embryol 192: 385-397, 1995.
Mundel P, Reiser J, Borja AZM, Pavenstadt H, Davidson GR, Kriz W, Zeller R. Rearrangements of the cytoskeleton, and cell contacts induce process formation during differentiation of conditionally immortalized mouse podocyte cell lines. Exp Cell Res 236: 248-258, 1997.
Pavenstadt H, Kriz W, Kretzler M. Cell biology of the glomerular podocyte. Physiol Rev 83: 253-307, 2003.
Saleem MA, O'Hare MJ, Reiser J, Coward RJ, Inward CD, Farren T, Xing CY, Ni L, Mathieson PW, Mundel P. A conditionally immortalized human podocyte cell line demonstrating nephrin and podocin expression. J Am Soc Nephrol 13: 630-638, 2002.
Sidaway JE, Davidson RG, McTaggart F, Orton TC, Scott RC, Smith GJ, Brunskill NJ. Inhibitors of 3-hydroxy-3-methylglutaryl-CoA reductase reduce receptor-mediated endocytosis in opossum kidney cells. J Am Soc Nephrol 15: 2258-2265, 2004.
Smithies O. Why the kidney glomerulus does not clog: a gel permeation/diffusion hypothesis of renal function. PNAS 100: 4108-4113, 2003.
Yamazaki H, Saito A, Ooi H, Kobayashi N, Mundel P, Gejyo F. Differentiation-induced cultured podocytes express endocytically active megalin, a Heymann nephritis antigen. Nephron Exp Nephrol 96: e52-e58, 2004.
Yoshika N, Ito H, Akamatsu R, Hazikano H, Okada S, Matsuo T. Glomerular podocyte vacuolation in focal segmental glomerulosclerosis. Arch Pathol Lab Med 110: 394-398, 1986.
Zhai XY, Nielsen R, Birn H, Drumm K, Mildenberger S, Freudinger R, Moestrup SK, Verroust PJ, Christensen EI, Gekle M. Cubilin, and megalin mediated uptake of albumin in cultured proximal tubule cells of opossum kidney. Kidney Int. 58: 1523-1533, 2000.
作者单位：1 Department of Infection, Immunity, and Inflammation and 3 Department of Cell Physiology and Pharmacology, University of Leicester Faculty of Medicine and Biological Sciences, 2 Department of Nephrology, Leicester General Hospital, Leicester, and 4 Academic and Children‘s Renal Unit, Universi