* Department of Pediatrics, Yoshida Hospital, Niigata; Department of Pathology, Institute of Nephrology, Niigata University, Niigata; and Sysmex Corporation, Kobe, Japan
Address correspondence to: Dr. Masanori Hara, Department of Pediatrics, Yoshida Hospital, Yoshida-machi, Nishikanbara-gun, 959-0242 Niigata, Japan. Phone: 81-256-92-5111; Fax: 81-256-92-2610; E-mail: mhara{at}yhp-unet.ocn.ne.jp
Previously it was shown that urine from patients with nephritiscontains podocytes and their fragments (podocalyxin [PCX]-positivegranular structures [PPGS]), reflecting the degree of podocyteinjury. The present study was designed to trace PPGS to theirorigin. Urine samples and renal biopsy specimens from 53 childrenwith nephrotic syndrome and nephritis were examined immunohistochemically.Immunofluorescence studies of kidney sections using an anti-PCXantibody demonstrated that PPGS originated from the glomerulusand flowed into the tubular lumen. Electron microscopic examinationrevealed that PPGS originated from microvillous or vesicle-likestructures on injured podocytes in the glomerulus. For examiningthe origin of the PPGS, apical, slit-diaphragmatic, and basalportions of the podocytes were specifically stained, revealingthat PPGS are composed primarily of apical podocyte membranes.Several newly developed antibodies that are reactive with varioussegments of the PCX molecule were used to analyze more detailedmembrane structures, and it was found that PPGS contained intactPCX molecules, indicating that cell membrane structures areexcreted in urine. The quantification of PCX content and podocytenumbers revealed that urinary sediment PCX (u-sed-PCX) contentper urinary podocyte was much higher than PCX content per podocytefrom isolated glomeruli of normal controls, suggesting thatu-sed-PCX are derived from sources other than just the celldebris of detached podocytes. Analysis of the correlation betweenu-sed-PCX and renal histology revealed that the presence ofPPGS reflects acute glomerular injury. In conclusion, podocyteapical cell membranes are shed into the urine from injured podocytes,indicating a previously unrecognized manifestation of podocyteinjury.
The podocyte is a highly differentiated cell that has characteristicinterdigitating foot processes that cover the outer surfaceof the glomerular basement membrane (GBM) in the kidney (1).The turnover rate of podocytes is very low under normal andvarious pathologic conditions compared with that of other glomerularcells (2,3). Meanwhile, podocytes contribute to the hydraulicpermeability of the glomerulus and play a crucial role as afilter for macromolecules (1). Because of these biologic andmorphologic characteristics of podocytes, injuries to podocytesare accompanied by characteristic changes in morphology, asobserved by electron microscopy (EM), including effacement offoot processes, microvillous transformation, and occasionaldetachment from the GBM (4–7). In several immunologicand nonimmunologic forms of glomerulonephritis, the podocyteis the primary target of injury (8,9). Podocyte injury is alsoa key event leading to glomerular sclerosis. Recent studieshave revealed that the denuded GBM left behind after a podocytebecomes detached and subsequently adheres to parietal epithelialcells, resulting in the formation of a synechia of the glomerulartuft to Bowman’s capsule, which represents the earlieststage of segmental sclerosis (10,11).
We recently demonstrated the presence of podocytes and theircell fragments in the urinary sediment of patients with glomerulardiseases, in an immunofluorescence (IF) study using a specificmonoclonal antibody against podocalyxin (PCX), a glycoproteinthat is prominently expressed on podocytes (12). Quantificationof urinary podocytes has clinical significance in its abilityto predict acute glomerular lesions (13). In addition to urinarypodocytes, urine sediments from nephritic patients contain PCX-positivegranular structures (PPGS) in or around the urine casts. Wehypothesized that these structures represent urinary podocytesand their cell debris. We subsequently found that PPGS are excretedin the urine in far greater numbers compared with urinary podocytes.However, because we also found PPGS in the urine of patientswithout any urinary podocytes, we questioned whether these structurestruly represent cell debris from detached podocytes. Thus, thepurpose of the present study was to trace PPGS to their originimmunohistochemically.
Patients, Urine Samples, and Kidney Specimens
Urine samples voided in the morning were obtained from 50 healthychildren and adolescents (25 male and 25 female; mean age, 12.3yr; range, 3 to 20 yr) and 53 patients with active glomerulonephritisor nephrotic syndrome (29 male and 24 female; mean age, 11.3yr; range, 3 to 23 yr) during the period 1997 to 2001. The renaldiseases included IgA nephropathy (IgAN; 20 cases; mean proteinuriclevel, 0.78 g/d; range, 0.31 to 2.56 g/d), nephrotic syndrome(15 cases, steroid sensitive; mean proteinuric level, 1.32 g/d;range, 0.89 to 5.2 g/d), lupus nephritis (LN; 6 cases; meanproteinuric level, 1.13 mg/dl; range, 0.56 to 5.6 g/d), andHenoch-Schoenlein purpura nephritis (HSPN; 12 cases; mean proteinuriclevel, 1.25 g/dl; range, 0.45 to 3.14 g/d). In addition to routineurinalysis, urine samples were processed for urinary podocyteanalysis, quantification of urinary sediment PCX (u-sed-PCX),IF, and immunoelectron microscopic (IEM) studies of urine sediments.Renal biopsies performed during this study period included 26patients: IgAN (n = 15), HSPN (n = 7), and LN (n = 4). As anormal control, the kidney specimens from autopsy or normalparts of nephrectomized kidney were used. This study was approvedby the ethics committee of the Yoshida Hospital. Informed consentwas obtained from the patients or the parents.
Immunofluorescence
The urine sediments and frozen sections from renal biopsy specimenswere stained using the various primary and secondary antibodiesin the same manner as described previously (12). The primaryantibodies used were as follows: anti-podocalyxin: monoclonal(mouse), PHM5 (Australian Monoclonal Development, Artarmon,Australia) (14); anti-CR1: monoclonal (mouse), clone; To5 (DAKO,Glostrup, Denmark); anti-GLEPP1: monoclonal, clone; 5C11, agift from Dr. Roger Wiggins (15); anti-nephrin: polyclonal (rabbit),supplied by Dr. Hiroshi Kawachi (16); anti–ZO-1: polyclonal(rabbit; Zymed Laboratories, South San Francisco, CA); anti-3integrin: polyclonal (rabbit; Chemicon, Temecula, CA); anti-ezrin:monoclonal (rat), clone; M11, (Sanko Junyaku, Tokyo, Japan);and anti–-actin: monoclonal (mouse), clone AC-15 (Abcam,Cambridge, UK). The secondary antibodies used were as follows:FITC-labeled anti-mouse IgG (Cappel, Chester, PA), TRITC-labeledanti-mouse IgG (Cappel), FITC-labeled anti-rat IgG (Cappel),FITC-labeled anti-rabbit IgG (Cappel), and Cy3 conjugated anti-mouseIgG+IgM (Chemicon). Alexa488-conjugated phalloidin (MolecularProbes, Eugene, OR) was used.
To determine whether PPGS come from a particular subcellularportion of podocytes, we performed IF study on urine sedimentsusing various antibodies that react with three distinct subcellularregions of podocytes: (1) the apical region of podocytes usinganti-PCX, anti-GLEPP1, and anti-CR1 antibodies; (2) the slitdiaphragm region using anti-nephrin and anti–ZO-1 antibodies;and (3) the basal region using the anti-3 integrin antibody.To examine the detailed relationship between PCX and cytoskeletalmolecules, we performed IF study to detect ezrin and actin inurine sediments using antibodies to ezrin, -actin, and Alexa488-labeledphalloidin.
Histologic Examination (LM, EM, IEM)
Twenty-six renal biopsy specimens were analyzed histologicallyfrom patients in whom u-sed-PCX levels were measured at thetime of renal biopsy. The following pathologic parameters wereanalyzed according to the criteria of Shigematsu et al. (17):extracapillary change (acute and chronic), intracapillary change(acute and chronic), and tubulointerstitial change (acute andchronic). Pathologists who were blinded to u-sed-PCX levelsperformed the pathologic examination. Among 26 renal biopsyspecimens, 10 specimens including five cases with IgAN, threecases with HSPN, and two cases with LN, were processed for EMexamination. The ultrastructural configuration of podocytesin the urine sediments obtained from five patients with HSPN(two cases) and IgAN (three cases) was examined by IEM examinationas reported previously (12).
Quantification of Urinary Podocytes (U-Podocyte Test)
The number of urinary podocytes was counted as reported previously(18). The number of podocytes in urine was expressed as cells/ml.
Production of Monoclonal Antibodies to PCX
cDNA coding for the human PCX was obtained from the human kidneycDNA library using reverse transcriptase–PCR based onthe method by Kershaw et al. (19). Two kinds of GST fusion protein(PC-46, whole portion of PCX; PC-35, extracellular portion)were obtained to produce monoclonal antibodies and characterizationof antibodies. The GST fusion protein that contained the intracellularportion of rabbit PCX (RBT-Intra) was provided by Dr. DavidKershaw (University of Michigan, Ann Arbor, MI).
Monoclonal antibodies against human PCX were produced from BALB/Cmice that were immunized with purified PC-46 by standard methods.The resulting hybridomas were grown in 96-well plates and selectedand subcloned on the basis of IF pattern assayed on cryostatsections of human renal cortex or ELISA using the polystyrenemultiwell plates coated with a wheat germ agglutinin-bindingfraction in a Triton X-100 glomerular lysate as previously reported(20). Among 12 monoclonal antibodies, two clones (70-4, No45)were chosen and further characterized by Western blotting andELISA. The monoclonal antibody (PHM5) that reacts with the carbohydrateportion of human podocalyxin was purchased.
The IgG fraction of PHM5 was affinity-purified using MEP HyperCel (Invitrogen, Carlsbad, CA). After purification, the IgGfraction was FITC-labeled using standard techniques (21).
ELISA
ELISA system was used to quantify PCX in various samples aspreviously reported (20). This system gave a linear plot overthe range of 12.5 to 800 ng/ml. The detection limit of thisELISA assay was 6.25 ng/ml.
For screening, polystyrene multiwell plates were coated witha wheat germ agglutinin-binding fraction in a Triton X-100 glomerularlysate served as a standard in the ELISA assay above at theconcentration of 5 µg/ml. After blocking of the plate,the hybridoma culture supernatant was added and processed tothe reaction of peroxidase (POD)-labeled anti-mouse IgG (Cappel)and the development of POD.
For the characterization of monoclonal antibodies, the polystyrenemultiwell plates were coated with PC-46, PC-35, and RBT-Intra(0.2 µg/ml, respectively). The plate was incubated with70-4, No45, and PHM5 (5 µg/well, respectively) at 37°Cfor 1 h. After washing, the plate was further incubated withPOD-labeled anti-mouse IgG (Cappel). The development of PODwas processed using standard methods.
Western Blot
Detection of protein in the human glomerular lysate was analyzedby Western blotting according to the procedures described previously(20).
Quantification of Podocyte Number per Glomerulus
The kidney specimens were obtained at the time of autopsy fromthree patients (72-yr-old man, 78-yr-old woman, and 82-yr-oldwoman) who died of diseases other than kidney diseases. Thenumber of podocytes per glomerulus was calculated on the basisof the method of Hishiki et al. (22).
Quantification of PCX of Isolated Glomeruli
Isolated glomeruli were obtained from the kidney specimens aboveusing mesh sieving. The number of glomeruli in the suspensionwas counted, and after centrifugation, the pellet was dissolvedin 0.2% Triton X-100 in PBS (23) and left to stand for 1 h atroom temperature. After centrifugation at 15,000 rpm for 5 min,the PCX content in the supernatant was measured in the samemanner as in quantification of PCX in urine sediments.
Statistical Analyses
All data were expressed as the mean ± SEM. A comparisonbetween the groups was made by unpaired t test. A comparisonbetween PCX content/podocyte from urinary sediments and PCX/podocytefrom isolated glomeruli was made by Mann-Whitney test becauseof a small number of samples. Differences between groups wereconsidered to be significant at P < 0.05.
PPGS Are Derived from Injured Podocytes
Urine sediments that stained with anti-PCX antibody were classifiedinto two structures: (1) PCX-positive cells (podocytes) and(2) PPGS in any cases with nephrotic syndrome, HSPN, IgAN, andLN, as previously reported (12). PCX-positive cells and PPGSwere frequently seen in the urine in cases during the acutephases of the diseases. A representative finding is shown inFigure 1. The location of the PPGS was frequently in the castsand occasionally around the cast in a scattered manner.
Figure 1. Immunofluorescence (IF) of urine sediments from a patient with Henoch-Schoenlein purpura nephritis (HSPN). Urine sediments that were cytospun on slide glass were stained with anti-podocalyxin (anti-PCX) antibody. Urine sediments contain numerous PCX-positive granular structures (PPGS) mainly in casts and also around casts. Several podocytes are seen (arrow). Magnification, x200.
When frozen sections from normal and nephritic kidneys werestained with anti-PCX antibody (PHM5) and carefully examined,a slightly more granular appearance of the glomerular stainingwas found in nephritic kidneys compared with the glomeruli ofnormal kidneys. A representative figure of normal and nephriticglomeruli is shown in Figure 2, A and B, respectively. In addition,fine granular structures similar to urinary PPGS were occasionallyfound in the tubular lumen in several cases with HSPN with severeglomerular lesions (Figure 2C).
Figure 2. IF of kidney section from a patient with HSPN (A) and normal control (B). Podocytes in nephritic glomeruli were more granular in appearance than those of the normal kidney. The fine granular structures similar to PPGS (arrow) were found in the tubular lumen in a case with HSPN (C). Magnification, x800 in A and B, x200 in C.
PPGS Have Cell Polarity
IF study was performed on urine sediments using various antibodiesthat react with three distinct subcellular regions of podocytes:(1) the apical region of podocytes using anti-PCX, anti-GLEPP1,and anti-CR1 antibodies; (2) the slit diaphragm region usinganti-nephrin and anti–ZO-1 antibodies; and (3) the basalregion, using the anti-3 integrin antibody. Urine sedimentsfrom various types of nephritis stained with anti-PCX intensely,with only occasional weak staining with anti-CR1 and GLEPP1.On the contrary, the urine sediments did not stain with anti-nephrin,anti–ZO-1, and anti-3 integrin, suggesting that only apicalportions of podocyte cell membranes ware excreted in the urine(Figure 3, A through C). Positive controls for all of theseantibodies in normal kidney sections confirmed good stainingwith each antibody (Figure 3, A' through C').
Figure 3. IF of urine sediment from a patient with IgA nephropathy using antibodies that react with different subcellular regions of podocyte: anti-PCX (reacting with apical cell membrane of podocyte), anti-nephrin (reacting with slit diaphragm region), and anti-3 integrin (reacting with basal region of podocyte). Casts in the sediment are stained with anti-PCX (A) but not with anti-nephrin (B) and anti-3 integrin (C). IF of kidney sections from normal control are in the lower panel, with anti-PCX (A'), anti-nephrin (B'), and anti-3 integrin (C'). Magnification, x400.
Urine Sediments Have Vesicles of Podocytes Similar to Microvilli
EM of biopsy specimens demonstrated protuberant apical cellmembranes on the podocyte surface, consistent with microvilloustransformation. These changes of apical cell membranes wereseen in several types of nephritis in this study and were mostprominently in cases with heavy proteinuria (Figure 4).
Figure 4. Electron microscopy (EM) of podocyte in the kidney section from a patient with HSPN. Numerous protuberant apical cell membranes on the podocyte surface (microvillous transformation) are seen. Size bar is indicated.
IEM of urine sediments showed that urinary podocytes have numerousmicrovillous structures or vesicle-like structures on theircell surfaces. Vesicles that contained different sizes of double-layeredcell membrane structures, similar to those seen near urinarypodocytes, were often observed in clusters in the urine sediments(Figure 5).
Figure 5. Immunoelectron microscopy (IEM) of urine sediments from a patient with IgA nephropathy. Numerous vesicles are seen on the cell surface of urinary podocytes (A). The vesicles with different size of double-layered cell membrane structures similar to those of urinary podocytes were observed as a cluster in the urine sediments (B). Size bars are indicated.
PPGS Have Whole-Cell Membrane Structures Antibody Characterization.
Among several monoclonal antibodies raised in this study, wedeveloped two kinds of antibodies (70-4 and No45). Western blotanalysis of 70-4, No45, and PHM5 showed the bands with the samemolecular weight, approximately 160 to 170 kD (Figure 6A). Theexamination using ELISA in which various antigens, includingfull-length podocalyxin (PC46), extracellular portion (PC35),and intracellular portion (RBT-Intra), were bound showed that(1) 70-4 are reactive with both PC46 and RBT-Intra and negativewith PC35 and (2) No45 are reactive with PC46 and PC35 and negativewith RBT-Intra, whereas PHM5 all are negative with PC46, PC35,and RBT-Intra (Figure 6B). These results clearly showed that70-4 is a monoclonal antibody that recognizes the intracellularregion of PCX, whereas No45 is a monoclonal antibody that recognizesthe extracellular region of PCX. PHM is a monoclonal antibodythat does not react with the protein portion of PCX. PHM5 isdemonstrated as a clone that reacts with carbohydrate of podocalyxin(14).
Figure 6. (A) Immunoblot analysis of glomerular lysates with PHM5 (recognizing sugar component of PCX), 70-4 (recognizing intracellular peptide of PCX), and No45 (recognizing extracellular peptide of PCX) showed the bands of PCX with molecular weight of approximately 160 to 170 kD. Antibody of control (RVG1, Cont. 1; and PBS, Cont. 2) are totally negative. (B) ELISA using plates coated with three kinds of antigens: full-length PCX (PC46), extracellular portion of PCX (PC35), and intracellular portion of PCX (RBT-Intra). Monoclonal antibody (70-4) is reactive with both PC46 and RBT-Intra and negative with PC35. Monoclonal antibody (No45) is reactive with PC45 and PC35 and negative with RBT-Intra. PHM5 are negative with PC46, PC35, and RBT-Intra. These results clearly showed that 70-4 is a monoclonal antibody that recognizes the intracellular portion of PCX, and No45 is a monoclonal antibody that recognizes the extracellular portion of PCX. PHM is a monoclonal antibody that does not recognize both protein portions of PCX.
IF Study of Urine Sediments Using Antibodies that React with Different Regions of PCX.
IF using different antibodies to PCX, including anti-sugar components(PHM5), anti-extracellular peptide (No45), and anti-intracellularpeptide (70-4), demonstrated nearly the same localization patternwith these three antibodies (Figure 7), indicating that PPGScontained whole-cell membrane structures.
Figure 7. IF of urine sediments from a patient with IgA nephropathy using antibodies that react with different portions of PCX. Upper channel: PHM5 (reacting with sugar components of PCX) and No45 (reacting with extracellular portion of PCX). The picture of merge shows nearly the same localization of both antibodies. Lower channel: PHM5 (reacting with sugar components of PCX) and 70-4 (reacting with intracellular portion of PCX). The picture of merge shows nearly the same localization of both antibodies. These findings clearly indicate that PPGS contain whole-cell membrane structures. Magnification, x400.
PPGS Are not Derived from Cell Debris of Detached Podocytes
Three kidneys were obtained at the time of autopsy. A portionof renal cortex was processed for light microscopic examinationto count the number of podocytes per glomerulus, and the remainingportion of the kidney was used for glomerular isolation. Theactual values of podocytes per glomerulus were 412, 388, and420, respectively. The numbers of glomeruli sieved were 120,000,200,000, and 170,000, respectively. The u-sed-PCX and the numberof urinary podocytes from 15 cases with HSPN, IgAN, and LN werealso examined.
The quantified values of PCX from isolated glomeruli were dividedby the total number of podocytes (cells) from isolated glomeruliin each kidney. The values of u-sed-PCX (ng/ml) were also dividedby the number of urinary podocyte (cells/ml). The mean ±SEM of PCX/podocytes from isolated glomeruli was 4.0 ±0.35 pg/podocyte, and PCX/urinary podocyte was 16.1 ±4.1 ng/cell (P < 0.01). The results are summarized in Table 1.
Table 1. Ratio of PCX to podocyte number in urine sediments and isolated glomerulia
PPGS Are not Associated with Actin Filaments
Recent studies demonstrated that the PCX molecule binds to actinfilaments via NERF 2 and ezrin. Therefore, an IF study was performedto detect ezrin and actin in urine sediments using antibodiesto ezrin, -actin, and Alexa488-labeled phalloidin. Urinary PPGSwere negative for ezrin, -actin, and phalloidin (Figure 8),suggesting against the presence of these PCX-binding proteinsin PPGS.
Figure 8. IF of urine sediments using antibodies that react with the molecules between PCX and actin filaments: anti-PCX (A), anti-ezrin (B), and phalloidin (C). Casts in the sediment are stained with anti-PCX but not with anti-ezrin and phalloidin. IF of kidney sections from normal control stained with anti-PCX (A'), anti-ezrin (B'), and phalloidin (C') in the lower panel. Magnification, x400 in A through C; x200 in A' through C'.
PPGS Reflect Acute Podocyte Injury Urinary Sediment PCX versus Urinary Podocytes and Proteinuria.
The levels of u-sed-PCX, proteinuria, and urinary podocytesin three groups of patients are shown in Table 2. Although asmall amount of u-sed-PCX was detected in the normal controlgroup, the levels of u-sed-PCX in both the nephrotic syndromeand nephritis groups were significantly higher than in the controlgroup (P < 0.0001 and <0.0002, respectively, versus control),and the level in the nephritis group was significantly higherthan the level in the nephrotic group (P < 0.03). The levelof proteinuria in the nephrotic group was significantly higherthan that in both the normal control and nephritic groups (P< 0.0001 and P < 0.006, respectively). Urinary podocyteswere detected only in the nephritis group, despite the findingof higher levels of proteinuria in the nephrotic group.
Table 2. U-sed-PCX, proteinuria, and u-podocyte levels in normal controls, nephrotic syndrome, and IgAN/HSPN/LNa
Urinary Sediment PCX versus Histology.
Twenty-six renal biopsies, including IgAN (n = 15), HSPN (n= 7), and LN (n = 4), were analyzed histologically. The caseswith acute extracapillary changes had a significantly higheru-sed-PCX level than those without these changes (37.41 ±6.11 versus 12.53 ± 6.52 ng/ml; P < 0.05). There wasno difference between the cases with or without other changes,such as chronic extracapillary changes and tubulointerstitialchanges. There was no correlation with the degree of mesangialproliferation.
Podocytes are injured in many forms of human and experimentalglomerular diseases. Independent of the underlying disease,if the early structural changes in podocytes are not reversed,then severe and progressive glomerular sclerosis develops. Thesechanges include podocyte vacuolization, pseudocyst formation,microvillous transformation, and detachment of podocytes fromthe GBM, resulting in podocyte loss from the glomeruli. Previousstudies of human diabetic nephropathy and IgAN have providedconvincing evidence for a correlation between the loss of podocytesand the progression of glomerular diseases (24,25). The detectionof urinary podocytes in various glomerular diseases, includingIgAN and diabetic nephropathy, is consistent with the podocyteloss from the glomeruli (13,26). Meanwhile, in addition to urinarypodocytes, numerous podocyte-related structures such as PPGSare found in urine. Appearance of these structures in the urineseems to reflect closely the events ongoing in the glomerulus(27). Therefore, we performed a detailed analysis of these podocyte-relatedsubstances in the urine in an attempt to improve our understandingof the mechanism(s) of podocyte injury during glomerular disease.
To trace PPGS to their origin immunohistochemically, we firstexamined PPGS morphologically, analyzing both urine sedimentsand kidney sections. We carefully examined kidneys sectionsfrom nephritic patients and normal controls stained with anti-PCXantibody and found the development of a fine granular appearanceon the outer surface of capillary walls in sections with nephritiscompared with controls, suggesting that PPGS originate fromthese granular structures on the capillary walls. Similar-sizedgranular structures to PPGS were also found in the luminal spaceof renal tubules, suggesting that the PPGS originated from theglomerulus and flowed into the tubular lumen. EM examinationof biopsy samples from patients with nephritis revealed finemicrovillous or vesicle-like structures on the apical surfaceof injured podocytes, in addition to other morphologic alterations.IEM of urine sediments from nephritic patients demonstratedthe microvillous or vesicle-like structures on the surface ofurinary podocytes that were similar to the changes found onpodocytes in situ. Clustering of vesicle-like structures withdifferent sizes was detected in urine sediments, findings alsovery similar to the structures on podocytes in the urine orglomerulus. The close similarity of these microvillous or vesicle-likestructures at both the light microscopic and EM levels stronglyindicated that PPGS in the casts in the urine sediments originatedfrom the microvillous or vesicle-like structures on injuredpodocytes in the glomerulus.
We further examined the subcellular portions of podocytes fromwhich these vesicle-like structures originated. Although ourfirst observations of urine sediments that immunostained withan anti-PCX antibody suggested that PPGS were derived from thecell debris or the destroyed cell membranes of detached podocytes,we subsequently found that the number of PPGS was disproportionalcompared with the number of podocytes, giving doubt to our speculationof the origin of PPGS in urine. To clarify this, we immunostainedurine sediments with antibodies that recognize different subcellularregions of podocytes, including the apical, slit diaphragm,and the basal regions. The IF study of urine sediments clearlydemonstrated that PPGS were positive only for anti-PCX antibody,indicating a definite polarity of PPGS. If the PPGS were derivedfrom the general cell debris of detached podocytes, then PPGSshould have stained with any of these region-specific antibodies.This finding indicated that PPGS are excreted into urine notas a result of passive cellular destruction during or afterthe detachment of podocytes from the GBM but as a result ofother mechanisms, such as active cell membrane shedding or vesiclerelease, that might be reflective of biologic activities ofinjured podocytes.
Next we examined whether the PPGS truly originated from thecell membranes of podocytes or not. To exclude the possibilitythat PPGS originate from the cell membranes of other cells suchas tubular cells on which PCX molecules are bound, we examinedPPGS using monoclonal antibodies that recognize three differentregions of the PCX molecule. The IF study revealed that PPGSincluded intact molecules of PCX, including the intracellular,extracellular, and sugar components, clearly indicating thatPPGS originated from the cell membranes of podocytes. Additionalevidence of cell membrane structures was based on morphologicfindings. IEM of urine sediments from patients with glomerulardiseases demonstrated double-layered structures, suggestiveof typical cell membranes.
Our next investigation of PPGS was to clarify their relationshipto the cytoskeleton of podocytes. Recent studies revealed thatthe podocyte actin cytoskeleton is connected to apical integralmembrane molecules, including PCX (28,29). We examined thisassociation using antibodies that recognize ezrin and -actin,and phalloidin to label actin. The results were negative foreach of these molecules. These results showed that PPGS containedonly membrane-bound podocyte proteins and apparently lackedcytoskeletal components. Takeda et al. (30) demonstrated thatthe PCX/NHERF2/ezrin/actin cytoskeleton association was disruptedin damaged podocytes. Thus, it is likely that the PPGS lackezrin and actin filaments in their structures.
Urinary PPGS are numerous and definitely disproportional comparedwith the number of urinary podocytes, so we also examined theratio of PCX content per podocyte in both urine sediments andisolated glomeruli. Calculation of the PCX/podocyte ratio demonstrateda markedly higher ratio of PCX/podocyte from urine than PCX/podocytefrom isolated glomeruli. If the PPGS in urine sediments originatedfrom the cell debris of detached podocytes, then the ratio ofPCX/urinary podocyte should be roughly equivalent to the PCX/podocytefrom isolated glomeruli. The markedly higher ratio of PCX/urinarypodocyte indicated that the PCX in urinary sediments is derivedfrom not only cell debris of detached podocytes but also othersource(s). In addition, significant amounts of PCX were detectedin urine sediments that did not contain urinary podocytes inpatients with nephrotic syndrome, which further supports theconcept that the PCX in urine sediments does not originate onlyfrom detached podocytes.
EM examination revealed a close similarity between the microvillousstructures on the podocyte surface in the kidney and the vesicle-likestructures in urine sediments, suggesting that these microvillousstructures in the glomeruli are excreted into the urine by sheddingfrom podocytes. By definition, shedding is the release of solubleor vesicle-associated cell surface constituents, without affectingcell viability (31). Shedding membrane vesicles from the cellsurface is generally a selective process that is widespreadin both normal and diseased cells. The excretion of vesicle-likestructures from injured podocytes found in this study is clearlyconsistent with the process of shedding in general. The ideaof shedding has been previously proposed by several authors.Wiggins et al. (32) observed a similar phenomenon in experimentalnephritis, although a specific marker for podocytes was notused in their studies. Pascual et al. (33) identified membrane-boundCR1 (CD35) in human urine and demonstrated the evidence forits release by podocytes. Lehto et al. (34) demonstrated urinaryexcretion of protectin (CD59) in membranous glomerulonephritis.The authors on both of these papers (33,34) used CR1 (CD35)or CD59 for a podocyte marker; however, the expression of CR1or CD53 could be altered depending on the level of complementactivation or phenotypical change of podocytes in diseases.These molecules might be less appropriate for the purpose ofexamining urinary excretion. On the contrary, PCX is expressedprimarily on the apical cell surface of podocytes, and the expressionof PCX in glomeruli was not altered in various kinds of nephritis(35,36). In the present study, we successfully observed extensiveurinary shedding of podocyte apical cell membranes using PCXas a podocyte marker. We believe that this shedding processrepresents an active biologic process that includes alteredmembrane dynamics of podocytes. Extensive microvillous transformationin the injured podocytes is also representative of similar membranedynamics (37).
In conclusion, we have found that podocyte apical cell membranesare shed into the urine after acute podocyte injury and thatthis represents a previously unrecognized manifestation of podocyteinjury.
Acknowledgments
Portions of this study were presented as a poster at the 2003American Society of Nephrology national meeting, November 12–17,2003, San Diego, CA.
We thank Dr. William E. Smoyer for thoughtful review of thismanuscript.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
Andrews P: Morphological alterations of the glomerular (visceral) epithelium in response to pathological and experimental situations.
J Electron Microsc Tech 9
: 115
–144, 1988[CrossRef][Medline]
Somlo S, Mundel P: Getting a foothold in nephrotic syndrome.
Nat Genet 24
: 333
–335, 2000[CrossRef][Medline]
Kerjaschki D: Caught flat-footed: Podocyte damage and the molecular bases of focal glomerulosclerosis.
J Clin Invest 108
: 1583
–1587, 2001[CrossRef][Medline]
Kriz W, Gretz N, Lemley KV: Progression of glomerular diseases: Is the podocyte the culprit?
Kidney Int 54
: 687
–697, 1998[CrossRef][Medline]
Kriz W: Progressive renal failure—Inability of podocytes to replicate and the consequences for development of glomerulosclerosis.
Nephrol Dial Transplant 11
: 1738
–1742, 1996[Free Full Text]
Hara M, Yamamoto T, Yanagihara T, Takada T, Itoh M, Adachi Y, Yoshizumi A, Kawasaki K, Kihara I: Urinary excretion of podocalyxin indicates glomerular epithelial cell injuries in glomerulonephritis.
Nephron 69
: 397
–403, 1995[Medline]
Hara M, Yanagihara T, Takada T, Itoh M, Matsuno M, Yamamoto T, Kihara I: Urinary excretion of podocytes reflects disease activity in children with glomerulonephritis.
Am J Nephrol 18
: 35
–41, 1998[CrossRef][Medline]
Hancock WW, Atkins RC: Monoclonal antibodies to human glomerular cells: A marker for glomerular epithelial cells.
Nephron 33
: 83
–90, 1983[Medline]
Wiggins RC, Wiggins JE, Goyal M, Wharram BL, Thomas PE: Molecular cloning of cDNAs encoding human GLEPP1, a membrane protein tyrosine phosphatase: Characterization of the GLEPP1 protein distribution in human kidney and assignment of the GLEPP1 gene to human chromosome 12p12–p13.
Genomics 27
: 174
–181, 1995[CrossRef][Medline]
Koop K, Eikmans M, Baelde HJ, Kawachi H, De Heer E, Paul LC, Bruijn JA: Expression of podocyte-associated molecules in acquired human kidney diseases.
J Am Soc Nephrol 14
: 2063
–2071, 2003[Abstract/Free Full Text]
Shigematsu H: Histological grading and staging of IgA nephropathy.
Pathol Int 47
: 194
–202, 1997[Medline]
Hara M, Yanagihara T, Kihara I: Urinary podocytes in primary focal segmental glomerulosclerosis.
Nephron 89
: 342
–347, 2001[CrossRef][Medline]
Kershaw DB, Beck SG, Wharram BL, Wiggins JE, Goyal M, Thomas PE, Wiggins RC: Molecular cloning and characterization of human podocalyxin-like protein. Orthologous relationship to rabbit PCLP1 and rat podocalyxin.
J Biol Chem 272
: 15708
–15714, 1997[Abstract/Free Full Text]
Kanno K, Kawachi H, Uchida Y, Hara M, Shimizu F, Uchiyama M: Urinary sediment podocalyxin in children with glomerular diseases.
Nephron Clin Pract 95
: c91
–c99, 2003[CrossRef][Medline]
Holmes K, Fowlkes BJ, Schmid I, Giorgi JV: Preparation of cells and reagents for flow cytometry. In:
Current Protocols in Immunology, edited by Coligan JE, Kruisbeek AM, Margulies DH, Shevach EM, Strober W, New York, John Wiley & Sons, 1994
, pp 5.3.1
–5.3.13
Hishiki T, Shirato I, Takahashi Y, Funabiki K, Horikoshi S, Tomino Y: Podocyte injury predicts prognosis in patients with IgA nephropathy using a small amount of renal biopsy tissue.
Kidney Blood Press Res 24
: 99
–104, 2001[CrossRef][Medline]
Kerjaschki D, Poczewski H, Dekan G, Horvat R, Balzar E, Kraft N, Atkins RC: Identification of a major sialoprotein in the glycocalyx of human visceral glomerular epithelial cells.
J Clin Invest 78
: 1142
–1149, 1986
Pagtalunan ME, Miller PL, Jumping-Eagle S, Nelson RG, Myers BD, Rennke HG, Coplon NS, Sun L, Meyer TW: Podocyte loss and progressive glomerular injury in type II diabetes.
J Clin Invest 99
: 342
–348, 1997[Medline]
Lemley KV, Lafayette RA, Safai M, Derby G, Blouch K, Squarer A, Myers BD: Podocytopenia and disease severity in IgA nephropathy.
Kidney Int 61
: 1475
–1485, 2002[CrossRef][Medline]
Nakamura T, Ushiyama C, Suzuki S, Hara M, Shimada N, Ebihara I, Koide H. Urinary excretion of podocytes in patients with diabetic nephropathy.
Nephrol Dial Transplant 15
: 1379
–1383, 2000[Abstract/Free Full Text]
Saleem MA: Urine—A mirror of the (glomerular) soul?
Nephron Clin Pract 95
: c75
–c76, 2003[CrossRef][Medline]
Takeda T: Podocyte cytoskeleton is connected to the integral membrane protein podocalyxin through Na+/H+-exchanger regulatory factor 2 and ezrin.
Clin Exp Nephrol 7
: 260
–269, 2003[CrossRef][Medline]
Orlando RA, Takeda T, Zak B, Schmieder S, Benoit VM, McQuistan T, Furthmayr H, Farquhar MG: The glomerular epithelial cell anti-adhesin podocalyxin associates with the actin cytoskeleton through interactions with ezrin.
J Am Soc Nephrol 12
: 1589
–1598, 2001[Abstract/Free Full Text]
Takeda T, McQuistan T, Orlando RA, Farquhar MG: Loss of glomerular foot processes is associated with uncoupling of podocalyxin from the actin cytoskeleton.
J Clin Invest 108
: 289
–301, 2001[CrossRef][Medline]
Taylor DD, Black PH: Shedding of plasma membrane fragments. In:
Developmental Biology, edited by Steinberg M, New York, Plenum Press, 1986
, pp 35
–57
Wiggins R, Glatfelter A, Kshirsagar B, Beals T: Lipid microvesicles and their association with procoagulant activity in urine and glomeruli of rabbits with nephrotoxic nephritis.
Lab Invest 56
: 264
–272, 1987[Medline]
Pascual M, Steiger G, Sadallah S, Paccaud JP, Carpentier JL, James R, Schifferli JA: Identification of membrane-bound CR1 (CD35) in human urine: Evidence for its release by glomerular podocytes.
J Exp Med 179
: 889
–899, 1994[Abstract/Free Full Text]
Lehto T, Honkanen E, Teppo AM, Meri S: Urinary excretion of protectin (CD59), complement SC5b-9 and cytokines in membranous glomerulonephritis.
Kidney Int 47
: 1403
–1411, 1995[Medline]
Hara M, Yanagihara T, Takada T, Itoh M, Adachi Y, Yoshizumi A, Kawasaki K, Yamamoto T, Kihara I: Podocalyxin on the glomerular epithelial cells is preserved well in various glomerular diseases.
Nephron 67
: 123
–124, 1994[Medline]
Moll S, Miot S, Sadallah S, Gudat F, Mihatsch MJ, Schifferli JA: No complement receptor 1 stumps on podocytes in human glomerulopathies.
Kidney Int 59
: 160
–168, 2001[CrossRef][Medline]
Le Hir M, Keller C, Eschmann V, Hahnel B, Hosser H, Kriz W: Podocyte bridges between the tuft and Bowman’s capsule: An early event in experimental crescentic glomerulonephritis.
J Am Soc Nephrol 12
: 2060
–2071, 2001[Abstract/Free Full Text]
Received for publication July 16, 2004.
Accepted for publication November 16, 2004.
This article has been cited by other articles:
Y. Li, Y. S. Kang, C. Dai, L. P. Kiss, X. Wen, and Y. Liu Epithelial-to-Mesenchymal Transition Is a Potential Pathway Leading to Podocyte Dysfunction and Proteinuria
Am. J. Pathol.,
February 1, 2008;
172(2):
299 - 308.
[Abstract][Full Text][PDF]
S. H. Sung, F. N. Ziyadeh, A. Wang, P. E. Pyagay, Y. S. Kanwar, and S. Chen Blockade of Vascular Endothelial Growth Factor Signaling Ameliorates Diabetic Albuminuria in Mice
J. Am. Soc. Nephrol.,
November 1, 2006;
17(11):
3093 - 3104.
[Abstract][Full Text][PDF]
M. Barton, J. J. Mullins, M. A. Bailey, and M. Kretzler Role of Endothelin Receptors for Renal Protection and Survival in Hypertension: Waiting for Clinical Trials
Hypertension,
November 1, 2006;
48(5):
834 - 837.
[Full Text][PDF]
Top
Abstract
Introduction
3 integrin: polyclonal (rabbit; Chemicon, Temecula, CA); anti-ezrin: monoclonal (rat), clone; M11, (Sanko Junyaku, Tokyo, Japan); and anti–
-actin: monoclonal (mouse), clone AC-15 (Abcam, Cambridge, UK). The secondary antibodies used were as follows: FITC-labeled anti-mouse IgG (Cappel, Chester, PA), TRITC-labeled anti-mouse IgG (Cappel), FITC-labeled anti-rat IgG (Cappel), FITC-labeled anti-rabbit IgG (Cappel), and Cy3 conjugated anti-mouse IgG+IgM (Chemicon). Alexa488-conjugated phalloidin (Molecular Probes, Eugene, OR) was used. 
