 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pathophysiology of Renal Disease and Progression |
,
* INSERM U652, Departments of Nephrology and Pathology, Hôpital Européen Georges Pompidou, University Paris 5, and Association pour l’Utilisation du Rein Artificiel, Paris, France
Address correspondence to: Dr. Patrick Bruneval, Department of Pathology, Hôpital Européen Georges Pompidou, 20 rue Leblanc, 75015 Paris, France. Phone: +33-1-56-09-38-60; Fax: +33-1-56-09-38-89; E-mail: patrick.bruneval{at}hop.egp.ap-hop-paris.fr
Received for publication April 7, 2006. Accepted for publication July 6, 2006.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
-actinin-4), podocyte synthesized proteins (vascular endothelial growth factor and novH), transcription factors (WT1 and PAX2), cyclin-dependent kinase inhibitor p57, and intermediate filaments (cytokeratins and vimentin) was tested. In addition, six normal fetal kidneys were studied to track the ontogeny of parietal podocytes. The podocyte protein labeling detected parietal podocytes in all of the kidneys, was found in 76.6% on average of Bowman’s capsule sections, and was prominent at the vascular pole. WT1 and p57 were expressed in some parietal cells, whereas PAX2 was present in all or most of them, so some parietal cells coexpressed WT1 and PAX2. Furthermore, parietal podocytes coexpressed WT1 and podocyte proteins. Cytokeratin-positive cells covered a variable part of the capsule and did not express podocyte proteins. Tuft-capsular podocyte bridges were present in 15.5 ± 3.7% of the glomerular sections. Parietal podocytes often covered the juxtaglomerular arterioles and were present within the extraglomerular mesangium. Parietal podocytes were present in fetal kidneys. Parietal podocytes that express the same epitopes as visceral podocytes do exist along Bowman’s capsule in the normal adult kidney. They are a constitutive cell type of the Bowman’s capsule. Therefore, their role in physiology and pathology should be investigated. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
The podocytes and PEC have the same embryonic origin (reviewed in references [7–9]). Glomeruli develop from the metanephric mesenchyme. This development passes through several stages (10): Vesicle, comma and S-shaped, glomerular capillary loop, and mature glomerulus. The distal extremities of the dividing branches of the ureteral bud (the future collecting ducts) penetrate into the loose undifferentiated mesenchyme and induce aggregates of mesenchymal cells that express vimentin. Then these cells acquire an epithelial phenotype, express cytokeratins (CK), and form renal vesicles. At the lower end of the S-shaped bodies, Bowman’s space begins to form, limited outside by a narrow band of PEC and inside by the crown of visceral epithelial cells, the future podocytes. When the glomerulus passes from the S-shaped body to the capillary loop stage, the podocytes acquire their definitive phenotype, arriving at the mature glomerulus stage. They take on their mature form with the formation of pedicels along the glomerular basement membrane (GBM), and the apical tight junctions are replaced by the slit diaphragms. They lose their mitotic activity and no longer express proliferation markers. They express de novo the cyclin-dependent kinase inhibitors (CKI) p27 and p57 (11,12), which prevent cellular division by blocking the cyclin-cyclin–dependent kinase complexes (13). The podocytes no longer express CK, an epithelial marker, but now reexpress vimentin, a protein that is characteristic of the intermediate filaments of mesenchymal cells. They begin to express the proteins that are specific to mature podocytes (14,15). These proteins include, among others, podocalyxin and glomerular epithelial protein-1 (GLEPP-1) on the plasma membranes; nephrin, podocin, and CD2-AP associated with slit diaphragms; and synaptopodin and -actinin-4 associated with pedicel cytoskeleton. During nephrogenesis and in mature glomeruli, the PEC express CK in a heterogeneous manner. Numerous transcription factors participate in nephrogenesis, particularly PAX2 and WT1 (9). PAX2 is strongly expressed in the mesenchymal cell aggregates, in the nuclei of the renal vesicles, and in the distal and collecting tubules. By the capillary loop stage, PAX2 is no longer expressed by the visceral epithelial cells at the capillary loop stage or in the podocytes of mature glomeruli but persists in the nuclei of the PEC and in the nuclei of the distal tubules and collecting ducts. The disappearance of the PAX2 from podocytes corresponds to their definitive loss of ability to divide. Nuclear expression of WT1 begins in the undifferentiated blastema, increases progressively in the condensed blastema, the renal vesicles and the S-shaped body stage, then is restricted to the mature podocytes, where it persists throughout life. Nevertheless, some investigators have reported that PEC could express WT1 (9), whereas others have not found this to be the case (16). PAX2 is indispensable to the transformation of mesenchymal cells into renal vesicles (17). PAX2 induces cellular proliferation (9) and also is antiapoptotic (18,19). The downregulation of PAX2 by WT1 seems to be a prerequisite that permits the WT1-controlled differentiation of podocytes (20). WT1 is indispensable in early renal development (21). Its persistence throughout life suggests that it plays a role in the homeostasis of the mature podocyte (22). WT1 regulates the expression of podocalyxin (23) and of nephrin (24,25).
The purpose of this study was to show that a subset of PEC express the proteins of mature podocytes and the transcription factor WT1. The results suggest that Bowman’s capsule contains parietal podocytes analogous to visceral podocytes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
Immunohistochemistry Markers
Podocytes were characterized using an anti-synaptopodin mAb, clone G1D4 (Biogen and Biotechnik, Heidelberg, Germany); an anti–GLEPP-1 mAb, clone 5C11 (Biogenex, San Ramon, CA); an anti-human podocalyxin mAb (MLC 48A8; a gift from Dr. Pierre Ronco, INSERM U 489, Hôpital Tenon, Paris, France); an anti-human -actinin-4 polyclonal antibody (pAb; ImmunoGlobe, Himmelstadt, Germany); an anti-human podocin pAb (a gift from Dr. M.C. Gubler, INSERM U574, Hôpital Necker, Paris, France); an anti-nephrin N-20 pAb and an anti-WT1 C19 pAb (both from Santa Cruz Biotechnology, Santa Cruz, CA). Two proteins that were synthesized by podocytes were identified: novH (26) using an anti-human novH pAb (a gift from Dr. Cecile Martinerie, INSERM U515, Hôpital Saint-Antoine, Paris, France) and vascular endothelial growth factor (VEGF) (27) using an anti-human VEGF-C mAb, clone C1 (Santa Cruz Biotechnology). PAX2 was characterized by an anti-PAX2 pAb (Zymed, San Francisco, CA). The CKI p57 was identified using C-20 pAb (Santa Cruz Biotechnology). CK were labeled using C2562 mAb cocktail (Sigma Aldrich, Saint Quentin Fallavier, France) that was directed against nine CK types and used as an epithelial marker. Normal podocytes are not labeled by C2562 mAb. Vimentin, labeled by V9 mAb (Dako, Trappes, France) was used as a mesenchymal marker. Normal podocytes are positive with this mAb. Immunohistochemistry procedures were performed as described previously (Table 1) (15).
|
Labeling Counts
For each adult normal renal sample, two independent observers assessed all of the glomerular sections for all of the markers. In each kidney, a minimum of 100 consecutive glomerular sections, including at least 40 long-axis sections through the vascular pole, were examined.
For each glomerulus, the extent of labeling along Bowman’s capsule was graded using six grades: 0 (no labeling), 1 to 25, 26 to 50, 51 to 75, 76 to 99, and 100%. The percentage of labeled glomerular sections was calculated. Results were evaluated first for the subset of long-axis glomerular sections that were cut through the vascular pole and second for all of the long-axis glomerular sections regardless of cut. Discontinuous labeling and that at a distance from the vascular pole on vascular-pole glomerular sections and discontinuous labeling on nonoriented glomeruli were counted. The number of podocyte bridges between the tuft and Bowman’s capsule also was counted. For nuclear markers (WT1, PAX2, and CKI p57), the number of positive nuclei also was counted to calculate a mean density per glomerulus.
For fetal kidneys, the same scoring system was used to assess the expression of WT1 and of specific podocyte proteins synaptopodin, GLEPP-1, and podocalyxin along the Bowman’s capsules in all capillary loop stage and mature glomeruli. A mean number of 14.9 ± 2.6 glomeruli were counted per fetal kidney. The low number of fetal glomeruli assessed precludes statistical comparison with the adult glomeruli data.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
|
|
-actinin-4, nephrin, and podocin) and the proteins that were synthesized and secreted by podocytes (novH and VEGF) marked the visceral podocytes that covered the glomerular capillaries. They also marked numerous cells in the Bowman’s capsules, which were identified as parietal podocytes (Figure 1, Panels 1 through 8). The percentage and the extent of Bowman’s capsules that were covered by parietal podocytes are outlined in Table 2. The parietal podocytes predominated in the region of the vascular pole, where they were seen to be in continuity with the visceral podocytes.
|
In the juxtaglomerular apparatuses (JGA), the marked cells often bordered the afferent and efferent arterioles (Figure 1, Panels 9 through 12). Some marked cells also were found within or covering the extraglomerular mesangium (Figure 1, Panels 9 and 12).
Considering all of the glomerular sections, 15.5 ± 3.7% of glomeruli showed cytoplasmic extensions, sometimes containing a nucleus, forming a bridge across the urinary space (Figure 2, Panels 13 through 15). In 10.1 ± 2.6% of sections, these bridges were situated in proximity to the vascular pole and linked the visceral podocytes with a section of Bowman’s capsule lined with parietal podocytes, whereas in 5.4 ± 1.4% of sections, these bridges were at some distance from the vascular pole.
|
PAX2 Labeling
No visceral podocyte expressed PAX2, but the nuclei of almost all or all of the cells in Bowman’s capsule expressed it (Figure 2, Panel 22). For vascular-pole sections, the extent of Bowman’s capsule that contained marked nuclei was 100% in 97.6% of Bowman’s capsules and 76 to 99% in the remaining 2.3% of capsules. The mean density of PAX2-positive nuclei on Bowman’s capsule was 16.0 ± 0.6 per glomerular section.
On serial sections, parietal podocytes exhibited both WT1 and PAX2 (Figure 2, Panels 23 and 24). On doubly marked sections, some cells of Bowman’s capsule coexpressed PAX2 and podocalyxin (Figure 3, Panel 25) or WT1.
|
Cytokeratin Labeling
No visceral podocytes were marked. The parietal cells were marked in a heterogeneous and inconstant manner. On vascular-pole sections, 28.0% of Bowman’s capsules were negative. The extent of Bowman’s capsule marked was 1 to 25% in 31.8% of Bowman’s capsules, 26 to 50% in 16.6%, 51 to 75% in 13.7%, 76 to 99% in 7.4%, and 100% in 2.6% of capsules. Labeling generally was away from the vascular pole (Figure 3, Panel 27), predominating at the tubular pole in 77.7% of the marked capsules.
On doubly marked sections, no parietal cell coexpressed CK and podocalyxin (Figure 3, Panel 28). Certain parietal cells expressed neither CK nor podocalyxin (Figure 3, Panel 28).
Vimentin Labeling
All of the visceral podocytes and numerous parietal cells expressed vimentin (Figure 3, Panel 29). On sections that passed through the vascular pole, 92.7% of capsules were marked. The extent of Bowman’s capsule labeling was 1 to 25% in 35.4% of Bowman’s capsules, 26 to 50% in 20.8%, 51 to 75% in 13.9%, 76 to 99% in 9.4%, and 100% in 13.3%.
Bowman’s Capsules without Tuft or with Retracted Tuft
Bowman’s capsules without recognizable tuft or with a markedly shrunken tuft were seen in most kidneys, often in the superficial cortex. These capsules often were frankly cystic, and their diameter exceeded 200 µm. None of these glomerular cysts had an identifiable tubular pole. Parietal podocytes that expressed the podocyte-specific proteins, WT1, and vimentin covered >75% and often 100% of these Bowman’s capsules (Figure 3, Panel 30). These parietal podocytes often expressed p57 but expressed neither CK nor PAX2.
Fetal Kidneys
From the S-shaped body and capillary loop stages onward, the visceral podocytes and parietal podocytes expressed the podocyte proteins tested, podocalyxin, GLEPP-1, synaptopodin, podocin, novH, and WT1 (Figure 3, Panels 31 through 34) and p57. CK was disappearing from the visceral podocytes during the capillary loop stage but was expressed in a heterogeneous manner by the parietal cells at this stage (Figure 3, Panel 35). PAX2 was expressed by the parietal cells from the S-shaped body stage onward but had disappeared from the visceral epithelial cells at the point where the S-shaped bodies transformed into the capillary loop stage (Figure 3, Panel 36).
Table 4 gives the mean percentages ± SEM of fetus Bowman’s capsules labeled and the extent of labeling. When one considers three-podocyte protein labeling (synaptopodin, GLEPP-1, and podocalyxin) as a whole in capillary loop and mature glomerular sections, 74.1% on average of Bowman’s capsules were marked. The extent of the Bowman’s capsules lined by parietal podocytes was on average 1 to 25% in 50.7% of Bowman’s capsules, 26 to 50% in 17.8%, 51 to 75% in 4.9%, and 76 to 99% in 0.6%. For WT1 in capillary loop and mature glomeruli, the extent of the Bowman’s capsules that contained marked nuclei was 26 to 50% in 1.9% of Bowman’s capsules, 51 to 75% in 19.7%, 76 to 99% in 42.6%, and 100% in 35.8%. The mean density of nuclei that were positive for WT1 was 18.1 ± 2.5 per glomerular section.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
-actinin-4), and (3) associated with the slit diaphragm (podocin and nephrin). Furthermore, they expressed proteins that were synthesized by podocytes (novH and VEGF) and the transcription factor WT1. Importantly, these cells did not coexpress CK. Parietal podocytes were frequent and predominated around the vascular pole, where they were in continuity with the visceral podocytes. Cells that expressed podocyte proteins were found in 76.6% of the sections of Bowman’s capsule that passed through the vascular pole. The extent of capsule that was lined by such cells was variable from one glomerulus to another: 1 to 25% in 55.5% of Bowman’s capsules but >50% in 6.8% of capsules. Islands of parietal podocytes were found apparently isolated from the vascular pole on the random sections used in our study. This finding does not exclude continuity of such islands on Bowman’s capsule with podocytes at the vascular pole, out of the plane of section. Indeed, SEM studies (6) have shown irregular-shaped areas of Bowman’s capsule lined by parietal podocytes extending far out from the vascular pole. Cytoplasmic extensions, sometimes containing a podocyte nucleus, formed bridges between the visceral podocytes and Bowman’s capsule in approximately 15.5% of glomerular sections, most often in proximity to the vascular pole. Some glomeruli, particularly those that were frankly cystic, in which the tuft was retracted or invisible and for which no tubular pole could be identified, had their Bowman’s capsules completely lined by parietal podocytes.
To our knowledge, our study is the first one published to describe parietal podocytes that were identified by podocyte markers in the parenchyma of normal kidneys. Several studies using SEM or TEM had shown (6) that in normal human kidneys cells that had all of the ultrastructural characteristics of podocytes lined a portion of the PBM of Bowman’s capsule, generally around the vascular pole. Their interdigitated pedicels, inserted on the PBM, were linked by one or two slit diaphragms. These cells and their pedicels covered the juxtaglomerular arterioles and the extraglomerular mesangium. In some cases, podocytes formed bridges across the urinary space and inserted by their pedicels on the GBM on the one hand and on the PBM on the other. SEM revealed that glomeruli with retracted tufts and cystic glomeruli were atubular glomeruli, whose Bowman’s capsule was lined by parietal podocytes (28). Our immunohistochemical results show that the parietal podocytes that had been recognized ultrastructurally express the same epitopes as visceral podocytes.
Parietal podocytes were originally described or illustrated by TEM in the normal rat kidney several decades ago (cited in reference [29]). Close relations between the JGA and parietal podocytes were described in the rat and other laboratory animals (30,31). There parietal podocytes line the extraglomerular mesangium and a segment of the afferent arteriole, including myo-epithelioid cells that contain renin secretory granules. Protrusions of the urinary space lined by parietal podocytes sometimes herniated into the contact zone between the afferent arteriole and the extraglomerular mesangium. These ultrastructural studies are in agreement with our immunohistochemical results that showed that labeled cells were adjacent to the juxtaglomerular arterioles and were found within the extraglomerular mesangium.
Distinctive Features of the Parietal Podocytes
Some parietal podocytes had characteristics that were different from those of the visceral podocytes. Some parietal podocytes did not express CKI p57, in contrast to visceral podocytes, which expressed it universally. In effect, on vascular-pole sections, 26.9% of Bowman’s capsules had occasional nuclei expressing p57, whereas 74.7% of Bowman’s capsules had nuclei expressing WT1. Moreover, the density of p57-positive nuclei was much lower than that for nuclei that expressed WT1: 0.47 versus 2.3 nuclei per glomerulus, respectively.
Certain parietal podocytes coexpressed WT1 and PAX2. In effect, in all of the Bowman’s capsules, virtually all nuclei expressed PAX2. This coexpression was confirmed by study of serial sections and slides that were doubly marked for WT1 and PAX2.
The expression of PAX2 and the nonexpression of CKI p57 in certain parietal podocytes might signify that these cells have retained the ability to divide. In fact, in fetal visceral podocytes, PAX2, theoretically proproliferative (9) and antiapoptotic (18,19), disappears with the appearance of CKI p57, which prevents cellular division by inhibiting the cyclin-cyclin dependent kinase (13). The disappearance of PAX2 and the appearance of CKI p57 in fetal visceral podocytes signal the loss of their ability to divide.
Origin of Parietal Podocytes
Nephrectomies for localized tumors of the kidney generate sufficient renal parenchyma for quantitative studies. Nonetheless, the part of the kidney that is considered histologically normal is not necessarily free of changes as a result of age, tumor, associated diseases, or simply the conditions surrounding its removal and preservation. However, the inability to correlate the parietal podocytes that were observed on TEM to any disease condition (29,32–34) had led to the speculative conclusion that parietal podocytes must exist in normal kidneys (6,33). Age does not seem to be the sole explanatory factor because we have seen parietal podocytes and podocyte bridges in renal biopsies from younger patients with nephrotic syndrome and minimal-change disease (personal observations).
Whether these parietal podocytes and podocyte bridges that join the tuft and Bowman’s capsule are labile or permanent and irreversible has not yet been documented. Several hypotheses have been advanced to explain the origin of the parietal podocytes: A process of migration of visceral podocytes to Bowman’s capsule forming bridges across the urinary space (6,35); metaplasia of the parietal epithelium, as has been demonstrated experimentally (36); and evagination of visceral podocytes into Bowman’s capsule at the vascular pole as it has been suggested in atubular glomeruli (28). The existence of cells that have podocyte phenotypes in maturing fetal kidneys strongly suggests that parietal podocytes are constitutive elements in the kidney of humans. In glomeruli at capillary loop and mature stages in 7- to 9.5-wk-old fetuses, our results suggest that parietal podocytes were present and covered percentages of Bowman’s capsule sections similar to that of adult kidneys. Furthermore, WT1 is actually expressed in Bowman’s capsule cells (9), and its expression markedly decreased from fetal to adult Bowman’s capsules.
Roles of Parietal Podocytes
The parietal podocytes have the same ultrastructural and molecular architecture as the visceral podocytes that are differentiated to filter. This suggests that the parietal podocytes could form a zone of permeability in Bowman’s capsule by which the glomerular filtrate could leave the glomerulus. This diffusion could take place from the urinary space into the periglomerular interstitium (or vice versa) but also from the urinary space into the JGA and thus could play a role in tubuloglomerular feedback (6,30,31).
The parietal podocytes and the podocyte bridges between tuft and capsule could play a role in some glomerular lesions. These bridges have been considered as possible precursors of the tuft-capsular adhesions that are seen in numerous glomerulopathies, particularly around the vascular pole (35). In models of anti-GBM glomerulonephritis in the mouse, Le Hir et al. (37) showed that during the initial phases of the glomerulonephritis, before the formation of crescents, podocytes formed bridges between the GBM and PBM. Moeller et al. (38) showed that cells that they identified as podocytes adhered to the PBM and divided on the internal aspect of the PBM to contribute to crescent formation. In human crescentic glomerulonephritis, the contribution of podocytes to crescent formation and the parietal origin of these podocytes were discussed (39). In membranous glomerulopathy in humans, a TEM study (40) showed that parietal podocytes behaved like visceral podocytes, and subepithelial deposits that were identical to those in the glomerular capillaries were observed between the parietal podocytes and the PBM.
|
Conclusion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E. Letavernier, P. Bruneval, S. Vandermeersch, J. Perez, C. Mandet, M.-F. Belair, J.-p. Haymann, C. Legendre, and L. Baud Sirolimus interacts with pathways essential for podocyte integrity Nephrol. Dial. Transplant., October 16, 2008; (2008) gfn574v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Achenbach, M. Mengel, I. Tossidou, I. Peters, J.-K. Park, M. Haubitz, J. H. Ehrich, H. Haller, and M. Schiffer Parietal epithelia cells in the urine as a marker of disease activity in glomerular diseases Nephrol. Dial. Transplant., October 1, 2008; 23(10): 3138 - 3145. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Albaqumi and L. Barisoni Current Views on Collapsing Glomerulopathy J. Am. Soc. Nephrol., July 1, 2008; 19(7): 1276 - 1281. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Ueda, Y. Miyazaki, T. Matsusaka, Y. Utsunomiya, T. Kawamura, T. Hosoya, and I. Ichikawa Bmp in Podocytes Is Essential for Normal Glomerular Capillary Formation J. Am. Soc. Nephrol., April 1, 2008; 19(4): 685 - 694. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Pouly, P. Bruneval, C. Mandet, S. Proksch, S. Peyrard, C. Amrein, V. Bousseaux, R. Guillemain, A. Deloche, J.-N. Fabiani, et al. Cardiac stem cells in the real world J. Thorac. Cardiovasc. Surg., March 1, 2008; 135(3): 673 - 678. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. S. Thorner, M. Ho, V. Eremina, Y. Sado, and S. Quaggin Podocytes Contribute to the Formation of Glomerular Crescents J. Am. Soc. Nephrol., March 1, 2008; 19(3): 495 - 502. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Barisoni, H. W. Schnaper, and J. B. Kopp A Proposed Taxonomy for the Podocytopathies: A Reassessment of the Primary Nephrotic Diseases Clin. J. Am. Soc. Nephrol., May 1, 2007; 2(3): 529 - 542. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |
