 |
||||
| 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 | |
*
INSERM U489, Tenon Hospital and St. Antoine Medical Faculty, Paris 6
University, Paris, France
INSERM U538, Tenon Hospital and St. Antoine Medical Faculty, Paris 6
University, Paris, France
Service Commun d'Imagerie Cellulaire (IFR65), Tenon Hospital and St.
Antoine Medical Faculty, Paris 6 University, Paris, France
University of East Anglia, Norwich, United Kingdom.
Address correspondence to Dr. Brigitte Lelongt, INSERM U489, Hôpital Tenon, 75020 Paris, France. Phone: 33-156-01-6719; Fax: 33-156-01-6217; E-mail: brigitte.lelongt{at}tnn.ap-hop-paris.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The developing kidney is the product of reciprocal inductive interactions between the ureter bud and its surrounding mesenchyme, the metanephric blastema (21). The metanephric mesenchyme is induced at the tips of the branching ureter bud and undergoes differentiation processes that lead a population of mesenchymal cells to change their phenotype into epithelial cells (22). These changes are associated with the expression of ECM components that play a crucial role in the establishment of cell polarity and epithelial phenotype (nephrogenesis) and in the branching morphogenesis of the ureter bud (2). Type IV collagenases MMP2 and MMP9, MT1-MMP (which activates MMP2), and TIMP have been identified in early stages of renal development (12,18,23), and several studies have shown their role in organotypic and cell culture models (17,19,24). Nevertheless, the renal ontogeny of these proteins is only partially known because most studies were performed from embryonic day 13 in the mouse (18,19) or in the rat (23,24). In addition, their persistence in the adult remains controversial: expression of MMP-2, MMP-9, and MT1-MMP proteins and/or mRNA was reported to cease with nephron maturation in the rat (23), whereas in the rabbit, both MMP2 and MMP9 continue to be expressed in the adult collecting duct (25).
The aim of this study was to describe in detail the expression pattern of type IV collagenases and MT1-MMP and their inhibitors (TIMP1, TIMP2, TIMP3) in mouse kidney from early metanephric stages to maturity. We paid special attention to the expression of these components in terminally differentiated tubule segments, where they had not been previously described. The results show that expression of the type IV collagenase system is finely regulated both chronologically and spatially. An important finding is the persistent expression of type IV collagenases and MT1-MMP at the apical pole in specific segments of the adult renal tubule, where these matrix metalloproteinases most likely cleave non-ECM substrates.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Antibodies and Specific Cell Markers
Characteristics of antibodies directed against MMP2, MMP9, MT1-MMP, and
TIMP are listed in Table 1.
Dolichos biflorus agglutinin (DBA) lectin conjugated to horseradish
peroxidase (HRP) or TRITC (10 µg/ml; Sigma Chemical Co., Saint-Quentin
Fallavier, France) was used to identify the ureteric bud and its derivatives
(26,27);
Helix pomatia agglutinin lectin conjugated to TRITC (10 µg/ml) was
co-incubated with DBA to reinforce the ureteric bud staining in whole-mount
kidneys analyzed by confocal microscopy
(17); goat anti—neural
cell adhesion molecule (NCAM) antibody (5 µg/ml; Santa Cruz Biotechnology,
Santa Cruz, CA) was used to detect induced metanephric mesenchyme cells
(28). Thick ascending limbs of
Henle's loop and proximal tubules were identified with rabbit anti-Tamm
Horsfall protein antibody (20 µg/ml)
(29) and sheep anti-megalin
antibody (1:1000) (30),
respectively.
|
|
Proteins were transferred to polyvinylidene fluoride membranes (Schleicher and Schuell, Darmstadt, Germany). Endogenous alkaline phosphatase activity and nonspecific staining due to primary antibodies were blocked by incubation for 1 h at 37°C in Tris-HCl buffer (pH 8.0) containing 10% fat dry milk, 0.05% Tween 20, and 10 mM Levamisole. Polyvinylidene fluoride membranes were incubated overnight at 4°C with primary antibodies in Tris-HCl buffer (pH 8.0), 0.05% Tween 20. Secondary alkaline phosphatase-conjugated anti-rabbit, anti-mouse, or anti-sheep IgG antibodies (Amersham, Les Ullis, France) diluted in the same buffer were used for immunodetection of MMP and TIMP. Alkaline phosphatase was revealed by nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate substrate (Sigma Chemical Co.) in 100 mM Tris-HCl, 100 mM NaCl, and 5 mM MgCl2 (pH 9.5). In control experiments, primary antibody was replaced by purified rabbit, mouse, or sheep nonimmune IgG.
Immunofluorescence Confocal Microscopy
Kidneys were dissected from 11.5- or 12.5-d embryos, immediately fixed in
4% paraformaldehyde in phosphate-buffered saline (PBS; pH 7.4) for 30 min at
room temperature, and incubated further for 15 min in PBS, 50 mM
NH4Cl (pH 7.4)
(17). They were saturated with
5% fat dry milk in PBS supplemented with 0.075% saponin. Kidneys were
incubated overnight at 4°C in primary antibodies or purified nonimmune IgG
from the same species as control, diluted in saponin-PBS supplemented with
2.5% fat dry milk. After several washes, they were incubated for 2 h at room
temperature in saponin-PBS containing donkey FITC-conjugated anti-rabbit IgG
or anti-sheep IgG antibody (dilution 1:100; Chemicon International, Temecula,
CA), and TRITC-conjugated Helix pomatia agglutinin and DBA lectins.
In some experiments, lectin staining was omitted and induced metanephric cells
were stained overnight at 4°C with anti-NCAM goat antibody followed by
donkey TRITC-conjugated anti-goat Ig antibody (dilution 1:50, 2 h at room
temperature; Amersham). Kidneys were washed, then mounted in DAKO fluorescence
mounting medium (Dako Corporation, Carpinteria, CA) and viewed under a Leica
TCS laser scanning confocal microscope (Lasertechnik, GmbH, Wetzlar, Germany).
Optical sections were generated and processed as described previously
(17).
Immunohistochemistry
Expression of MMP2, MMP9, MT1-MMP, and TIMP was analyzed during
nephrogenesis and in adult mice by standard light microscopy using a two-step
immunoperoxidase labeling procedure. Newborn and adult kidneys were perfused
with PBS for 1 min before dissection. Kidneys dissected from 16.5-d embryos,
3-d newborn mice, and adult mice were washed in PBS (pH 7.2). Whole kidneys
(E16.5) or half kidneys (newborn and adult) were immediately fixed in 4%
paraformaldehyde in PBS (pH 7.2) at 4°C for 3 h (day 16.5), 6 h (newborn),
or 12 h (adult). They were incubated in 30% sucrose PBS (pH 7.4) for 12 to 18
h at 4°C, rinsed twice in PBS, and snap-frozen in liquid nitrogen for
storage at -80°C until sectioning. Samples from adult kidneys were also
dehydrated in a graded series of ethanol followed by xylene and embedded in
paraffin.
Five-µm frozen sections were applied to pretreated glass slides (Superfrost Plus; Menzel Glaser, Braunsthweig, Germany). Five-µm sections from paraffin-embedded kidneys were deparaffinized in xylene, rehydrated through graded ethanol washes, and rinsed in deionized water, then in PBS (pH 7.4). All sections were incubated in 50 mM NH4Cl PBS (pH 7.4) for 15 min at room temperature before staining. They were first incubated for 20 min with Immunopure peroxidase suppressor (Pierce, Rockford, IL) to quench endogenous peroxidase activity. They were then saturated with 1% fat dry milk and 10 µg/ml donkey Ig diluted in PBS (pH 7.4) for 1 h at room temperature. They were incubated overnight at 4°C with primary antibodies (Table 1) or control nonimmune IgG, diluted in DAKO antibody diluent to reduce background. After several washes in PBS, secondary anti-rabbit, anti-sheep, or anti-mouse antibodies conjugated to HRP (Chemicon International) were reacted with sections for 2 h at room temperature. Slides were washed in PBS, then red staining was developed for 10 min at room temperature with AEC substrate (9-amino-3-ethylcarbazole solubilized in N,N-dimethylformamide and diluted in 0.1 M sodium acetate buffer, pH 5.0, supplemented with 0.015% H2O2 before use). Sections were washed in deionized water for 5 min before counterstaining and mounting (see below).
In most experiments, a second staining step was carried out to identify on the same section, proximal tubule segments with sheep anti-megalin antibody, ascending limbs of Henle's loop with rabbit anti-Tamm Horsfall protein antibody, or collecting ducts with HRP-conjugated DBA lectin. We used DBA lectin because binding sites are expressed throughout ontogeny in the ureteric bud and its derivatives (27) and on intercalated cells in the adult mouse collecting duct (31). Sections were first incubated in Immunopure peroxidase suppressor for 20 min. They were then reacted for 2 h at room temperature with primary antibodies or DBA lectin diluted in DAKO antibody diluent. HRP-conjugated donkey anti-rabbit or anti-sheep antibodies were used to detect primary antibodies bound to Tamm Horsfall protein and megalin, respectively. Black staining was developed for 1 to 2 min at room temperature with DAB-NiCl2 substrate (diaminobenzidine 10 mg, NiCl2 30 mg diluted in 40 ml PBS [pH 7.4], supplemented with 0.015% H2O2 before use). Sections were washed in PBS between each incubation step and finally counterstained in blue with Harris or Meyer hematoxylin for 45 s at room temperature. After additional washes in deionized water, slides were mounted in Tris-glycerol buffer (pH 7.4) and viewed under a Zeiss Axiophot microscope (Carl Zeiss S.A., Le Pecq, France).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Localization of each antigen in embryonic kidney was then analyzed by laser confocal microscopy at days 11.5 and 12.5 p.c. (Figures 2,3,4, Table 2). All proteins were expressed in metanephric mesenchyme at all dates, and only MT1-MMP and to a lesser extent MMP-2 were detected in the ureter bud.
|
|
|
|
At day 12.5, MMP9 (Figure 2, A through C), MMP2 (Figure 2, D through F), and MT1-MMP (Figure 2, G through I) were co-localized with NCAM in the induced mesenchyme (Figure 2, C, F, and I). In addition, MMP2 (Figure 2, E and F) and particularly MT1-MMP (Figure 2, H and I) were found in the ureter bud.
Figure 3 shows expression of TIMP1 (Figure 3, A and B), TIMP2 (Figure 3, D and E), and TIMP3 (Figure 3, G and H) in mesenchyme at day 11.5. No staining was observed in the ureter bud (Figure 3, B, E, and H). At day 12 in the mouse T-shaped branching of the ureter bud is associated with epithelial induction of the mesenchyme. At day 12.5, TIMP1, TIMP2, and TIMP3 were mainly detected in the recently induced mesenchyme, where they were co-localized with NCAM (yellow fluorescence, Figure 3, C, F, and I). Because prominent staining of the ureter bud, particularly at branches and clefts, had been reported previously at day 13 p.c. in the rat kidney (24), we paid special attention to TIMP2 distribution in the mouse at day 12.5 p.c. (Figure 4). Although mesenchymal cells, including those in contact with the ureter bud, were brightly stained, no labeling of the ureter bud and its basement membrane was detected, even at branches and clefts (Figure 4, A and B). As shown in Figure 4C, TIMP2 was expressed in cytoplasmic vesicles and at the cell membrane of mesenchymal cells. A dotted yellow fluorescence was visible at the tip of ureter bud branches and consistent with binding of TIMP2 to the cell membrane of ureter bud cells (Figure 4C).
Expression Pattern of Type IV Collagenases and Their Inhibitors
during Renal Tubule Segmentation
Immunohistochemical studies were performed at day 16.5 of gestation, at
birth, and in adult mice (Figures
5,6,7,
Table 2). Expression of the
components of the type IV collagenase system was detected in immature tubule
segments as early as day 16.5 p.c. and persisted in adulthood, although
changes in the topography of expression occurred along the renal tubule and at
the subcellular level. Because the level of expression of MMP and their
inhibitors is low and because access of antibody to epitopes may depend on
ontogenic stage and tissue processing, we used different immunohistochemical
techniques on frozen and paraffin sections. Sensitivity of paraffin
immunohistochemistry was usually greater for adult samples, except for TIMP1
and TIMP3 detection, which required frozen sections.
|
|
|
Figure 5 compares the expression of MMP9 (left column) and TIMP1 (right column) at day 16.5 p.c. and in the adult. MMP9 was expressed at all ages in proximal tubule cells (Figure 5, A and G). At day 16.5 p.c., proximal tubule staining was cytoplasmic with a diffuse, microvesicular pattern (Figure 5, A and C), whereas in the adult, it was mostly apical (Figure 5G). Staining was also detected in collecting duct sections (Figure 5, E and H) and predominated in the adult at the apical pole of principal cells (Figure 5H). MMP9 was also transiently expressed in the glomerular capillary loops and in extraglomerular vascular structures in newborn mice (data not shown).
TIMP1 expression pattern (Figure 5, B, D, F, I, and J) was different from that of MMP9. TIMP1 was weakly expressed in the proximal tubule at day 16.5 p.c. (Figure 5B) and disappeared later (Figure 5I). In contrast, a clear staining was noted in the ascending limb of Henle's loop from embryonic day 16.5 (Figure 5D) to adulthood (data not shown). TIMP1 was also found in distal tubule sections (Figure 5, D and I) and in collecting ducts (Figure 5J). It appeared at the basal aspect of collecting duct cells in newborn mice (data not shown), and its expression increased in the adult, where it was expressed predominantly at the basal aspect of intercalated cells (Figure 5J). TIMP1 could not be detected in glomerulus, in vessels, and in ECM.
Figure 6 shows the expression of components of the MMP2/MT1-MMP/TIMP2 system at day 16.5 p.c. (Figure 6, A through F) and in the adult (Figure 6, G through L). At day 16.5 p.c., MMP2 was detected mainly in the cytoplasm of proximal tubule cells (Figure 6A) and in collecting duct sections (Figure 6D). In newborn and adult mice, an intense apical staining was observed in cortical tubules that showed morphologic and immunomorphologic characteristics of the distal tubule (Figure 6G) or the connecting tubule (Figure 6H, double arrow). A clear staining was observed between proximal tubules (Figure 6G), distal tubules (Figure 6G), and collecting ducts (Figure 6H, triple arrow). MMP2 could also be detected in glomerulus in newborn and adult mice. Staining predominated in capillary loops but was also noted in some mesangial cells (not shown).
MT1-MMP had an expression pattern different from that of MMP2 (Figure 6, B, E, I, and J). At day 16.5 p.c., it was expressed in the proximal tubule (Figure 6B) but predominated at the apex of proximal cells (Figure 6E). In the adult, it persisted in the brush border (Figure 6I) and at the apical pole of collecting duct cells (Figure 6J). MT1-MMP was not expressed in glomerulus at any time.
TIMP2 was identified in cytoplasmic vesicles of proximal tubule cells in 16.5-d p.c. (Figure 6, C and F) and newborn (not shown) kidneys. In adult mice, TIMP2 was mainly expressed at the apical pole of a majority of cells in connecting tubules and collecting ducts (Figure 6, K and L). It was also weakly detected in the brush border (Figure 6K).
TIMP3 was expressed in most tubule segments of embryonic kidney. At day 16.5 p.c., staining was observed in the proximal tubule (Figure 7A), in ascending limb cells (Figure 7C), and in collecting ducts (Figure 7E). In adult mice, staining persisted at the basal pole of cells in all segments but predominated in thick ascending limb (Figure 7D), distal tubule (Figure 7B), and collecting duct (Figure 7F).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A role for metalloproteinases in early stages of renal development is likely for at least two reasons: first, phenotypic conversion of mesenchymal cells is associated with changes in the composition of the ECM; second, a constant remodeling of the ECM is needed at the growing tips of the invading ureter bud to allow further branchings in the metanephric mesenchyme. We reported previously that MMP2 and MMP9 were expressed at day 11 p.c. in the noninduced mesenchyme but not in the ureter bud and that MMP9 was required for renal organogenesis in vitro (17). We now show that all proteins of the type IV collagenase system are detected in the noninduced mesenchyme at day 11.5, although MT1-MMP is localized mainly in the ureter bud. One day later, proteins are preferentially expressed by the induced mesenchymal cells, but MT1-MMP and to a lesser extent MMP2 are also localized in the ureter bud. These results are in keeping with our previous zymography studies on MMP2 and MMP9 secretion at day 11 (17) and day 12 (B. Lelongt, personal communication) in isolated ureter buds and mesenchymes. They differ from those of Ota et al. (18) at day 13, suggesting that MT1-MMP was produced by the ureteric bud only, but are consistent with their finding that MMP2 was synthesized by the mesenchyme and expressed in the ureteric bud as a ligand of MT1-MMP. Alternatively, MMP2 could be produced by ureteric bud cells from day 12 as suggested by our zymographic data. In our study, TIMP2 was almost exclusively detected in the mesenchyme at days 11.5 and 12.5 p.c., whereas the most intense staining described by Barasch et al. (24) was in the ureteric bud, particularly at branches and clefts. Differences could be because Barasch et al. used rat kidneys dissected at embryonic day 13 and cultured for 48 h before staining, whereas we directly stained mouse kidneys collected in vivo at days 11.5 and 12.5. Although the antibody that we used should be able to detect MMP2-TIMP2 complexes according to manufacturer warranties, its specificity against the C-terminal portion of TIMP2, which is implicated in the MMP2-TIMP2 interaction, could also explain a partial failure to recognize MMP2-TIMP2 complexes in the ureteric bud. Figure 4C indeed suggests a weak expression of TIMP2 in the ureteric bud tips. In both studies, TIMP2 was detected in the induced mesenchyme, a finding in keeping with its antiapoptotic role and its stimulating effect on mesenchymal growth (24).
The expression pattern of type IV collagenases and their inhibitors in vivo suggests that a regulated balance of their activities is required to facilitate migration of the ureteric bud branches into the loose embryonic mesenchyme and to control early ECM changes during phenotypic conversion of the induced mesenchyme. In addition, the coexpression in the mesenchyme of the two gelatinases, MT1-MMP, and TIMP, and their partial functional redundancy may explain why MMP2-and MMP9-deficient mice do not have a major renal phenotype (32,33,34,35).
Most renal cell types derive from induced mesenchymal cells that further undergo various differentiation processes. It is not surprising, therefore, that MMP and their related proteins continue to be expressed during nephrogenesis. By semiquantitative Western blot analysis, we showed that their expression peaked at day 16.5 p.c. and persisted in adulthood but at lower levels, although MMP2 was less modulated. These results are in keeping with our immunohistochemical data showing that proteins of the type IV collagenase system are expressed in various nephron segments from most immature stages to most differentiated structures. Tanney et al. (23) showed a clear-cut expression of MMP2, MT1-MMP, and TIMP2 in differentiating nephron structures but failed to detect their persistence in fully differentiated nephrons. We also found MMP2, MT1-MMP, TIMP2, and TIMP3 in the most immature structures, including primary condensates, vesicles, and comma-shaped and S-shaped bodies, at the basolateral pole of the cells (data not shown). MMP9 but not TIMP1 was also expressed at these stages. Expression of these proteins in more mature structures, particularly in the different tubule segments, had not been reported until the present study.
The expression pattern of MMP9 and TIMP1 in differentiating and mature tubule segments was divergent. At all stages, MMP9 was localized predominantly in proximal tubules and collecting ducts, but its subcellular localization changed from cytoplasmic at day 16.5 p.c. to apical in mature structures. In contrast, TIMP1 was not detected or was weakly detected in proximal tubule sections but was expressed in thick ascending limbs of Henle's loop and in distal tubules. These results suggest that under physiologic conditions, MMP9 may not be inhibited by TIMP1 at the apical pole of adult renal tubule segments.
The expression pattern of MMP2, MT1-MMP, and TIMP2 showed similarities during tubule maturation. At day 16.5, all three partners were strongly expressed in proximal tubule cells. In the adult kidney, the distribution of TIMP2 and MT1-MMP was almost superimposable. Contrary to MT1-MMP and TIMP2, MMP2 was not detected at the apex of proximal tubules and collecting ducts, but it is conceivable that MMP2 released into the urine by distal tubule cells can bind to its ligand MT1-MMP at the urinary pole of principal cells.
The changing, basolateral to apical pattern of expression of gelatinases during tubule segmentation suggests that they could be implicated in the regulation of basement membrane remodeling at early stages of nephrogenesis (23), whereas at later stages, and particularly in adult mice, they could cleave substrates in the tubule lumen or in the apical cell domain. This hypothesis is supported by the recent demonstration that arginine vasopressin and epidermal growth factor are potent regulators of MMP9 expression in collecting duct principal cells (25).
This study reports the first description of TIMP3 ontogeny. Its expression in earliest stages (E11.5, E12.5) is superimposable to that of other TIMP. Later, TIMP3 is remarkable for its diffuse expression in most tubule segments, with a prominent location at the basal aspect of the cell. The role of TIMP3 remains to be established. As in other differentiating tissues, it could be involved in the maintenance of differentiated state or in the control of proliferation and apoptosis (5,36,37,38,39).
In conclusion, we carried out a detailed ontogenic study, at the protein level, of the type IV collagenase system in the mouse kidney. Components of this system undergo both spatial and temporal regulation. They can be considered as differentiation markers of embryonic cell populations and, later, of differentiating tubule segments. A growing body of evidence indicates that they play a critical role in earlier stages of kidney branching morphogenesis, but they may also be involved in the renal tubule differentiation process and in specific tubular functions in mature nephrons. Additional studies are needed to identify these functions.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  N. E. Hellman, J. Spector, J. Robinson, X. Zuo, S. Saunier, C. Antignac, J. W. Tobias, and J. H. Lipschutz Matrix Metalloproteinase 13 (MMP13) and Tissue Inhibitor of Matrix Metalloproteinase 1 (TIMP1), Regulated by the MAPK Pathway, Are Both Necessary for Madin-Darby Canine Kidney Tubulogenesis J. Biol. Chem., February 15, 2008; 283(7): 4272 - 4282. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Rubera, C. Poujeol, G. Bertin, L. Hasseine, L. Counillon, P. Poujeol, and M. Tauc Specific Cre/Lox Recombination in the Mouse Proximal Tubule J. Am. Soc. Nephrol., August 1, 2004; 15(8): 2050 - 2056. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. L. Steer and S. K. Nigam Developmental approaches to kidney tissue engineering Am J Physiol Renal Physiol, January 1, 2004; 286(1): F1 - F7. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Schumacher, R. Strehl, U. de Vries, H. J. Groene, and W. W. Minuth SBA-Positive Fibers between the CD Ampulla, Mesenchyme, and Renal Capsule J. Am. Soc. Nephrol., October 1, 2002; 13(10): 2446 - 2453. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |
