Alterations in cell-adhesive and migratory properties of proximal tubule and collecting duct cells from bcl-2 -/- mice

【摘要】  Bcl-2 protects cells from apoptosis initiated by a variety of stimuli including loss of cell adhesion. Bcl-2 -/- mice develop renal hypoplastic/cystic dysplasia with renal cyst formation coinciding with renal maturation in normal mice. To gain a better understanding of the role cell-adhesive mechanisms play during renal maturation, we generated proximal tubule and collecting duct cell lines from postnatal day 10 ( P10 ) and P20 bcl-2 +/+ and bcl-2 -/- mice. Very little is known about the role cell-adhesive and migratory mechanisms play during renal maturation. We observed that modulation of cell-adhesive properties, which normally occur in a nephron segment-specific manner during renal maturation, and cell migration were altered in cells from bcl-2 -/- mice. Enhanced migration of bcl-2 -/- proximal tubule cells in a scratch wound assay was completely inhibited by incubation with PP1 (Src inhibitor) and moderately affected by incubation with SB-203580 (p38 inhibitor). These cells expressed increased levels of fibronectin and had numerous central focal adhesions. P20 bcl-2 -/- proximal tubule cells adhered to fibronectin but adhered poorly to collagen, vitronectin, or laminin. Collecting duct cells, similar to proximal tubule cells from bcl-2 -/- mice, demonstrated enhanced migration in a scratch wound assay that was inhibited by incubation with PP1. Migration of these cells was moderately affected by incubation with PD-98059 (MEK inhibitor) or LY-294002 (PI3 kinase inhibitor), whereas incubation with SB-203580 had no effect. P10 bcl-2 -/- collecting duct cells also expressed increased levels of fibronectin but decreased levels of thrombospondin-1 and demonstrated precocious binding to fibronectin and vitronectin compared with bcl-2 +/+ cells. The ability of P20 bcl-2 +/+ collecting duct cells to adhere to fibronectin and vitronectin corresponded with a decline in thrombospondin-1 expression. Therefore, alterations in cell-adhesive and migratory characteristics may be an early indicator of aberrant renal epithelial cell differentiation.

【关键词】  thrombospondin fibronectin focal adhesion renal maturation

CELL MIGRATION, PROLIFERATION, differentiation, and apoptosis play integral roles during kidney development. The formation of nephron segments requires that a complex series of interactions occur to allow proper migration and positioning of specific cell types. After induction of the metanephros and condensation of the metanephric blastema, tubulogenesis proceeds through a histologically defined sequence of stages during which developing nephrons assume comma and S-shapes ( 2, 12 ). The rodent is born with immature kidneys in which nephrons continue to develop from the nephrogenic zone located at the periphery of the kidney. This period of time is termed renal maturation and is complete by postnatal day 16 ( P16 ) ( 7 ). However, the molecular and cellular processes that are essential during renal maturation remain largely unknown.

Proper regulation of apoptosis is essential for normal kidney development and to maintain normal renal function in the adult. Bcl-2 plays an important role in regulating apoptosis during kidney development. Bcl-2 is highly expressed in the ureteric bud and metanephric mesenchyme during early development and gradually declines to low levels in the postnatal kidney ( 10 ). The physiological role bcl-2 plays during kidney development is not well understood. We have shown that cystic kidneys from bcl-2 -/- mice demonstrate nuclear localization of -catenin, loss of apical brush-border actin staining, and sustained activation of focal adhesion kinase (FAK) and paxillin, suggesting that altered cell-adhesive properties may contribute to the cystic process ( 15, 18 ). We also showed that bcl-2 interacts with paxillin and FAK in metanephroi from normal mice ( 18 ). The physiological consequences of these interactions and the effect loss of bcl-2 has on such interactions, perhaps contributing to renal cyst formation, require further investigation. We presently lack essential information as to what steps are critical for cyst formation and how to modulate this process. The loss of bcl-2 during early development affects the environment in which renal epithelial cells are required to survive, differentiate, and mature. Thus bcl-2-deficient renal epithelial cells may not receive the appropriate signals, form appropriate signaling complexes, or have such an excessive loss of progenitor cells that initiation of terminal differentiation may not be achieved.

Previous work from this laboratory demonstrated significant cyst formation in both proximal tubules and collecting ducts in P20 bcl-2 -/- mice ( 16 ). In the studies described here, we investigated cell-adhesive and migratory properties of proximal tubule and collecting duct cells from P10 (before renal maturation) and P20 (following renal maturation in normal mice) bcl-2 +/+ and bcl-2 -/- mice. Very little is known about the processes required for renal maturation. We demonstrate that modulation of cell-adhesive properties occurs in a nephron segment-specific manner and is aberrant as early as P10 in bcl-2 -/- mice. This is the earliest indication of renal maturation gone awry in bcl-2 -/- mice. We observed increased secretion of fibronectin, decreased secretion of thrombospondin-1, increased migration in a scratch wound assay, altered adhesion to extracellular matrices, and integrin expression. The Src kinase inhibitor, PP1, effectively inhibited migration of bcl-2 -/- proximal tubule and collecting duct cells. Therefore, altered cell-adhesive and migratory properties may be an earlier indicator of aberrant differentiation leading to renal cyst formation.

MATERIALS AND METHODS

Cell preparation. Immortomice expressing a temperature-sensitive SV40 large T antigen were obtained from Charles River Laboratories (Wilmington, MA). Bcl-2 +/- mice were crossed with the Immortomouse and screened as previously described ( 16 ). The kidneys from P10 and P20 bcl-2 +/+ and bcl-2 -/- Immortomice were harvested from two to three mice of each genotype and processed individually. The experiments described here were performed with two separate isolations of cells with similar results.

The tissue was minced into small pieces in a 60-mm tissue culture dish using sterile razor blades and digested in 5 ml of collagenase type I (1 mg/ml in serum-free DMEM; Worthington, Lakewood, NJ) for 30-45 min at 37°C. After digestion, DMEM containing 10% FBS was added, and the cells were pelleted and rinsed twice in DMEM containing 10% FBS. The cells were resuspended in growth medium DMEM:F-12 (Invitrogen, Carlsbad, CA) supplemented with 1% FBS, 5 x MITO (BD BioSciences, Franklin Lakes, NJ), 44 U/ml -interferon (R&D, Minneapolis, MN), 2 mM glutamine, 50 µg/ml streptomycin/50 U/ml penicillin (Sigma, St. Louis, MO), 50 µg/ml gentamicin (Invitrogen), and 50 U/ml nystatin (Sigma) and plated on a 100-mm dish precoated with Matrigel (1:400 in serum-free DMEM:F-12 medium). The cells were plated and grown to near confluence. One 100-mm dish of cells was harvested by incubation with 2 mM EDTA in Tris-buffered saline (TBS) containing 0.05% BSA for 10 min and scraping. The cells were rinsed with serum-free DMEM:F-12 and incubated with magnetic beads (Dynel, Biotech, Brown Deer, WI) precoated with Dolichos biflorus agglutinin (DBA) or Lotus tetragonolobus agglutinin (LTA). After being bound, the magnetic beads were washed six times with DMEM containing 10% FBS and the bound cells were plated into a single well of a 24-well plate precoated with Matrigel (1:400) in growth medium. The cells were maintained at 33°C with 5% CO 2. Cells were progressively passed to larger plates, maintained, and propagated on Matrigel (1:400)-coated 60-mm plates. The selection process was repeated twice.

Western blot analysis. For fibronectin and thrombospondin-1 analysis, cells were plated at 2 x 10 4 per 60 mm on Matrigel-coated dishes and allowed to reach 90% confluence in 2 days. The cells were then rinsed once with serum-free medium and incubated with serum-free growth medium (complete medium without serum) for 2 days. Then, conditioned medium was collected and clarified by centrifugation. The samples were mixed with appropriate volume of 6 x SDS buffer and analyzed by 4-20% SDS-PAGE (Invitrogen). The proteins were transferred to a nitrocellulose membrane, and the membrane was incubated with a rabbit anti-rat fibronectin (Invitrogen) or anti-thrombospondin-1 (Clone A6.1, Neo Marker, Fremont, CA). The blot was washed, incubated with appropriate secondary antibody, and developed using ECL (Amersham, Piscataway, NJ).

Indirect immunofluorescence staining. Proximal tubule or collecting duct cells were plated on chamber slides (CC1:Falcon, Franklin Lakes, NJ) coated with Matrigel as described above. Cells were then rinsed with PBS, fixed with 3% of paraformaldhyde for 10 min on ice, washed two times with PBS, and incubated with anti-vinculin (1:100; Sigma), paxillin (1:200; BD Biosciences), -catenin (1:600; Sigma), aquaporin-1 (1:100; Santa Cruz Biotechnology, Santa Cruz, CA), aquaporin-2 (1:100; Santa Cruz Biotechnology), calbindin (1:100; Santa Cruz Biotechnology), DBA-FITC (1:40; Vector, Burlingame, CA), or LTA-FITC (1:40; Vector) for 30 min at 37°C. After being washed three times with TBS, cells were incubated with appropriate CY3-conjugated secondary antibodies (Jackson ImmunoResearch, West Grove, PA) at 37°C for 30 min. Cells were washed three times with TBS, mounted, and photographed using a Zeiss fluorescence microscope (Axiophot, Zeiss, Germany) equipped with a digital camera.

Scratch wound assays. The migration of cells was assessed in scratch wound assays. Confluent monolayers of proximal tubule or collecting duct cells were wounded using a micropipette tip (1-ml blue tip), rinsed with growth medium, and wound closure was monitored for 48 h and photographed using a digital camera. The cells were grown on Matrigel- or fibronectin (2 µg/ml)-coated plates. Similar assays were performed in the presence of 5-flurouracil (10 µg/ml; Sigma) to rule out the potential contribution of differences in cell proliferation. These experiments were repeated at least twice with similar results using different isolations of cells. In some cases, the cells were wounded and incubated in medium containing DMSO (control), PP1 (Src inhibitor; 10 µM), PD-98059 (MEK inhibitor; 50 µM), LY-294002 (PI3 kinase inhibitor; 3.25 mM), or SB-203580 (p38 inhibitor; 10 µM) for 48 h (all from Alexis, San Diego, CA).

Cell adhesion assays. Varying concentrations of fibronectin, laminin, vitronectin, and type I collagen (all from BD BioSciences) were coated on 96-well plates. As a control, wells were coated with BSA (1% BSA) overnight at 4°C. After being blocked with 1% BSA, 5 x 10 4 cells were removed by EDTA, washed in serum-free medium, added to each well, and the cells were allowed to adhere to the plate for 1-2 h at 37°C. The plate was washed with TBS containing 1 mM Ca +2 and 1 mM Mg +2 until no cells remained in the wells coated with BSA. The number of cells in each well was quantitated by measuring the cellular phosphatase activity as previously described ( 11 ). All samples were done in triplicate.

FACScan analysis. FACScan analysis was performed essentially as previously described ( 11 ). The cells were washed once with PBS containing 0.04% EDTA and incubated with 3 ml of dissociation solution (Sigma) to remove the cells from the plate. The cells were washed with TBS, blocked in TBS containing 1% goat serum on ice for 20 min, and incubated with the appropriate dilution of primary antibody, anti- 3 (AB1920; Chemicon, Temecula, CA), 5 (AB1949;Chemicon), 6 (MAB1378; Chemicon), 1 (SC-8978; Santa Cruz Biotechnology), or control IgG (Chemicon). The cells were washed with TBS containing 1% BSA and then incubated with the appropriate secondary antibody on ice for 30 min. After the incubation, the cells were washed twice with TBS containing 1% BSA and resuspended in 0.5 ml of TBS containing 1% BSA. FACScan analysis was performed on a FACScan caliber flow cytometer (Becton-Dickinson, Franklin Lakes, NJ).

Immunohistochemistry. As previously described ( 18 ), fresh frozen kidney sections were fixed in acetone, blocked, and incubated with anti-rat fibronectin (1:500; Invitrogen). The sections were then incubated with indocarbocyanin (CY3)-labeled secondary antibody (Jackson ImmunoResearch). An equal timed photograph was then taken of the sections. Photographs were taken using a Zeiss fluorescence microscope (Axiophot, Zeiss) equipped with a digital camera.

RESULTS

Bcl-2 -/- proximal tubule and collecting duct cells secrete increased amounts of fibronectin. The cell migration and adhesion characteristics of renal epithelial cells during postnatal renal maturation remain largely unknown. Here, we used cell lines from bcl-2 +/+ and bcl-2 -/- mice to examine their cell-adhesive and migratory properties. We compared and contrasted proximal tubule and collecting duct cells prepared from P10 (before renal maturation) and P20 (following renal maturation) mice. Proximal tubule and collecting duct cells were prepared as described in MATERIALS AND METHODS. The identity of the proximal tubule cells was verified by positive staining for LTA and aquaporin-1, whereas collecting duct cells were positive for DBA, aquaporin-2, and calbindin (data not shown).

One characteristic of dedifferentiated cells or cells that have not completed terminal differentiation is that they tend to display increased expression of fibronectin compared with their differentiated counterparts ( 20 ). Serum-free medium was placed on P10 and P20 proximal tubule and collecting duct cells and collected 48 h later. Figure 1 A is a Western blot of the conditioned medium blotted for fibronectin. We observed that proximal tubule and collecting duct cells from P10 and P20 bcl-2 -/- mice secrete increased amounts of fibronectin into the medium compared with bcl-2 +/+ cells.

Fig. 1. Bcl-2 -/- cells secrete an increased amount of fibronectin. A : postnatal day 10 ( P10 ) and P20 proximal tubule and collecting duct cells from bcl-2 +/+ and bcl-2 -/- mice were grown for 2 days in serum-free medium. The medium was harvested, clarified, and Western blotted for fibronectin. These experiments were repeated 3 times with similar results. B : fibronectin deposition was examined in kidneys from P20 bcl-2 -/- and bcl-2 +/+ mice. Frozen sections were stained with anti-fibronectin. Photographs with equally timed exposures and magnification were taken. The arrows point to the renal capsule, and the arrowheads point to glomeruli. The staining was repeated with kidneys from 4 mice of each genotype with similar results.

We next examined deposition of fibronectin in kidneys from P20 bcl-2 +/+ and bcl-2 -/- mice. Kidneys from bcl-2 +/+ display faint fibronectin staining in the cortex ( Fig. 1 B ). The arrow points to the renal capsule. In contrast, kidneys from bcl-2 -/- mice display increased deposition of fibronectin ( Fig. 1 B ). We observed similar increased fibronectin staining in kidneys from bcl-2 -/- mice regardless of the degree of cyst formation (data not shown). Therefore, increased production of fibronectin also occurs in kidneys from bcl-2 -/- mice.

Collecting duct cells secrete a decreased amount of thrombospondin-1. Thrombospondin-1 is a secreted extracellular matrix protein that can modulate cell-adhesive and migratory properties through its interactions with a number of cell-surface receptors and cytokines ( 1, 13 ). Serum-free medium was placed on P10 and P20 proximal tubule and collecting duct cells and collected 48 h later. Figure 2 is a Western blot of the conditioned medium blotted for thrombospondin-1. Both bcl-2 +/+ and bcl-2 -/- P10 proximal tubule cells secreted a large amount of thrombospondin-1. This level decreased substantially in P20 proximal tubule cells.

Fig. 2. Bcl-2 -/- collecting duct cells secrete a decreased amount of thrombospondin-1. P10 and P20 proximal tubule and collecting duct cells from bcl-2 +/+ and bcl-2 -/- mice were grown for 2 days in serum-free medium. The medium was harvested, clarified, and Western blotted for thrombospondin-1. These experiments were repeated 3 times with similar results.

The regulation of thrombospondin-1 in proximal tubule and collecting duct cells was not identical. P10 bcl-2 +/+ collecting duct cells secrete substantial amounts of thrombospondin-1, which declines to low levels at P20. In contrast, bcl-2 -/- collecting duct cells secreted consistently lower levels of thrombospondin-1 at the times examined. Thus thrombospondin-1 is inappropriately downregulated in P10 bcl-2 -/- collecting duct cells. These data suggest that cell adhesion and migratory properties may be affected in cells from bcl-2 -/- mice.

Bcl-2 -/- proximal tubule cells demonstrate increased migration in a scratch wound assay. We examined migration of proximal tubule and collecting duct cells using a scratch wound assay. Proximal tubule cells were wounded and grown in culture for 48 h ( Fig. 3 ). At 48 h, only a few P10 bcl-2 +/+ cells had migrated into the wound (note arrows). In contrast, P10 proximal tubule bcl-2 -/- cells spread across the wound ( Fig. 3 A ). Wounded P20 bcl-2 +/+ proximal tubule cells (postrenal maturation) do not migrate into the wound even after 7 days (data not shown). In contrast, P20 bcl-2 -/- proximal tubule cells migrate with complete wound closure by 48 h. To rule out the potential contribution of differences in cell proliferation, similar assays were performed in the presence of 5-flurouracil with identical results. These experiments were repeated at least twice with similar results. Therefore, bcl-2 -/- cells were more migratory.

Fig. 3. Proximal tubule cells from bcl-2 -/- mice displayed increased migration in a scratch wound assay. A : P10 and P20 proximal tubule cells were grown to confluence and wounded with a P1000 pipet tip. Photomicrographs were taken after 48 h. Under these conditions, bcl-2 -/- proximal tubule cells migrate to close the wound during the 48 h in culture. B : P20 proximal tubule cells were plated on Matrigel- or fibronectin (2 µg/ml)-coated plates. The cells were grown to confluence and wounded. Photomicrographs were taken after 48 h. Please note migration of bcl-2 +/+ proximal tubule cells on fibronectin-coated plates. Note arrow points to wound. C : P20 bcl-2 -/- proximal tubule cells were wounded and incubated with PP1 (Src inhibitor; 10 µM), PD-98059 (MEK inhibitor; 50 µM), LY-294002 (PI3 kinase inhibitor; 3.25 mM), or SB-203580 (p38 inhibitor; 10 µM). Photomicrographs were taken following 48 h. These experiments were repeated 3 times with similar results.

We observed that P20 bcl-2 -/- proximal tubule cells secreted increased amounts of fibronectin compared with their wild-type counterparts ( Fig. 1 ). We next examined whether plating proximal tubule cells on fibronectin (2 µg/ml)-coated plates would influence their migration in a scratch wound assay. Figure 3 B is a scratch wound assay of P20 bcl-2 +/+ and bcl-2 -/- proximal tubule cells plated on Matrigel- or fibronectin-coated plates. As observed previously, little if any migration of P20 bcl-2 +/+ proximal tubule cells plated on Matrigel-coated plates was noted ( Fig. 3, A and B ). However, when these same cells are plated on fibronectin-coated plates, P20 bcl-2 +/+ proximal tubule cells nearly fill the wound in 48 h. P20 bcl-2 -/- cells plated on fibronectin behave similar in a scratch wound assay to those plated on Matrigel with wound closure within 48 h. Thus the presence of fibronectin may influence cell migration.

Several signaling pathways including PI3 kinase, Src, Erks, and p38 MAPK have been implicated in cell migration ( 8, 14 ). Altered signaling through these pathways could contribute to renal cyst formation due, in part, to sustained migration and altered adhesion. We next examined whether inhibitors of these signaling pathways could block migration of P20 bcl-2 -/- proximal tubule cells using the scratch wound assay. P20 bcl-2 -/- proximal tubule cells were wounded and grown in culture for 48 h in medium containing DMSO (control; data not shown), PP1 (Src inhibitor; 10 µM), PD-98059 (MEK inhibitor; 50 µM), LY-294002 (PI3 kinase inhibitor; 3.25 mM), or SB-203580 (p38 inhibitor; 10 µM). P20 bcl-2 -/- proximal tubule cells incubated with DMSO (control), LY-294002 (PI3 kinase inhibitor), or PD-98059 (MEK inhibitor) readily migrated into and closed the wound ( Fig. 3 C ). In contrast, cells incubated with PP1 (Src inhibitor) exhibited little to no migration into the wound, whereas incubation with SB-203580 (p38 inhibitor) had a moderate effect on migration ( Fig. 3 C ).

Bcl-2 -/- collecting duct cells demonstrate increased migration in a scratch wound assay. Migration of collecting duct cells was analyzed using a scratch wound assay. Similar to proximal tubule cells, collecting duct cells from P10 and P20 bcl-2 +/+ mice demonstrated little migration into the wounded area. Collecting duct cells from P10 and P20 bcl-2 -/- mice spread readily across the wound to fill the area ( Fig. 4 A ).

Fig. 4. Collecting duct cells from bcl-2 -/- mice display increased migration. A : P10 and P20 collecting duct cells were grown to confluence and wounded. Photomicrographs were taken after 48 h. Under these conditions, bcl-2 -/- collecting duct cells migrate to close the wound during the 48 h in culture. B : P20 bcl-2 -/- collecting duct cells were wounded and incubated with PP1 (10 µM), PD-98059 (50 µM), LY-294002 (3.25 mM), or SB-203580 (10 µM). Photomicrographs were taken following 48 h. These experiments were repeated 3 times with similar results.

We next incubated collecting duct cells with PI3 kinase, MEK, Src, and p38 MAPK inhibitors to determine whether these signaling pathways were involved in migration of P20 bcl-2 -/- collecting duct cells. The P20 bcl-2 -/- collecting duct cells were wounded and grown in culture for 48 h in medium containing DMSO (control; data not shown), PP1 (Src inhibitor; 10 µM), PD-98059 (MEK inhibitor; 50 µM), LY-294002 (PI3 kinase inhibitor; 3.25 mM), or SB-203580 (p38 inhibitor; 10 µM). P20 bcl-2 -/- collecting duct cells incubated with DMSO (control; data not shown), and SB-203580 readily migrated into and closed the wound ( Fig. 4 B ). In contrast, addition of PP1 inhibited migration into the wound while decreased migration into the wound was observed in collecting duct cells incubated with PD-98059 and LY-294002 ( Fig. 4 B ). These data suggest that differences in signal transduction pathway preferences may exist between tubular segments. This could indicate that adhesive characteristics may be unique to specific tubular segments at different stages of development.

Adhesion to extracellular matrices is development and nephron specific. The differences in migration observed in the proximal tubule and collecting duct cells suggested that altered adhesive mechanisms may exist. Adhesion to collagen, fibronectin, laminin, and vitronectin was assessed in proximal tubule and collecting duct cells from P10 and P20 bcl-2 +/+ and bcl-2 -/- mice ( Fig. 5 A ). P10 proximal tubule cells did not adhere to collagen even at 10 µg/ml ( Fig. 5 A ). After renal maturation, P20 bcl-2 +/+ proximal tubule cells adhere to collagen with saturation occurring at 1-2 µg/ml of collagen. No significant adhesion of P20 bcl-2 -/- proximal tubule cells to collagen was observed. Similarly, P10 bcl-2 +/+ and bcl-2 -/- proximal tubule cells do not adhere well to laminin, although modest adhesion was noted at the highest concentration of laminin (10 µg/ml) in P10 bcl-2 -/- cells. In contrast, P20 bcl-2 +/+ proximal tubule cells adhere well to laminin, whereas P20 bcl-2 -/- cells did not adhere. Therefore, adhesion to collagen and laminin increases following renal maturation ( P20 ) in bcl-2 +/+ proximal tubule cells.

Fig. 5. Altered adhesion to matrices was observed in bcl-2 -/- cells. Adhesion to type I collagen, fibronectin, laminin, and vitronectin was assessed in proximal tubule and collecting duct cells from P10 and P20 bcl-2 +/+ and bcl-2 -/- mice. A : adhesion of proximal tubule cells to matrices. B : adhesion of collecting duct cells to the various matrices. In all cases, represents bcl-2 +/+ cells and represents bcl-2 -/- cells (means ± SD). These adhesion assays were repeated twice with similar results. All samples were performed in triplicate. OD, outer diameter.

We examined the ability of proximal tubule cells to adhere to fibronectin and found that both bcl-2 +/+ and bcl-2 -/- proximal tubule cells adhere to fibronectin ( Fig. 5 A ). Bcl-2 +/+ P10 and P20 proximal tubule cells demonstrated similar adhesion to fibronectin. Although P20 bcl-2 -/- proximal tubule cells retain their ability to adhere to fibronectin, it is significantly lower than that observed in P10 cells. A slightly different trend is observed for adhesion to vitronectin. P10 bcl-2 +/+ cells exhibited moderate adhesion to vitronectin with P10 bcl-2 -/- cells being even more adherent to vitronectin. However, a decline in the ability of P20 bcl-2 +/+ and bcl-2 -/- proximal tubule cells occurred with significant adhesion only noted at the highest concentration of vitronectin (10 µg/ml).

We next examined how adhesion to various matrices was affected in postnatal collecting duct cells ( Fig. 5 B ). P10 bcl-2 +/+ collecting duct cells did not adhere well to collagen, laminin, fibronectin, or vitronectin. Although P10 bcl-2 -/- cells did not adhere well to collagen or laminin, these cells did adhere to vitronectin and fibronectin. P20 bcl-2 +/+ and bcl-2 -/- collecting duct cells demonstrated similar adhesive characteristics. Neither P20 bcl-2 +/+ or bcl-2 -/- cells adhered well to collagen or laminin, although both cells displayed a similar ability to adhere to fibronectin and vitronectin. The level of adhesion to fibronectin and vitronectin in P20 collecting duct cells was similar to that observed in the P10 bcl-2 -/- collecting duct cells.

Aberrant integrin expression in P20 bcl-2 -/- proximal tubule cells. P20 bcl-2 -/- proximal tubule cells exhibit lack of adhesion to collagen and laminin. We next examined integrin expression of the 3 -, 5 -, 6 -, and 1 -subunits in proximal tubules by FACScan analysis ( Fig. 6 ). Bcl-2 +/+ proximal tubule cells did not express detectable levels of 5 -integrin; however, bcl-2 -/- proximal tubule cells demonstrate significant 5 -integrin expression. Similar expression of 3 - and 6 -integrin as well as a slight decrease in 1 -integrin expression were observed in bcl-2 -/- cells compared with their wild-type counter parts.

Fig. 6. P20 Bcl-2 -/- proximal tubule cells demonstrate increased expression of 5 -integrin. FACScan analysis was used to examine integrin expression in P20 proximal tubule cells. Expression of 3 -, 5 -, 6 -, and 1 -integrin (open lines) was examined compared with IgG control (filled lines). Please note increased expression of 5 -integrin in P20 bcl-2 -/- proximal tubule cells. These experiments were repeated with identical results.

Formation of focal adhesions and cell-cell contacts. Vinculin staining was used to visualize focal adhesions in proximal tubule and collecting duct cells. P10 bcl-2 +/+ proximal tubule cells demonstrated vinculin staining of peripheral focal adhesions, whereas bcl-2 -/- proximal tubule cells displayed vinculin staining of central focal adhesions ( Fig. 7 A ). This is consistent with the enhanced migration of these cells. Similar results were observed with P20 proximal tubule cells. In contrast, vinculin staining of P10 and P20 bcl-2 +/+ and bcl-2 -/- collecting duct cells demonstrated similarly stained focal adhesions on the cell periphery. Staining for paxillin, another component of focal adhesions, yielded similar results (data not shown).

Fig. 7. Bcl-2 -/- proximal tubule cells display focal adhesions over the entire surface of the cell. A : to visualize focal adhesions, P10 bcl-2 +/+ and bcl-2 -/- proximal tubule and collecting duct cells were stained with antivinculin. B : to visualize cell-cell contacts, P10 collecting duct cells were stained with anti- -catenin. These studies were repeated 6 times with similar results.

We next stained these cells with anti- -catenin to visualize cell-cell contacts ( Fig. 7 B ). Staining of P10 bcl-2 +/+ collecting duct cells demonstrate smooth symmetrical -catenin staining at sites of cell-cell contact, whereas bcl-2 -/- collecting duct cells demonstrated a more discontinuous pattern of -catenin staining. Similar results were observed in proximal tubule cells (data not shown).

DISCUSSION

Cell migration, proliferation, differentiation, and apoptosis play integral roles during nephrogenesis. Unfortunately, examining cell migration and adhesive mechanisms in an intact organ can be a difficult undertaking. Preparation of cell lines was an important step in understanding the cell-adhesive characteristics of proximal tubule and collecting duct cells during renal maturation and how altered regulation of these processes may contribute to renal cystic disease. Other laboratories have prepared immortalized kidney cell lines in this manner ( 9, 21 ) that have adhesive mechanisms similar to nonimmortalized primary kidney cell lines ( 23, 24 ). Very little is known about regulation of cell-adhesive and migratory properties during renal maturation. However, our data emphasize the need to use appropriate age-matched and nephron segment-matched controls when studying normal and diseased kidneys. We observed that modulation of cell-adhesive properties which normally occur in a segment-specific manner was altered in renal epithelial cells from bcl-2 -/- mice. Here, we show that bcl-2 -/- proximal tubule cells demonstrated 1 ) enhanced migration in a scratch wound assay that was inhibited by incubation with PP1 (Src inhibitor) and moderately affected by SB-203580 (p38 inhibitor) ( Fig. 3 C ); 2 ) increased expression of fibronectin and numerous central focal adhesions; and 3 ) decreased ability of P20 bcl-2 -/- proximal tubule cells to adhere to fibronectin and an inability to adhere to collagen or laminin. Collecting duct cells from bcl-2 -/- mice demonstrated 1 ) enhanced migration in a scratch wound assay that was inhibited by incubation with PP1 (Src kinase inhibitor) and slowed by incubation with PD-98059 (MEK inhibitor) or LY-294002 (PI3 kinase inhibitor); 2 ) increased levels of fibronectin and decreased levels of thrombospondin-1 expression; and 3 ) precocious binding to fibronectin and vitronectin at P10. The ability of collecting duct cells to bind to fibronectin and vitronectin corresponded with a decline in thrombospondin-1 levels in these cells. This is consistent with anti-adhesive activity associated with thrombospondin-1 ( 5 ).

Bcl-2 plays an important role in regulating apoptosis during nephrogenesis. Bcl-2 expression is highest during early nephrogenesis and gradually declines to low levels in the postnatal kidney ( 3, 10 ). In the proximal tubule and collecting duct cell lines used in the studies presented here, bcl-2 protein expression was at undetectable levels. The important role bcl-2 plays during early nephrogenesis is underscored by the excessive apoptosis at embryonic day 12 ( E12 ) in bcl-2 -/- mice ( 17 ) and resulting renal hypoplasia. It is likely that the lack of terminal differentiation in kidneys from bcl-2 -/- mice results from changes that occurred in an earlier stage of development. Loss of bcl-2 during early development would dramatically affect the environment in which renal epithelial cells are required to survive, differentiate, and mature. It is tempting to speculate that different cells types may adapt differently to these adverse circumstances. Thus loss of bcl-2 early during nephrogenesis may affect the ability of these kidneys to differentiate/mature perhaps as a result of inappropriate signal transduction in the postnatal kidney.

Renal cyst formation has been described as a benign neoplasm ( 6 ). As seen in many disease states, renal cysts typically contain dedifferentiated epithelial cells. Understanding the process of development including postnatal renal maturation is essential to understanding various renal diseases. Here, we demonstrate that cell-adhesive properties are impacted in a nephron and developmental stage-specific fashion. After renal maturation, bcl-2 +/+ proximal tubule cells acquire the ability to adhere to collagen and laminin. We also observed that following renal maturation, proximal tubule cells (bcl-2 +/+ and bcl-2 -/-) lose their ability to adhere to vitronectin. In contrast, P20 bcl-2 -/- proximal tubule cells demonstrate an inability to adhere to collagen and laminin.

Collecting duct cells demonstrate a different pattern of adhesion to these extracellular matrices. Before renal maturation ( P10 ), bcl-2 +/+ collecting duct cells did not adhere to collagen, fibronectin, laminin, or vitronectin. These cells also expressed very low levels of fibronectin and significant levels of thromobspondin-1. After renal maturation, bcl-2 +/+ collecting duct cells gain the ability to adhere to fibronectin and vitronectin. This corresponded with a decline in thrombospondin-1 levels and a modest increase in fibronectin expression. Interestingly, both P10 and P20 bcl-2 -/- collecting duct cells adhere to fibronectin and vitronectin but not collagen or laminin. These cells also expressed greater amounts of fibronectin but low levels of thrombospondin-1. In addition, altered adhesion to extracellular matrices was observed in cells from postnatal polycystic kidneys from cpk mice ( 22 ). Therefore, gaining a better understanding of the processes required for normal renal maturation will give us important insight into which deregulated processes contribute to renal cyst formation.

Proximal tubule and collecting duct cells from postnatal bcl-2 -/- mice migrate more rapidly into a wound than their wild-type counterparts and secrete more fibronectin. We also observed upregulation of 5 -integrin in P20 bcl-2 -/- proximal tubule cells. It is tempting to speculate that fibronectin plays a role in the enhanced migration because P20 bcl-2 +/+ proximal tubule cells plated on fibronectin-coated plates lose their nonmigratory phenotype and instead migrate to fill a wound within 48 h ( Fig. 3 B ). Although the trigger for increased secretion of fibronectin is elusive, excessive fibronectin may play an important role in a less dedifferentiated phenotype. An environment where fibronectin expression is high could favor focal adhesion formation disrupting cell-cell interactions. In addition, bcl-2 -/- collecting duct cells express low levels of thromobspondin-1. Thrombospondin-1 levels are high at P10 and decrease to low levels at P20 in bcl-2 +/+ collecting duct cells. Increased levels of thrombospondin-1 at P10 could increase transforming growth factor- levels promoting differentiation and leading to a decrease in proliferation ( 4 ). Thus the appropriate expression of fibronectin and thrombospondin-1 in the postnatal kidney may be an essential part of renal epithelial cell differentiation.

Our previous work in intact cystic kidneys from P20 bcl-2 -/- mice demonstrated sustained activation of Src and Erks but no alteration of p38 MAPK or PI3 kinase activation ( 19 ). Migration of bcl-2 -/- proximal tubule and collecting duct cells is inhibited by the Src kinase inhibitor PP1. The inhibitor studies presented here also suggest that differences in signaling pathways usage may exist between different nephron segments in cystic kidneys from bcl-2 -/- mice. Work from other laboratories suggests that Src activity correlates with migration ( 8, 14 ), whereas activation of the MAP kinase/Erks pathway correlates with cell survival and proliferation ( 8 ). Although this appears to be the case in proximal tubule cells, collecting duct cells from bcl-2 -/- mice behave somewhat differently. Migration of P20 bcl-2 -/- collecting duct cells was inhibited by PD-98059 (MEK inhibitor) and LY-294002 (PI3 kinase inhibitor), suggesting that migration in this segment may be more tightly regulated.

In summary, modulation of cell-adhesive properties occurs in nephron segment-specific manner. Although both bcl-2 -/- proximal tubule and collecting duct cells demonstrated similar alterations in migration and secretion of fibronectin, differences in adhesion to extracellular matrices were consistent with their nephron segment of origin. These data imply that treatment of a renal disease in which multiple nephron segments are involved may be a complex undertaking. Therefore, understanding the processes involved during terminal differentiation of renal epithelial cells may allow us to identify their trigger(s), pointing the way for effective treatments of renal cystic disease.

GRANTS

This work and C. M. Sorenson were supported by a Beginning Grant-in-Aid from the American Heart Association Northland Affiliate. J. Ziehr was supported by a Hilldale Undergraduate Research Fellowship. N. Sheibani was supported by National Institutes of Health Grants AR-45599 and EY-13700. N. Sheibani is the recipient of a career development award from the Research to Prevent Blindness Foundation.

ACKNOWLEDGMENTS

The authors thank Dr. A. Friedman for critically reviewing the manuscript, R. Gordon for figure preparation, Dr. S. Wang for fibronectin staining of kidney sections, and E. Scheef for FACS analysis.

【参考文献】
  Adams JC. Thrombospondins: multifunctional regulators of cell interactions. Annu Rev Cell Dev Biol 17: 25-51, 2001.

Bard JBL, McConnell JE, and Davies JA. Towards a genetic basis for kidney development. Mech Dev 48: 3-11, 1994.

Chandler D, El-Naggar AK, Brisbay S, Redline RW, and McDonnell TJ. Apoptosis and expression of the bcl-2 proto-oncogene in the fetal and human adult kidney: evidence for the contribution of bcl-2 expression to renal carcinogenesis. Hum Pathol 25: 789-796, 1994.

Crawford SE, Stellmach V, Murphy-Ullrich JE, Ribeiro SMF, Lawler J, Hynes RO, Boivin GP, and Bouck N. Thrombospondin-1 is a major activator of TGF- 1 in vivo. Cell 93: 1159-1170, 1998.

Goicoechea S, Pallero MA, Eggleton P, Michalak M, and Murphy-Ullrich JE. The anti-adhesive activity of thrombospondin is mediated by the n-terminal domain of cell surface calreticulin. J Biol Chem 277: 37219-37228, 2002.

Grantham JJ. Polycystic kidney disease: neoplasia in disguise. Am J Kidney Dis 15: 110-116, 1990.

Koseki C, Herzlinger D, and Al-Awqati Q. Apoptosis in metanephric development. J Cell Biol 119: 1327-1333, 1992.

LaVallee T, Prudovsky I, McMahon G, Hu X, and Maciag T. Activation of the MAP kinase pathway by FGF-1 correlates with cell proliferation induction while activation of the Src pathway correlates with migration. J Cell Biol 141: 1647-1658, 1998.

Loghman-Adham M, Nauli SM, Soto CE, Kariuki B, and Zhou J. Immortalized epithelial cells from human autosomal dominant polycystic kidney cysts. Am J Physiol Renal Physiol 285: F397-F412, 2003.

Novak DV and Korsmeyer SJ. Bcl-2 protein expression during murine development. Am J Pathol 145: 61-73, 1994.

Rothermel TA, Engelhardt B, and Sheibani N. Polyoma virus middle-T-transformed PECAM-1 deficient mouse brain endothelial cells proliferate rapidly in culture and form hemangiomas in mice. J Cell Physiol. In press.

Saxen L and Sariola H. Early organogenesis of the kidney. Pediatr Nephrol 1: 385-392, 1987.

Sheibani N and Frazier WA. Thrombospondins, PECAM-1 and regulation of angiogenesis. Histol Histopathol 14: 285-294, 1999.

Slack J, Adams R, Rovin J, Bissonette E, Stoker C, and Parsons J. Alterations in the focal adhesion kinase/Src signal transduction pathway correlate with increased migratory capacity of prostate carcinoma cells. Oncogene 20: 1152-1163, 2001.

Sorenson CM. Nuclear localization of -catenin and loss of apical brush border actin in cystic tubules of bcl-2 -/- mice. Am J Physiol Renal Physiol 276: F210-F217, 1999.

Sorenson CM, Padanilam B, and Hammerman MR. Abnormal postpartum renal development and cystogenesis in the bcl-2 -/- mouse. Am J Physiol Renal Fluid Electrolyte Physiol 271: F184-F193, 1996.

Sorenson CM, Rogers SA, Korsmeyer SJ, and Hammerman MR. Fulminant metanephric apoptosis and abnormal kidney development in bcl-2-deficient mice. Am J Physiol Renal Fluid Electrolyte Physiol 268: F73-F81, 1995.

Sorenson CM and Sheibani N. Focal adhesion kinase, paxillin and bcl-2: analysis of expression, phosphorylation and association during morphogenesis. Dev Dyn 215: 371-382, 1999. <a href="/cgi/external_ref?access_num=10.1002/(SICI)1097-0177(199908)215:4

Sorenson CM and Sheibani N. Sustained activation of MAPK/ERK's signaling pathway in cystic kidneys from Bcl-2 -/- mice. Am J Physiol Renal Physiol 283: F1085-F1090, 2002.

Su X, Sorenson CM, and Sheibani N. Isolation and characterization of murine retinal endothelial cells. Mol Vis 9: 171-178, 2003.

Sweeney W Jr, Kusner L, Carlin C, Chang S, Futey L, Cotton C, Dell K, and Avner E. Phenotypic analysis of conditionally immortalized cells isolated from the BPK model of ARPKD. Am J Physiol Cell Physiol 281: C1695-C1705, 2001.

Van Adelsberg J. Murine polycystic kidney epithelial cell lines have increased integrin-mediated adhesion to collagen. Am J Physiol Renal Fluid Electrolyte Physiol 267: F1082-F1093, 1994.

Wang Z, Symons JM, Goldstein SL, McDonald A, Miner JH, and Kreidberg JA. 3 1 Integrin regulates epithelial cytoskeleton organization. J Cell Sci 112: 2925-2935, 1999.

Woolf AS, Kolatsi-Joammou M, Hardman P, Andermarcher E, Moorby C, Fine LG, Jat PS, Noble MD, and Gherardi E. Roles of hepatocyte growth factor/scatter factor and the Met receptor in the early development of the metanephros. J Cell Biol 128: 171-184, 1995.

作者单位：Departments of 1 Pediatrics and 2 Ophthalmology and Visual Sciences, University of Wisconsin Medical School, Madison, Wisconsin 53792