The extracellular matrix (ECM) acts as a scaffold for kidney cellular organization. Local secretion of the ECM allows kidney cells to readily adapt to changes occurring within the kidney. In addition to providing structural support for cells, the ECM also modulates cell survival, migration, proliferation, and differentiation. Although aberrant regulation of ECM proteins can play a causative role in many diseases, it is not known whether ECM production, cell adhesion, and migration are regulated in a similar manner in kidney epithelial and endothelial cells. Here, we demonstrate that lack of BIM expression differentially impacts kidney endothelial and epithelial cell ECM production, migration, and adhesion, further emphasizing the specialized role of these cell types in kidney function. Bim −/− kidney epithelial cells demonstrated decreased migration, increased adhesion, and sustained expression of osteopontin and thrombospondin-1 (TSP1). In contrast, bim −/− kidney endothelial cells demonstrated increased cell migration, and decreased expression of osteopontin and TSP1. We also observed a fivefold increase in VEGF expression in bim −/− kidney endothelial cells consistent with their increased migration and capillary morphogenesis. These cells also had decreased endothelial nitric oxide synthase activity and nitric oxide bioavailability. Thus kidney endothelial and epithelial cells make unique contributions to the regulation of their ECM composition, with specific impact on adhesive and migratory properties that are essential for their proper function.
- capillary morphogenesis
- extracellular matrix proteins
bcl-2 is the founding member of a family of proteins that influence apoptosis. Family members contain conserved regions denoted as Bcl-2 homology (BH) domains. Proapoptotic members are divided into those that only contain a BH3 domain and those that contain multiple BH domains (14). BIM is a BH3 only-containing proapoptotic protein. Anoikis is a form of cell death resulting from matrix detachment. BIM is a critical mediator of anoikis in epithelial cells, acting as a sensor of integrin and growth factor signals to the Erk pathway (17). Although it is well established that extracellular matrix (ECM) expression can impact cell survival, less is known as to whether modulation of proteins that influence apoptosis can impact ECM production and tissue homeostasis in a cell type-specific manner.
The mammalian kidney is a complex organ that contains over 25 different cell types. It is a highly vascularized organ in which the various segments of the vascular tree accomplish specialized regional functions (5). The microenvironment of the kidney consists of epithelial, vascular, fibroblast, and smooth muscle cells embedded in a complex network of ECM proteins, which enhances the complexity of this microenvironment. The ECM is locally secreted and acts as a scaffold for tissue organization, regulating growth factor and cytokine availability to aid tissue homeostasis. ECM composition adapts to the changing conditions within the organ, including injury. In addition to providing structural support for cells, the ECM also modulates several cell functions including cell survival, migration, proliferation, and differentiation (3). Thus changes in the ECM milieu can affect kidney structure/function through aberrant modulation of cell function such as cell survival.
Cell-cell and cell-matrix interactions impact cell proliferation, migration, differentiation, and apoptosis. The ability of the cell to sense their three-dimensional location through interaction with the ECM and neighboring cells is essential for tissue homeostasis. Altered ECM expression can impact cell-adhesive mechanisms influencing tissue architecture and function. Disruption of this delicately balanced microenvironment can also lead to many disease states. However, it is not well understood whether this delicate balance is regulated similarly in all cell types housed within this microenvironment or how they may vary in their responses. Previous work from this laboratory demonstrated that bcl-2 not only regulates apoptosis but also influences the ECM milieu, with significant impact on adhesion and migration characteristics of kidney epithelial cells. Loss of bcl-2 expression resulted in precocious downregulation of thrombospondin-1 (TSP1) and osteopontin, increased cell migration, and decreased cell adhesion (28).
To begin to address whether loss of a specific pro-apoptotic protein, BIM, in kidney epithelial and endothelial cells would have a similar impact on the microenvironment, we prepared and characterized kidney cells from weanling wild-type and bim −/− mice. Kidney epithelial cells demonstrated sustained expression of TSP1 and osteopontin, while kidney endothelial cells demonstrated decreased expression. These changes corresponded with decreased migration and increased adhesion to fibronectin, vitronectin and collagen IV in bim −/− kidney epithelial cells. In contrast, bim −/− kidney endothelial cells demonstrated increased migration, enhanced capillary morphogenesis, decreased phosphorylated endothelial nitric oxide synthase (p-eNOS) expression, a twofold decrease in nitric oxide (NO) production, and a fivefold increase in VEGF expression. Thus loss of bim expression differentially impacts kidney endothelial and epithelial cell function through modulation of their responses to their distinct local microenvironment.
MATERIALS AND METHODS
Experimental animals and cell cultures.
The mice used for these studies were maintained and treated in accordance with our protocol approved by the University of Wisconsin Animal Care and Use Committee. Immortomice expressing a temperature-sensitive SV40 large T antigen were obtained from Charles River Laboratories (Wilmington, MA). As previously described (7, 28), bim −/− mice (Jackson Laboratory, Bar Harbor, ME) were crossed with the Immortomouse and screened. To isolate kidney endothelial cells, kidneys from two to three pups (4 wk-old wild-type and bim −/− Immortomice) were dissected out aseptically and placed in serum-free DMEM containing penicillin/streptomycin (Sigma, St. Louis, MO). The kidneys were pooled, rinsed with DMEM, minced into small pieces in a 60-mm tissue culture dish using sterilized 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. Following digestion, DMEM with 10% FBS was added and cells were pelleted. The cellular digests were then filtered through a double layer of sterile 40-μm nylon mesh (Sefar America, Hanover Park, IL), centrifuged at 400 g for 10 min to pellet cells, and the cells were then washed twice with DMEM containing 10% FBS. The cells were resuspended in 1.5 ml medium (DMEM with 10% FBS) and incubated with sheep anti-rat magnetic beads precoated with anti-platelet endothelial cell adhesion molecule (PECAM)-1 antibody (MEC13.3, BD Biosciences, Bedford, MA), as described previously (22). After affinity binding, magnetic beads were washed six times with DMEM with 10% FBS, and the bound cells were plated into a single well of a 24-well plate precoated with 2 μg/ml of human fibronectin (BD Biosciences) in endothelial growth medium. Endothelial cells were grown in DMEM containing 10% FBS, 2 mM l-glutamine, 2 mM sodium pyruvate, 20 mM HEPES, 1% nonessential amino acids, 100 μg/ml streptomycin, 100 U/ml penicillin, 55 U/ml heparin (Sigma), 100 μg/ml endothelial growth supplement (Sigma), and murine recombinant interferon-γ (R&D Systems, Minneapolis, MN) at 44 U/ml. Cells were maintained at 33°C with 5% CO2. Cells were progressively passed to larger plates, maintained, and propagated in 1% gelatin-coated 60-mm dishes. Kidney endothelial cells were positive for B4-lectin (a mouse endothelial cell-specific lectin) and expressed PECAM-1 and vascular endothelial (VE)-cadherin as previously described (7, 10). The experiments described here were performed with three separate isolations of cells with similar results.
To isolate collecting duct epithelial cells (referred subsequently to as kidney epithelial cells), both kidneys from 4 wk-old wild-type and bim −/− Immortomice were 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) for 30–45 min at 37°C (28). Following 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:F12, Invitrogen, Carlsbad, CA) supplemented with 1% FBS, 5× MITO (BD Biosciences, Franklin Lakes, NJ), 44 U/ml γ-interferon (R&D Systems), 2 mM glutamine, 50 μg/ml streptomycin/50 U/ml penicillin (Sigma), 50 μg/ml gentamicin (Invitrogen), and 50 U/ml nystatin (Sigma) and plated on a 35-mm dish precoated with Matrigel (1:400 in serum-free DMEM:F12). The cells were plated, grown to near confluence, and expanded in 60-mm dishes coated with Matrigel. Cells from two 60-mm dishes were harvested by incubation with 2 mM EDTA in Tris-buffered saline containing 0.05% BSA for 10 min and scraping. The cells were rinsed with serum-free DMEM:F12 and incubated with magnetic beads precoated with Dolichos biflorus agglutinin (DBA) (28). After binding, 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% CO2. Cells were progressively passed to larger plates, maintained, and propagated on Matrigel (1:400)-coated 60-mm plates. The selection process was repeated twice. Collecting duct epithelial cells expressed aquaporin 2 and calbindin and were DBA positive (mouse collecting duct-specific lectin) as we previously described (20, 28).
Cell apoptosis assays.
As an apoptotic stimulus, cells were incubated with 5-fluorouracil (5-FU; 1 mM for epithelial cells, 5 mM for endothelial cells) or growth medium for 48 h at 37°C. The rate of apoptotic cells was determined by in situ monitoring of caspase activity using the CaspACE FITC-VAD-FMK in situ marker (Promega, Madison, WI) or Caspase 3/7 Glo (Promega), as recommended by the supplier (20).
Cells (4 × 105) were plated in 60-mm tissue culture dishes and allowed to reach confluence (2–3 days). After aspiration of the medium, cell layers were wounded using a 1-ml micropipette tip. Plates were then rinsed with PBS, fed with growth medium containing 100 ng/ml of 5-FU to rule out potential contribution of differences in cell proliferation, and incubated at 37°C for the duration of the experiment. The wounds were observed and photographed up to 72 h. The distance migrated was determined as the percentage of total distance for quantitative assessments as described previously (7). These experiments were repeated at least twice with two different isolations with similar results.
Capillary morphogenesis in Matrigel.
Matrigel (10 mg/ml; BD Biosciences) was applied at 0.5 ml/35-mm tissue culture dish and incubated at 37°C for at least 30 min to harden. Cells were removed using trypsin-EDTA, washed with growth medium once, and resuspended at 1 × 105 cells/ml in serum-free growth medium. Cells (2 ml) were gently added to the Matrigel-coated plates, incubated at 37°C, monitored for 6–24 h, and photographed using a Nikon microscope equipped with a digital camera. For quantitative assessment of the data, the mean number of branch points in 10 high-power fields (×100) was determined after 24 h. A longer incubation of the cells did not result in further branching morphogenesis (22).
Cell adhesion assays.
Cell adhesion to various matrix proteins was performed as previously described (18). Briefly, varying concentrations of fibronectin, vitronectin, collagen I, collagen IV, and laminin (BD Biosciences) prepared in TBS with Ca2+ and Mg2+ (2 mM each; TBS with Ca/Mg) were coated on 96-well plates (50 μl/well; Nunc Maxisorbe plates, Fisher Scientific) overnight at 4°C. As a control, wells were coated with 1% BSA. Plates were rinsed four times with 200 μl of TBS with Ca/Mg and blocked with 200 μl of 1% BSA prepared in TBS with Ca/Mg for at least 1 h at room temperature. Cells were removed by the dissociation solution (Sigma), washed with TBS, and resuspended at 5 × 108 cells/ml in HBS (20 mM HEPES, 150 mM NaCl, pH 7.6, and 4 mg/ml BSA). After blocking, plates were rinsed with TBS with Ca/Mg once, 50 μl of cell suspension was added to each well containing 50 μl of TBS with Ca/Mg, and the cells were allowed to adhere to the plate for 1.5 h at 37°C. The nonadherent cells were removed by gently washing the plate four times with TBS with Ca/Mg, or until no cells were left in wells coated with BSA. The number of adherent cells in each well was quantified by measuring the cellular phosphatase activity as previously described (28). All samples were done in triplicate.
Western blot analysis.
Cells were plated at 4 × 105 in 60-mm dishes coated with 1% gelatin (endothelial cells) or Matrigel (epithelial cells) and allowed to reach ∼90% confluence in 2 days. The cells were then rinsed once with serum-free medium and incubated with serum-free DMEM for 48 h at 37°C. Then, conditioned medium (3.5 ml) was collected and clarified by centrifugation. The 40-μl of sample was mixed with appropriate volume of 6× SDS buffer and analyzed by SDS-PAGE (4–20% Tris glycine gel; Invitrogen). In some cases, total protein lysates were prepared from these cells in a modified RIPA buffer [142.5 mM KCl, 5 mM MgCl2 , 10 mM HEPES, pH 7.4, 2 mM orthovanadate, and 2 mM sodium difluoride, 1% Nonidet P-40, and a complete protease inhibitor cocktail (Roche, Mannheim, Germany)]. The proteins were transferred to a nitrocellulose membrane, and the membrane was incubated with an anti-fibronectin (Sigma), rabbit anti-chicken tenascin C polyclonal antibody (AB19013; Millipore, Billerica, MA), anti-TSP1 monoclonal antibody (clone A6.1; Neo Marker, Fremont, CA), anti-collagen IV (AB756P; Millipore), anti-osteopontin (R&D Systems), anti-β-catenin (Sigma), anti-heat shock protein (HSP) 90 (Cell Signaling Technology), anti-Akt (Cell Signaling Technology), anti-phospho-Akt (Cell Signaling Technology), anti-β-actin (Sigma), anti-phospho-eNOS (Cell Signaling Technology), and anti-eNOS (Santa Cruz Bioechnology, Santa Cruz, CA). The blot was washed, incubated with appropriate secondary antibody and developed using ECL (Amersham, Piscataway, NJ) (7, 10).
FACScan analysis was performed as previously described (10). The cells were washed once with PBS containing 0.04% EDTA and incubated with 2 ml of dissociation solution (Sigma) to remove the cells from the plate. The cells (106) 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-PECAM-1 (BD Pharmingen), anti-VE-cadherin (Alexis Biochemical, San Diego, CA), B4-lectin (Sigma), anti-calbindin (Cell Signaling Technology, Danvers, MA), anti-aquaporin-2 (Cell Signaling Technology), DBA (Vector), anti-β1 (Millipore), anti-α5 (MAB1949; Millipore), anti-αv (01521 D; BD Pharmingen), anti-α1 (BD Pharmingen), anti-β3 (MAB1957; Millipore), anti-αvβ3 (MAB1976Z; Millipore), or control IgG (Millipore). For antibodies that required cell permeabilization, cells were removed from the dish, washed with PBS, fixed with 2% paraformaldehyde on ice for 30 min, washed with PBS, and resuspended in PBS containing 0.1% Triton X-100 and 0.1% BSA containing the appropriate dilution of primary antibody. The cells were washed with TBS containing 1% BSA and then incubated with the appropriate secondary antibody (1:200) on ice for 30 min. After 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).
Transwell filters (Costar 3422) were coated with 1% gelatin (EC) or 200 μg/ml of Matrigel (CD), rinsed with PBS, and then blocked with 2% BSA in PBS. Five hundred microliters of serum-free DMEM:F12 medium was added to the bottom of each well, and 1 × 105 cells in 100 μl of medium was added to the top of each well. Each condition was done in duplicate. Following 4 (endothelial cells) or 16 h (epithelial cells) in a 33°C tissue culture incubator, the cells and medium were aspirated and the upper side of the membrane was wiped with a cotton swab. The cells that had migrated through the membrane were fixed with 2% paraformaldehyde and stained with hematoxylin and eosin. Ten fields of cells were counted for each condition, and the average and SD were determined.
VEGF protein levels were determined from conditioned medium prepared from kidney endothelial or epithelial_cells, utilizing a mouse VEGF Immunoassay kit (R&D Systems). Briefly, kidney endothelial cells were grown for 2 days in serum-free medium at 37°C. The conditioned medium (50 μl) was used in the VEGF immunoassay, which was performed in triplicate as recommended by the manufacturer and was normalized to the number of cells. The assay was repeated twice using two different isolations of endothelial cells with similar results.
Nitric oxide analysis.
Kidney endothelial cells were plated in black-wall clear bottom Microtest TM 96-well plates (35 3948; 5 × 103 cells in 100 μl; BD). The next morning, the medium was changed to EC medium containing 30 μM DAF-FM diacetate (D-23842; Invitrogen) and 5 μg/ml of Cell trackerRed (C34552; Invitrogen). Following a 40-min incubation at 33°C, fresh EC medium was placed on the cells and the incubation continued for 20 min. The wells were washed with TBS, and the cells in each well were resuspended in 100 μl of TBS. The absorbance was read at 495/515 nm using a fluorescence plate reader (7). These experiments were performed in triplicate and repeated twice with similar results.
RNA purification and qPCR.
The total RNA from cells was extracted by a mirVana PARIS kit (Ambion) according to the manufacturer's instructions. cDNA synthesis was performed from 1 μg of total RNA using a Sprint RT Complete-Double PrePrimed kit from (Clontech). One microliter of each cDNA (dilution 1:10) was used as a template in qPCR assays, performed in triplicate of three biological replicates on Mastercycler Realplex (Eppendorf) using SYBR qPCR Premix (Clontech). Amplification parameters were as follows: 95°C for 2 min; 40 cycles of amplification (95°C for 15 s, 60°C for 40 s); dissociation curve step (95°C for 15 s, 60°C for 15 s, 95°C for 15 s).
Standard curves were generated from known quantities for each target gene of linearized plasmid DNA. Ten times dilution series were used for each known target, which were amplified using SYBR-Green qPCR. The linear regression line for nanograms of DNA was determined from relative fluorescent units (RFU) at a threshold fluorescence value (Ct) to quantify gene targets from cell extracts by comparing the RFU at the Ct to the standard curve, normalized by the simultaneous amplification of RpL13A, which was used as a housekeeping gene to normalize all samples. The primer sequences are listed in Table 1.
Processing of kidneys for histological studies and immunochemistry.
Following surgical removal from mice, kidneys from P20 wild-type and bim −/− mice were fixed with formalin overnight and processed for paraffin sectioning. For immunohistochemical staining, paraffin sections were deparaffinized with xylene and rehydrated. Antigen unmasking was performed using antigen-unmasking solution (Vector Laboratories, Burlingame, CA) according to the manufacturer's instructions. The sections were then washed in PBS and incubated for 15 min in PBS blocking buffer (PBS containing 1% bovine serum albumin, 0.3% Triton X-100, and 0.2% skim milk powder). The sections were incubated overnight with anti-PECAM-1 (1:150; R&D Systems). The sections were then incubated with indocarbocyanine (CY3)-labeled secondary antibody (Jackson ImmunoResearch, West Grove, PA) and photographed. Vascular density was determined by counting the number of capillaries and tubules on at least 10 high-magnification fields (×400). The data are represented as the number of capillaries per tubule.
Statistical differences between groups were evaluated with Student's unpaired t-test (two-tailed). Values are shown as means ± SD. P values <0.05 were considered significant.
Bim −/− kidney endothelial and epithelial cells show decreased apoptosis when challenged.
To determine the role BIM plays in kidney function, we isolated kidney epithelial and endothelial cells from wild-type and bim −/− mice as previously described (7, 10, 11, 20, 28). Both kidney endothelial and epithelial cells express significant levels of BIM with epithelial cells expressing the highest level (Fig. 1A). We next examined expression of epithelial and endothelial markers to ensure these cells retained normal characteristics. Wild-type and bim −/− kidney epithelial cells were DBA positive (mouse collecting duct-specific lectin) and expressed modest levels of aquaporin-2 and calbindin (Fig. 1B), while endothelial cells expressed VE-cadherin and PECAM-1 and were positive for B4-lectin (a mouse microvascular EC-specific lectin; Fig. 1C) as we previously described (10, 11, 20, 28).
The morphology of wild-type and bim −/− kidney epithelial cells were similar when plated on Matrigel-coated plates (Fig. 1D). The morphology of wild-type and bim −/− kidney endothelial cells also had a similar appearance when plated on gelatin-coated plates (Fig. 1E). Since BIM is a proapoptotic bcl-2 family member, we next addressed whether loss of BIM expression affected the level of apoptosis. Bim −/− kidney epithelial (Fig. 2A) and endothelial cells incubated with 5-FU for 48 h (Fig. 2B) demonstrated decreased numbers of apoptotic cells compared with wild-type cells. There were no significant differences in the basal rates of apoptosis in the absence of 5-FU.
Sustained osteopontin and TSP1 expression in bim −/− epithelial cells.
Altered ECM expression can influence cell adhesion and migration. We next examined whether lack of BIM differentially impacts ECM production in kidney epithelial and endothelial cells. Serum-free conditioned medium was prepared and evaluated by Western blot analysis. We observed sustained expression of TSP1 and osteopontin in bim −/− epithelial cells, while collagen IV and fibronectin expression was similar (Fig. 3A). In contrast, TSP1 and osteopontin expression was downregulated in bim −/− endothelial cells compared with wild-type, while fibronectin and tenascin C expression was similar (Fig. 3B).
Loss of bim expression differentially impacts migration of kidney epithelial and endothelial cells.
Cell migratory and adhesive properties impact the ability of epithelial and endothelial cells to form branched structures. We next examined cell migration characteristics using a scratch-wound assay. A confluent monolayer of wild-type and bim −/− kidney epithelial and endothelial cells were wounded and returned to 37°C in the presence of 5-FU (100 ng/ml) to prevent cell proliferation. Bim −/− epithelial cells migrated more slowly than their wild-type counterparts (Fig. 4A). In contrast, bim −/− endothelial cells migrated more rapidly than their wild-type counterparts, completely covering the wounded area (Fig. 4B). The quantitative assessment of the data is shown in Fig. 4, C and D. Similar results were obtained using a Transwell migration assay. Bim −/− epithelial cells demonstrated a 2.5-fold decrease in the number of cells that migrated through the membrane compared with their wild-type counterparts (Fig. 4E). In contrast, we observed a twofold increase in the number of bim −/− endothelial cells that migrated through the membrane compared with their wild-type counterparts (Fig. 4F). Thus loss of BIM expression differentially impacts migration of kidney epithelial and endothelial cells.
Bim −/− epithelial cells are more adherent.
Changes in migration of bim −/− kidney epithelial and endothelial cells could be due to altered cell adhesion. We next examined the ability of wild-type and bim −/− cells to adhere to various ECM proteins including fibronectin, laminin, collagen I, collagen IV, and vitronectin (Fig. 5, A and B). Bim −/− epithelial cells displayed increased adhesion to fibronectin, vitronectin, and collagen IV compared with wild-type cells, while wild-type and bim −/− endothelial cells adhered similarly well to fibronectin, vitronectin, and collagen IV. Minimal collagen I or laminin adhesion was observed for wild-type or bim −/− cells. Thus lack of BIM expression differentially influenced adhesion of kidney epithelial and endothelial cells.
We next analyzed integrin expression on the surface of kidney epithelial and endothelial cells by FACScan analysis (Fig. 6, A and B). Wild-type and bim −/− epithelial cells expressed similar levels of α2-, α6- β1-, and αvβ3-integrins. Wild-type and bim −/− endothelial cells expressed similar levels of α4-, α5-, αvβ3-, β1-, and β3-integrins on their surface. Thus the increased adhesion noted in bim −/− epithelial cells may be independent of significant changes in the expression levels of integrins and may be due to alterations in the affinity and/or avidity of these integrins.
Wild-type and bim −/− cells undergo tubular morphogenesis in Matrigel.
We next determined whether the changes observed in cell migration and adhesion impacted the ability of kidney epithelial and endothelial cells to undergo tubular morphogenesis, in the absence of BIM (Fig. 7, A and B). Wild-type and bim −/− kidney epithelial and endothelial cells plated on Matrigel formed an extensive network within 24 h. Longer incubation of the cells did not result in further branch formation. The quantitative assessment of the data shown in Fig. 7, C and D, demonstrated that lack of bim expression in endothelial cells resulted in a twofold increase in branch points. Similar numbers of branch points were noted in wild-type and bim −/− epithelial cells.
Decreased p-eNOS expression in bim −/− kidney endothelial cells.
VEGF promotes angiogenesis through activation of Akt1 and eNOS (1, 4, 9). Here, we examined expression and phosphorylation of eNOS in kidney endothelial cells, as well as its associated protein, HSP90. HSP90 expression was similar in wild-type and bim −/− kidney endothelial cells. However, bim −/− kidney endothelial cells demonstrated a significant decrease in levels of p-eNOS expression compared with wild-type cells (Fig. 8A). Total eNOS levels were similar in wild-type and bim −/− endothelial cells. Consistent with the decreased p-eNOS expression, NO production decreased approximately twofold in bim −/− kidney endothelial cells (Fig. 8B). VEGF expression increased fivefold in bim −/− kidney endothelial cells compared with wild-type kidney endothelial cells (Fig. 8C). VEGF expression also increased in bim −/− kidney epithelial cells compared with their wild-type counterparts (Fig. 8D). We next examined the expression of Akt1 and phosphorylated Akt1 in lysates from wild-type and bim −/− endothelial cells by Western blot analysis (Fig. 8A). Expression of Akt1 and phospho-Akt1 was similar among the endothelial cells examined here. Thus enhanced migration of bim −/− kidney endothelial cells may be mediated by increased production of VEGF, concomitant with decreased eNOS activity and NO bioavailability.
Increased kidney vascular density in the absence of bim.
To determine whether increased capillary morphogenesis in vitro correlated with increased renal vascular density, we immunostained kidney sections from P28 wild-type and bim −/− mice with anti-PECAM-1. Wild-type and bim −/− mice demonstrated significant PECAM-1 staining (Fig. 9). However, calculation of the ratio of peritubular capillaries per tubule demonstrated an increased number of peritubular capillaries in kidneys from bim −/− mice compared with their wild-type counterparts (Fig. 9). Thus enhanced capillary morphogenesis of bim −/− kidney endothelial cells correlated with increased numbers of capillaries in kidneys from these mice.
The kidney is a highly vascularized organ in which the various segments of the vascular tree accomplish specialized regional functions, potentially influencing renal epithelial cells in the surrounding area (5). Here, we examined whether loss of bim expression had a similar impact on kidney endothelial and epithelial cell function. Bim −/− kidney endothelial cells demonstrated increased migration and capillary morphogenesis. This proangiogenic phenotype of bim −/− kidney endothelial cells was associated with a fivefold increase in VEGF expression and downregulation of TSP1. Phosphorylation of eNOS and increased NO production can mediate the proangiogenic activity of VEGF. However, bim −/− kidney endothelial cells exhibited decreased p-eNOS and NO production, indicating that the proangiogenic activity of VEGF was not mediated through eNOS in these cells and may be responsible for suppression of TSP1 expression (23). These results are consistent with our data in bim −/− lung endothelial cells, in which increased VEGF expression and a proangiogenic phenotype was independent of changes in eNOS/NO activity (7). TSP1 has recently been shown to inhibit NO-mediated angiogenesis through a cGMP-dependent and -independent manner (8). Thus in the absence of TSP1 angiogenesis may proceed without a need for excess NO.
Vascular homeostasis is maintained by a tightly balanced production of pro- and antiangiogenic factors. The rigid control of vascular homeostasis, however, is abrogated in many disease states, resulting in an array of vasculopathies. This may be accomplished by upregulation of proangiogenic and/or downregulation of angiostatic factors. TSP1 is an endogenous inhibitor of angiogenesis (6) which inhibits angiogenesis in vivo and migration of capillary endothelial cells in vitro. The molecular mechanisms by which TSP1 and/or its angiostatic peptides inhibit vascularization in vivo or endothelial cell migration in vitro are evolving. Mutant mice which lack TSP1 exhibit a significant increase in blood vessel density in many organs (12, 13), thus providing further evidence for the important role of this molecule in the regulation of angiogenesis in vivo.
TSP1 is synthesized and secreted by a number of renal cell types, including glomerular mesangial cells and renal tubule cells (15, 16, 27). TSP1 activates latent transforming growth factor (TGF)-β and is thought to play a role in the pathogenesis of diabetic tubular hypertrophy, glomerular expansion, and fibrotic renal disease. Peptide antagonists of TSP1-mediated TGF-β activation block glucose-induced activation of TGF-β and expression of fibronectin, type IV collagen, and osteopontin (15). Renal tubule hypertrophy and mesangial expansion are inhibited by interference with this pathway (2, 19, 29). Although increased TSP1 expression is associated with tubular hypertrophy and mesangial expansion, which has a negative impact on kidney function (15, 27), its expression in the endothelium is essential for maintenance of the differentiated state of the vasculature. Thus loss of TSP1 expression in the endothelium can be associated with activation of endothelial cells and vasculopathies. Although suppression of TSP1 expression may improve renal tubular hypertrophy and mesangial expansion, renal vascular rarefaction could be an unwanted side effect. Thus understanding how manipulation of a targeting molecule impacts its various cellular functions is essential for effective development of therapeutic regimens.
Osteopontin, like TSP1, is a matricellular protein that plays important roles in normal physiology and in pathological states such as fibrosis. Osteopontin mediates cell adhesion, migration, and survival of several cell types including endothelial, renal epithelial, smooth muscle, and inflammatory cells. Increased osteopontin expression in bim −/− kidney epithelial cells may contribute to increased adhesion to ECM proteins and decreased migration. In bim −/− kidney endothelial cells, decreased expression of osteopontin and TSP1 correlated with increased migration and capillary morphogenesis. Thus, in the absence of BIM, modulation of TSP1 and osteopontin expression occur in opposing fashion in kidney endothelial and epithelial cells, concurrent with opposing changes in migration. We had previously shown that decreased endothelial cell migration correlated with decreased capillary morphogenesis and vascular density (7, 11, 24). In contrast, decreased migration of bim −/− kidney epithelial cells did not affect tubular morphogenesis in Matrigel. Decreased migration of bim −/− epithelial cells correlated with increased TSP1 and osteopontin expression and cell adhesion. Thus our data suggest that modulation of ECM expression, migration, and adhesion differentially impacts endothelial and epithelial branching morphogenesis.
Kidney tubules have been proposed to be an angiogenic soup (26). Our data support this supposition with wild-type kidney epithelial cells expressing fivefold higher levels of VEGF than their endothelial cell counterparts. Interestingly, loss of BIM expression in kidney endothelial cells increased VEGF levels to that observed in wild-type kidney epithelial cells. Kidney epithelial cells in vitro also express antiangiogenic factors such as TSP1 and osteopontin in a developmentally regulated fashion (Sorenson CM, unpublished observations and Ref. 28). Thus it is tempting to speculate that epithelial cell dysfunction due to disease may influence epithelial cell secretion of growth factors and ECM proteins, further impacting kidney endothelial cell and vascular function.
Studies from our laboratories have consistently shown that when endothelial cell migration is decreased, capillary morphogenesis and vascular development are significantly compromised (7, 11, 24). Enhanced endothelial cell migration does not negatively impact capillary morphogenesis but instead leads to enhanced vascular density (7, 22, 25). Here, we also show increased migration, capillary morphogenesis, and vascular density in the absence of BIM. Unlike endothelial cells, decreased cell migration of bim −/− epithelial cells did not adversely impact tubular morphogenesis in Matrigel. Previously, studies from our laboratories demonstrated that enhanced kidney epithelial migration led to an inability to undergo tubular morphogenesis in Matrigel and branching morphogenesis in vivo (20). The impact of the modulation of migration has on capillary/tubular morphogenesis in Matrigel appears to be regulated differently in endothelial and epithelial cells. In summary, the studies presented here emphasize the importance of considering all cell types within an organ in designing treatment modalities with minimal off-target effects.
This work was supported by grants from the University of Wisconsin Department of Pediatrics Research and Development Fund and University of Wisconsin Medical School Research Committee (C. M. Sorenson). C. M. Sorenson was funded, in part, by National Institutes of Health (NIH) Grant DK067120 and American Heart Association Research Award 0950057G. N. Sheibani is supported by NIH grants EY016995, EY018179, and RC4EY021357, P30 CA014520 UW Paul P. Carbone Cancer Center support grant, P30 EY016665, and an unrestricted departmental award from Research to Prevent Blindness. N. Sheibani is the recipient of a Research Award from the American Diabetes Association (1-10-BS-160) and the Retina Research Foundation. M. E. Morrison is a recipient of a Senior Thesis Grant from the College of Letters and Science at the University of Wisconsin-Madison. S. Y. Park is a recipient of a predoctoral studentship from AstraZeneca.
No conflicts of interest, financial or otherwise, are declared by the authors.
Author contributions: N.S. and C.M.S. provided conception and design of research; N.S., M.E.M., Z.G., S.P., and C.M.S. performed experiments; N.S., M.E.M., Z.G., S.P., and C.M.S. analyzed data; N.S. and C.M.S. interpreted results of experiments; N.S. and C.M.S. edited and revised manuscript; N.S. and C.M.S. approved final version of manuscript; M.E.M., Z.G., S.P., and C.M.S. prepared figures; C.M.S. drafted manuscript.
The authors thank Robert Gordon for assistance with graphics.
- Copyright © 2012 the American Physiological Society