Vascular Endothelial Growth Factor Induces Nephrogenesis and Vasculogenesis

Transcription

1 J Am Soc Nephrol 10: , 1999 Vascular Endothelial Growth Factor Induces Nephrogenesis and Vasculogenesis ALDA TUFRO,* VICTORIA F. NORWOOD,* ROBERT M. CAREY, and R. ARIEL GOMEZ* Departments of *Pediatrics and Internal Medicine, University of Virginia School of Medicine, Charlottesville, Virginia. Abstract. The expression of vascular endothelial growth factor (VEGF) and its receptors Flt-1 and Flk-1 in the rat kidney was examined during ontogeny using Northern blot analysis and immunocytochemistry. In prevascular embryonic kidneys (embryonic day 14 [E14]), immunoreactive Flt-1 and Flk-1 were observed in isolated angioblasts, whereas VEGF was not detected. Angioblasts aligned forming cords before morphologically differentiating into endothelial cells. In late fetal kidneys (E19), immunoreactive VEGF was detected in glomerular epithelial and tubular cells, whereas Flt-1 and Flk-1 were expressed in contiguous endothelial cells. To determine whether VEGF induces endothelial cell differentiation and vascular development in the kidney, the effect of recombinant human VEGF (5 ng/ml) was examined on rat metanephric organ culture, a model known to recapitulate nephrogenesis in the absence of vessels. After 6 d in culture in serum-free, defined media, metanephric kidney growth and morphology were assessed. DNA content was higher in VEGF-treated explants ( g/kidney, n 9) than in paired control explants ( g/kidney, n 9) (P 0.05). VEGF induced proliferation of tubular epithelial cells, as indicated by an increased number of tubules and tubular proliferating cell nuclear antigen-containing cells. VEGF induced upregulation of Flk-1 and Flt-1 expression, as assessed by Western blot analysis. Developing endothelial cells were identified and localized using immunocytochemistry and electron microscopy. Flt-1, Flk-1, and angiotensin-converting enzyme-containing cells were detected in VEGF-treated explants, whereas control explants were negative. These studies confirmed previous reports indicating that the expression of VEGF and its receptors is temporally and spatially associated with kidney vascularization and identified angioblasts expressing Flt-1 and Flk-1 in prevascular embryonic kidneys. The data indicate that VEGF expression is downregulated in standard culture conditions and that VEGF stimulates growth of embryonic kidney explants by expanding both endothelium and epithelium, resulting in vasculogenesis and enhanced tubulogenesis. These data suggest that VEGF plays a critical role in renal development by promoting endothelial cell differentiation, capillary formation, and proliferation of tubular epithelia. Metanephric kidneys develop as a result of reciprocal inductive interactions between the metanephric blastema and the ureteric bud (1,2). Renal blood vessels develop simultaneously with nephron epithelial differentiation by angiogenesis, i.e., sprouting from preexisting vessels, and/or by vasculogenesis, i.e., differentiation in situ from angioblasts (1,3,4). Glomerular endothelial cells could differentiate in situ within the vascular cleft of the S-shaped body (4). Alternatively, glomerular endothelial cells may sprout from preexisting vessels in the surrounding intrarenal or extrarenal mesenchyme and migrate into the forming nephron (1,5,6). Experimental data support the hypothesis that renal vascularization occurs by both vasculogenic and angiogenic mechanisms and that renal endothelial cells derive from the blastema or from extrarenal sources depending on the environment (7 9). Received December 16, Accepted April 12, Correspondence to Dr. Alda Tufro, Department of Pediatrics/Nephrology, University of Virginia, MR4 Bldg./Room 2017, Charlottesville, VA Phone: ; Fax: ; at4w@virginia.edu / Journal of the American Society of Nephrology Copyright 1999 by the American Society of Nephrology The molecular basis of kidney vascularization is largely unknown. A role for soluble angiogenic factors has been postulated (1,10). Vascular endothelial growth factor (VEGF) may play such a role because it is a direct-acting specific endothelial cell mitogen that stimulates angiogenesis and regulates embryonic vessel development in a gene dosage-dependent manner (11 13). VEGF tyrosine kinase receptors Flk-1 and Flt-1 are expressed in endothelial cells and angioblasts, their mesodermic precursor cells (14 16). VEGF receptors are required for the normal development of the vasculature, as demonstrated by the failure of vasculogenesis and organization of the embryonic vasculature in Flk-1- and Flt-1-deficient mice, respectively (17,18). Spatial and temporal expression of VEGF, Flt-1, and Flk-1 suggest that VEGF may be a physiologic stimulus for endothelial cell differentiation and proliferation leading to the development of capillaries (16,19,20). In fetal human kidneys, VEGF localizes to epithelial cells in S-shaped bodies and collecting ducts (21). In adult kidneys, VEGF is expressed in glomerular visceral epithelial cells and in various tubular cells (21). Flt-1 and Flk-1 are expressed in glomerular endothelial cells in developing and adult kidneys (16,19,21), suggesting a role for the system in renal vascular development and the maintenance of vascular phenotype. Because disruption of

2 2126 Journal of the American Society of Nephrology J Am Soc Nephrol 10: , 1999 VEGF, Flk-1, and Flt-1 genes is lethal before metanephric development (12,13,17,18), the role of VEGF and its receptors in renal morphogenesis is unknown. In vitro organ culture of embryonic kidneys recapitulates nephrogenesis in vivo (1). After 5 to 7 d in culture, S-shaped bodies and avascular glomeruli develop within kidney explants (5,6). Hence, metanephric kidney culture provides an excellent model to induce vasculogenesis/angiogenesis in vitro. The absence of vessels in embryonic kidney cultures has been attributed to the putative extrametanephric origin of glomerular vessels (1). We hypothesized that standard metanephric organ culture conditions do not support expression of angiogenic factors required for endothelial cell differentiation and proliferation. We demonstrated that low oxygen stimulates vasculogenesis in the metanephric organ culture model by upregulating VEGF (8). In the present study, we examined the expression of VEGF and its receptors during ontogeny in the rat and we determined that exogenous recombinant human VEGF (rhvegf) induces endothelial cell differentiation and capillary formation as well as proliferation of tubular epithelia in the metanephric organ culture. Materials and Methods Tissue Isolation Rats were mated for 6 h, and the following day was considered day 1 for staging of the embryos. Metanephric kidneys were microdissected at day 14 of gestation (E14, n 44), day 15 of gestation (E15, n 6), day 17 of gestation (E17, n 8), day 19 of gestation (E19, n 24), day 20 of gestation (E20, n 24), day 1 of postnatal life (NB1, n 12), and 60 d of life (A, n 6). Kidneys were frozen ( 70 C) for protein and RNA extraction or fixed for morphologic studies as detailed below. Northern Analysis Total RNA was extracted from E14, NB1, and A kidneys by standard technique (22). Ten micrograms of total RNA samples was resolved in 1.2% agarose/formaldehyde gels and transferred to nylon membranes (Zetaprobe, Bio-Rad, Richmond, CA). Northern blots were hybridized with VEGF cdna fragment (204 bp), Flt-1 cdna (510 bp), and Flk-1 cdna (230 bp) 32 P-labeled by PCR. cdna were amplified by reverse transcription-pcr from rat kidney total RNA using the following primers: VEGF: 5 -CGCGGATCCAGGAGTAC- CCTGATGAG-3 and 5 -CCGGAATTCACATTTGTTGTGCTGT- 3 ; Flt-1: 5 -TGTGGAGAAACTTGGTGACCT-3 and 5 -TGGAGAA- CAGCAGGACTCCTT-3 ; Flk-1: 5 -CGTGGATCCACCAAAGGGG- CACGATTC CGTC-3 and 5 -CTCGAATTCTGTAACAGATGAGAT- GCTCCAAGG-3. They were sequenced at the University of Virginia sequencing facility to verify their identity to rat VEGF, Flt-1, and Flk-1 cdna (23 25). To verify equal loading, Northern blots were also hybridized to GAPDH cdna 32 P-labeled by random priming (26). Blots were exposed 2 to 4 d to XAR film at 70 C for autoradiography. Hybridization signals were quantified by densitometry (LBK Ultrascan XL laser densitometer, Bromma, Sweden). Metanephric Kidney Organ Culture Metanephric kidney organ culture was performed as described (8). Intact metanephroi were microdissected from Sprague Dawley rat embryos at 14 d gestation, and were cultured in defined, serum-free media (Dulbecco s modified Eagle s medium-f12) for 6 d at 37 C in a mix of 95% air/5% CO 2 in a 95% humidity incubator. Media was supplemented with Hepes 10 mm, insulin 5 g/ml, transferrin 5 g/ml, selenium 2.8 nm, prostaglandin E 1 25 ng/ml, T3 32 pg/ml, penicillin 50 U/ml, gentamycin 5 g/ml, and mycostatin 50 U/ml. Media was changed daily. Explants (n 242) were paired for control or VEGF treatment as follows: From each embryo, one kidney was assigned to control (media alone, n 121) and the other was assigned to VEGF (Sigma, St. Louis, MO) (media rhvegf 5 ng/ml, n 121). rhvegf optimal concentration was determined in preliminary studies assessing explant growth as a function of VEGF dose (data not shown). Ten experiments were performed using 20 L. In each experiment, four to six explants were used for DNA determinations and six to eight were used for histologic studies. Western Blot Analysis E20 rat kidneys, control, and VEGF-treated explants after 6 d in culture were homogenized in lysis buffer (1% Nonidet-P40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 100 g/ml phenylmethylsulfonyl fluoride, 40 g/ml aprotinin, 12.5 g/ml leupeptin in phosphate-buffered saline). Protein samples (50 to 80 g) from four experiments were electrophoresed in 10 to 15% sodium dodecyl sulfate-polyacrylamide gels and transferred to nitrocellulose membranes (Hybond ECL; Amersham, Buckinghamshire, United Kingdom) by electroblotting. Western blots were blocked with 5% nonfat dry milk and 0.5% Tween 20 in phosphate-buffered saline and exposed to 1 g/ml anti-flk-1, 2 g/ml anti-flt-1, or 2 g/ml anti- VEGF polyclonal antibodies (SC# 315, SC#316, and SC#507; Santa Cruz Biotechnology, Santa Cruz, CA), followed by horseradish peroxidase-conjugated anti-rabbit IgG. Horseradish peroxidase activity was detected by chemiluminescence using the enhanced chemiluminescence system (ECL; Amersham) following the manufacturer s protocol. Protein expression was quantified by densitometric analysis using ImageQuant (Molecular Dynamics, Sunnyvale, CA). DNA Measurements DNA measurements were performed using a fluorometric method as described by Labarca and Paigen (27). Nine pools of three to four explants from each experimental group were assessed in duplicate. Histologic Studies Kidneys and explants were fixed in Bouin s fixative, paraffinembedded, stained with hematoxylin-eosin or processed for immunocytochemistry, and examined by light microscopy. For electron microscopy, explants (n 8) were fixed in 2% paraformaldehyde 2.5% glutaraldehyde, rinsed in 50 mm sodium cacodylate buffer, post-fixed in 1% osmium tetroxide, and stained en bloc with 3% uranyl acetate in veronal acetate buffer. Then explants were dehydrated and embedded in PolyBed 812 resin (Polysciences, Warrington, PA) at 60 C for 2 d. Ultrathin sections were cut on an LKB Ultratome V (LKB Instruments, Gaithersburg, MD), counterstained with saturated uranyl acetate followed by lead citrate, examined, and photographed on a Zeiss EM 10CA transmission electron microscope at 60 kv. Immunocytochemistry VEGF, proliferating cell nuclear antigen (PCNA), and the following endothelial cell markers were localized using specific antibodies: Flt-1, Flk-1, angiotensin-converting enzyme (ACE) (14,24,28,29). Sections were deparaffinized, microwaved in 10 mm sodium citrate, and incubated with anti-vegf (1:20) (Oncogene, Cambridge, MA), anti-flt-1 (1:250), or anti-flk-1 (1:100) antibody (Santa Cruz Bio-

3 J Am Soc Nephrol 10: , 1999 VEGF Induces Nephrogenesis and Vasculogenesis 2127 technology), or anti-pcna (1:150) (Novocastra, Newcastle upon Tyne, United Kingdom) for 60 min at room temperature. Secondary antibody was biotinylated anti-rabbit IgG or anti-mouse IgG, as appropriate. Reactions were detected by immunoperoxidase using the ABC technique (30,31). Negative controls were absence of primary or secondary antibodies and competition with specific peptide. Sections were counterstained with hematoxylin and examined by light microscopy. ACE was localized in whole mount explants. To provide spatial orientation for labeled endothelial cells, peanut lectin labeling of glomeruli was performed as well (32). Cultured explants were fixed in 4% paraformaldehyde, incubated with neuraminidase 0.1 U/ml for 60 min at 37 C, washed, and incubated with rhodamine-labeled peanut lectin (Vector, Burlingame, CA) (7.5 g/ml) and anti-ace polyclonal antibody (1:200) for 90 min at 37 C. Explants were washed, incubated with anti-goat fluorescein-labeled secondary antibody for 60 min, mounted, and examined using a confocal microscope (Leitz, Glenwood, NJ). Morphometric Analysis A total of 16 explants from five separate experiments was examined using a video microscope system (Olympus AH-2; Sony CCD- Iris, San Jose, CA). Two nonconsecutive sections per explant were digitized. Tubular cross sections, ureteric bud, and total areas were measured and counted using image analysis software (Mocha, Jaendel, Inc., San Rafael, CA). Statistical Analyses DNA content was expressed as mean SEM. Groups were compared using unpaired t test because several explants were pooled for DNA measurement. Densitometric data from Western blot analysis of four or more independent experiments were compared by paired t test (control versus VEGF-treated) or unpaired t test (control versus E20 embryonic kidneys) and were expressed as mean SEM foldchanges in protein expression levels. Statistical significance was defined as P In ontogeny studies, three nonconsecutive sections from six E14, six E15, six E17, six E19, four NB1, and three adult kidneys were examined for each immunocytochemical study. In organ culture studies, three nonconsecutive sections from nine to 12 explants per experimental group were examined in each immunohistochemical study. Results of the morphometric analysis of tubular, ureteric bud, and mesenchymal areas were expressed as mean SEM. Control and VEGF groups were compared using unpaired t test. Results Expression of VEGF, Flt-1, and Flk-1 during Ontogeny VEGF mrna is expressed in prevascular kidneys (E14), and steady-state mrna levels increased 70% during vascularization (NB1) and persisted until adulthood (Figure 1A). VEGF peptides were not detectable by immunocytochemistry in E14 kidneys (Figure 2A). Immunoreactive VEGF was observed in E19 kidneys localized to visceral and parietal glomerular epithelial cells, glomerular capillaries, and developing tubular cells (Figure 2B). In newborn and adult kidneys, VEGF immunostaining was observed in glomerular epithelial cells, endothelial cells lining vessels, and tubular epithelia (Figure 2C). Flt-1 and Flk-1 mrna were detected by Northern blot in E14 as well as in newborn and adult kidneys (Figure 1). Flt-1 mrna increased threefold during development, whereas Flk-1 was slightly downregulated. Immunoreactive Flt-1 and Flk-1 Figure 1. Representative Northern blots showing kidney vascular endothelial growth factor (VEGF), Flt-1, and Flk-1 mrna steadystate levels during ontogeny. Lane 1, embryonic day 14 (E14); lane 2, 1-d-old newborn; lane 3, 60-d old rat. were observed in large, isolated cells within the avascular metanephric blastema in E14 (Figure 3, A and B). Flk-1 immunostaining localized to the cell membrane and the nucleus of these cells. Flk-1-containing cells lined up in a cordlike manner in E15 (Figure 4A), before acquiring the typical endothelial cell shape. As vascularization proceeded, immunoreactive Flt-1 and Flk-1 were observed in glomerular and vascular endothelial cells (Figure 4, B and C). In addition, Flt-1 immunostaining was observed in developing tubular cells (Figure 5). Effect of VEGF on Metanephric Organ Culture rhvegf induced metanephric organ culture growth by cell proliferation, as indicated by a higher DNA content in VEGFtreated than in paired control explants (Figure 6) and by a clear increase in PCNA-containing cells in VEGF-treated explants (Figure 7). A nonrelevant protein (bovine serum albumin) at identical concentration did not alter explant growth (data not shown). Flk-1 and Flt-1 protein expression, assessed by Western blot analysis, was fold and fold higher, respectively, in VEGF-treated explants than in controls (Figure 8), indicating that rhvegf induced upregulation of both receptors. VEGF protein expression was fold lower in control explants than in E20 kidneys, indicating that VEGF downregulates in control culture conditions (Figure 9). Three VEGF isoforms (approximately 26, 21, and 19 kd) were detected in E20 and control explants, and the downregulation was similar for all isoforms. Light microscopy revealed that explant morphology was altered by rhvegf. Distinctive features of VEGF-treated explants included the expansion of tubular epithelia and the presence of endothelial cells (Figure 10, A and B). Morphometric analysis showed that the number of tubular cross sections was higher in VEGF-treated explants (80 16) than in control explants (56 7; P 0.05), whereas the tubular cross section area was similar. Endothelial cells containing immunoreactive Flt-1 and Flk-1 were detected in VEGF-treated explants. Endothelial cells were consistently observed in the vascular cleft of S-shaped bodies, and surrounding tubules and

4 2128 Journal of the American Society of Nephrology J Am Soc Nephrol 10: , 1999 Figure 3. E14 kidney sections showing isolated angioblasts expressing Flk-1 (A) and Flt-1 (B) in the metanephric blastema. ureteric bud branches (Figure 11, A and B). This was confirmed by the detection of ACE-expressing cells next to and within developing glomeruli by fluorescence immunolabeling (Figure 11C). Control explants contained occasional Flt-1- and Flk-1-expressing cells, and no ACE- or VEGF-containing cells were detected (Figure 11, D through F). Developing endothelial cells forming capillary lumina were identified by electron microscopy within VEGF-treated explants (Figure 12), whereas they were absent from control explants. Figure 2. Immunocytochemical localization of VEGF. (A) Absence of immunoreactive VEGF in E14 metanephric mesenchyme. (B) Immunoreactive VEGF in glomerular epithelial cells (arrows) and endothelial cells (arrowheads) in an E19 kidney section. (C) VEGF immunostaining in tubules (arrowheads) and in parietal and visceral glomerular epithelial cells (arrows) in a 1-d-old newborn kidney section. Discussion We report that the avascular metanephric blastema contains angioblasts expressing Flk-1 and Flt-1, suggesting that vasculogenesis is involved in renal vascular development. Contiguous expression of VEGF and its receptors occurs during the development of the kidney vasculature in the rat. In the metanephric organ culture, VEGF expression is downregulated. Addition of rhvegf induces the differentiation and proliferation of angioblasts into endothelial cells. The phenotype of these developing endothelial cells is defined by the expression of VEGF receptors and ACE. Identification of capillary lumina confirms their endothelial nature. Thus, our results demonstrate that VEGF induces vasculogenesis in the metanephric organ culture. We showed that Flt-1 and Flk-1 are expressed in isolated cells before any morphologic evidence of vascular development in the metanephric blastema, indicating the presence of angioblasts in situ. Flk-1 immunostaining localized to the cell membrane and the nucleus of angioblasts. The basis for nuclear

5 J Am Soc Nephrol 10: , 1999 VEGF Induces Nephrogenesis and Vasculogenesis 2129 Figure 5. E17 kidney section showing that immunoreactive Flt-1 localizes to developing tubules in addition to endothelial cells. Figure 4. Immunolocalization of VEGF receptors during metanephric vascularization. (A) E15 kidney section showing alignment of Flk-1- containing cells forming a cord-like structure. (B) E17 kidney section illustrating Flt-1 expression in endothelial cells within an S-shaped body, a larger vessel, and developing tubules. (C) E19 kidney section illustrating Flk-1 expression in endothelial cells within the developing kidney. Figure 6. DNA content ( g/kidney) from control (o, n 7) and VEGF-treated (f, n 7) explants after 6dinculture. staining is unclear, but it is unlikely due to nonspecific peroxidase staining since it is not present in negative controls. Within 24 h, Flk-1-containing cells align to form cord-like structures, then they acquire the typical endothelial cell phenotype in the following 2 d of gestation. As renal vascularization proceeds, VEGF is detected in glomerular epithelial cells and developing tubules; meanwhile, Flt-1 and Flk-1 are ex-

6 2130 Journal of the American Society of Nephrology J Am Soc Nephrol 10: , 1999 Figure 7. VEGF-induced explant cell proliferation detected by proliferating cell nuclear antigen immunostaining (brown). (A) Control explant. (B) VEGF-treated explant. Magnification, 80. Figure 8. Western blot analysis of Flk-1 (left panel) and Flt-1 (right panel) expression in VEGF-treated and control explants. M r, 180 to 200 kd and 160 to 180 kd, respectively, correspond to those reported (34,46). Recombinant human VEGF (rhvegf) induces upregulation of Flk-1 and Flt-1 expression in the explants. pressed in contiguous endothelial cells as reported previously in other species (16,21,33). We also observed Flt-1-containing cells in developing tubules of E17 and E19 kidneys. We cannot exclude that the anti-flt-1 antibody may detect, in addition to Flt-1, an epitope from another protein localized to tubular epithelial cells. We report that the normal changes in VEGF, Flk-1, and Flt-1 expression occurring during renal morphogenesis do not take place in culture conditions. Immunoreactive VEGF, Flk-1, and Flt-1 were not detected by immunocytochemistry, and were barely detectable by Western blot analysis in control explants after 6 d inculture. Furthermore, VEGF protein expression was downregulated in control explants compared to kidneys of equivalent gestational age (E20). These results and previous Figure 9. Western blot analysis of VEGF expression in E20 kidneys and control explants. Eighty micrograms of protein was loaded in each lane. Three isoforms of approximately 26, 21, and 19 kd were detected in both lanes. All three isoforms were downregulated approximately eightfold in control explants. data (8) demonstrate that VEGF expression is very low in control conditions, and suggest that VEGF receptors downregulate in the absence of their ligand during kidney development, as has been described in the developing brain (15,34). Flk-1 and Flt-1 downregulation was associated morphologically with a lack of endothelial cell differentiation. Since expression of Flk-1 and Flt-1 is required for normal endothelial cell differentiation and assembly (17,18), VEGF receptor downregulation may prevent vascularization of metanephric organ culture in control conditions. A threshold VEGF concentration may be necessary to sustain endothelial cell differentiation and proliferation, as indicated by the dependence of embryonic vessel development on the number of copies of the VEGF gene (12,13). This threshold may not be attained in control culture conditions. The mechanism(s) for altered VEGF expression in metanephric organ culture are unknown. It is possible that a lower, physiologic oxygen tension is required for normal VEGF expression in the developing kidney.

7 J Am Soc Nephrol 10: , 1999 VEGF Induces Nephrogenesis and Vasculogenesis 2131 Figure 10. Representative paired control (A) and VEGF-treated (B) explants. Note extensive tubulogenesis in VEGF-treated explant. Magnification, 33. In response to exogenous VEGF, upregulation of Flt-1 and Flk-1 expression occurred as indicated by Western blot data and by the presence of numerous Flt-1- and Flk-1-containing endothelial cells localized to the vascular cleft of S-shaped bodies, surrounding tubules, and ureteric bud branches. Endothelial cells also expressed ACE, a marker of developing and mature endothelial cells (35). Distinctive morphologic features of developing endothelial cells included formation of capillary lumina. Our results did not determine whether endothelial cells differentiated from mesenchymal cells. However, they clearly established that endothelial cells differentiate and proliferate in situ from angioblasts enclosed in the avascular metanephric blastema, and therefore that vasculogenesis takes place during metanephric development. We showed previously that upregulation of VEGF associated with explant exposure to low oxygen resulted in endothelial cell differentiation and proliferation (8). In the avian embryo, overexpression of VEGF induced Flk-1 upregulation, hypervascularization, and increased vascular permeability (33). In the newborn mice, anti-vegf-neutralizing antibodies impaired glomerular capillary development (36). VEGF may regulate the extent of vascularization by controlling its receptor expression in angioblasts and endothelial cells. It is possible that VEGF acts as a survival factor for angioblasts, thus recruited to differentiate into endothelial cells in addition to its endothelial cell mitogenic role (12,13,37). The origin of kidney vasculature is controversial. Classic organ culture studies and interspecies transplantation experiments supported the hypothesis of an extrarenal origin of glomerular endothelial cells and an angiogenic mechanism for renal vascularization (1,6,38 40). However, earlier studies and recent transplantation experiments using transgenic mice suggested that glomerular endothelial cells develop by vasculogenesis (4,7,9,41). We have shown that low oxygen stimulates endothelial cell differentiation and vasculogenesis in metanephric organ culture (8). The present data demonstrate the ability of VEGF to promote endothelial cell differentiation and formation of capillaries in the organ culture in a manner similar to that observed during normal development. Taken together, our data suggest that VEGF may induce vasculogenesis during normal renal development. Whether VEGF plays an additional role in the angiogenic process as a chemoattractant for endothelial cells remains to be established (16). Thus, vasculogenesis and angiogenesis may contribute to renal vascularization in a coordinated manner, as has been described in organs of endodermal origin (42). VEGF induced explant growth by cell proliferation. Because the vast majority of metanephric kidney cells are epithelial or mesenchymal, the increased DNA content in VEGF-treated explants suggests that the mitogenic effect of VEGF may not be limited to endothelial cells in the metanephric organ culture. VEGF induced a clear increase in tubular and endothelial cell proliferation as indicated by PCNA immunostaining. Morphometric analysis revealed that the total number of tubular cross sections was increased by 40% in VEGF-treated explants compared with controls, demonstrating an increase in the number of tubules, in their length, or both. Thus, VEGF induced tubulogenesis in the metanephric organ culture. These data are in agreement with our previous studies showing that low oxygen induced tubulogenesis associated with upregulation of VEGF expression in this model (8). The presence of immunoreactive Flt-1 in tubular cells of E17 embryonic kidneys is consistent with this nephron segment being a target for VEGF. Taken together, these data suggest that VEGF may be mitogenic for developing tubular cells. Flt-1 expression has not

8 2132 Journal of the American Society of Nephrology J Am Soc Nephrol 10: , 1999 Figure 11. (A) VEGF-treated explant showing immunoreactive Flt-1 localized to endothelial cells and tubular epithelial cells. (B) VEGF-treated explant showing immunoreactive Flk-1 localized to endothelial cells. (C) Whole mount VEGF-treated explant showing glomeruli labeled with peanut lectin (red) and FITC-labeled immunoreactive angiotensin-converting enzyme (ACE) (green and yellow) staining endothelial cells next to and within glomeruli. (D) Control explant showing no immunoreactive Flt-1. (E) Control explant showing no Flk-1 immunoreactivity. (F) Whole mount control explant showing a peanut lectin-labeled glomerulus (red) and no ACE immunostaining. Magnification: 80 in A, C, and D; 160 in B, E, and F. been reported previously in epithelial cells. However, Flt-1 has been localized to angioblasts, endothelial cells, uterine smooth muscle cells, spongiotrophoblast, extraembryonic and intraembryonic mesoderm, ectoplacental cone, leukemia, and teratocarcinoma cell lines (16,43 45). We observed immunoreactive Flt-1 in developing tubular epithelial cells in E17 kidneys and VEGF-treated explants. Although Flt- 1 immunostaining in VEGF-treated explants was not as distinct as in E17, it was clearly absent in control explants, suggesting that the immunostaining was not an artifact. Consistently, Western analysis showed a clear upregulation of Flk-1 and Flt-1 expression in VEGF-treated explants. Flt-1 expression in tubular cells undergoing rapid proliferation both in E17 and VEGF-treated explants suggests that tubular epithelial cells are a target for VEGF at least transiently, and that Flt-1 may mediate the VEGF tubulogenic effect. Alternatively, tubulogenesis may be

9 J Am Soc Nephrol 10: , 1999 VEGF Induces Nephrogenesis and Vasculogenesis 2133 Figure 12. VEGF-treated explant. Electron microscopy showing an endothelial cell forming a capillary. Magnification, an indirect effect of VEGF, mediated by other genes. Additional studies are currently under way to evaluate this novel VEGF effect. In summary, we observed a close temporal and spatial association of VEGF and its receptors expression with the development of the kidney vasculature in the rat. The presence of angioblasts expressing Flt-1 and Flk-1 in the prevascular metanephric blastema suggests that vasculogenesis contributes to renal vascularization. VEGF expression was downregulated in standard culture conditions compared with native embryonic kidneys. Addition of rhvegf to the explants induced differentiation and proliferation of endothelial cells as well as proliferation of tubular epithelial cells, resulting in vasculogenesis and tubulogenesis within metanephric kidneys in culture. These data provide evidence supporting a key role for VEGF and its receptors in renal development by promoting endothelial cell differentiation, capillary formation, and proliferation of tubular epithelia. Acknowledgments This study was supported by an American Heart Association (AHA) Virginia Affiliate Grant-in-Aid (VA-95-G11) and National Institutes of Health Grant K08 DK Dr. Norwood is supported by the Child Health Research Center (HL28810), AHA, Virginia affiliate (VA-95-G25), and the University of Virginia Children s Medical Center Research Fund. Dr. Gomez is supported by the Child Research Center (HL28810), the O Brien Center for Kidney and Urologic Research (DK-45179), and the Center of Excellence in Pediatric Nephrology and Urology (DK-44756). References 1. Saxen L: Organogenesis of the Kidney, Cambridge, Cambridge University Press, Herzlinger D: Renal stem cells and the lineage of the nephron. Annu Rev Physiol 56: , Coffin DJ, Harrison J, Schwartz S, Heimark R: Angioblast differentiation and morphogenesis of the vascular endothelium in the mouse embryo. Dev Biol 148: 51 62, Hyink DP, Tucker DC, St. John P: Endogenous origin of glomerular endothelial and mesangial cells in grafts of embryonic kidneys. Am J Physiol 270: F886 F889, Ekblom P: Renal development. In: The Kidney: Physiology and Pathophysiology, edited by Seldin DW, Giebisch G, New York, Raven, 1992, pp Pinson Hyink D, Abrahamson DR: Origin of the glomerular vasculature in the developing kidney. Semin Nephrol 15: , Robert B, St. John P, Hyink DP, Abrahamson DR: Evidence that embryonic kidney cells expressing Flk-1 are intrinsic, vasculogenic angioblasts. Am J Physiol 271: F744 F753, Tufro-McReddie A, Norwood VF, Aylor KW, Botkin SJ, Carey RM, Gomez RA: Oxygen regulates vascular endothelial growth factor-mediated vasculogenesis and tubulogenesis. Dev Biol 183: , Abrahamson DR, Robert B, Hyink DP, St. John PL, Daniel TO: Origins and formation of microvasculature in the developing kidney. Kidney Int 54: S7 S11, Kloth S, Ebenbeck C, Kubitza M, Schmidbauer A, Rockl W, Minuth WW: Stimulation of renal microvascular development under organotypic culture conditions. FASEB J 9: , Ferrara N, Houck K, Jakeman L, Leung DW: Molecular and biological properties of the vascular endothelial growth factor family of proteins. Endocrine Rev 13: 18 32, Carmeliet P, Ferreira V, Breier G: Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature 380: , Ferrara N, Carver-Moore K, Chen H: Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 380: , Yamaguchi TP, Dumont DJ, Conlon RA, Breitman ML, Rossant J: flk-1, an flt-related receptor tyrosine kinase, is an early marker for endothelial precursors. Development 118: , Risau W: Differentiation of endothelium. FASEB J 9: , Dumont D, Fong G, Puri M, Gradwohl G, Alitalo K, Breitman M: Vascularization of the mouse embryo: A study of flk-1, tek, tie, and vascular endothelial growth factor expression during development. Dev Dyn 203: 80 92, Shalaby F, Rossant J, Yamaguchi TP: Failure of blood-island formation and vasculogenesis in Flk-1-deficient mice. Nature 376: 62 66, Fong G-H, Rossant J, Gertsenstein M, Breitman ML: Role of the Flt-1 receptor tyrosine kinase in regulating the assembly of vascular endothelium. Nature 376: 66 70, Peters KG, de Vries C, Williams LT: Vascular endothelial growth factor receptor expression during embryogenesis and tissue repair suggests a role in endothelial differentiation and blood vessel growth. Proc Natl Acad Sci USA 90: , Jakeman LB, Armanini M, Phillips HS, Ferrara N: Developmental expression of binding sites and messenger ribonucleic acid vascular endothelial growth factor suggests a role for this protein in vasculogenesis and angiogenesis. Endocrinology 133: , Simon M, Grone HJ, Johren O: Expression of vascular endothelial growth factor and its receptors in human renal ontogenesis and adult kidney. Am J Physiol 268: F240 F250, Chomczynski P: A reagent for the single-step simultaneous isolation of RNA, DNA and proteins from cell and tissue samples. Biotechniques 15: 67 69, 1993

10 2134 Journal of the American Society of Nephrology J Am Soc Nephrol 10: , Conn G, Bayne ML, Soderman DD: Amino acid and cdna sequences of a vascular endothelial cell mitogen that is homologous to platelet-derived growth factor. Proc Natl Acad Sci USA 87: , Shibuya M, Yamaguchi S, Yamane A: Nucleotide sequence and expression of a novel human receptor-type tyrosine kinase gene (flt) closely related to the fms family. Oncogene 5: , Terman BI, Carrion ME, Kovacs E, Rasmussen BA, Eddy RL, Shows TB: Identification of a new endothelial cell growth factor receptor tyrosine kinase. Biochem Biophys Res Commun 187: , Feinberg A, Vogelstein B: A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132: 6 13, Labarca C, Paigen K: A simple, rapid and sensitive DNA assay procedure. Anal Biochem 102: , Caldwell PRB, Seegal BC, Hsu KC, Das M, Soffer RL: Angiotensin-converting enzyme: Vascular endothelial localization. Science 191: , Foley JF, Dietrich DR, Swenberg JA, Maronpot RR: Detection and evaluation of proliferating cell nuclear antigen (PCNA) in rat tissue by an improved immunohistochemical procedure. J Histotechnol 14: , Hsu SM, Raine L, Fanger H: Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: A comparison between ABC and unlabelled antibody (PAP) procedures. J Histochem Cytochem 29: , Tufro-McReddie A, Arrizurieta E, Brocca S, Gomez R: Dietary protein intake modulates the intrarenal distribution of renin and its mrna during development. Am J Physiol 263: F427 F435, Gilbert T, Gaonach S, Moreau E, Merlet-Benichou C: Defect of nephrogenesis induced by gentamicin in rat metanephric organ culture. Lab Invest 70: , Flamme I, von Reutern M, Drexler HCA, Syed-Ali S, Risau W: Overexpression of vascular endothelial growth factor in the avian embryo induces hypervascularization and increased vascular permeability without alterations of embryonic pattern formation. Dev Biol 171: , Millauer B, Wizigmann-Voos S, Schnurch H: High affinity VEGF binding and developmental expression suggest Flk-1 as a major regulator of vasculogenesis and angiogenesis. Cell 72: , Mounier F, Hinglais N, Sich M: Ontogenesis of angiotensin-i converting enzyme in human kidney. Kidney Int 32: , Kitamoto Y, Tokunaga H, Tomita K: Vascular endothelial growth factor is an essential molecule for mouse kidney development: Glomerulogenesis and nephrogenesis. J Clin Invest 99: , Alon T, Hemo I, Itin A, Pe er J, Stone J, Keshet E: Vascular endothelial growth factor acts as a survival factor for newly formed retinal vessels and has implications for retinopathy of prematurity. Nat Med 1: , Bernstein J, Cheng F, Roszka J: Glomerular differentiation in metanephric culture. Lab Invest 45: , Ekblom P, Sariola H, Karkinen-Jaaskelainen M: The origin of the glomerular endothelium. Cell Differ 11: 35 39, Sariola H, Ekblom P, Lehtonen E: Differentiation and vascularization of the metanephric kidney grafted on the chorioallantoic membrane. Dev Biol 96: , Potter EL: Development of the human glomerulus. Arch Pathol 8: , Noden DM: Embryonic origins and assembly of blood vessels. Am Rev Respir Dis 140: , Plate KH, Breier G, Millauer B, Ullrich A, Risau W: Upregulation of vascular endothelial growth factor and its cognate receptors in a rat glioma model of tumor angiogenesis. Cancer Res 53: , Neufeld G, Tessler S, Gitay-Goren H, Cohen T, Levi BZ: Vascular endothelial growth factor and its receptor. Prog Growth Factor Res 5: 89 97, Brown LF, Detmar M, Tognazzi K, Abu-Jawdeh G, Iruela- Arispe ML: Uterine smooth muscle cells express functional receptors (flt-1 and KDR) for vascular permeability factor/vascular endothelial growth factor. Lab Invest 76: , Ferrara N, Davis-Smyth T: The biology of vascular endothelial growth factor. Endocrine Rev 18: 4 25, 1997