Expression of vascular endothelial growth factor and its receptors in ...

1 downloads 0 Views 6MB Size Report
Institute of Pathology, Philipps University of Marburg, D-35043 Marburg; German Primate Center,. Giittingen; and Max Planck Institute, W. G. Kerckhoff Institute, ...
Expression of vascular endothelial and its receptors in human renal in adult kidney

growth factor ontogenesis and

M. SIMON, H.-J. GRiiNE, 0. JOHREN, J. KULLMER, K. H. PLATE, W. RISAU, AND E. FUCHS Institute of Pathology, Philipps University of Marburg, D-35043 Marburg; German Primate Center, Giittingen; and Max Planck Institute, W. G. Kerckhoff Institute, Bad Nauheim, Germany Simon, M., H.-J. Griine, 0. Jahren, J. Kullmer, K. H. Plate, W. Risau, and E. Fuchs. Expression of vascular endothelial growth factor and its receptors in human renal ontogenesis and in adult kidney. Am. J. PhysioZ. 268 (RenaZ Fluid Electrolyte Physiol. 37): F240-F250, 1995.Vascular endothelial growth factor (VEGF) may modulate vascular permeability, chemotaxis for monocytes, and protease activity. In addition, VEGF may play a role in embryonic and tumor angiogenesis. In fetal mouse kidney, VEGF mRNA and protein expression have been demonstrated. This finding led to the hypothesis that VEGF might be involved in renal growth and development. To further elucidate the role of VEGF in human kidney, expression of VEGF and its receptors, the specific tyrosine kinase receptors, flt-1 and KDR, were studied. In fetal (6-24 gestational wk; mesonephros and metanephros) and adult kidney, VEGF mRNA and protein could be colocalized in glomerular epithelia and collecting duct cells by in situ hybridization and immunohistology. By reverse transcriptionpolymerase chain reaction, mRNA of three VEGF isoforms, VEGFlzi, VEGFiG5, and VEGFiB9, were found in fetal kidney and cortex, isolated glomeruli, and medulla of adult human kidney. KDR and flt-1 mRNA were coexpressed in endothelia of glomeruli and in peritubular capillaries in fetal and adult kidney. These data support the assumption that VEGF and its receptors may influence renal ontogenesis. We speculate that the constitutive expression of VEGF in adult kidney may be required for the function of VEGF receptor positive-fenestrated endothelia in glomeruli and postglomerular vessels. The expression of VEGF in collecting duct and of its receptors in medullary capillaries may in addition be relevant for maintaining medullary osmolality.

ENDOTHELIAL GROWTH factor (VEGF) or vascular permeability factor are dimeric glycoproteins of 2M, = 34-43 kDa with multiple vascular effects (7-9, 25, 38-40, 47). By differential splicing of mRNA, four proteins, VEGFZo6, VEGFlsS, VEGF165, and VEGF1zl, are known in humans (16, 22, 44). The latter two proteins are secreted by fetal and adult epithelial and mesenchymal cells (e.g., vascular smooth muscle cells) (10, 17, 19, 20, 28, 35); they have heparin binding affinity (15) and exert specific mitogenic effects on endothelial but not on other cells in vitro and in vivo (9, 27). VEGFZo6 and VEGFlsg are cell associated, probably due to their high affinity for proteoglycans, and may also exert angiogenic activity like the two smaller proteins (2, 16, l&32). VEGF increase microvascular permeability in serosal linings, subcutis, and skin and induce a dose-dependent

relaxation of small muscular arteries (26, 39). This dilatation is dependent on vascular nitric oxide synthesis and probably mediated by a phospholipase C activation (26). In cultured human umbilical vein endothelial cells, VEGF stimulates inositol1,4,5trisphosphate accumulation, a rise in intracellular Ca2+, and release of von Willebrand factor (3). VEGF, in addition, has chemotactic properties for monocytes (7). Two receptor tyrosine kinases (RTK), flt-1 and KDR, designated as flk-1 in mouse, have been identified as receptors for VEGF. These RTK constitute a subgroup within the class III tyrosine kinases with seven immunoglobulin-like loops in their extracellular domain (13, 41-43). During murine development, high levels of VEGF mRNA were found in most fetal organs, including the kidney (2). It was also reported that, in the adult mouse and rat kidney, a constitutive expression of VEGF mRNA and protein can be detected in the glomeruli (2, 30). Jakeman et al. (23) localized binding sites for 1251-labeledVEGF to the capillaries of glomeruli and postglomerular vessels in the rat kidney. Observations in humans, focusing on the glomerular capillary network, have documented a constitutive expression of VEGF in adult kidneys (5). The described functional properties make VEGF an attractive candidate for ontogenetic and pathophysiological processes in the kidney. VEGF may play a role in renal vascular development. The constitutive expression of VEGF in glomeruli has been linked to the permselectivity of the glomerular filter. Several immunologic and nonimmunologic glomerular diseasesare associated with endothelial cell proliferation and monocyte infiltration into the glomerulus and tubulointerstitium. The knowledge of the sites of expression of a growth factor and its receptors may help to elucidate its physiological role and may aid to interpret pathophysiological phenomena. As studies have demonstrated, the possibility of species differences in the expression of vasoactive proteins and their receptors, an extensive characterization of VEGF synthesis, and receptor localization in human kidney seemsappropriate (l&21). The aim of this study was to analyze in detail the mRNA and protein expression of VEGF and mRNA expression of its receptors flt-1 and KDR in the human fetal and adult kidney. Early (mesonephros) stages of renal organogenesis were included to gain some insight into the role of VEGF in the development of renal parenchyma.

F240

the American

vascular endothelial growth factor; vascular endothelial ability factor; human fetal; adult kidney

perme-

VASCULAR

0363-6127/95

$3.00

Copyright

o 1995

Physiological

Society

VEGF

AND

ITS

RECEPTORS

IN HUMAN

KIDNEY

Fig. 1. Photomicrographs of whole kidney sections after hybridization using [35SlmRNA for vascular endothelial growth factor (VEGF), flt-1, and KDR. A-D: VEGF mRNA in fetal (A and B, 17-wk gestation) and adult (C and D) human kidney using antisense (A and C) and sense probes (B and D). A clear depiction of positive glomeruli (G) in fetal and adult kidney in A and C. Intense signals over fetal collecting ducts (CD, Al. E-H: flt-1 mRNA in fetal (E and 3’) and adult (G andH) human kidneys using antisense (E and G) and sense probes (F andH1. Glomeruli (G, E and G) are positive, already in nephrogenic zone of fetal kidney and in cortex of adult kidney. A weak binding was seen in inner and outer medulla of adult kidney’(for localization, see Fig. 4K). Z-L: KDR mRNA in fetal (Z and J) and adult (K and L) human kidney using antisense (I and K) and sense probes LZ and L). Accentuated mRNA positivity in glomeruli (G) of subcapsular nephrogenic zone and inner cortex of fetal kidney. Glomeruli remain positive in adult kidney. Also, signals in peritubular capillary (V) remain positive in adult kidney, with strongest intensity in outer medulla.

F241

F242

VEGF

AND

ITS

RECEPTORS

Table I. Gender and age of adult and fetal human kidneys

studied Tissue

Gender

Adult

kidney

Regular Regular Regular Regular Regular AA-amyloidosis Fetal Regular Regular Regular Regular Regular Regular Regular For gender: f, female; kidney, age in gestational

MATERIALS

Age

f f m f m f

42 57 46 65 77 63

m m m f f f f

6 17 19 20 22 22 24

kidney

m, male. For adult kidney, wk. AA, amyloid A.

age in yr; for fetal

AND METHODS

Adult human kidneys (n = 6) were obtained from patients undergoing nephrectomy due to renal cell carcinoma. The age of the patients was 58 5 12 yr (mean -+ SD). The tissue was taken from tumor-free parts of the kidneys immediately after surgical removal. Part of the tissue was frozen in liquid nitrogen and stored at -70°C until processed further. Tissue slices were immersion fixed in phosphate-buffered formaldehyde (PFA, 4%), pH 7.2, for 24 h and paraffin embedded. Fetal kidneys were obtained during autopsies performed shortly after abortion. The age of the fetuses was 6, 17, 19,20, 22, and 24 wk of gestation. The tissue was partly snap frozen, partly cut into slices, fixed, and paraffin embedded as described above. All cases studied are summarized in Table 1. For in situ hybridization and immunohistochemistry, coronal sections of fetal kidneys and sections with cortex, outer, and inner medulla of adult kidneys were taken. In situ hybridization. Five-micrometer paraffin sections were mounted on 0.5% gelatin-coated glass slides and allowed to dry for 2 h at 42°C. The sections were taken for in situ hybridization on the next day or stored in airtight boxes at -20°C up to 4 wk without a change in results. To generate riboprobes by in vitro transcription, pBluescript KS+ plasmids (Stratagene; Heidelberg, Germany) containing cDNA inserts for VEGF (520 bp), flt-1 (1,080 bp), and KDR (1,400 bp) were used to generate run-off transcripts (13, 43, 48). The cDNA insert for human VEGF codes for the putative amino acid signal sequence of 26 amino acids and the following region of 115 amino acids common to all four isoforms (48). In vitro transcription was carried out using a Trans-Probe-T-Kit (Pharmacia; Freiburg, Germany) and 200 PCi [35S]UTP (Amersham; Braunschweig, Germany) for each transcription reaction under conditions described by the manufacturer. The vectors were cut with BamH I (VEGF), Not I

IN HUMAN

KIDNEY

@t-l>, and X&a I (KDR) and transcribed each with T3-RNA polymerase to yield antisense probes. To yield sense probes, the plasmids were cut with EcoR I (VEGF and KDR) andXho I (fit-l), and 1inearized DNAs were transcribed each with T7RNA polymerase. The integrity of the probes was checked on a 6% denaturing polyacrylamide gel. To obtain transcripts with a length of N 500 bp the synthesized flt-1 and KDR mRNAs were processed by limited alkaline hydrolysis (11). For in situ hybridization, [35S]mRNAs were diluted from 2 x lo4 to 1 x lo5 countsminla ~1~l in hybridization buffer containing 2 mM EDTA pH 7.5, 20 mM tris(hydroxymethyl)aminomethane (Tris), pH 7.5, 0.6 M NaCl, 2~ Denhardt’s solution, 20% dextran sulfate, 0.1 J-Lg/pl tRNA, and 0.2 M dithiothreitol. After deparaffinization, kidney sections were digested with 20 pg/ml proteinase K (Boehringer; Mannheim, Germany) in phosphate-buffered saline (PBS) for 12 (fetal) and 16 min (adult). Sections were postfixed for 5 min in PFA and acetylated (0.25% acetic anhydride in 0.1 M triethanolamine, 10 min). Twenty to fifty microliters of hybridization mixture was placed on each section and covered with a siliconized glass coverslip. Hybridization was carried out in moist chambers at 55°C for 16 h. Coverslips were removed by washing in 4x saline sodium citrate (SSC) and 2~ SSC for 10 min each at 37°C. Slides were then washed in 0.5~ SSC for 15 min at 37°C followed by a washing step in 0.5~ SSC for 30 min at 60°C. After slides were rinsed in STE buffer (containing 10 mM Tris, 10 mM NaCl, 1 mM EDTA) for 15 min at 37°C ribonuclease A (Sigma; Deisenhofen, Germany) was added at a concentration of 20 pg/ml and incubated for 30 min at 37°C. Slides were washed in STE for 15 min, 0.5~ SSC for 30 min at 37°C in 0.5~ SSC for 15 min and 0.1~ SSC for 15 min at room temperature, and dehydrated. Sections were dried for 1 h at room temperature and exposed to Hyperfilm 3H (Amersham) at 4°C for 5-7 days. Emulsion autoradiography was performed using Kodak NTB2 emulsion (Eastman; Rochester, NY), and coated slides were exposed at 4°C for lo-14 days. Slides were developed in Kodak D19, fixed in Kodak Unifix, and counterstained with hemalaun and eosin. Reverse transcription-polymerase chain reaction. RNA was isolated from cortex and medulla of adult human kidney and from whole, fetal human kidney by the method described by Chomczynski and Sacchi (6). Five micrograms total RNA were reverse transcribed using Moloney murine leukemia virus reverse transcriptase (12). Polymerase chain reaction (PCR) was carried out according to Gene Amp DNA amplification reagent kit instructions (Perkin-Elmer Cetus; Norwalk, CT) with modifications as described (12). Thirty cycles were performed. During each cycle, the samples were heated to 94°C for 30 s, cooled to 56°C for 60 s, and heated to 72°C for 60 s. Reverse transcription (RT)-PCR reagent blank yielded no detectable products. Also, PCR carried out with total RNA to exclude DNA contamination of the RNA preparation was negative for VEGF and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). A 16-~1 sample of each 50 ~1 of PCR solution was fractionated by electrophoresis in a 2% agarose gel. Gels were photographed using a Polaroid 665 film. The following oligonucleotide primers were used: VEGF 5’ oligonucleotide,

Fig. 2. Expression of VEGF and VEGF receptor KDR in mesonephros of 6th gestational wk. A-D: dark-field (A and C) and bright-field (B and 0) overviews of mesonephros (Mes) with Wolffian duct (WD) and embryonic gonad (EG) for VEGF (A and B) and VEGF receptor KDR mRNA (C and 0). In A and B, outer rim of glomerulus is intensely stained for VEGF (arrowheads); in higher power, in E and F, a labeling of podocytes (arrowheads) is apparent in dark-field (E) and bright-field (F) photographs. In contrast, KDR mRNA label (C, D, G, and H) is in a ringlike pattern in inner part of glomerulus (arrowheads) and also at inner surface of capillaries and vessels (V) in Mes; details can be seen in G (dark field) and H (bright field) with silver grains on capillary endothelia (arrowheads). Mesangial cells (M) and podocytes are not labeled.

VEGF

AND

ITS

RECEPTORS

IN HUMAN

KIDNEY

F243

F244

VEGF

AND

ITS

RECEPTORS

TGGATCCATGAACTTTCTGCT; VEGF 3’ oligonucleotide, CACCGCCTTGGCTTGTCACAT. To allow for quantitative comparison of VEGF expression in the different samples, a PCR for GAPDH was carried out under identical conditions with the following oligonucleotide primers: GAPDH 5’ oligonucleotide, CGATGCTGGCGCTGAGTAC; GAPDH 3’ oligonucleotide, CGTTCAGCTCAGGGATGACC. Immunohistochemistry. Immunohistochemistry was performed on 5-pm frozen sections and 4-pm paraffin sections of fetal and adult human kidney. Frozen sections were thaw mounted on gelatin-coated glass slides and fixed in acetone for 7 min at -20°C. Paraffin sections were dewaxed, incubated in 6 M urea, and processed by microwave for 20 min. Both tissue preparations were incubated with an antipeptide antibody directed against the NHz-terminal 20 amino acids common to the four isoforms of human VEGF for 60 min at 37°C. The antibody was raised in rabbits and affinity purified (34). Slices were washed in PBS and processed further by the conventional APAAP method. For blocking of renal endogenous alkaline phosphatase in frozen sections, levamisole (0.3 mg/ml; Sigma) was added. Anti-humanVEGF antibodies, previously saturated with recombinant human VEGF, served as a negative control. Slides were counterstained with hematoxylin. RESULTS

In situ hybridization. Autoradiograms are shown for fetal and adult kidneys for the mRNA of VEGF (Fig. 1, A-D), of flt-1 (Fig. 1, E-H), and of KDR (Fig. 1, I-L). Hybridization with sense riboprobes gave no signals over the renal parenchyma (Fig. 1, B, D, F, H, J, and L). In contrast, glomeruli demonstrated strong labeling for VEGF, flt-1, and KDR mRNA in fetal and adult kidneys with the antisense riboprobes (Fig. 1, A, C, E, G, I, and K). In emulsion-coated slides, VEGF could be localized to glomerular visceral epithelial cells in the mesonephros (Fig. 2, A, B, E, and F) and the metanephros (Fig. 3, A and B); here, VEGF mRNA could be first seen in S-shaped bodies in the outer epithelial rim, whereas ureteric bud epithelia were negative. In adult kidneys, visceral epithelial cells maintained their VEGF mRNA positivity (Fig. 4, A and B). Because of the complex capillary convolute of adult glomeruli, some positive epithelial cells seemed to lie within the glomerulus. But, in all cases in which endothelial and mesangial cells could be identified, these cells were negative for VEGF mRNA. To enhance the cellular localization of expression of VEGF mRNA in glomeruli, a kidney with glomerular amyloid A-amyloidosis was additionally studied. The glomeruli had wide mesangial areas with adjoining wide-open capillaries; only podocytes at the border of the glomerular tuft had a positive labeling;

IN HUMAN

KIDNEY

mesangial cells and endothelia were negative (Fig. 4E). In contrast, the mRNA label for the VEGF receptors flt-1 and KDR could be depicted in endothelia of the glomerulus. In the mesonephros and metanephros, the capillary lumina were surrounded by silver grains (Figs. 2, C, D, G, and H, and 3, C and D). Also, in the adult kidney, glomerular endothelia were positive (Fig. 4, C and D); this was especially apparent in the amyloid glomerulus of Fig. 4F with negative mesangial cells. Nevertheless, as the scatter of label around a positive source is N 15 pm for 35S-labeled riboprobes, it cannot be definitely excluded that some mesangial cells may have also demonstrated a positivity for VEGF receptors. The distribution of KDR and flt-1 label was identical in glomeruli and in pre- and postglomerular vessels. In fetal and adult kidneys, postglomerular vessels in cortex and medulla showed a positivity for endothelial cells, often closely surrounding erythrocytes (Figs. 3, G and H, and 4, I and J). Large preglomerular vessels had a faint endothelial positivity for both VEGF receptors KDR and flt-1 (data not shown). In adult kidneys, collecting ducts in cortex and medulla could be seen to express a weak label for VEGF mRNA (Fig. 4, A, B, G, and H) that could not be depicted in proximal tubules (Fig. 4, A and B). In fetal kidney, collecting ducts had a very strong VEGF mRNA label (Fig. 3, E and F) that could already be clearly discerned as broad black stripes in the autoradiograms (Fig. lA). RT-PCR. By RT-PCR, the three isoforms of VEGFIZ1, VEW65, and VEGFlsS could be shown to be present in total tissue, cortex, isolated glomeruli, and medulla of adult and in whole fetal kidney of humans. With consideration of GAPDH message intensity, the quantity of all three isoforms was less in medulla than in cortex of adult kidney. The isoforms VEGFlG5 and VEGFiZ1 were equally abundant, and VEGFlB9 was the least abundant in all tissues tested (Fig. 5). Immunohistochemistry. With frozen sections, immunohistological staining with affinity-purified rabbit antibodies to human VEGF demonstrated the protein in the glomerular extracellular matrix and in glomerular cells of both fetal and adult human kidneys (Fig. 6A). A more detailed cytoplasmic staining in the glomeruli could be discerned using microwave processed paraffin sections. In these, the VEGF protein could be clearly detected in the glomerular epithelial cells. Endothelial and mesangial cells did not stain (Fig. 6, A and C). An intense expression of VEGF protein was found in collecting duct epithelial cells of fetal kidneys (Fig. 6B, inset), and this staining was still detectable in the collecting duct cells of

Fig. 3. VEGF and VEGF receptors KDR and flt-1 mRNA in fetal kidney of 19th gestational wk in cortex (A-D) and medulla (E-H). A-D: dark field (A) and bright field (B) for VEGF staining shows positivity for glomerular epithelia (P) in early (Al and B1) and late (A2 and B2) phase of glomerular development. Ureteric bud (UB, Al and Bl > is negative as well as cells in glomerular stalk around erythrocytes. Mesenchyme (ME) and vessels (V) are without VEGF signal. Dark field (C) and bright field (D) for KDR staining shows positivity in capillaries, sometimes around erythrocytes in glomerular stalk in early (C1 and 01) and late (CZ and 02) phase of glomerular development (arrowheads) and in extraglomerular mesenchyme. Epithelial cells of glomeruli (P) and tubular epithelia (EP) are negative. Same staining pattern was seen for ‘flt-1 mRNA (not shown). E-H: dark field (E) and bright field (F) for VEGF staining shows positivity in collecting ducts (CD) without positive label for neighboring vessels with erythrocytes (EY). Dark field (G) and bright field (H) for flt-1 mRNA staining shows positivity in medullary blood vessels (arrowheads) around tubules that are sometimes filled with erythrocytes. Same staining pattern was seen for KDR mRNA.

VEGF

AND

ITS

RECEPTORS

IN HUMAN

KIDNEY

F245

F246 the adult human cytoplasmic.

VEGF

kidney

(Fig. 6B).

AND

ITS

The staining

RECEPTORS

was

DISCUSSION

VEGF has been shown to act as a growth factor for endothelial cells (27). It binds with high affinity to specific receptors, which are expressed on capillary endothelial cells in different tumors and fetal organs (14,29,31, 33,34). The data presented here show specific expression of VEGF and its two receptors, the tyrosine kinases flt-1 and KDR, in fetal and adult human kidney. The VEGF mRNA and protein could be colocalized on glomerular visceral epithelia. This is consistent with studies in mice, rat, and human kidney (1, 2, 24, 30). Recently, it has been reported that bovine glomerular endothelial cells in culture express VEGF mRNA and protein. mRNA of VEGF could be induced by fetal bovine serum and 12-O-tetradecanoylphorbol-13-acetate (45). Although the radioactive in situ hybridization method used has the inherent problem that a signal may erroneously extend to a neighboring cell, a clear-cut positivity of glomerular endothelial cells for VEGF mRNA could not be detected. Immunohistologically, VEGF protein was not present in endothelia of the glomerulus. On the other hand, an expression of mRNA of the VEGF receptors flt-1 and KDR on mesangial cells, in addition to the signal on endothelia, could not definitely be excluded, although it is not probable. In addition, we colocalized VEGF mRNA and protein on the collecting duct epithelia. This has not been described so far. The result of medullary VEGF expression could be confirmed by RT-PCR, which demonstrated VEGF mRNA of the three isoforms VEGFIB1, VEGFlG5, and VEGFla9 in medulla of adult human kidney (Fig. 5). Human kidney development starts in the fifth gestational week. The ureteric bud invades the nephrogenic mesenchyme and induces glomerular and tubular structures. Vascularization of the kidney takes place simultaneously. A pair of vessels sprouts from the dorsal aorta and invades the nephroblastema. The endothelial cells migrate directly into the glomerular anlage when S-shaped bodies are formed. It has been speculated that induction and subsequent differentiation of the mesenthyme produces angiogenic factors (36). VEGF has been shown to be an angiogenic mitogen; two of its isoforms are secreted (2, 22). In this study, we detected VEGF

IN HUMAN

KIDNEY

and its two receptors in the fetal human kidney in different cell types. Differentiating glomerular epithelia produced VEGF and may attract the endothelial cells into the glomeruli or modulate their inward growth. The glomerular endothelial cells expressed abundant flt- 1 and KDR mRNA in fetal kidney from the beginning of glomerular capillary development and at the same time as VEGF appeared in glomerular podocytes. This study shows that, already in the mesonephros, VEGF was transcribed in glomeruli that also expressed KDR receptor mRNA like the surrounding capillaries and veins. A gradual enhancement of KDR expression to more developed glomeruli in inner cortical parenchyma, as described by others (24), was not seen in the seven fetal kidneys studied. The flt-1 and KDR mRNA were found in invading capillaries of comma- and S-shaped stages of the glomeruli in the outer cortex (Fig. 3, C and D). Similarly, the epithelia of the collecting duct that were positive for VEGF mRNA may generate VEGF for angiogenesis of the peritubular capillaries, which develop from efferent arterioles descending into the medulla. mRNA of the VEGF receptors could be demonstrated on peritubular capillary endothelial cells in fetal cortex and medulla. An expression pattern similar to that described here for VEGF and its two receptors on neighboring cells has been reported for other fetal organs (heart, lung, liver, and brain) in mouse and humans (2, 24). In the fetal mouse brain, VEGF was expressed on choroid plexus epithelium and in cells of the ventricular layer. The mouse analogue of human KDR, flk-1, was highly expressed on endothelial cells invading the neuroectoderm from the perineural plexus covering the brain (29). We (Grijne and Simon, unpublished observations) and others (24) found flt-1 and KDR mRNAs on endothelial cells and VEGF mRNA in bronchial epithelia in fetal lung. These findings, therefore, provide additional evidence of a role for VEGF as a specific regulator of angiogenesis in several fetal organs. The cortical and medullary expression pattern of VEGF, flt-1, and KDR was maintained in adult human kidney. These observations are consistent with a study of Jakeman et al. (23) in the rat. Binding sites for 1251-VEGF were shown on the fenestrated endothelia of cortical and glomerular capillaries and medullary vasa recta in the adult rat kidney (23). However, in a recent study using in situ hybridization, flt-1 and KDR mRNA have been shown on capillary endothelial cells of renal

Fig. 4. A-J: VEGF and VEGF receptors KDR and flt-1 mRNA in adult kidneys of 46-yr-old male (A-D and G-J) and 63-yr-old female with amyloidosis (E and F) in cortex (A-F) and medulla (G-J). A-F: cortex; dark field (A) and bright field (B and E) for VEGF staining with positive label in glomeruli. Cells at outer rim of capillaries corresponding to visceral epithelia (arrowheads) show silver-grain positivity. Positivity for mesangial cells cannot definitely be excluded, although recognizable mesangial cells (open arrowheads) in normal (A and B) and amyloid kidney (E) do not demonstrate silver-grain staining. A cortical collecting duct (CD) is slightly stained. Dark field (C) and bright field (D and F) for KDR mRNA staining with positive label in glomerular cells surrounding lumina of glomerular capillaries (arrowheads and C in C, D, and F). Also, endothelial cells of peritubular capillaries (open arrowheads) are labeled. Epithelial glomerular (P, F) and tubular cells are negative. Mesa&al cells (M) in amyloid kidney (F) are negative; this cannot certainly be shown for regular glomerulus with condensed capillary convolute. The flt-1 showed same staining pattern as VEGF receptor KDR (not shown). G-J: medulla; dark field (G) and bright field (H) for medullary VEGF staining with positive label in collecting duct (CD) epithelia and negative peritubular interstitium. Dark field (I) and bright field (J) for medullary fit-1 mRNA staining with positive label in endothelia (arrowheads) of peritubular capillaries, sometimes containing erythrocytes. No positivity in epithelia of tubules.

VEGF

AND

ITS

RECEPTORS

IN HUMAN

KIDNEY

F247

F248

EGF

I

AND

ITS

RECEPTORS

VEGF

GAPDH Fig. 5. Reverse transcription-polymerase chain reaction (PCR) of VEGF and glyceraldehyde-3-phosphate dehydrogenese (GAPDH) in 2 adult kidneys 146 (lanes C and M) and 57 yr (lanes G and A) I and in 2 fetal kidneys 119- (lane Fi) and 20. (large Fz) wk gestation]. The 3 isoforms VEGFizl (450 nt), VEGF 1ss (582 nt), and VEGFis9 (654 nt) can be seen. VEGFizi and VEGFies are expressed at equivalent amounts clearly exceeding famt expression of VEGFisa. In cortex (lane C) of adult kidney, expression of VEGF is stronger than in medulla (lane M). Glomeruli (lane G) show a distinct signal for 3 isoforms as does total adult kidney parenchyma (lane A). GAPDH expression demonstrates equal amounts of DNA taken for reactions in all 6 lanes. Reagent blank and total RNA PCR gave negative results.

cell carcinoma but not in normal adult kidneys of humans (4). This discrepancy to our results may be due to different hybridization techniques. In this and other studies, VEGF expression has been demonstrated on cells adjacent to endothelial cells of brain, lung, liver, spleen, and kidney that are not actively dividing (30). Thus the physiological function of VEGF does not seem to be limited to angiogenesis. Because VEGF may increase vascular permeability with a molar potency 50,000 times greater than histamine, it has been suggested that it modulates microvascular permeability (39). In adult human kidney, VEGF mRNA was still detectable in glomerular and collecting duct epithelial cells. The flt-1 and KDR mRNA were coexpressed on glomerular endothelia and on renal peritubular endothelial cells. The glomerular endothelium is a specialized endothelial cell with multiple fenestrae and is involved in function of the semipermeable glomerular basement membrane (GBM) by producing a negatively charged glycocalyx on the plasma membrane and fenestrae. Likewise, the glomerular podocytes are contributing to the synthesis of the GBM. Thus one may speculate that generation of VEGF in podocytes, paracrine secretion, and consecutive binding to VEGF receptors expressed on glomerular endothelium may be involved in a local regulatory mechanism necessary for the interaction of these two cell types to provide integrity and normal function of the GBM. Endothelia of peritubular capillaries of the cortex and outer medulla and ascending vasa recta of the inner

Fig 6. Immunohistology of VEGF protein m fetal and adult human kidney. A: fetal kidney at 17-wk gestation. VEGF protein in matrix of comma-shaped glomerular structure (Lnset) with rather weak positivity in cytoplasmic rim of epithelia. In developed glomeruli, positivity is concentrated in matrix (MX) with weak staining of podocytes and focal positivity of parietal epithelia (PE). B: fetal (znset) and adult epithelia of collecting duct have VEGF protein positivity. C: in formaldehydefixed paraffin-embedded adult kidney, a stronger positivity in podocytes (P) than observed in A is apparent after microwave pretreatment of sections. Endothelial and mesangial cells are negative for VEGF.

IN HUMAN

KIDNEY

VEGF

AND

ITS

RECEPTORS

medulla are endowed with fenestrae (37, 46). It may be postulated that paracrine secretion of VEGF from collecting duct epithelium and consecutive binding to receptors on surrounding capillaries could influence the permeability of these capillaries and thereby, perhaps, the osmotic gradient of the kidney medulla. In summary, intense expression of VEGF and its two receptors, flt-1 and KDR, in fetal human kidney provide evidence for a specific role of this growth factor in angiogenesis during human kidney organogenesis starting early on in mesonephric development. Constitutive expression of VEGF on glomerular epithelium and its recep tors on glomerular endo thelial cells in adult h Uof man kidney may suggest a role in maintenance glomerular function and permselectivity. Similarly, expression of VEGF on collecting duct epithelia and VEGF receptors on neighboring medullary capillaries may perhaps serve as a local regulatory system for medullary interstitial homeostasis. Functional studies are necessary to gain insight into the role and regulation of VEGF and its receptors in vivo in physiological and pathophysiological states of adult kidney. We thank Herbert A. Weich (Genexpression; Braunschweig, Germany) for the generous gift of human vascular endothelial growth factor (VEGF) cDNA and the affinity-purified anti-VEGF antibody. We also thank Georg Breier (Max-Planck-Institute; Bad Nauheim, Germany) and Bruce Terman (Lederle Laboratories; New York, NY) for their gift of human flt-1 and KDR cDNA. This work was supported by a grant from the Deutsche Forschungsgemeinschaft (SFB330 C7) to H.-J. Grone. Address for reprint requests: H.-J. Grone, Institute of Pathology, Philipps University of Marburg, Lahnberge Clinic, D-35043 Marburg, Germany. Received

21 March

1994;

accepted

in final

form

2 August

IN HUMAN

9.

10.

11.

12.

13.

14.

15.

16.

17.

1994. 18.

REFERENCES 1. Berse, B., L. F. Brown, L. Van De Water, H. F. Dvorak, and D. R. Senger. Vascular permeability factor (vascular endothelial growth factor) gene is expressed differentially in normal tissues, macrophages, and tumors. Mol. BioZ. CeZZ. 3: 211-220, 1992. 2. Breier, G., U. Albrecht, S. Sterrer, and W. Risau. Expression of vascular endothelial growth factor during embryonic angiogenesis and endothelial cell differentiation. DeueZopment 114: 521532,1992. 3. Brock, T. A., H. F. Dvorak, and D. R. Senger. Tumor secreted vascular permeability factor increases cytosolic Ca2+ and von Willebrand factor release in human endothelial cells. Am. J. PuthoZ. 138: 213-221, 1991. 4. Brown, L. F., B. Berse, R. W. Jackman, K. Tognazzi, E. J. Manseau, H. F. Dvorak, and D. R. Senger. Increased expression of vascular permeability factor (vascular endothelial growth factor) and its receptors in kidney and bladder carcinomas. Am. J. PczthoZ. 143: 1255-1262, 1993. 5. Brown, L. F., B. Berse, K. Tognazzi, E. J. Manseau, L. Van De Water, D. R. Senger, H. F. Dvorak, and S. Rosen. Vascular permeability factor mRNA and protein expression in human kidney. Kidney Int. 42: 1457-1461,1992. 6. Chomczynski, P., and N. Sacchi. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. AnaZ. Biochem. 162: 156-159, 1987. 7. Clauss, M., M. Gerlach, H. Gerlach, J. Brett, F. Wang, P. C. Familletti, Y.-C. E. Pan, J. V. Olander, D. T. Conolly, and D. Stern. Vascular permeability factor: a tumor-derived polypeptide that induces endothelial cell and monocyte procoagulant activity, and promotes monocyte migration J. Exp. Med. 172: 1535-1545, 1990. 8. Conn, G., M. L. Bayne, D. D. Soderman, P. W. Kwok, K. A. Sullivan, T. M. Palisi, D. A. Hope, and K. A. Thomas. Amino

19.

20.

21.

22.

23.

24.

25.

26.

KIDNEY

F249

acid and cDNA sequences of a vascular endothelial cell mitogen that is homologous to platelet-derived growth factor. Proc. NcztZ. Acad. Sci. USA 87: 2628-2632,199O. Conolly, D. T., D. M. Heuvelman, R. Nelson, J. V. Olander, B. L. Eppley, J. J. Delfino, N. R. Siegel, R. M. Leimgruber, and J. Feder. Tumor vascular permeability factor stimulates endothelial cell growth and angiogenesis. J. CZin. Invest. 84: 1470-1478,1989. Conolly, D. T., J. V. Olander, D. M. Heuvelman, R. Nelson, R. Monsell, N. Siegel, B. L. Haymore, R. Leimgruber, and J. Feder. Human vascular permeability factor: isolation from U937 cells. J. BioZ. Chem. 264: 20017-20024, 1989. Cox, K. H., D. V. DeLeon, L. M. Angerer, and R. C. Angerer. Detection of mRNAs in sea urchin embryos by in situ hybridization using asymmetric RNA probes. Dev. BioZ. 101: 485-502, 1984. Cullinan-Bove, K., and R. D. Koos. Vascular endothelial growth factor/vascular permeability factor expression in the rat uterus: rapid stimulation by estrogen correlates with estrogeninduced increases in uterine capillary permeability and growth. EndocrinoZogy 133: 829-837,1993. DeVries, C. D., J. A. Escobedo, H. Ueno, K. Houck, N. Ferrara, and L. T. Williams. The fms-like tyrosine kinase, a receptor for vascular endothelial growth factor. Science Wash. DC 255: 989-991,1992. Eichmann, A., C. Marcelle, C. Breant, and N. M. Le Douarin. Two molecules related to the VEGF receptor are expressed in early endothelial cells during avian embryonic development. Mech. Dev. 42: 33-48, 1993. Ferrara, N., and W. J. Henzel. Pituitary follicular cells secrete a novel heparin-binding growth factor specific for vascular endothelial cells. Biochem. Biophys. Res. Commun. 161: 851-858, 1989. Ferrara, N., K. Houck, L. Jakeman, and D. W. Leung. Molecular and biological properties of the vascular endothelial growth factor family of proteins. EndocrinoZ. Rev. 13: 18-32, 1992. Ferrara, N., J. Winer, and T. Burton. Aortic smooth muscle cells express and secrete vascular endothelial growth factor. Growth Factors 5: 141-148, 1991. Fuchs, E., H.-J. Griine, M. Simon, and A. Laue. Distribution of 1251-endothelin-l, -2, -3 binding sites in mammalian kidneys (Abstract). Comp. Biochem. Physiol. 101A: 775-778, 1992. Gospodarowicz, D., J. A. Abraham, and J. Schilling. Isolation and characterization of a vascular cell mitogen produced by pituitary-derived folliculo stellate cells. Proc. A&Z. Acad. Sci. USA 86: 7311-7315,1989. Gospodarowicz, D., and K. Lau. Pituitary follicular cells secrete both vascular endothelial growth factor and follistatin. Biochem. Biophys. Res. Commun. 165: 292-298,1989. Griine, H.-J., A. Laue, and E. Fuchs. Localization and quantification of [1251]-endothelin binding sites in human fetal and adult kidneys: relevance to renal ontogeny and pathophysiology. KZin. Wochenschr. 68: 758-767,199O. Houck, K. A., N. Ferrara, J. Winer, G. Cachianes, B. Li, and D. W. Leung. The vascular endothelial growth factor family: identification of a fourth molecular species and characterization of alternative splicing of RNA. MOL. EndocrinoZ. 5: 1806-1814,199l. Jakeman, L. B., J. Winer, G. L. Bennett, C. A. Altar, and N. Ferrara. Binding sites for vascular endothelial growth factor are localized on endothelial cells in adult rat tissues. J. CZin. Invest. 89: 244-253, 1992. Kaipainen, A., J. Korhonen, K. Pajusola, 0. Aprelikova, M. G. Persico, B. I. Terman, and K. Alitalo. The related flt4, fltl, and KDR receptor tyrosine kinases show distinct expression patterns in human fetal endothelial cells. J. Exp. Med. 178: 2077-2088,1993. Keck, P. J., S. D. Hauser, G. Krivi, K. Sanzo, T. Warren, J. Feder, and D. T. Conolly. Vascular permeability factor, an endothelial cell mitogen related to PDGF. Science Wash. DC 246: 1309-1312,1989. Ku, D. D., J. K. Zaleski, S. Liu, and T. A. Brock. Vascular endothelial growth factor induces EDRF-dependent relaxation in coronary arteries. Am. J. PhysioZ. 265 (Heart Circ. Physiol. 34): H586-H592,1993.

F250

VEGF

AND

ITS

RECEPTORS

27. Leung, D. W., G. Cachianes, W.-J. Kuang, D. V. Goeddel, and N. Ferrara. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science Wash. DC 246: 1306-1309, 1989. 28. Levy, A. P., R. Tarmargo, H. Brem., and D. Nathans. An endothelial cell growth factor from the mouse neuroblastoma cell line NB 41. Growth Factors 2: 9-19, 1989. 29. Millauer, B., S. Wizigman-Voos, H. Schniirch, R. Martinez, N.-P. H. Moller, W. Risau, and A. Ullrich. High affinity VEGF binding and developmental expression suggests flk-1 as a major regulator of vasculogenesis and angiogenesis. CeZZ 72: 835-846, 1993. 30. Monacci, W. T., M. J. Merrill, and E. H. Oldfield. Expression of vascular permeability factor/vascular endothelial growth factor in normal rat tissues. Am. J. PhysioZ. 264 (CeZZ Physiol. 33): c995-c1002,1993. 31 Oelrichs, R. B., H. H. Reid, 0. Bernard, A. Ziemiecki, and A. F. Wilks. NYK/FLK-1: a putative receptor protein tyrosine kinase isolated from E 10 embryonic neuroepithelium is expressed in endothelial cells of the developing embryo. Oncogene 8: 11-18, 1993. 32. Park, J. E., G. A. Keller, and N. Ferrara. The vascular endothelial growth factor (VEGF) isoforms: differential deposition into the subepithelial extracellular matrix and bioactivity of extracellular matrix-bound VEGF. MOL. BioZ. CeZZ 4: 1317-1326, 1993. 33. Plate, K. H., G. Breier, B. Millauer, A. Ullrich, and W. Risau. Up-regulation of vascular endothelial growth factor and its cognate receptors in a rat glioma model of tumor angiogenesis. Cancer Res. 53: 5822-5827, 1993. 34. Plate, K. H., G. Breier, H. A. Weich, and W. Risau. Vascular endothelial growth factor is a potential tumour angiogenesis factor in human gliomas in vivo. Nature Lond. 359: 845-847, 1992. 35. Plot&, J., J. Schilling, and D. Gospodarowicz. Isolation and characterization of a newly identified endothelial cell mitogen produced by AtT-20 cells. EMBO J. 8: 3801-3806,1989. 36. Saxen, L. Vascularization of the nephron. In: Organogenesis of the Kidney, edited by P. W. Barlow, P. B. Green, and C. C. Wylie. Cambridge, UK: Cambridge Univ. Press, 1987, p.129-146. 37. Schwartz, M. M., M. J. Karnovsky, and M. A. Venkatachalam. Ultrastructural differences between rat inner medullary descending and ascending vasa recta. Lab. Invest. 35: 161-170, 1976.

IN HUMAN 38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

KIDNEY

Senger, D. R., D. T. Conolly, L. Van De Water, J. Feder, and H. F. Dvorak. Purification and NHZ-terminal amino acid sequence of guinea pig tumor-secreted vascular permeability factor. Cancer Res. 50: 1774-1778, 1990. Senger, D. R., S. J. Galli, A. M. Dvorak, C. A. Perruzzi, V. S. Harvey, and H. F. Dvorak. Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites fluid. Science Wash. DC 219: 983-985,1983. Senger, D. R., C. A. Perruzzi, J. Feder, and H. F. Dvorak. A highly conserved vascular permeability factor secreted by a variety of human and rodent tumor cell lines. Cancer Res. 46: 56295632,1986. Shibuya, M., S. Yamaguchi, A. Yamane, T. Ikeda, A. Tojo, H. Matsushime, and M. Sato. Nucleotide sequence and expression of a novel human receptor-type tyrosine kinase gene (fit) closely related to the fms family. Oncogene 5: 519-524, 1990. Terman, B. I., M. E. Carrion, E. Kovacs, B. A. Rasmussen, R. L. Eddy, and T. B. Shows. Identification of a new endothelial cell growth factor receptor tyrosine kinase. Oncogene 6: 16771683,199l. Terman, B. I., M. Dougher-Vermazen, M. E. Carrion, D. Dimitrov, D. C. Armellino, D. Gospodarowicz, and P. Biihlen. Identification of the KDR tyrosine kinase as a receptor for vascular endothelial cell growth factor. Biochem. Biophys. Res. Commun. 187: 1579-1586,1992. Tischer, E., R. Mitchell, T. Hartmen, M. Silva, D. Gospodarowicz, J. C. Fiddles, and J. A. Abraham. The human gene for vascular endothelial growth factor: multiple protein forms are encoded through alternative exon splicing. J. BioZ. Chem. 266: 11947-11954,1991. Uchida, K., S. Uchida, K. Nitta, W. Yumura, F. Marumo, and H. Nihei. Glomerular endothelial cells in culture express and secrete vascular endothelial growth factor. Am. J. PhysioZ. 266 (RenaZ Fluid EZectroZyte Physiol. 35): FBl-F86, 1994. Venkatachalam, M. A., and M. J. Karnovsky. Extravascular protein in the kidney. An ultrastructural study of its relation to renal peritubular capillary permeability using protein tracers. Lab. Invest. 27: 435-444, 1972. Wahl, S. M., H. Wong, and N. McCartney-Francis. Role of growth factors in inflammation and repair. J. CeZZ. Biochem. 40: 193- 199,1989. Weindel, K., D. Marme, and H. Weich. Aids-associated Kaposi’s sarcoma cells in culture express vascular endothelial growth factor. Biochem. Biophys. Res. Commun. 183: 1167-1174, 1992.