Temporally Compartmentalized Expression of Ephrin-B2 during Renal Glomerular Development

J Am Soc Nephrol 12: 2673–2682, 2001

Temporally Compartmentalized Expression of Ephrin-B2 during Renal Glomerular Development TAKAMUNE TAKAHASHI,* KEIKO TAKAHASHI,* SEBASTIAN GERETY,† HAI WANG,† DAVID J. ANDERSON,† and THOMAS O. DANIEL* *Nephrology Division and Center for Vascular Biology, Vanderbilt University Medical Center, Nashville, Tennessee; and †Department of Biology, California Institute of Technology, Pasadena, California.

Abstract. Glomerular development proceeds through the spatially ordered and sequential recruitment, proliferation, assembly, and differentiation of endothelial, mesangial, and epithelial progenitors. The molecular determinants of cell-cell recognition and targeting in this process have yet to be defined. The Eph/ephrin family of membrane receptors and counterreceptors are critical participants of developmental vascular assembly in extrarenal sites. Renal expression patterns of ephrin-B2 and EphB4 were investigated using mice expressing ␤-galactosidase under control of ephrin-B2 or EphB4 promoters. The earliest glomerular expression of ephrin-B2 was identified in a subset of differentiating comma-stage glomerular epithelial cells (podocyte progenitors) adjacent to the vascular cleft where endothelial progenitors are subsequently recruited. Epithelial ephrin-B2 expression was accompanied by expression in endothelial and mesangial cells as capillary assembly

progressed. At or near completion of glomerular maturation, epithelial ephrin-B2 expression was extinguished, with persistence in glomerular endothelial cells. Throughout development, one of several ephrin-B2 receptors, EphB4, was persistently and exclusively expressed in endothelial cells of venous structures. The findings show sequential ephrin-B2 expression across glomerular lineages, first in a distinct subset of podocyte progenitors and subsequently in endothelial cells of the developing glomerulus. Given targeting functions for Eph/ephrin family proteins, the findings suggest that ephrin-B2 expression marks podocyte progenitors at the site of vascular cleft formation, where expression may establish an “address” to which endothelial and mesangial progenitors are recruited. Thus, the present results suggest that ephrin-B2 and EphB interactions play an important role in glomerular microvascular assembly.

Developmental assembly of the renal glomerular microcirculation is a precise and coordinated process that forms a highly specialized vascular filtering apparatus (1,2). Glomerular endothelial cells are initially dispersed throughout the metanephric mesenchymal tissue (3,4) and are subsequently recruited to the clefts of comma-shaped glomeruli, where they proliferate and assemble into glomerular capillaries adjacent to mesangial and glomerular epithelial cells. A distinct population of endothelial progenitors, recruited from the extrarenal para-aortic region, assembles a basket-like network on the superficial surface of the developing kidney, where they interconnect to integrate large vessel and glomerular microvascular circulations (5). Available data implicate both vasculogenic and angiogenic processes in this integrated assembly. Several growth factors and their receptors mediate critical steps in glomerular capillary development. Vascular endothelial growth factor is expressed in developing podocytes and seems to promote differentiation and recruitment of endothelial progenitors expressing the vascular endothelial growth factor

receptor flk-1 from the adjacent metanephric mesenchyme (6,7). Endothelial production of platelet-derived growth factor–BB (PDGF-BB) and mesangial progenitor expression of platelet-derived growth factor receptor–␤ (PDGFR-␤) are required for the coordinated recruitment, proliferation, and assembly of mesangial cells (8). The endothelial tie-2 receptor and its angiopoietin ligands, which mediate endothelial-pericyte interactions and microvascular maturation (9 –11), have been detected in endothelial and mesangial cells of developing glomeruli (12,13). Transforming growth factor-␤1 actions are implicated at several steps during glomerulogenesis, most prominently in the maturation phase required to stabilize vascular structure (14). In addition to these examples of paracrine intercellular communication, cell-cell contact is implicated in correct targeting and assembly of the glomerulus, yet little is known about the molecular regulation of targeting and assembly. The Eph receptor tyrosine kinases are transmembrane proteins that are engaged by contact with their counter-receptors, ephrins, upon cell-cell contact (15,16). Both Eph and ephrin proteins contribute to developmental patterning and neuronal targeting (17). Among ephrins, the ephrin-A subclass proteins (ephrin-A1 to -A5) are membrane attached through glycosyl-phosphatidyl-inositol linkages, whereas the ephrin-B subclasses (ephrin-B1 to -B3) are transmembrane proteins with highly conserved cytoplasmic domains. Ephrin-As interact with the A subset of Eph receptors

Received October 27, 2000. Accepted July 11, 2001. Correspondence to Dr. Thomas O. Daniel, Immunex Corporation, 51 University Street, Seattle, WA 98101. Phone: 206-470-4875; Fax: 206-933-3733; E-mail: [email protected] 1046-6673/1212-2673 Journal of the American Society of Nephrology Copyright © 2001 by the American Society of Nephrology

2674

Journal of the American Society of Nephrology

(EphA1 to A8) and Ephrin-B ligands with the B subset of Eph receptors (EphB1 to EphB6). Mice homozygous for ephrin-B2 or EphB4 allele mutation display early embryonic vascularization defects at the developmental stage when interconnection of ephrin-B2 expressing arterial limb vascular networks normally links with EphB4 expressing venous limb networks (18,19). Lethal vascular defects are also detected in EphB2/EphB3 double mutants (20). EphB2 expression in mesenchymal cells adjacent to vessels has suggested that Ephrin-B/EphB signaling may participate in intercellular interactions between endothelial cells and adjacent mesenchymal cells (20). As for renal expression of ephrin-B and EphB, early studies have shown that glomerular endothelial progenitors display EphB1 immunoreactivity both before and during recruitment to comma- and S-shaped glomeruli (21). Moreover, specific oligomerized forms of ephrin-B1 stimulate in vitro assembly of human renal microvascular endothelial cells, expressing EphB1, into capillary-like structures (21,22), and endothelial cells derived from different vascular beds display striking differences in capillary assembly responses to specific members of the ephrin-A and ephrin-B families (21). Furthermore, more recent studies have demonstrated a high level of ephrin-B2 expression in adult glomeruli (23,24). These findings suggested an important role of ephrin-B and EphB engagement in glomerular microvascular assembly. In the present study, we explored the spatial and temporal pattern of ephrin-B2 and EphB4 expression to define their potential roles in renal glomerular development. The data demonstrate a sequential, compartmentalized pattern of ephrin-B2 expression through distinct cell lineages and suggest important roles of ephrin-B2 and EphB interactions during glomerular microvascular assembly.

Materials and Methods Animals The ephrin-B2tlacZ/⫹ or EphB4tlacZ/⫹ mice were generated as described previously (18,19). Disruption of the ephrin-B2 or EphB4 gene by Tau-␤-galactosidase insertion drives expression under control of the endogenous ephrin-B2 or EphB4 promoter. Animals were genotyped at weaning by PCR detecting ␤-galactosidase sequences or targeted exon as described previously (19). Flk1tm1Jrt mice and wildtype C57Bl6/J mice were purchased from the Jackson Laboratory (Bar Harbor, ME) and genotyped by PCR amplifying ␤-galactosidase or Neo-resistant gene sequences.

␤-Galactosidase Development Procedures Timed pregnant mice were killed at the indicated gestational day. Kidney tissues were removed from dissected embryos and fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) for 5 to 60 min at 4°C. The tissues were rinsed twice with cold PBS and permeabilized with PBS containing 0.02% NP-40, 0.01% sodium deoxycholate, and 2 mM MgCl2 for 15 min at 4°C. Color development was carried out for at least 6 h at room temperature in solution containing 0.02% NP-40, 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, 2 mM MgCl2, and 1 mg/ml 5-bromo-4-chloro-3-indolyl-Dgalactopyranoside (X-gal; Sigma, St. Louis, MO) in PBS. Tissues were postfixed with 4% paraformaldehyde in PBS for 60 min at 4°C,

J Am Soc Nephrol 12: 2673–2682, 2001

washed in PBS, dehydrated through graded ethanol series and isopropanol or Histo-Clear (National Diagnostics, Atlanta, GA), and embedded in paraffin. Serial sections, 5 ␮m thick, were cut, dehydrated, mounted, and examined by light microscopy using Nomarski optics (Zeiss Axiophot; Carl Zeiss, Thornwood, NY).

Combined Immunostaining with ␤-Galactosidase Histochemistry

Immunohistochemistry with anti–␣ smooth muscle actin (␣SMA) and anti–Tamm Horsfall protein and histochemistry with the lectin from Dolichos biflorus agglutinin (DBA) were superimposed on ␤-galactosidase histochemistry. The kidneys subjected to ␤-galactosidase development were embedded in paraffin, and the paraffin sections were dewaxed, dehydrated, and blocked with 5% donkey serum/PBS or 0.1% bovine serum albumin/PBS for 30 min at room temperature. Then, the sections were incubated with the following: (1) horseradish peroxidase (HRP)-labeled anti-␣SMA monoclonal (1:200, clone1A4; DAKO, Carpinteria, CA), (2) anti–Tamm Horsfall protein goat antiserum (1:400; ICN Biomedicals Inc., Costa Mesa, CA), and (3) HRP-conjugated DBA (5 ␮g/ml, Sigma) for 60 min at room temperature. Tamm Horsfall protein immunostaining was then conducted by reaction with HRP-conjugated anti-goat IgG (3.2 ␮g/ml, Jackson Immunoresearch Laboratories, West Grove, PA) for 60 min at room temperature. Thereafter, the slides were rinsed three times with PBS, and color reaction was developed using diaminobenzidine chromogen solution (Liquid DAB, DAKO), dehydrated, mounted, and examined by light microscopy (Zeiss Axiophot).

Immunohistochemistry Procedures Kidney tissues were snap-frozen in a dry ice–acetone bath. Cryostat sections (5 ␮m) were fixed in acetone for 10 min at ⫺20°C, washed three times with cold PBS, and blocked with 5% normal donkey serum or with 5% normal donkey serum plus 5% mouse or rat serum for 20 min. The sections were incubated with the following: (1) rabbit anti–␤-galactosidase (2.1 ␮g/ml; 5 Prime Inc., Boulder, CO), (2) phycoerythrin-conjugated rat anti-CD31 monoclonal (10 ␮g/ml, clone MEC13.3; Pharmingen, Sanprego, CA), (3) Cy3-conjugated mouse anti-␣SMA monoclonal (1:200, clone1A4; Sigma), or (4) sheep antilaminin (20 ␮g/ml, provided by Dale Abrahamson, University of Kansas Medical Center, Kansas City, KS) in the combinations indicated in the figure legends. Tissue sections were incubated for 60 min at room temperature, washed with cold PBS, and, when necessary, incubated with FITC- or rhodamine-conjugated donkey anti-rabbit IgG (6.5 ␮g/ml; Jackson Immunoresearch Laboratories) or rhodamine-conjugated anti-sheep IgG (6.5 ␮g/ml Jackson Immunoresearch Laboratories) for 30 min at room temperature. Washed sections were mounted (Vectashield; Vector Laboratories, Burlingame, CA) and analyzed by confocal microscopy (Zeiss LSM410). For the doublelabeling study with anti-LacZ and anti–thiazide-sensitive Na-Cl cotransporter (TSC-1), anti-LacZ directly labeled with FITC was prepared as described previously (25). The frozen section was first incubated with an affinity-purified anti-rat TSC1 rabbit polyclonal (provided from Steven Hebert, Yale University, New Haven, CT) (26) at a dilution of 1:1000 for 60 min at room temperature and then with rhodamine-conjugated donkey anti-rabbit IgG (15 ␮g/ml, Jackson Immunoresearch Laboratories) for 30 min. Subsequently, the section was incubated with rabbit IgG (20 ␮g/ml) for 30 min. After rinsing in PBS, the section was reacted with FITC-conjugated anti-LacZ (0.2 mg/ml) for 60 min.

J Am Soc Nephrol 12: 2673–2682, 2001

Results The pattern of ephrin-B2 expression in adult and developing mouse kidney was evaluated in heterozygous mice carrying a homologous recombinant ephrin-B2 allele expressing a Tau␤-galactosidase fusion protein under control of the endogenous ephrin-B2 promoter (18). Tau sequences promote association of the expressed ␤-galactosidase with microtubules (27), permitting clear definition of the cellular patterns of ephrin-B2 promoter activity. At low-power resolution in Figure 1, ␤-galactosidase expression is prominent in vascular structures, including the branch of renal artery, interlobular artery (Figure 1, A and B), arterioles and glomeruli (Figure 1A), peritubular capillaries (Figure 1A), and subsets of medullary vascular bundles (Figure 1, C, G, and I) of adult mouse kidney. The ephrin-B2 promoter-driven ␤-galactosidase activity was prominent in arterial but not venous endothelial cells, as previously reported (18). On closer inspection, we determined that ephrin-B2 expression was not limited to endothelial cells but extended to vascular smooth muscle cells (VSMC) in muscular arteries, including the renal artery, interlobular artery, and afferent arterioles (Figure 1, E and F). The staining pattern was distinct from the endothelial-restricted ␤-galactosidase expression driven by the endogenous flk-1 promoter (Figure 1E). Ephrin-B2 promoter activity was detected also in medullary collecting ducts labeled by DBA and adjacent distal tubules (Figure 1, C and G) but not in ascending limb of Henle labeled by anti–Tamm Horsfall protein (Figure 1I) and distal convoluted tubules identified by TSC1 expression (Figure 1H), consistent with medullary collecting duct derivation from the ureteric bud epithelium that stains at early developmental stages. Motivated by the prominent vascular and particularly glomerular expression of ephrin-B2, we set out to define the temporal and spatial pattern of ephrin-B2 promoter activity and to ascertain its potential roles in glomerular vascular assembly. We examined ephrin-B2 expression from embryonic day 12.5, the earliest developmental stage at which capillaries develop in metanephric stroma, to postnatal day 7, when the overall functional pattern of mouse kidney has been established. Shown in the E12.5 metanephric tissue of Figure 2A, ␤-galactosidase activity was detected in arterial endothelial cells sprouting from dorsal aorta and in the ureteric bud branches extending into the metanephric mesenchyme. Weak ␤-galactosidase activity was also detected in undifferentiated mesenchymal cells surrounding metanephric ducts. At E13.5, glomerular epithelial cells of comma-shaped glomeruli as well as branching arteries and ureteric buds showed prominent ␤-galactosidase activity (Figure 2, B and C). At E16.5, high-level ␤-galactosidase activity was detected in developing vessels, including those in muscular arteries, arterioles, glomerular capillaries, and both collecting ducts and adjacent connecting tubules (Figure 2, D through F). This vascular and distal tubular expression pattern persists from renal development to maturity. The high level of ephrin-B2 promoter activity in the developing glomeruli led us to analyze further the identity of the expressing cell types. Shown in Figure 3A-a, ephrin-B2 sur-

Ephrin-B2 in Renal Glomerular Development

2675

rogate ␤-galactosidase activity was restricted to the epithelial cells from which visceral epithelial cells are derived in commashaped developing glomeruli. Weak ephrin-B2 expression was seen in parietal glomerular epithelial sites with some extension to proximal tubules (not shown). This expression pattern is distinct from that of heterozygous mice expressing ␤-galactosidase under control of the flk-1 promoter (Figure 3A-e). Flk-1 has been used to define the origin and pattern of assembly of glomerular endothelial progenitors in previous studies (5). Flk-1 promoter-driven ␤-galactosidase activity was detected in endothelial precursors that are discontinuously distributed in metanephric stroma. These cells are thought to migrate to vascular clefts of comma-shaped glomeruli (Figure 3A-e). Low level and less frequent expression of ephrin-B2 was detected in these endothelial precursors (Figure 3A-a). As glomerular development progressed to the S-shaped stage (Figure 3A-b) and capillary loop stage (Figure 3A-c), prominent ephrin-B2– driven ␤-galactosidase expression was detected in glomerular vascular cells, including not only visceral epithelial cells but also endothelial and mesangial cells. Emphasizing the temporal distinctions in the pattern of ephrin-B2 expression during glomerular development, the cellular expression pattern changed markedly by completion of glomerular maturation. Ephrin-B2 expression in glomerular epithelial cells was reduced by postnatal day 7 (not shown), and no ␤-galactosidase activity was detected in glomerular epithelial cells of adult mouse kidney (Figure 3A-d). The shifting pattern of expression led us to evaluate more closely the relationship of ␤-galactosidase activity to the glomerular basement membrane as a means of discriminating expression in visceral epithelial from expression in endothelial cells. Our targeting strategy with Tau-LacZ protein enabled us to analyze the lineage expression using anti–␤-galactosidase immunohistochemistry. In double-labeling experiments that evaluated ␤-galactosidase and laminin immunoreactivities in the same sections by confocal microscopy (Figure 3B), anti– ␤-galactosidase initially labeled visceral glomerular epithelial cells (Figure 3B-a). At the next stages of glomerular development, ephrin-B2 expression was evident in all glomerular cell components, including epithelial, endothelial, and mesangial cells. (Figure 3, B-b and B-c). Subsequently, at full maturity, ephrin-B2 expression was restricted to glomerular endothelial cells (Figure 3B-d). This progression of expression through distinct cell lineages led us to evaluate ephrin-B2 expression in vascular cells of larger arteries as well as of the glomerulus. For this purpose, we conducted simultaneous detection of ␤-galactosidase immunoreactivity and ␣SMA immunoreactivity. In E14.5 kidneys, ephrin-B2– driven ␤-galactosidase was detected only in sprouting and branching arterial endothelial cells but not in surrounding VSMC labeled by anti-␣SMA (Figure 4A). As arteriogenesis progressed to the neonatal stage, ephrin-B2 expression was apparently detected in VSMC of the renal arterial circulation (Figure 4B). In adulthood, VSMC dominantly express ephrinB2, with some attenuation in smooth muscle cells of afferent arteriole and glomerular capillaries (Figure 4C). To explore the distribution of cells expressing receptors with

2676

Journal of the American Society of Nephrology

J Am Soc Nephrol 12: 2673–2682, 2001

Figure 1. Ephrin-B2 expression in adult mouse kidney. (A through E) Renal tissues from ephrin-B2 heterozygous mice were stained by ␤-galactosidase histochemistry. The staining marks ephrin-B2 promoter activity in adult kidney vasculature, including the branch of renal artery (B), interlobular arteries (A), afferent and efferent arterioles (aa, ea), glomerulus (Glom), peritubular capillaries (PCap), vascular bundles of the outer medulla (arrows), and tubular components including connecting tubules (CT) and collecting ducts (CD). Staining is not detected in the sections from wild-type mice (D). It is noteworthy that ephrin-B2 promoter is active in vascular smooth muscle cells (VSMC) of large muscular artery (B) and interlobular artery (E-a and E-b). This pattern apparently differs from the endothelial-restricted staining of flk-1

J Am Soc Nephrol 12: 2673–2682, 2001

Ephrin-B2 in Renal Glomerular Development

2677

Figure 2. Ephrin-B2 expression in developing mouse kidney. The ephrin-B2 promoter activity of developing kidney was visualized by ␤-galactosidase histochemistry. (A) At embryonic day 12.5 (E12.5), ephrin-B2 expression is apparent in the ureteric bud epithelium (UB), in the early arterial vessels (Art) entering into metanephros, and in the vascular networks (arrowhead) surrounding the metanephric blastema and assembled around the ureteric bud. (B through F) At E13.5 (B) and E14.5 (C), staining is evident in the developing glomeruli, including comma-shaped (C), S-shaped (S), and capillary loop stage (CL) glomeruli. Similar expression pattern of ephrin-B2 is observed in E16.5 (D), E18.5 (E), and postnatal day 7 (F) kidneys. The combined DBA/␤-galactosidase histochemistry demonstrates the ephrin-B2 expression in the connecting tubules (CT) to be continuous with collecting ducts (CD) (E, insert). Magnification, ⫻150.

heterozygous mice at postnatal day 7 (E-c). The outer surface of the vessels are indicated by hatched lines and circles. (F) Cryosections of ephrin-B2 heterozygous mice were immunostained with anti–␤-galactosidase. In renal arteries, ephrin-B2 expression is detected in endothelial cells (ECs) and VSMC. Endothelial expression of ephrin-B2 is higher in afferent arteriole and glomerular capillaries than in the branch of renal artery. Ephrin-B2 expression of glomerular mesangial cells (MCs) is obscure. (G and H) The paraffin sections of adult kidney, subjected to ␤-galactosidase development, were stained with Dolichos biflorus agglutinin (DBA) peroxidase conjugate (brown). The staining defines ephrin-B2 expression in collecting ducts (CD), adjacent connecting tubules (CT), and vascular bundles (arrows). The double-labeling study for anti-LacZ (green) and anti-TSC1 (red) demonstrates lack of ephrin-B2 expression in distal convoluted tubules (DCT) (H, insert). (I) The immunostaining with anti–Tamm Horsfall protein (brown) was combined with ␤-galactosidase histochemistry. Ephrin-B2 expression is not observed in ascending limb (AL) of distal tubules. It is noteworthy that vascular bundles are composed of ephrin-B2–positive (arrowheads) and –negative (arrows) vessels (insert). The thin limbs of Henle (TL) are negative for ephrin-B2 expression. Magnifications: ⫻250 in A through E; ⫻400 in F (right); ⫻600 in F (middle and left); ⫻80 in G; ⫻200 in H; ⫻100 in I.

2678

Journal of the American Society of Nephrology

J Am Soc Nephrol 12: 2673–2682, 2001

Figure 3. Progression of glomerular ephrin-B2 expression in distinct lineages. (A) Embryonic and adult kidneys from ephrin-B2 (a through d) or flk-1 (e through h) heterozygous mice were examined by ␤-galactosidase histochemistry (a and e, E14.5; b, c, f, and g, E16.5; d and h, adult mice). Ephrin-B2 expression is initially detected in precursors of the visceral glomerular epithelial cells (VE) adjacent to the vascular cleft (VC) of comma-shaped glomeruli (a) and subsequently in glomerular capillary tufts of S-shaped (b) and capillary loop stage glomeruli (c). Ephrin-B2 and flk-1 expression is seen in endothelial progenitors (arrowheads) distributed around comma-shaped glomeruli (a and e). In the glomeruli of adult animals (d), glomerular epithelial ephrin-B2 expression is markedly attenuated. In contrast, flk-1 expression is initially observed in endothelial progenitors migrating into the vascular cleft (VC) (e) and remains restricted to glomerular endothelial cells during glomerulogenesis (f and g). In adult kidney, flk-1 is continuously expressed in glomeruli, whereas arterial (not shown) and arteriolar flk-1 are markedly diminished (h). (B) Frozen tissue sections of neonatal (a through c) and adult (d) kidneys were double-immunolabeled with anti–␤galactosidase (ephrin-B2, green) and anti-laminin (red). In comma- and S-shaped glomeruli (a), ephrin-B2 is predominantly expressed in glomerular epithelium (Epi). Subsequently, definitive expression of ephrin-B2 can be seen in glomerular endothelial (En) and mesangial (Mes) cells as well as in glomerular epithelial cells (Epi) of capillary stage glomeruli (b and c). Finally, ephrin-B2 remains to be expressed in glomerular endothelial cells of adult kidney (d). Magnifications: ⫻400 in A; ⫻800 in B-a; ⫻1000 in B-b; ⫻1000 in B-c; ⫻1000 in B-d.

J Am Soc Nephrol 12: 2673–2682, 2001

Ephrin-B2 in Renal Glomerular Development

2679

Figure 4. Lineage progression of ephrin-B2 expression in arteries. Immunofluorescence stainings were conducted on frozen kidney tissue sections with anti–␤-galactosidase (green) and anti-CD31 (red) or anti–␣ smooth muscle actin (␣SMA; red) in combinations indicated in each panel. (A) In E14.5 stage kidney, ephrin-B2 is expressed in endothelial cells (ECs) of developing arterial vessels (overlapping staining for ␤-galactosidase and CD31) but not in VSMC labeled by anti-␣SMA. (B) In neonatal kidney, ephrin-B2 expression is detected in VSMC as well as in endothelial cells. High-level expression of endothelial ephrin-B2 is revealed by overlapping staining (yellow) for ␤-galactosidase and CD31 (right). (C) In adult arteries, ephrin-B2 is dominantly expressed in VSMC, and endothelial ephrin-B2 is reduced. Magnification, ⫻600

which ephrin-B2 may interact, we examined patterns of EphB4 expression. EphB4 was shown previously to be expressed in venous limb endothelial cells during vascular development in extrarenal sites, and mice genetically deficient in EphB4 were shown to have a phenotype complementary to that of ephrin-B2 null mice (19). Using recombinant mice expressing tau-␤-galactosidase under control of the EphB4 promoter (19), we detected ␤-galactosidase activity in endothelial cells of the renal vein extending to metanephric tissue at E14.5 (Figure 5A). Definitive EphB4 promoter activity was not observed in developing glomeruli at comma-shaped, S-shaped, or later stages of development

(Figure 5, A, B, D, and E). Renal venous endothelial EphB4 expression was dramatically reduced after postnatal day 7 stage (not shown) and was not detectable in adult kidney using optimized conditions (Figure 5C). Low levels of expression were identified in the endocapillary space of a small subset (⬍10%) of glomeruli, as shown in Figure 5C. The unfavorable signal-tonoise ratio and the infrequent glomerular expression made further definition of the responsible cell population impossible. It is noteworthy that the consistent and stable EphB4 expression in venous endothelium contrasts sharply with the progression of ephrin-B2 expression through multiple cell lineages.

2680

Journal of the American Society of Nephrology

J Am Soc Nephrol 12: 2673–2682, 2001

Figure 5. Venous expression of EphB4 in developing mouse kidney. The renal tissues from EphB4 heterozygous mice were stained by ␤-galactosidase histochemistry. (A through C) At E14.5 (A), the staining visualizes EphB4 promoter activity in renal veins (V) extending to metanephros. No apparent signals are detected in developing glomeruli, including S-shaped (S) and capillary loop stage (CL) glomeruli. In neonate kidney (B), EphB4 expression is detected in venous endothelial cells (V) constructing postglomerular circulations. EphB4 expression is dramatically reduced in adult kidney (C). It is noteworthy that a few significant EphB4 signals are observed in mature glomeruli (arrows) and in some blood cells (arrowhead). (D and E) Anti-␣SMA immunohistochemistry (brown) was combined with ␤-galactosidase histochemistry to define venous endothelial expression of EphB4. EphB4 expression is exclusively observed in venous endothelial cells in E14.5 (D) and neonate (E) kidneys. No stainings can be seen in arterial endothelial cells. Magnifications: ⫻100 in A through C; ⫻200 in D; ⫻150 in E.

Discussion A highly organized glomerular structure is formed through orchestrated intercellular connections between three distinct cell populations (1). However, little is known about cell targeting receptors that assemble glomerular cells to appropriate positions upon cell-cell contact. Recently, it was shown that ephrin-B and its EphB receptor participate in developmental vascular cell assembly. Here we show the sequential, compartmentalized expression of ephrin-B2 through different cell lineages during glomerular development. The initial studies of ephrin-B2 demonstrated that its expression marks arterial endothelial cells, whereas its binding partner, EphB4, is restricted in expression to venous endothelial cells in early embryonic angiogenesis (18,19). This spatial segregation, coupled with the developmental timing of vascularization failure in both ephrin-B2–and EphB4-deficient mice, suggests a role for ephrin-B2/EphB4 engagement in the coor-

dinated targeting and interconnection of venous and arterial limb structures. In formation of the glomerulus, endothelial progenitors seem to develop from metanephric mesenchyme, migrate to the developing glomerulus, and assemble into glomerular capillaries through a vasculogenic process (3,4). Ultimately they are integrated to arterial and venous circulations through afferent and efferent arterioles. On the basis of developmental restriction of ephrin-B2 expression to arterial and EphB4 to venous endothelial cells, we sought to define the boundary of ephrinB2/EphB4 interaction. We demonstrated that arterial and glomerular endothelial cells express ephrin-B2, whereas EphB4 is indeed expressed in venous limb vessels branched from renal vein. These findings raise two points. First, early endothelial progenitors, identified by recruitment to the vascular cleft and expression of Flk-1, express ephrin-B2, similar to arterial en-

J Am Soc Nephrol 12: 2673–2682, 2001

Ephrin-B2 in Renal Glomerular Development

dothelial cells. This arterial “programming” of glomerular endothelium is maintained even in the efferent end of glomerular endothelial capillaries and efferent arteriole (Figure 1A). Second, the distinct site at which ephrin-B2– expressing efferent arterioles interconnect with EphB4-expressing peritubular capillaries remains ill-defined at present. It is noteworthy that the medullary vascular bundle contains two distinct populations of capillaries, including ephrin-B2– expressing and –nonexpressing endothelium. Although a rare cell expresses EphB4 in mature glomeruli (Figure 5, B and C), this seems most likely to represent hematopoietic-derived cells (28 –30). We were surprised to find that during renal development, ephrin-B2 expression is first prominent in glomerular epithelial, not endothelial, cells. This expression is restricted to a subpopulation of epithelial cells adjacent to the vascular cleft, representing podocyte progenitor cells. Developmental studies in Xenopus demonstrated that ephrin-B1 and ephrin-B2 are expressed in the somites, whereas EphB4 is present on the posterior cardinal vein and intersomitic vein (31). A recent study demonstrated that either ectopic expression of ephrin-B1 or disruption of EphB4 signaling leads to aberrant projection of intersomitic vessels (32). Consistent with these findings, ephrin-B2 null mice develop aberrant sprouting of intersomitic vessels (33). These findings suggest that sprouting intersomitic vessels are guided through migration routes by somitic ephrin-B and endothelial EphB engagement (32). By analogy, we speculate that glomerular epithelial ephrin-B2 may guide endothelial progenitors expressing EphB1 to glomerular sites (21). Ephrin-B2 expression was identified in developing glomerular mesangial cells, as defined by ␣SMA expression (data not shown). Mesangial expression persisted through the early perinatal period but was extinguished in mature glomeruli (Table 1). This stage-restricted expression of mesangial ephrin-B2

2681

appears analogous to its expression in periendothelial mesenchymal cells in developing intersomitic vessels at E9.5. In early extrarenal embryonic angiogenesis, arterial limb endothelial cells express ephrin-B1, ephrin-B2, and EphB3, whereas mesenchymal cells that surround these vessels express ephrin-B2 and EphB2 (20). Some (30%) of EphB2/EphB3 double knockout mice experience embryonic vascularization failure, implicating other EphB receptors in cell-cell interactions necessary for proper vascular assembly. By extrapolation, one may expect that mesangial EphB3 (34) and/or other EphB proteins engage the endothelial ephrin-B2 to direct or modulate endothelial-mesangial cell assembly. We noted prominent expression of ephrin-B2 in VSMC of mature large arteries, in addition to its endothelial expression (Table 1). This vascular smooth muscle expression is not restricted to specific developmental stages and continues to maturity. On the basis of evidence that ephrin-B2 and EphB6 engagement arrests tumor cell growth (35), ephrin-B2 may participate in stabilizing and maintaining the vascular smooth muscle layer. Similarly, persistent ephrin-B2 expression in glomerular and arterial endothelial cells may be involved in the maintenance of endothelial integrity. Expression of ephrin-B2 in medullary and cortical collecting ducts and connecting tubules of developing and adult kidneys (Figure 1, C and G) suggests possible roles in developmental tubular/endothelial interactions. By analogy to the glomerular epithelial/endothelial interactions, tubule epithelial ephrin-B2 may direct the spatial organization of peritubular capillaries at these sites in cortex and medulla. Although earlier studies focused on functions for ephrin-B2 and EphB4 in embryonic arterial-venous vascular assembly, spatial- and timing-restricted expression of ephrin-B2 and EphB4 during glomerular development implicates these proteins in the integration of glomerular cell assembly.

Acknowledgments Table 1. Temporal and spatial expression of ephrin-B2 in renal vascular developmenta Large Arteries

Afferent Arterioles

Stage

E14 E16 P0 Adult

EC

VSMC

EC

VSMC

⫹⫹ ⫹⫹ ⫹⫹ ⫹

⫺ ⫹ ⫹⫹ ⫹⫹

⫹⫹ ⫹⫹ ⫹⫹ ⫹⫹

⫺ ⫹/⫺ ⫹ ⫹

Glomeruli

Comma S CL Adult a

EC

Mes

Epi

⫹/⫺ ⫹⫹ ⫹⫹ ⫹⫹

⫺ ⫹⫹ ⫹⫹ ⫺

⫹⫹ ⫹⫹ ⫹⫹ ⫺

EC, endothelial cells; VSMC, vascular smooth muscle cells; Mes, mesangial cells; Epi, visceral epithelial cells; Comma, comma-shaped; S, S-shaped; CL, capillary loop stage.

This work was supported by National Institutes of Health Grant R01-DK-47078 (T.O.D.). Analysis was performed in part through use of the VUMC Cell Imaging Resource supported by CA68485 and DK20593. The authors thank Raymond Harris and Matthew Breyer for support and advice, Dale Abrahamson for providing the antilaminin antibody, Steven Hebert for providing the anti-rTSC1 antibody, and John Schlueter and Amy Hsu for technical assistance.

References 1. Takahashi T, Huynh-Do U, Daniel TO: Renal microvascular assembly and repair: Power and promise of molecular definition. Kidney Int 53: 826 – 835, 1998 2. Daniel TO, Abrahamson DR: Endothelial signal integration in vascular assembly. Annu Rev Physiol 62: 649 – 671, 2000 3. Abrahamson DR: Glomerular endothelial cell development [Editorial]. Kidney Int 56: 1597–1598, 1999 4. Abrahamson DR, Robert B, Hyink DP, St John PL, Daniel TO: Origins and formation of microvasculature in the developing kidney. Kidney Int Suppl 67: S7–S11, 1998 5. Robert B, St John PL, Abrahamson DR: Direct visualization of renal vascular morphogenesis in Flk1 heterozygous mutant mice. Am J Physiol 275: F164 –F172, 1998

2682

Journal of the American Society of Nephrology

6. 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: 2351–2357, 1997 7. Tufro A, Norwood VF, Carey RM, Gomez RA: Vascular endothelial growth factor induces nephrogenesis and vasculogenesis. J Am Soc Nephrol 10: 2125–2134, 1999 8. Lindahl P, Hellstrom M, Kalen M, Karlsson L, Pekny M, Pekna M, Soriano P, Betsholtz C: Paracrine PDGF-B/PDGF-R␤ signaling controls mesangial cell development in kidney glomeruli. Development 125: 3313–3322, 1998 9. Holash J, Wiegand SJ, Yancopoulos GD: New model of tumor angiogenesis: Dynamic balance between vessel regression and growth mediated by angiopoietins and VEGF. Oncogene 18: 5356 –5362, 1999 10. Suri C, Jones PF, Patan S, Bartunkova S, Maisonpierre PC, Davis S, Sato TN, Yancopoulos GD: Requisite role of angiopoietin-1, a ligand for the TIE2 receptor, during embryonic angiogenesis. Cell 87: 1171–1180, 1996 11. Suri C, McClain J, Thurston G, McDonald DM, Zhou H, Oldmixon EH, Sato TN, Yancopoulos GD: Increased vascularization in mice overexpressing angiopoietin-1. Science 282: 468 – 471, 1998 12. Yuan HT, Suri C, Landon DN, Yancopoulos GD, Woolf AS: Angiopoietin-2 is a site-specific factor in differentiation of mouse renal vasculature. J Am Soc Nephrol 11: 1055–1066, 2000 13. Yuan HT, Suri C, Yancopoulos GD, Woolf AS: Expression of angiopoietin-1, angiopoietin-2, and the Tie-2 receptor tyrosine kinase during mouse kidney maturation. J Am Soc Nephrol 10: 1722–1736, 1999 14. Liu A, Dardik A, Ballermann BJ: Neutralizing TGF-␤1 antibody infusion in neonatal rat delays in vivo glomerular capillary formation 1. Kidney Int 56: 1334 –1348, 1999 15. Holder N, Klein R: Eph receptors and ephrins: Effectors of morphogenesis. Development 126: 2033–2044, 1999 16. Wilkinson DG: Eph receptors and ephrins: Regulators of guidance and assembly. Int Rev Cytol 196: 177–244, 2000 17. Gale NW, Yancopoulos GD: Ephrins and their receptors: A repulsive topic? Cell Tissue Res 290: 227–241, 1997 18. Wang HU, Chen ZF, Anderson DJ: Molecular distinction and angiogenic interaction between embryonic arteries and veins revealed by ephrin-B2 and its receptor Eph-B4. Cell 93: 741– 753, 1998 19. Gerety SS, Wang HU, Chen ZF, Anderson DJ: Symmetrical mutant phenotypes of the receptor EphB4 and its specific transmembrane ligand ephrin-B2 in cardiovascular development. Mol Cell 4: 403– 414, 1999 20. Adams RH, Wilkinson GA, Weiss C, Diella F, Gale NW, Deutsch U, Risau W, Klein R: Roles of ephrinB ligands and EphB receptors in cardiovascular development: Demarcation of arterial/venous domains, vascular morphogenesis, and sprouting angiogenesis. Genes Dev 13: 295–306, 1999 21. Daniel TO, Stein E, Cerretti DP, St John PL, Robert B, Abrahamson DR: ELK and LERK-2 in developing kidney and microvascular endothelial assembly. Kidney Int Suppl 57: S73–S81, 1996 22. Stein E, Lane AA, Cerretti DP, Schoecklmann HO, Schroff AD, Van Etten RL, Daniel TO: Eph receptors discriminate specific ligand oligomers to determine alternative signaling complexes,

J Am Soc Nephrol 12: 2673–2682, 2001

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

attachment, and assembly responses. Genes Dev 12: 667– 678, 1998 Gale NW, Baluk P, Pan L, Kwan M, Holash J, DeChiara TM, McDonald DM, Yancopoulos GD: Ephrin-B2 selectively marks arterial vessels and neovascularization sites in the adult, with expression in both endothelial and smooth-muscle cells. Dev Biol 230: 151–160, 2001 Shin D, Garcia-Cardena G, Hayashi SI, Gerety S, Asahara T, Stavrakis G, Isner J, Folkman J, Gimbrone MA, Anderson DJ: Expression of ephrinB2 identifies a stable genetic difference between arterial and venous vascular smooth muscle as well as endothelial cells, and marks subsets of microvessels at sites of adult neovascularization. Dev Biol 230: 139 –150, 2001 Takahashi T, Takahashi K, Mernaugh R, Drozdoff V, Sipe C, Schoecklmann H, Robert B, Abrahamson DR, Daniel TO: Endothelial localization of receptor tyrosine phosphatase, ECRTP/ DEP- 1, in developing and mature renal vasculature. J Am Soc Nephrol 10: 2135–2145, 1999 Plotkin MD, Kaplan MR, Verlander JW, Lee WS, Brown D, Poch E, Gullans SR, Hebert SC: Localization of the thiazide sensitive Na-Cl cotransporter, rTSC1 in the rat kidney. Kidney Int 50: 174 –183, 1996 Mombaerts P, Wang F, Dulac C, Chao SK, Nemes A, Mendelsohn M, Edmondson J, Axel R: Visualizing an olfactory sensory map. Cell 87: 675– 686, 1996 Inada T, Iwama A, Sakano S, Ohno M, Sawada K, Suda T: Selective expression of the receptor tyrosine kinase, HTK, on human erythroid progenitor cells. Blood 89: 2757–2765, 1997 Suda T, Iwama A, Hashiyama M, Sakano S, Ohno M: Receptor tyrosine kinases involved in hematopoietic progenitor cells. Leukemia 11[Suppl 3]:451– 453, 1997 Sakano S, Serizawa R, Inada T, Iwama A, Itoh A, Kato C, Shimizu Y, Shinkai F, Shimizu R, Kondo S, Ohno M, Suda T: Characterization of a ligand for receptor protein-tyrosine kinase HTK expressed in immature hematopoietic cells. Oncogene 13: 813– 822, 1996 Helbling PM, Saulnier DM, Robinson V, Christiansen JH, Wilkinson DG, Brandli AW: Comparative analysis of embryonic gene expression defines potential interaction sites for Xenopus EphB4 receptors with ephrin-B ligands. Dev Dyn 216: 361–373, 1999 Helbling PM, Saulnier DM, Brandli AW: The receptor tyrosine kinase EphB4 and ephrin-B ligands restrict angiogenic growth of embryonic veins in Xenopus laevis. Development 127: 269 –278, 2000 Adams RH, Diella F, Hennig S, Helmbacher F, Deutsch U, Klein R: The cytoplasmic domain of the ligand ephrinB2 is required for vascular morphogenesis but not cranial neural crest migration. Cell 104: 57– 69, 2001 Takahashi T, Shirasawa T, Miyake K, Yahagi Y, Maruyama N, Kasahara N, Kawamura T, Matsumura O, Mitarai T, Sakai O: Protein tyrosine kinases expressed in glomeruli and cultured glomerular cells: Flt-1 and VEGF expression in renal mesangial cells. Biochem Biophys Res Commun 209: 218 –226, 1995 Tang XX, Zhao H, Robinson ME, Cohen B, Cnaan A, London W, Cohn SL, Cheung NK, Brodeur GM, Evans AE, Ikegaki N: Implications of EPHB6, EFNB 2, and EFNB 3 expressions in human neuroblastoma. Proc Natl Acad Sci USA 97: 10936 – 10941, 2000