Differentiation 81 (2011) 81–91

Contents lists available at ScienceDirect

Differentiation

0301-46

Join the

doi:10.1

n Corr

E-m1 Pr

03550,2 Pr

Biomed

Univers

journal homepage: www.elsevier.com/locate/diff

Sema4C-Plexin B2 signalling modulates ureteric branching indeveloping kidney

Nina Perala a,n, Madis Jakobson a, Roxana Ola a, Pietro Fazzari b,1, Junia Y. Penachioni b,Mariann Nymark a,2, Tiina Tanninen a, Tiina Immonen a, Luca Tamagnone b, Hannu Sariola a,c

a Institute of Biomedicine/Medical Biochemistry and Developmental Biology, Biomedicum Helsinki, P.O. Box 63, FI-00014 University of Helsinki, Finlandb Institute for Cancer Research and Treatment, University of Turin Medical School, Candiolo, Turin, Italyc HUCH Laboratory Diagnostics, Paediatric Pathology, P.O. Box 400, FI-00029 Helsinki University Central Hospital, Finland

a r t i c l e i n f o

Article history:

Received 24 May 2010

Received in revised form

27 September 2010

Accepted 9 October 2010Available online 30 October 2010

Keywords:

Plexin B2

Sema4C

Branching morphogenesis

Kidney development

81/$ - see front matter & 2010 International

International Society for Differentiation (ww

016/j.diff.2010.10.001

esponding author. Tel.: +358 9 19125148; fax

ail address: nina.perala@helsinki.fi (N. Perala)

esent address: Instituto de Neurociencias de

Spain.

esent address: Folkhalsan Institute of Genetics

icum Helsinki and Division of Nephrology, Dep

ity Central Hospital, Helsinki, Finland.

a b s t r a c t

Semaphorins, originally identified as axon guidance molecules, have also been implicated in angiogenesis,

function of the immune system and cancerous growth. Here we show that deletion of Plexin B2 (Plxnb2), a

semaphorin receptor that is expressed both in the pretubular aggregates and the ureteric epithelium in

the developing kidney, results in renal hypoplasia and occasional double ureters. The rate of cell

proliferation in the ureteric epithelium and consequently the number of ureteric tips are reduced in the

kidneys lacking Plexin B2 (Plxnb2� /�). Semaphorin 4C, a ligand for Plexin B2, stimulates branching of the

ureteric epithelium in wild type and Plxnb2+/� kidney explants, but not in Plxnb2� /� explants. As

shown by co-immunoprecipitation Plexin B2 interacts with the Ret receptor tyrosine kinase, the receptor

of Glial-cell-line-derived neurotrophic factor (Gdnf), in embryonic kidneys. Isolated Plxnb2� /� ureteric

buds fail to respond to Gdnf by branching, but this response is rescued by Fibroblast growth factor 7 and

Follistatin as well as by the metanephric mesenchyme. The differentiation of the nephrogenic

mesenchyme, its morphology and the rate of apoptosis in the Plxnb2� /� kidneys are normal. Plexin

B2 is co-expressed with Plexin B1 (Plxnb1) in the kidney. The double homozygous Plxnb1–Plxnb2-

deficient mice show high embryonic lethality prior to onset of nephrogenesis. The only double

homozygous embryo surviving to E12 showed hypoplastic kidneys with ureteric branches and

differentiating mesenchyme. Taken together, our results show that Sema4C-Plexin B2 signalling regulates

ureteric branching, possibly through modulation of Gdnf signalling by interaction with Ret, and suggest

non-redundant roles for Plexin B1 and Plexin B2 in kidney development.

& 2010 International Society of Differentiation. Published by Elsevier Ltd. All rights reserved.

1. Introduction

The mammalian permanent kidney or metanephros develops byreciprocal interactions between the Wolffian duct-derived uretericbud (UB) and the metanephric mesenchyme (MM). The tips of theUB induce the surrounding MM to undergo mesenchyme-to-epithelium transition to become secretory nephrons, the functionalunits of the kidney. The branching ureteric epithelium thendifferentiates into collecting ducts (Costantini, 2006; Dressler,2006; Saxen, 1987). If the branching is retarded, the number ofnephrons remains small causing renal hypoplasia (Schedl, 2007).

Society of Differentiation. Publishe

w.isdifferentiation.org)

: +358 9 19125235.

.

Alicante, Sant Joan d’Alacant

, Folkhalsan Research Center,

artment of Medicine, Helsinki

Budding of the UB from the Wolffian duct takes place atembryonic day (E) 11 in mouse. Both the formation of the UBand its subsequent epithelial branching are essentially controlledby Glial-cell-line-derived neurotrophic factor (Gdnf) that isexpressed by the cap condensates of the MM. Gdnf binds to andactivates the dimeric complex of Ret receptor tyrosine kinase andits co-receptor Gdnf family receptor a1 (Gfra1) on the Wolffianduct and at the tips of the UB (Costantini and Shakya, 2006).Ureteric branching proceeds in a stereotypic manner and isregulated by several signalling molecules besides Gdnf, includingmembers of the Wnt and fibroblast growth factor (Fgf) families.Negative regulators of branching include Sprouty-1 and membersof the bone morphogenetic protein (Bmp) and TGFb families(Dressler, 2006).

We have now studied the role of Plexin B2 in kidney development,as it is expressed in the ureteric epithelium and condensing MM (Peralaet al., 2005). Plexins (Plxn) are the receptors for the semaphorins (Sema)(Tamagnone et al., 1999; Winberg et al., 1998), which were originallydescribed as axon guidance cues (Kolodkin et al., 1992). Plexins are also

d by Elsevier Ltd. All rights reserved.

N. Perala et al. / Differentiation 81 (2011) 81–9182

involved in the development of the cardiovascular system (Torres-Vazquez et al., 2004; Toyofuku et al., 2004a, 2004b), in the function ofthe immune system (Wong et al., 2003; Yamamoto et al., 2008) and inthe invasive growth of cancer cells (Giordano et al., 2002). Plexins aredivided in four subfamilies based on structural criteria (Tamagnoneet al., 1999) and the B-subfamily consists of Plexin B1, B2 and B3. Plxnb3

is expressed at only low levels during embryogenesis (Artigiani et al.,2004; Perala et al., 2005). Plxnb1 and Plxnb2 are widely expressed byboth neuronal and non-neuronal embryonic tissues, including thedeveloping kidney, in partly overlapping patterns (Perala et al., 2005;Worzfeld et al., 2004). Plexin B1-deficient (Plxnb1� /�) mice in theC57BL/6 background are healthy and viable (Deng et al., 2007; Fazzariet al., 2007), although they exhibit a transitory renal phenotype duringembryogenesis (Korostylev et al., 2008). From E13.5 to E14.5 thekidneys of Plxnb1� /� embryos show more ureteric branches thanthose of heterozygous and wild type littermates. The ligand of PlexinB1, Semaphorin 4D (Sema4D), inhibits ureteric branching in vitro

(Korostylev et al., 2008). Plexin B2-deficient (Plxnb2� /�) mice diebefore birth in the C57BL/6 background. They show defects in olfactorybulb development, neuronal proliferation and differentiation as well asneural tube closure (exencephaly) (Deng et al., 2007; Friedel et al.,2007; Hirschberg et al., 2009). In the CD1 background, Plxnb2� /�micesurvive but exhibit a profoundly altered layering and foliation of thecerebellum (Friedel et al., 2007). Defects outside of the nervous systemhave not been reported.

In vitro, both Plexin B1 and Plexin B2 act as receptors for Sema4D(Masuda et al., 2004). The preferable ligand for Plexin B2 is Sema4C(Deng et al., 2007). The binding of Sema4D and Sema4C to Plexin B1and Plexin B2, respectively, leads to the phosphorylation of ErbB-2and the plexin protein itself (Deng et al., 2007; Swiercz et al., 2004).Ligand binding to plexins also modifies the adhesion of cells to theextracellular matrix through conformational change and activationof the cytoplasmic GAP domain, leading to the inactivation of R-Rasand thereby integrins (Oinuma et al., 2004; Oinuma et al., 2006).Plexins regulate the actin cytoskeleton through interactionswith PDZ-RhoGEF and leukemia-associated Rho GEF (LARG) andactivation of RhoA, causing for example the collapse of the axonalgrowth cone (Aurandt et al., 2002; Driessens et al., 2002; Hirotaniet al., 2002; Perrot et al., 2002; Swiercz et al., 2002).

In this study we show that Plxnb2� /� embryos in the C57BL/6background exhibit renal hypoplasia with occasional doubleureters. In Plxnb2� /� kidneys, the branching of the uretericepithelium and cell proliferation in the ureteric tips are reduced.The response of the isolated ureteric epithelium to Gdnf iscompromised and co-immunoprecipitation experiments show aninteraction between Plexin B2 and Ret. Sema4C enhances branch-ing of E11.5 Plxnb2+/+ and Plxnb2+/� kidneys in culture, but notof Plxnb2� /� kidneys. Our data show that Sema4C-Plexin B2signalling is a new regulator of the ureteric branching and apotential regulator of Gndf-Ret signalling, and that the functionsof the closely related plexins, Plexin B1 and Plexin B2, seem to bedifferent during kidney development.

2. Materials and methods

2.1. Animals

Wild type mouse (Mus musculus) embryos were collected fromNMRI and C57BL/6 mice. The day of vaginal plug appearance wasdefined as the embryonic day 0.5 (E0.5). The Plexin B2-deficientmice (Plxnb2tmlMat1, here defined as Plxnb2� /�) in the C57BL/6background (Friedel et al., 2007) were genotyped as previouslydescribed. Plexin B1-deficient (Plxnb1� /�) mice in the C57BL/6strain were genotyped as previously described (Fazzari et al., 2007).Embryos were collected at stages E11.5–E16.5. The permit for the

animal experiments was received from the State Provincial Office ofSouthern Finland.

2.2. Tissue culture

Kidney explants were dissected from E11.5 or E12.5 embryosand cultured on Nuclepore filters (0.1 mm, Whatman) in DMEMsupplemented with 10% fetal bovine serum, Glutamax andpenicillin–streptomycin as previously described (Sainio, 2003).For induction of supernumerary budding, E11.5 explants werecultured in the presence of either 15 or 50 ng/ml of recombinant ratGdnf (R&D Systems).

To isolate the UB epithelium, E12.5 kidneys were treated withpancreatin-trypsin solution for 1 min at room temperature (Sainio,2003). Thereafter the UB epithelium was mechanically separatedfrom the mesenchyme in DMEM+10% FCS, fragments of UBepithelium were collected under microscope and transferredinto growth media. For mesenchyme-free 3D cultures, UBs weretransferred into polymerized Matrigel (BD Biosciences) matrixusing pulled glass capillaries. Matrigel matrix was prepared onice by 1:2 dilution of Matrigel with UB culture medium, whichcomposed of DMEM/F12 supplemented with 10% charcoal/dextrantreated FCS (Hyclone), 200 nM all-trans- and 9-cis retinoic acid(Sigma) and either 50 ng/ml recombinant rat Gdnf, 100 ng/ml Fgf7(R&D Systems) and 100 ng/ml Follistatin (R&D Systems) or Gdnfonly or Fgf7 and Follistatin. 50 ml of the matrix was used in 96-wellplates and 150 ml of UB culture medium was added to the wells thenext day. The adhesion of Plxnb2� /� UB cells to the extracellularmatrix molecules was analyzed by letting the isolated E12.5 UBcells adhere onto chamber-slides coated with laminin 0.5 mg/cm2

(Sigma), 0.1% fibronectin (Sigma), collagen (Millipore) or Matrigel.The cells were fixed after 24 h with 4% PFA and stained.

In vitro culture of metanephric mesenchymes from E11.5embryos was done as described previously (Kuure et al., 2007).The isolated MMs were differentiated by a transient (48 h) pulse of5 mM of 50-bromoindirubin-30-oxime (BIO; Calbiochem) andsubsequent culture in standard medium 3 days before fixationwith ice-cold methanol.

2.3. Production of Plexin B2 antibody

A polyclonal antibody, TI-2B, was raised in rabbits against apeptide specific for the intracellular part of mouse Plexin B2 (1572-EDSQQDLPGERHALLEEENR-1591) (Washington Biotechnology,Inc.) and affinity purified. For immunohistochemistry, TI-2B wasused at 1:100 and in Western blotting at 1:1000.The specificity ofthe antibody was tested by Western blotting with lysates of COS7cells transfected with cDNA expression constructs (Artigiani et al.,2003) to overexpress either Plexin B1 or Plexin B2.

2.4. Production of purified recombinant Sema4C

The secreted form of Sema4C was purified from the conditionedmedia of cells transfected with a cDNA expression construct, a kindgift of Dr. Steven Strittmatter (Yale University). Sema4C wasaffinity purified using metal-ion chromatography (Ni-NTA agarose,Qiagen), then analyzed and quantified by SDS-PAGE followed by gelstaining with Coomassie blue.

2.5. Antibodies and in situ hybridization probes

The primary antibodies used for immunohistochemistry wereHRP-conjugated rabbit anti-Actin (Santa Cruz) (1:1000), mouseanti-pan-cytokeratin (Sigma) (1:400), rabbit anti-brush border

N. Perala et al. / Differentiation 81 (2011) 81–91 83

(BB) (a gift from Dr. Aaro Miettinen) (1:2000), rat anti-mouse-Endomucin (Santa Cruz) (1:500), rabbit anti-phospho-ErbB-2(Y1248) (Abcam) (1:1000), rabbit anti-phospho-Met (Y1230/1234/1235) (Biosource, Invitrogen) (1:300), rabbit anti-mousePlexin-B2 Biotin (eBioscience) (1:1000), goat anti-Ret (SantaCruz) (1:1000), rabbit anti-Sox9 (Stolt et al., 2003) (1:500),mouse anti-vinculin (BD) (1:200), rabbit anti-Wilms tumourgene 1 (Santa Cruz) (1:200). Rhodamine-conjugated peanutagglutinin lectin (PNA) (Vector Laboratories) (1:200) andrhodamine phalloidin (Invitrogen) (1:100) were used directly tostain the tubules forming from the mesenchyme and F-actinbundles, respectively. Secondary antibodies were HRP-conjugated polyclonal anti-rabbit and anti-rat immunoglobulins(IgG) (DAKO) (1:1000) and anti-rat and anti-mouse Alexa Fluor 488and anti-rabbit Alexa Fluor 488 and 568 (Molecular Probes)(1:500).

The template for the DIG-labeled Sema4C probe for in situhybridization was IMAGE:6836945 (nt 2377–3715 of mouseSema4C, GenBank Acc:XM_898566.2). The other probes havebeen published earlier: Crlf1 (Alexander et al., 1999), Cxrc4 (Zouet al., 1998), Etv4 (Pea3) (Lin et al., 1998), Etv5 (Erm) (Trokovic et al.,2005), Gdnf, a gift from Dr. JG Pichel, Gfra1 (Suvanto et al., 1997),Myb (Lu et al., 2009), Pax2 (Dressler et al., 1990), Plxnb2 and Plxnb1

(Perala et al., 2005), Raldh2 (Niederreither et al., 1997), Ret (Pachniset al., 1993), Sprouty-1 (Zhang et al., 2001), Wnt9b (Carroll et al.,2005) and Wnt11 (Kispert et al., 1996).

2.6. Histology, and sectional and whole mount

immunohistochemistry

Whole embryos were fixed with 4% PFA overnight at 4 1C,embedded to paraffin and sectioned at 5 mm and stained byhematoxylin–eosin. Immunohistochemistry on paraffin sectionswas done with Powervision+ Poly HRP histostaining kit (LeicaMicrosystems) and the antibodies were diluted in Powervisionantibody blocking solution. For detecting Plexin B2 with the TI-2Bantibody, a combination of tyramine signal amplification (TSA) andHISTOMOUSE-kit (Zymed) was used (Adams, 1992). The sectionswere counterstained with hematoxylin and mounted by 20%Mowiol 4–88 (Calbiochem).

For whole mount immunohistochemistry, the microdissectedkidneys were fixed with ice-cold methanol. For more detailedanalysis, the whole mount samples were further fixed with 4% PFAand embedded in 5% low melting temperature agarose (BioWhit-taker Molecular Applications, USA) in PBS for cutting of 8–10 mmsections with Leica VT1000S vibratome (Leica Microsystems,Germany). The immunolabelings were performed as previouslydescribed (Sariola et al., 1988). The sections were analyzed aftermounting in Vectashield with DAPI (Vector laboratories).

Whole mount in situ hybridization was performed by InSituProAutomate (Intavis, Cologne, Germany) on PFA-fixed samples asdescribed previously (Kuure et al., 2005). Non-radioactive sectionalin situ hybridization was done with Ventana Discovery in situhybridization automate (Roche Diagnostics) on paraffin sections aspreviously described (Perala et al., 2005).

2.7. Quantification of the ureteric tips

Each embryo was photographed before microdissection. Forthe branching assay the litters were stage matched by the devel-opmental stage of the hindlimb. To analyze the number of ureterictips in vivo, the E12.5 kidneys were cultured overnight (16 h)on Nuclepore filters to flatten the tissue and then fixed withcold methanol. To analyze the effect of soluble Sema4c on

ureteric branching in vitro, the E11.5 urogenital blocks were cutinto halves and the halves were cultured in media supplementedwith the elution buffer with or without 100 ng/ml soluble purifiedSema4C for 48 h and fixed with cold methanol. The number ofureteric tips was counted after staining with the anti-pan-cytoker-atin antibody. The TIFF-images were analyzed with ImageJ (http://rsb.info.nih.gov/ij/) (Rasband, 2009), inverted to a negative imageand then converted to binary image and a skeleton (Fig. 2C0) usingdefault settings. The number of tips was calculated as well asthe length of the kidneys was measured from the skeleton andthe results of the same embryonic stage were pooled. Fordetermination of the effect of Sema4C, the results werecalculated as the difference in the percentage of number ofureteric tips between kidneys treated with elution buffer with orwithout soluble Sema4c for 48 h.

Statistical analysis of the results of three groups was performedwith one-way analysis of variance followed by Tukey’s MultipleComparison Test (GraphPad Prism, GraphPad Software Ink, CA).The statistical significance (p-value) of the effect of Sema4C wasdetermined using Student’s t-test. All results are shown as averagesand standard error of mean (sem).

2.8. Proliferation and apoptosis assays

Cell proliferation was analyzed by bromodeoxyuridine (BrdU)incorporation. The pregnant female mice were injected intraper-itoneally with 100 ml/10 g BrdU (Amersham, GE Healthcare) atE12.5 and sacrificed 4 h after injection. The embryos were fixed by4% PFA, embedded in paraffin and serially sectioned. The incor-poration of BrdU was analyzed by indirect immunofluorescenceusing a monoclonal antibody to BrdU (Amersham, GE Healthcare).The samples were mounted by Vectashield containing DAPI (Vectorlaboratories) to label the nuclei. The number of BrdU-positive cellsin the tips of the ureteric epithelium was compared with the totalnumber of tip cells as revealed by DAPI staining.

Apoptosis was analyzed by ApopTag Fluorescein In SituApoptosis Detection Kit (TUNEL assay) (Millipore) according tomanufacturer’s protocol. The sections were mounted by Vecta-shield containing DAPI and photographed. A grid was placed onthe photographed sections in the Adobe Photoshop program (onesquare: 10 mm�10 mm) and the level of apoptosis was calculatedas the number of TUNEL-positive squares compared with thetotal number of squares. Statistical analysis was performed asdescribed above.

2.9. Protein isolation and immunoprecipitation

Proteins were isolated from embryonic kidneys using lysisbuffer (1% Igepal CA-630, 1% Triton-X, 10% glycerol, 2 mM EDTA/TBS pH 7.5) supplemented with 100 mM NaVO3 and Complete MiniProtease Inhibitor Cocktail Tablet (Roche). The tissue (25–30 E14.5kidneys) was homogenized in 500 ml lysis buffer using a syringeand needles (first 22 G, then 26 G). The lysate was kept on ice for30 min and then centrifuged at 4 1C to pellet the cell debris. 30 ml ofthe supernatant was used directly in SDS-PAGE and the rest wasused for immunoprecipitation. For immunoprecipitation the totallysate was precleared in solubilization buffer and 25 ml streptavi-dine beads (Cell Signalling) and the immunoprecipitation wascarried out according to the manufacturer’s instructions using theanti-mouse Plexin-B2 Biotin antibody. Proteins were eluted withLaemmli sample buffer at 100 1C and separated with 7.5% SDS-PAGE along with the aliquots of total lysate and transferred to aHybond-ECL membrane (Amersham, GE Healthcare). Immunode-tection was performed by ECL kit (Amersham TM, GE Healthcare)according to manufacturer’s instructions.

N. Perala et al. / Differentiation 81 (2011) 81–9184

3. Results

3.1. Expression of Plexin B2 protein

A polyclonal rabbit antibody, TI-2B, was produced to theintracellular domain of Plexin B2. It was shown to be specific forPlexin B2 by Western blotting and immunohistochemistry, andnot to recognize the closest family member, Plexin B1 (Suppl.Fig. 1). The expression pattern of the Plexin B2 protein inthe developing kidney recapitulates the previously publishedexpression pattern of the Plxnb2 mRNA (Perala et al., 2005).At E12.5 Plexin B2 is expressed by the ureteric epithelium, theS-shaped bodies, developing glomeruli and the cap condensates inthe MM (Fig. 1).

3.2. Plxnb2� /� kidneys are hypoplastic and express markers for

different renal structures

The deletion of the Plexin B2 gene results in exencephaly (inapproximately 90% of the homozygotes) and neonatal lethality inC57BL/6 mouse strain (Deng et al., 2007; Friedel et al., 2007). Wenow analyzed kidney morphogenesis in this strain. The kidneys ofthe Plxnb2� /� mice were consistently smaller than those of thePlxnb2+/� and wild type littermates. However, the overall size ofthe Plxnb2� /� embryos and their developmental stage did notdiffer from the Plxnb2+/� and Plxnb2+/+ littermates, as judged bythe stage of limb bud development at E12–E16.5. The difference inrenal size was first seen at E12 and remained at least to E16.5, thelast stage studied (Fig. 2A).

The Plxnb2� /� kidneys were morphologically normal with alldifferent segments of the developing secretory nephrons at E15.5(Fig. 2B–B00) and normal expression patterns of stalk and tipmarkers of the ureteric epithelium. We analyzed by immunohis-tochemistry and whole mount and sectional in situ hybridizationthe expression of markers of different renal mesenchymal, epithe-lial, vascular and stromal structures in the Plxnb2� /� kidneys,including Wilms tumour gene 1 (Wt1), Pax2, cytokeratins, VanGogh-like 2, pErk, Wnt11, Wnt9b, Sox9, Foxd1 and Retinaldehydedehydrogenase 2 (Suppl. Fig. 2 and data not shown).

3.3. Impaired ureteric branching in Plexin B2-deficient mice

To examine the size and ureteric branching of the kidneys ofPlxnb2� /� embryos, the length of the kidneys and the number oftips were quantified from skeleton images (Fig. 2C0) derived from

Fig. 1. Expression of Plexin B2 in the developing kidney. A, A0: Antibody staining wi

undifferentiated mesenchyme (arrowhead in A0) and developing glomeruli (star in A0)

A0: 50 mm.

E12.5 kidneys cultured overnight (Fig. 2C). Results from embryos ofthe same developmental stage (Fig. 2C00) were pooled together. ThePlxnb2� /� kidneys (555729.5 mm) (average7sem) (n¼9) were25% shorter along their longitudinal axis compared to the kidneysof wild type (746717.6 mm) (n¼25) (po0.001) (Fig. 2D). Plxnb2+/�

kidneys were also slightly smaller (693716.0 mm) (n¼20) thanthe wild type kidneys, but the difference was not statisticallysignificant. In Plxnb2� /� kidneys the number of ureteric tips wasdecreased by over 40% (po0.001) (Fig. 2E). The number of ureterictips in Plxnb2� /� kidneys was 1670.9, in Plxnb2+/� kidneys28.871.1 and in the wild type kidneys 28.271.2 (Fig. 2E).

Out of the 118 Plxnb2� /� embryos (E11.5–E16.5) studied,11 (9.3%) showed unilateral double ureters and kidneys (Suppl.Fig. 3), which was not seen in any of the Plxnb2+/� or wild typelittermates. The size and extent of ureteric branching of eachindividual kidney in the double kidney complex was comparable tothat of the contralateral kidney. Double ureters with double pelviswere also found in E16.5 Plxnb2� /� embryos.

The pathogenesis of the ureteric branching defect was analyzedby proliferation and apoptosis assays. Based on the number ofBrdU-positive cells in sagittal sections of E12.5 kidneys, the cells inthe tips of the ureteric epithelium in Plxnb2� /� kidneys prolifer-ated 10% less than in the Plxnb2+/� (po0.01) and 9% less than inwild type kidneys (po0.05) (Fig. 2F). The proportion of BrdU-positive cells versus the total number of ureteric tip cells was50.671.3% (n¼21) in Plxnb2� /� kidneys, 56.471.4% (n¼21) inPlxnb2+/� and 55.671.1% (n¼20) in the wild types (Fig. 2F).

To assess the degree of apoptosis in Plxnb2� /� kidneys,sections of E12.5 kidneys were analyzed by TUNEL assay. Therewere no significant differences in the level of apoptosis betweenthe wild type (10.270.5% of TUNEL-positive squares) (n¼11),Plxnb2+/� (9.770.7%) (n¼11) and Plxnb2� /� (9.470.6%)(n¼12) kidneys (Fig. 2G). In all genotypes there was no or verylittle apoptosis in the ureteric epithelium.

3.4. The branching defect is intrinsic to the ureteric epithelium

We next examined whether the branching defect in Plxnb2� /�

kidneys was due to an intrinsic defect of the ureteric epithelium orthe mesenchyme by culturing isolated UBs and separated MMs indifferentiation promoting conditions. The UBs of E12.5 Plxnb2� /�

kidneys were grown in three-dimensional Matrigel cultures sup-plemented with different growth factors for 4–5 days. Gdnf alonefailed to induce branching of Plxnb2� /� UBs or the response to theinduction was very rudimentary as compared to wild type kidneys

th TI-2B. Plexin B2 protein is expressed in the ureteric epithelium (arrow in A0),

at E12.5. B: Control staining with no primary antibody. Scale bars: A, B: 100 mm;

Fig. 2. Plxnb2� /� kidneys are hypoplastic and the ureteric branching is retarded. A: Plxnb2� /� kidneys are smaller than in wild type (+/+) littermates at E16.5. B–B00:

Histology of E15.5 Plxnb2+/+ (B), Plxnb2+/� (B0) and Plxnb2� /� (B00) kidneys. Scale bars: 100 mm. C–C00: The ureteric epithelium of kidney explants of stage-matched (C00)

E12.5 wild type, Plxnb2+/� and Plxnb2� /� embryos were cultured for 16 h and stained for pan-cytokeratin (C). The images were converted to skeletons (C0) and the number of

ureteric tips was calculated. D: The length of the kidneys (orange dashed line in C0). The Plxnb2� /� kidneys are significantly smaller than the Plxnb2+/� and wild type kidneys

(average of kidney length +/� sem). Also the Plxnb2+/� kidneys are slightly smaller than the wild type kidneys. E: The Plxnb2� /� kidneys have 40% less UB tips than the

Plxnb2+/+ or Plxnb2+/� kidneys. F: BrdU-incorporation shows that in the Plxnb2� /� kidneys cell proliferation in the tips is decreased by 9% as compared to wild type and

heterozygous kidneys. G: There is no change in the level of apoptosis in the Plxnb2� /� kidneys based on the TUNEL-assay (*po0.05; **po0.01; ***po0.001).

(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

N. Perala et al. / Differentiation 81 (2011) 81–91 85

(Fig. 3A–C). However, the response of Plxnb2� /� UBs to Gdnf wasrescued by the addition of Follistatin (an activin antagonist) andFibroblast growth factor 7 (Fgf7). In the presence of Fgf7 andFollistation, the UBs of all genotypes, including Plxnb2� /� , formedflat, ellipsoid branches (Fig. 3D–F). When Gdnf was added togetherwith Fgf7 and Follistatin, the UB branches were rounder and moreballoon-shaped than the UBs induced by Fgf7 and Follistatin alone,also in the Plxnb2� /� UBs (Fig. 3G–I).

Isolated MMs at E11.5 undergo epithelial differentiation in vitro

after inhibition of GSK3 with either 50-bromoindirubin-30-oxime(BIO) or lithium (Kuure et al., 2007). The isolated MMs from E11.5Plxnb2� /� mice responded normally to BIO by formation ofsecretory nephrons as analyzed by the nephron marker peanutagglutinin lectin (PNA), and expressed brush border antigens of the

proximal tubules (Suppl. Fig. 4C). In summary, the results fromisolated UB and MM cultures suggest that the impaired uretericbranching is due to an intrinsic defect of the epithelium.

3.5. Plexin B2 interacts with the Gdnf receptor Ret

Because the experiments with the isolated UBs showed that theresponsiveness of the isolated Plxnb2� /� UBs to Gdnf is affected, theresponse of the whole Plxnb2� /� kidneys to Gdnf was analyzed.When E11.5 urogenital explants were grown with 15 or 50 ng/mlexogenous Gdnf, the Wolffian ducts formed supernumerary buds inall genotypes (Fig. 4A–C). Thus, the response to ectopic Gdnf wasnormal in the Wolffian duct of whole Plxnb2� /� kidneys.

Fig. 3. The isolated Plxnb2� /� UBs fail to branch normally. The ureteric epithelium was isolated from the mesenchyme at E12.5 and grown in vitro in a three-dimensional

Matrigel culture. A–C: In the presence of Gdnf the wild-type (A) and Plxnb2+/� (B) UBs are able to branch, whereas the response of Plxnb2� /� UB (C) is severely impaired, even

after 4 days in culture. D–F: Fgf7 and Follistatin induce comparable branching in all genotypes after 5 days in culture. G–I: When Gdnf is combined with Fgf7 and Follistatin the

response to induction is similar in both the wild-type (G), heterozygous (H) and homozygous (I) samples at day 5 in culture. Scale bars: 100 mm.

N. Perala et al. / Differentiation 81 (2011) 81–9186

To further study Gdnf signalling in the Plxnb2� /� kidneys, themRNA expression of Gdnf, its receptors and downstream targetswere analyzed. The expression pattern of Gdnf was normal in theMM of E11.5 (Fig. 4D–F) and E12.5 (data not shown) Plxnb2� /�

embryos. Both Ret receptor tyrosine kinase (Fig. 4G–I) and the co-receptor Gfra1 (Fig. 4J–L) were expressed in the ureteric epitheliumof Plxnb2� /� kidneys at E12.75. Wnt11, a downstream target ofGdnf/Ret signalling and marker of the ureteric tips, showed anormal pattern of expression in the ureteric epithelium at E12.75(Fig. 4M–O). Additional downstream targets such as Etv5 (Fig. 4P–R), Myb (Fig. 4S–U), Etv4, Crlf1 and Cxcr4 (data not shown)were all expressed normally. Furthermore, an inhibitor of theGdnf/Ret signalling, Sprouty1, showed normal expression patternin the Plxnb2� /� urogenital area at E11.5 and E12.5 (data notshown).

As the components of Gdnf/Ret signalling did not show anydetectable alterations on the expression level, we analyzed whetherthere were some interactions at the protein level between Plexin B2and the Gdnf receptors. We immunoprecipited lysates of E14.5embryonic kidneys with a biotinylated antibody against Plexin B2.The antibody immunoprecipitated Plexin B2 from total lysate(Fig. 4V). When blotting the Plexin B2 immunoprecipitate with anantibody against Ret, a protein of 170 kDa co-immunoprecipitatedwith Plexin B2. This 170 kDa band corresponded to the glycosylated,mature form of Ret present at the cell membrane (Fig. 4V). In the totallysate, also the 150 kDa immature Ret protein was present but thisform did not co-immunoprecipitate with Plexin B2.

3.6. Sema4C increases ureteric branching through Plexin B2 in vitro

The expression of the ligand for Plexin B2, Sema4C, as well asthe expression of Plxnb2 were analyzed by whole mount in situhybridization in wild type E11.5 kidneys cultured overnightin vitro. The receptor and ligand were expressed in an overlappingpattern in the Wolffian duct, ureteric epithelium and MM(Fig. 5A).

To analyze the role of Sema4C, E11.5 wild type, Plxnb2+/� andPlxnb2� /� kidneys were cultured with 100 ng/ml of purified,soluble recombinant Sema4C for 48 h. The contralateral kidneyfrom the same embryo served as a control and was cultured in thepresence of either mock serum or the elution buffer that was usedto purify the protein. The mock serum or elution buffer did notaffect the growth of the explants. Sema4C increased the number ofureteric tips in the explants (Fig. 5B). The tip number of controlkidney was normalized to 100% because of the variation betweenstarting stages of individuals from different E11.5 litters. Theincrease of tip number in the kidney grown with Sema4C ascompared to the control kidney was calculated in percentages.Sema4C increased the number of ureteric tips in wild type kidneyexplants to 11676.7% (po0.05) (n¼8). In the Plxnb2+/� explantsthe increase in branching by Sema4C was even greater, 40% (to14078.4%) (po0.001) (n¼14). In contrast, the Plxnb2� /� kidneyexplants failed to respond to Sema4C, and the number of ureterictips compared to the control kidney did not change significantly(from 100% to 9577.9%) (n¼7) (Fig. 5B).

Fig. 4. Plexin B2 is associated with Ret in the embryonic kidney. A–C: The Wolffian

duct of E11.5 Plxnb2+/+ (A), Plxnb2+/� (B) and Plxnb2� /� (C) kidneys form

supernumerary buds in response to 50 ng/ml Gdnf. D–F: The expression pattern of

Gdnf in E11.5 Plxnb2� /� kidneys (F). Also Ret (G–I) and Gfra1 (J–L) are expressed

normally in all genotypes at E12.75. Expression patterns of Wnt11 (M–O), Etv5 (P–R)

and Myb (S–U), downstream targets of Gdnf/Ret signalling, in the Plxnb2+/+ ,

Plxnb2+/� and Plxnb2� /� kidneys at E12.5. Scale bars: 100 mm. V: Immunopre-

cipitation (IP) with an antibody against Plexin B2 (PB2) and Western blot (WB) with

an anti-Ret antibody show that Plexin B2 associates with the glycosylated, mature

form of Ret (170 kDa) in the developing E14.5 kidney. The 150 kDa Ret protein seen

in the total lysate did not coimmunoprecipitate with Plexin B2.

Fig. 5. Sema4C is expressed in the ureteric epithelium and MM and stimulates

ureteric branching through Plexin B2. A: Sema4C and Plxnb2 are both expressed in

the Wolffian duct (arrow head), ureteric epithelium (arrow) and MM (star) in E11.5

kidneys grown overnight on filters. B: E11.5 kidneys were cultured with either

control elution buffer or purified recombinant soluble Sema4C for 48 h in vitro. The

number of ureteric tips in the control kidneys was normalized to correspond to 100%

(red line). Sema4C increases the number of tips in wild type kidneys by 16%, whereas

the increase in the heterozygous kidneys is 40%. The Plxnb2� /� kidneys did not

respond to Sema4C (*po0.05; ***po0.001). (For interpretation of the references to

colour in this figure legend, the reader is referred to the web version of this article.)

N. Perala et al. / Differentiation 81 (2011) 81–91 87

3.7. Downstream targets of Plexin B2 in Plxnb2� /� kidneys

We analyzed whether the expression of Plxnb1 was altered inthe E13.5 Plxnb2� /� kidneys by sectional mRNA in situ hybridiza-tion. The expression pattern of Plxnb1 in Plxnb2� /� kidneys wasidentical to that in wild type and Plxnb2+/� kidneys (Fig. 6A–C).

Activation of Plexin B2 has been shown to result in phosphor-ylation of Met (Conrotto et al., 2004) and ErbB-2 (Deng et al., 2007;Swiercz et al., 2004). Neither the expression pattern of pMet(Fig. 6D–F) nor of pErbB-2 (Fig. 6G–I) was changed in the E13.5Plxnb2� /� kidneys compared to Plxnb2+/� or wild typelittermates.

To test whether the Plxnb2� /� UB cells are able to adhere toand spread on extracellular matrix molecules, we plated isolatedE12.5 UB cells into wells coated with either laminin (Fig. 6J–L),fibronectin, collagen or matrigel (data not shown) and let themadhere for 24 h to the wells. The spreading and adhesion ofPlxnb2� /� UB cells to the different substrates did not differfrom that of the controls. Furthermore, the structure of the actincytoskeleton stained with rhodamine phalloidin and distribution ofVinculin-positive focal adhesions in the Plxnb2� /� UB cells werenormal (Fig. 6J–L). We also analyzed the organization of the actincytoskeleton in E12.5 kidney sections. The organization of the actincytoskeleton in the stalks (data not shown) and in the Sox9-positivetips of the ureteric epithelium of Plxnb2� /� kidneys was notaltered as compared to the wild type and Plxnb2+/� controls(Fig. 6M–O).

3.8. Embryo deficient of both Plexin B1 and Plexin B2 develops kidneys

To study the possible redundant role of Plexin B1 and Plexin B2in kidney development, we analyzed embryos that lacked both

Fig. 6. The expression patterns of Plxnb1, pMet and pErbB-2 and the organization of the actin cytoskeleton in Plxnb2� /� kidneys. A–C: Expression of Plxnb1 mRNA in E13.5

kidney (left panel Plxnb2+/+ , middle Plxnb2+/� , and right panel Plxnb2� /�). D–F: The expression pattern of pErbB-2 at E13.5. G–I: The expression pattern of pMet at E13.5. J–L:

During 24 h culture, isolated UB cells of E12.5 Plxnb2� /� kidneys adhere to and spread on laminin. They show a normal pattern of actin cytoskeleton (red, stained with

rhodamine phalloidin) and focal adhesions (green, stain for Vinculin). (Blue colour indicates nuclei.) M–O: The actin cytoskeleton (red, visualized by rhodamine phalloidin) in

the tip cells (green, Sox9-positive) of the ureteric epithelium displays normal organization in E12.5 Plxnb2� /� kidneys. (Blue colour indicates nuclei). Scale bars: A–I 50 mm, J–O

20 mm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

N. Perala et al. / Differentiation 81 (2011) 81–9188

alleles of the genes or different combinations of the alleles. First westudied the kidneys of Plxnb1� /� embryos to verify the resultspublished previously (Korostylev et al., 2008). At E11.5 and E12.5the Plxnb1� /� kidneys did not differ significantly in size ordevelopmental stage from the Plxnb1+/� and wild type controls.However, at E13.5 the kidneys of Plxnb1� /� embryos were biggerwith more ureteric branches than the heterozygous littermatecontrol, which confirms the earlier results (Korostylev et al., 2008).

The double heterozygous mice (Plxnb1+/� Plxnb2+/�) and themice lacking both alleles of Plxnb1 and one allele of Plxnb2

(Plxnb1� /� Plxnb2+/�) were viable and able to produceoffspring although the litters were very small. Plxnb2� /�

combined with Plxnb1+/� as well as double homozygousdeletions were embryonic lethal. The Plxnb1+/� Plxnb2+/�

mice were crossed with Plxnb1� /�Plxnb2+/� mice and wewere able to harvest only one double homozygote (Plxnb1� /�

Fig. 7. Plxnb1/Plxnb2 double homozygote develops kidneys. Sections of a double

heterozygous (A) and double homozygous (B) kidneys at E12. The ureteric tips are

marked with arrows while the comma-shaped bodies of differentiating MM are

marked by asterisks. Note that the ureteric stalks seem to be longer in the compound

homozygote than heterozygote. Scale bars: 50 mm.

N. Perala et al. / Differentiation 81 (2011) 81–91 89

Plxnb2� /�) embryo surviving to E12 out of 19 embryos. It showedhypoplastic kidneys with ureteric branches and differentiatingmesenchyme (Fig. 7B).

4. Discussion

We show here that lack of Plexin B2 results in renal hypoplasiaand occasional double ureters and that the ligand for Plexin B2,Sema4C, increases branching of the ureteric epithelium in vitro

through Plexin B2. The renal hypoplasia of Plxnb2� /� mice iscaused by an intrinsic defect of the ureteric epithelium, which leadsto both reduced branching and proliferation. Finally, the isolatedPlxnb2� /� UBs fail to respond to Gdnf and co-immunoprecipita-tion experiments show an interaction between Plexin B2 and Ret.

Plexin B2 is expressed by both the ureteric epithelium and thecondensing MM in the developing kidney (Perala et al., 2005). ThePlxnb2� /� kidneys were smaller and had fewer branches of theureteric epithelium than the control kidneys, even though theywere morphologically normal and expressed normally markers formesenchymal, epithelial, vascular and stromal structures, as wellas Gdnf pathway components, the principal regulator of uretericbudding (Moore et al., 1996; Pichel et al., 1996; Sanchez et al., 1996)and branching (Qiao et al., 1999a). The Plxnb2� /� kidneys had 40%fewer ureteric tips than kidneys of the wild type littermates atE12.5 and the reduction in the number of tips and the smallerkidney size remained until E16.5, the last stage studied. Cellproliferation defects have been previously reported in thedeveloping central nervous system of Plxnb2� /� embryos,where neuronal precursors in the ventricular zone andcerebellum as well as in the cortical plate show reducedproliferation rate (Deng et al., 2007; Hirschberg et al., 2009).Proliferation in the ureteric epithelium in Plxnb2� /� kidneyswas reduced by 10%, which is statistically highly significant andapparently contributes to the reduction in the number of branches.As the number of ureteric tips is directly linked to the number ofinduced secretory nephrons (Michael and Davies, 2004), thiscontributes to the renal hypoplasia in Plxnb2� /� embryos.

The reduction in ureteric branching of the Plxnb2� /� kidneyscould be due to a defect in either the renal epithelium or MM. Thisquestion was addressed by culturing isolated UBs and MMs indefined conditions in vitro. Unlike isolated wild type and Plxnb2+/�

UBs, the isolated Plxnb2� /� UBs failed to branch and grow inMatrigel in the presence of Gdnf. The isolated Plxnb2� /� MM wasable differentiate into secretory nephrons when induced by theGSK3 inhibitor BIO. Taken together, these results imply that theimpaired branching in the Plxnb2� /� kidneys is due to an intrinsicdefect of the ureteric epithelium.

The Plxnb2� /� UBs responded to Fgf7 and Follistatin, alter-native inducers of UB branching and growth (Maeshima et al.,2007; Qiao et al., 1999b; Qiao et al., 2001). Fgf7 and Follistatin alsorestored the response of the isolated Plxnb2� /� UBs to Gdnf.Moreover, the Wolffian duct of whole Plxnb2� /� kidneys showednormal response to Gdnf by formation of supernumerary budssimilarly to wild type controls. These results suggest that in vivo, inthe presence of the MM expressing Fgf7 and Follistatin amongother factors, the response of the ureteric epithelium to Gdnf isintact, even in the Plxnb2� /� kidneys. The MM has recently beenshown to be an indispensable factor for the elongation anddifferentiation of the ureteric tree and to even correctdevelopmental errors in ureteric branching (Shah et al., 2009).This could be the case also in Plxnb2� /� kidneys.

The immunoprecipitation experiments suggest that Plexin B2associates with the mature, glycosylated form of Ret, implying thatPlexin B2 may contribute to the branching of the ureteric epithe-lium by directly modulating the activation or signal transduction ofRet. Thus, Ret is a new interaction partner for Plexin B2. Furtherstudies are needed to reveal the functional significance of thePlexin B2-Ret interaction. Previously, B-subfamily plexins havebeen shown to associate with and upon ligand binding to cause thephosphorylation and activation of receptor tyrosine kinases such asMet and ErbB-2 (Conrotto et al., 2004; Deng et al., 2007; Giordanoet al., 2002; Swiercz et al., 2004), which are also expressed in theureteric epithelium.

Approximately 10% of the Plxnb2� /� embryos showed uni-lateral double ureters or renal pelvises. Such structures are seen inmany genetic models with defects in ureteric branching, such as Ret

hypomorphs (de Graaff et al., 2001) and Sprouty-1-deficient mice(Basson et al., 2005) or when the MM domain expressing Gdnf isexpanded as a result of the loss of transcription factor Foxc1 (Kumeet al., 2000). The expression patterns of Gdnf, Ret, Pax2 and Sprouty-

1 were, however, normal in Plxnb2� /� kidneys, and more studiesare needed to identify the mechanism responsible for theoccasional double ureters.

Plxnb1 and Plxnb2 are expressed in overlapping patterns in theembryonic kidney (Korostylev et al., 2008; Perala et al., 2005).However, the functions of Plexin B1 and Plexin B2 during kidneydevelopment are apparently opposite. Our results verify the recentdata showing that at E13.5 the kidneys of Plxnb1� /� embryos arelarger than those of the control littermates and that they have anenhanced number of branches of the ureteric epithelium. FromE15.5 onwards they are again indistinguishable from heterozygousand wild type kidneys (Korostylev et al., 2008). In contrast, thedeletion of Plexin B2 results in small kidneys and retarded uretericbranching. The reduction in renal size is already seen in the micewith one Plxnb2 null allele. The branching defect in Plxnb2� /�

kidneys is apparent from the very first branches at E12.5 and lastsat least to E16.5, the last time point that we analyzed. The oppositeeffect of these two receptors is somewhat unexpected becausePlexin B1 and Plexin B2 share several common downstreamsignalling pathways and ligands (Conrotto et al., 2004; Denget al., 2007; Driessens et al., 2002; Masuda et al., 2004; Perrotet al., 2002; Swiercz et al., 2004). The downstream targets ofdifferent plexins deserve further studies. Due to the high

N. Perala et al. / Differentiation 81 (2011) 81–9190

embryonic lethality of the Plxnb1� /� Plxnb2� /� mice, we have sofar been able to collect only one double homozygous embryo atE12. This embryo had kidneys, which had differentiating epithelialand mesenchymal structures present, furthermore suggesting non-redundant roles for these plexins in kidney morphogenesis. Tofurther study the function of Plexin B1 and Plexin B2 in kidneydevelopment, a conditional mouse mutant of Plexin B2 is needed.

Two semaphorins, Sema3A and Sema4D, have been implicatedin the regulation of ureteric branching during kidney development.Exogenous Sema3A modulates ureteric branching in a negativefashion (Tufro et al., 2008) while exogenous Sema4D also impairsthe branching of the ureteric epithelium. The effect of Sema4D isabolished in Plxnb1� /� kidney explants, suggesting that Plexin B1is the main receptor for Sema4D in embryonic kidney (Korostylevet al., 2008). We show that exogenous soluble Sema4C increases theureteric branching in wild type and Plxnb2+/� kidney explants, butnot in the Plxnb2� /� kidneys. This indicates that in kidneydevelopment, like in granule cell progenitors of the developingcerebellum (Deng et al., 2007), Sema4C is the primary ligand forPlexin B2. The Plxnb2+/� kidneys are supersensitive to exogenousSema4C and form significantly more branches in the presence ofSema4C than the wild type kidneys. The increased sensitivity toSema4C could be explained by an altered negative feedback loop;however, further studies are needed to reveal the mechanism ofhaploinsufficiency of the Plexin B2 gene. Plexin B1 expression inPlxnb2� /� and Plxnb2+/� kidneys was normal, suggesting thatcompensatory overexpression of Plexin B1 does not explainSema4C hyperresponsiveness in Plexin B2 heterozygotes.

Signalling through plexins of the B-subfamily activates and phos-phorylates receptor tyrosine kinases Met and ErbB-2 (Conrotto et al.,2004; Deng et al., 2007; Giordano et al., 2002; Swiercz et al., 2004). Wefailed to detect any differences in the expression of phosphorylated Metor ErbB-2 by immunohistochemistry with phospho-specific antibodies,indicating that the activation of these receptors is not abolished in thePlxnb2� /� kidneys. However, we cannot exclude minor changes in theactivation of these receptors. Plexins of the B-subfamily have also beenshown to regulate the actin cytoskeleton by interacting with PDZ-RhoGEF and LARG, which subsequently activate RhoA and causeformation of actin bundles and cell rounding (Aurandt et al., 2002;Driessens et al., 2002; Hirotani et al., 2002; Perrot et al., 2002; Swierczet al., 2002). Plexins regulate the activity of integrins through R-Ras andadhesion to the extracellular matrix (Oinuma et al., 2004, 2006). Ingranule cell precursors isolated from the cerebellum, the activation ofPlexin B2 by Sema4C increases significantly the adhesion of the cells tolaminin in vitro, and this effect is abolished in granule cell precursorscollected from Plxnb2� /� embryos (Deng et al., 2007). We analyzedthe actin cytoskeleton, adhesion and formation of focal adhesionplaques in Plxnb2� /� UB cells when grown on different substrates,such as laminin and fibronectin. The UB cells of all genotypes formedfocal adhesions, adhered to and spread equally well on the differentsubstrates. This suggests that the adhesive properties of the UB cells donot depend on Plexin B2. The actin cytoskeleton in the Plxnb2� /� UBcells was also indistinguishable from the heterozygous and wild typecontrol. We analyzed the organization of the actin cytoskeleton inPlxnb2� /� E12.5 kidneys. The normal organization of the actincytoskeleton in the ureteric tip cells, especially in the apical domain,is crucial for branching morphogenesis to proceed normally (Michaelet al., 2005). In the absence of Plexin B2, the organization of theactin cytoskeleton in the tips and stalks was similar to wild type andPlxnb2+/� kidneys. This suggests that perturbations in the regulationand maintenance of the actin cytoskeleton are apparently notresponsible for the reduced epithelial branching in Plxnb2� /�

kidneys. Further studies are needed to determine what downstreampathway components in Plxnb2� /� kidneys are affected.

In conclusion, the lack of Plexin B2 in kidney developmentaffects the branching morphogenesis of the ureteric epithelium,

leading to hypoplastic kidneys and occasional double ureters, but itdoes not affect the differentiation of the nephrogenic mesenchyme.In the presence of Plexin B2, soluble Sema4C increases uretericbranching in vitro. In summary, Sema4C-Plexin B2 signallingcontributes to the ureteric branching morphogenesis, possiblythrough modulation of Ret function.

Role of the funding sources

None of the sponsors had a role in the study design; in thecollection, analysis, and interpretation of data; in the writing of thereport; or in the decision to submit the paper for publication.

Acknowledgements

We thank German Research Center for Environmental Health(GSF), Dr. Roland Friedel (Helmholtz Center Munich), Dr. MarcTessier-Lavigne (Stanford University and Genentech) and Dr. AlainChedotal (Universite Pierre et Marie Curie-Paris) for the Plxnb2� /�

mice, Dr. Aaro Miettinen (University of Helsinki) for the anti-BBantibody, Dr. Michael Wegner (University of Erlangen-Nurnberg)for the anti-Sox9 antibody and Dr. Steven Strittmatter (YaleUniversity) for the Sema4C cDNA expression construct. We thankAgn�es Vihera and Lea Armassalo for excellent technical assistanceand the staff of the animal facilities at the Meilahti and Viikkicampuses for the exceptional service. This work was financiallysupported by Sigrid Juselius Foundation, the Academy of Finland,University of Helsinki, Helsinki Graduate School in Biotechnologyand Molecular Biology (GSBM), K. Albin Johansson Foundation, TheFinnish Kidney Foundation, The Finnish Cultural Foundation andthe Swedish Cultural Foundation in Finland.

Appendix A. Supplementary data

Supplementary data associated with this article can be found inthe online version at doi:10.1016/j.diff.2010.10.001.

References

Adams, J.C., 1992. Biotin amplification of biotin and horseradish peroxidase signalsin histochemical stains. J. Histochem. Cytochem. 40, 1457–1463.

Alexander, W.S., Rakar, S., Robb, L., Farley, A., Willson, T.A., Zhang, J., Hartley, L.,Kikuchi, Y., Kojima, T., Nomura, H., Hasegawa, M., Maeda, M., Fabri, L., Jachno, K.,Nash, A., Metcalf, D., Nicola, N.A., Hilton, D.J., 1999. Suckling defect in micelacking the soluble haemopoietin receptor NR6. Curr. Biol. 9, 605–608, S1.

Artigiani, S., Barberis, D., Fazzari, P., Longati, P., Angelini, P., van de Loo, J.W.,Comoglio, P.M., Tamagnone, L., 2003. Functional regulation of semaphorinreceptors by proprotein convertases. J. Biol. Chem. 278, 10094–10101.

Artigiani, S., Conrotto, P., Fazzari, P., Gilestro, G.F., Barberis, D., Giordano, S.,Comoglio, P.M., Tamagnone, L., 2004. Plexin-B3 is a functional receptor forsemaphorin 5A. EMBO Rep. 5, 710–714.

Aurandt, J., Vikis, H.G., Gutkind, J.S., Ahn, N., Guan, K.L., 2002. The semaphorinreceptor plexin-B1 signals through a direct interaction with the Rho-specificnucleotide exchange factor, LARG. Proc. Natl. Acad. Sci. USA 99, 12085–12090.

Basson, M.A., Akbulut, S., Watson-Johnson, J., Simon, R., Carroll, T.J., Shakya, R., Gross,I., Martin, G.R., Lufkin, T., McMahon, A.P., Wilson, P.D., Costantini, F.D., Mason,I.J., Licht, J.D., 2005. Sprouty1 is a critical regulator of GDNF/RET-mediatedkidney induction. Dev. Cell 8, 229–239.

Carroll, T.J., Park, J., Hayashi, S., Majumdar, A., McMahon, A.P., 2005. Wnt9b plays acentral role in the regulation of mesenchymal to epithelial transitions under-lying organogenesis of the mammalian urogenital system. Dev. Cell 9, 283–292.

Conrotto, P., Corso, S., Gamberini, S., Comoglio, P.M., Giordano, S., 2004. Interplaybetween scatter factor receptors and B plexins controls invasive growth.Oncogene 23, 5131–5137.

Costantini, F., 2006. Renal branching morphogenesis: concepts, questions, andrecent advances. Differentiation 74, 402–421.

Costantini, F., Shakya, R., 2006. GDNF/Ret signaling and the development of thekidney. Bioessays 28, 117–127.

N. Perala et al. / Differentiation 81 (2011) 81–91 91

de Graaff, E., Srinivas, S., Kilkenny, C., D’Agati, V., Mankoo, B.S., Costantini, F., Pachnis,V., 2001. Differential activities of the RET tyrosine kinase receptor isoformsduring mammalian embryogenesis. Genes Dev. 15, 2433–2444.

Deng, S., Hirschberg, A., Worzfeld, T., Penachioni, J.Y., Korostylev, A., Swiercz, J.M.,Vodrazka, P., Mauti, O., Stoeckli, E.T., Tamagnone, L., Offermanns, S., Kuner, R.,2007. Plexin-B2, but not Plexin-B1, critically modulates neuronal migration andpatterning of the developing nervous system in vivo. J. Neurosci. 27, 6333–6347.

Dressler, G.R., 2006. The cellular basis of kidney development. Annu. Rev. Cell Dev.Biol. 22, 509–529.

Dressler, G.R., Deutsch, U., Chowdhury, K., Nornes, H.O., Gruss, P., 1990. Pax2, a newmurine paired-box-containing gene and its expression in the developingexcretory system. Development 109, 787–795.

Driessens, M.H., Olivo, C., Nagata, K., Inagaki, M., Collard, J.G., 2002. B plexins activateRho through PDZ-RhoGEF. FEBS Lett. 529, 168–172.

Fazzari, P., Penachioni, J., Gianola, S., Rossi, F., Eickholt, B.J., Maina, F., Alexopoulou, L.,Sottile, A., Comoglio, P.M., Flavell, R.A., Tamagnone, L., 2007. Plexin-B1 plays aredundant role during mouse development and in tumour angiogenesis. BMCDev. Biol. 7, 55.

Friedel, R.H., Kerjan, G., Rayburn, H., Schuller, U., Sotelo, C., Tessier-Lavigne, M.,Chedotal, A., 2007. Plexin-B2 controls the development of cerebellar granulecells. J. Neurosci. 27, 3921–3932.

Giordano, S., Corso, S., Conrotto, P., Artigiani, S., Gilestro, G., Barberis, D., Tamagnone,L., Comoglio, P.M., 2002. The semaphorin 4D receptor controls invasive growthby coupling with Met. Nat. Cell Biol. 4, 720–724.

Hirotani, M., Ohoka, Y., Yamamoto, T., Nirasawa, H., Furuyama, T., Kogo, M., Matsuya,T., Inagaki, S., 2002. Interaction of plexin-B1 with PDZ domain-containing Rhoguanine nucleotide exchange factors. Biochem. Biophys. Res. Commun. 297,32–37.

Hirschberg, A., Deng, S., Korostylev, A., Paldy, E., Costa, M.R., Worzfeld, T., Vodrazka,P., Wizenmann, A., Gotz, M., Offermanns, S., Kuner, R., 2009. Gene deletionmutants reveal a role for semaphorin receptors of the Plexin-B family inmechanisms underlying corticogenesis. Mol. Cell. Biol. 30, 764–780.

Kispert, A., Vainio, S., Shen, L., Rowitch, D.H., McMahon, A.P., 1996. Proteoglycans arerequired for maintenance of Wnt-11 expression in the ureter tips. Development122, 3627–3637.

Kolodkin, A.L., Matthes, D.J., O’Connor, T.P., Patel, N.H., Admon, A., Bentley, D.,Goodman, C.S., 1992. Fasciclin IV: sequence, expression, and function duringgrowth cone guidance in the grasshopper embryo. Neuron 9, 831–845.

Korostylev, A., Worzfeld, T., Deng, S., Friedel, R.H., Swiercz, J.M., Vodrazka, P., Maier,V., Hirschberg, A., Ohoka, Y., Inagaki, S., Offermanns, S., Kuner, R., 2008. Afunctional role for semaphorin 4D/plexin B1 interactions in epithelial branchingmorphogenesis during organogenesis. Development 135, 3333–3343.

Kume, T., Deng, K., Hogan, B.L., 2000. Murine forkhead/winged helix genes Foxc1(Mf1) and Foxc2 (Mfh1) are required for the early organogenesis of the kidneyand urinary tract. Development 127, 1387–1395.

Kuure, S., Popsueva, A., Jakobson, M., Sainio, K., Sariola, H., 2007. Glycogen synthasekinase-3 inactivation and stabilization of beta-catenin induce nephron differ-entiation in isolated mouse and rat kidney mesenchymes. J. Am. Soc. Nephrol.18, 1130–1139.

Kuure, S., Sainio, K., Vuolteenaho, R., Ilves, M., Wartiovaara, K., Immonen, T., Kvist, J.,Vainio, S., Sariola, H., 2005. Crosstalk between Jagged1 and GDNF/Ret/GFRal-pha1 signalling regulates ureteric budding and branching. Mech. Dev. 122,765–780.

Lin, J.H., Saito, T., Anderson, D.J., Lance-Jones, C., Jessell, T.M., Arber, S., 1998.Functionally related motor neuron pool and muscle sensory afferent subtypesdefined by coordinate ETS gene expression. Cell 95, 393–407.

Lu, B.C., Cebrian, C., Chi, X., Kuure, S., Kuo, R., Bates, C.M., Arber, S., Hassell, J., MacNeil,L., Hoshi, M., Jain, S., Asai, N., Takahashi, M., Schmidt-Ott, K.M., Barasch, J.,D’Agati, V., Costantini, F., 2009. Etv4 and Etv5 are required downstream of GDNFand Ret for kidney branching morphogenesis. Nat. Genet. 41, 1295–1302.

Maeshima, A., Sakurai, H., Choi, Y., Kitamura, S., Vaughn, D.A., Tee, J.B., Nigam, S.K.,2007. Glial cell derived neurotrophic factor independent ureteric bud outgrowthfrom the wolffian duct. J. Am. Soc. Nephrol. 18, 3147–3155.

Masuda, K., Furuyama, T., Takahara, M., Fujioka, S., Kurinami, H., Inagaki, S., 2004.Sema4D stimulates axonal outgrowth of embryonic DRG sensory neurones.Genes Cells 9, 821–829.

Michael, L., Davies, J.A., 2004. Pattern and regulation of cell proliferation duringmurine ureteric bud development. J. Anat. 204, 241–255.

Michael, L., Sweeney, D.E., Davies, J.A., 2005. A role for microfilament-basedcontraction in branching morphogenesis of the ureteric bud. Kidney Int. 68,2010–2018.

Moore, M.W., Klein, R.D., Farinas, I., Sauer, H., Armanini, M., Phillips, H., Reichardt,L.F., Ryan, A.M., Carver-Moore, K., Rosenthal, A., 1996. Renal and neuronalabnormalities in mice lacking GDNF. Nature 382, 76–79.

Niederreither, K., McCaffery, P., Drager, U.C., Chambon, P., Dolle, P., 1997. Restrictedexpression and retinoic acid-induced downregulation of the retinaldehydedehydrogenase type 2 (RALDH-2) gene during mouse development. Mech. Dev.62, 67–78.

Oinuma, I., Katoh, H., Negishi, M., 2006. Semaphorin 4D/Plexin-B1-mediated R-RasGAP activity inhibits cell migration by regulating beta(1) integrin activity. J. CellBiol. 173, 601–613.

Oinuma, I., Katoh, H., Negishi, M., 2004. Molecular dissection of the semaphorin 4Dreceptor plexin-B1-stimulated R-Ras GTPase-activating protein activity andneurite remodeling in hippocampal neurons. J. Neurosci. 24, 11473–11480.

Pachnis, V., Mankoo, B., Costantini, F., 1993. Expression of the c-ret proto-oncogeneduring mouse embryogenesis. Development 119, 1005–1017.

Perala, N.M., Immonen, T., Sariola, H., 2005. The expression of plexins during mouseembryogenesis. Gene Expr. Patterns 5, 355–362.

Perrot, V., Vazquez-Prado, J., Gutkind, J.S., 2002. Plexin B regulates Rho through theguanine nucleotide exchange factors leukemia-associated Rho GEF (LARG) andPDZ-RhoGEF. J. Biol. Chem. 277, 43115–43120.

Pichel, J.G., Shen, L., Sheng, H.Z., Granholm, A.C., Drago, J., Grinberg, A., Lee, E.J.,Huang, S.P., Saarma, M., Hoffer, B.J., Sariola, H., Westphal, H., 1996. Defects inenteric innervation and kidney development in mice lacking GDNF. Nature 382,73–76.

Qiao, J., Sakurai, H., Nigam, S.K., 1999a. Branching morphogenesis independent ofmesenchymal–epithelial contact in the developing kidney. Proc. Natl. Acad. Sci.USA 96, 7330–7335.

Qiao, J., Uzzo, R., Obara-Ishihara, T., Degenstein, L., Fuchs, E., Herzlinger, D., 1999b.FGF-7 modulates ureteric bud growth and nephron number in the developingkidney. Development 126, 547–554.

Qiao, J., Bush, K.T., Steer, D.L., Stuart, R.O., Sakurai, H., Wachsman, W., Nigam, S.K.,2001. Multiple fibroblast growth factors support growth of the ureteric bud buthave different effects on branching morphogenesis. Mech. Dev. 109, 123–135.

Rasband, W.S., 2009. ImageJ Available at: /http://rsb.info.nih.gov/ij/S.Sainio, K., 2003. Experimental methods for studying urogenital development. In: Vize,

P., Woolf, A., Bard, J. (Eds.), The Kidney. Academic Press, London, UK, pp. 327.Sanchez, M.P., Silos-Santiago, I., Frisen, J., He, B., Lira, S.A., Barbacid, M., 1996. Renal

agenesis and the absence of enteric neurons in mice lacking GDNF. Nature 382,70–73.

Sariola, H., Holm, K., Henke-Fahle, S., 1988. Early innervation of the metanephrickidney. Development 104, 589–599.

Saxen, L., 1987. Organogenesis of the Kidney. Cambridge University Press,Cambridge.

Schedl, A., 2007. Renal abnormalities and their developmental origin. Nat. Rev.Genet. 8, 791–802.

Shah, M.M., Tee, J.B., Meyer, T., Meyer-Schwesinger, C., Choi, Y., Sweeney, D.E., Gallegos,T.F., Johkura, K., Rosines, E., Kouznetsova, V., Rose, D.W., Bush, K.T., Sakurai, H., Nigam,S.K., 2009. The instructive role of metanephric mesenchyme in ureteric budpatterning, sculpting, and maturation and its potential ability to buffer uretericbud branching defects. Am. J. Physiol. Renal Physiol. 297, F1330–1341.

Stolt, C.C., Lommes, P., Sock, E., Chaboissier, M.C., Schedl, A., Wegner, M., 2003. TheSox9 transcription factor determines glial fate choice in the developing spinalcord. Genes Dev. 17, 1677–1689.

Suvanto, P., Wartiovaara, K., Lindahl, M., Arumae, U., Moshnyakov, M., Horelli-Kuitunen, N., Airaksinen, M.S., Palotie, A., Sariola, H., Saarma, M., 1997. Cloning,mRNA distribution and chromosomal localisation of the gene for glial cell line-derived neurotrophic factor receptor beta, a homologue to GDNFR-alpha. Hum.Mol. Genet. 6, 1267–1273.

Swiercz, J.M., Kuner, R., Behrens, J., Offermanns, S., 2002. Plexin-B1 directly interactswith PDZ-RhoGEF/LARG to regulate RhoA and growth cone morphology. Neuron35, 51–63.

Swiercz, J.M., Kuner, R., Offermanns, S., 2004. Plexin-B1/RhoGEF-mediated RhoA activa-tion involves the receptor tyrosine kinase ErbB-2. J. Cell Biol. 165, 869–880.

Tamagnone, L., Artigiani, S., Chen, H., He, Z., Ming, G.I., Song, H., Chedotal, A.,Winberg, M.L., Goodman, C.S., Poo, M., Tessier-Lavigne, M., Comoglio, P.M., 1999.Plexins are a large family of receptors for transmembrane, secreted, and GPI-anchored semaphorins in vertebrates. Cell 99, 71–80.

Torres-Vazquez, J., Gitler, A.D., Fraser, S.D., Berk, J.D., Van, N.P., Fishman, M.C., Childs,S., Epstein, J.A., Weinstein, B.M., 2004. Semaphorin-plexin signaling guidespatterning of the developing vasculature. Dev. Cell. 7, 117–123.

Toyofuku, T., Zhang, H., Kumanogoh, A., Takegahara, N., Suto, F., Kamei, J., Aoki, K.,Yabuki, M., Hori, M., Fujisawa, H., Kikutani, H., 2004a. Dual roles of Sema6D incardiac morphogenesis through region-specific association of its receptor,Plexin-A1, with off-track and vascular endothelial growth factor receptortype 2. Genes Dev. 18, 435–447.

Toyofuku, T., Zhang, H., Kumanogoh, A., Takegahara, N., Yabuki, M., Harada, K., Hori,M., Kikutani, H., 2004b. Guidance of myocardial patterning in cardiac develop-ment by Sema6D reverse signalling. Nat. Cell Biol. 6, 1204–1211.

Trokovic, R., Jukkola, T., Saarimaki, J., Peltopuro, P., Naserke, T., Vogt Weisenhorn,D.M., Trokovic, N., Wurst, W., Partanen, J., 2005. Fgfr1-dependent boundary cellsbetween developing mid- and hindbrain. Dev. Biol. 278, 428–439.

Tufro, A., Teichman, J., Woda, C., Villegas, G., 2008. Semaphorin3a inhibits uretericbud branching morphogenesis. Mech. Dev. 125, 558–568.

Winberg, M.L., Noordermeer, J.N., Tamagnone, L., Comoglio, P.M., Spriggs, M.K.,Tessier-Lavigne, M., Goodman, C.S., 1998. Plexin A is a neuronal semaphorinreceptor that controls axon guidance. Cell 95, 903–916.

Wong, A.W., Brickey, W.J., Taxman, D.J., van Deventer, H.W., Reed, W., Gao, J.X.,Zheng, P., Liu, Y., Li, P., Blum, J.S., McKinnon, K.P., Ting, J.P., 2003. CIITA-regulatedplexin-A1 affects T-cell-dendritic cell interactions. Nat. Immunol. 4, 891–898.

Worzfeld, T., Puschel, A.W., Offermanns, S., Kuner, R., 2004. Plexin-B family membersdemonstrate non-redundant expression patterns in the developing mousenervous system: an anatomical basis for morphogenetic effects of Sema4Dduring development. Eur. J. Neurosci. 19, 2622–2632.

Yamamoto, M., Suzuki, K., Okuno, T., Ogata, T., Takegahara, N., Takamatsu, H., Mizui, M.,Taniguchi, M., Chedotal, A., Suto, F., Fujisawa, H., Kumanogoh, A., Kikutani, H., 2008.Plexin-A4 negatively regulates T lymphocyte responses. Int. Immunol. 20, 413–420.

Zhang, S., Lin, Y., Itaranta, P., Yagi, A., Vainio, S., 2001. Expression of Sprouty genes 1,2 and 4 during mouse organogenesis. Mech. Dev. 109, 367–370.

Zou, Y.R., Kottmann, A.H., Kuroda, M., Taniuchi, I., Littman, D.R., 1998. Function of thechemokine receptor CXCR4 in haematopoiesis and in cerebellar development.Nature 393, 595–599.