The guanine nucleotide exchange factor C3G is necessary for the … · hypomorphic C3G allele,...

INTRODUCTION

Cells receive cues from the extracellular matrix and fromgrowth factor receptors to activate cellular functions such ascell division, cell migration, cell differentiation and cellsurvival. These signals can be transmitted through adaptorproteins and guanine nucleotide exchange factors to smallGTPases. Small GTPases activate, through mediators, anumber of downstream targets, including mitogen-activatedprotein kinase (MAPK) and Jun kinases. These signallingpathways are often perturbed in cancer cells causing excessivecell proliferation, abnormal cell migration and aberrant celladhesive properties.

C3G is a guanine nucleotide exchange factor transmittingsignals from the extracellular matrix or growth factors tomembers of the Ras family of GTPases. Specifically C3G hasbeen shown to respond to B and T cell receptor activation(Reedquist et al., 1996; Smit et al., 1996), insulin and EGF(Okada and Pessin, 1997), integrin binding (Arai et al., 1999;Uemura and Griffin, 1999), erythropoietin and interleukin 3(Nosaka et al., 1999), and interferon γ (Alsayed et al., 2000).Activation of C3G can lead to the activation of Jun kinase(Mochizuki et al., 2000; Tanaka et al., 1997) or MAP kinase(Nosaka et al., 1999). C3G stimulated guanine nucleotideexchange predominantly on Rap1 and, to a lesser extend, onR-Ras (Ichiba et al., 1997; Ohba et al., 2001). However, C3G-mediated activation of Jun kinase appears to occur solelythrough activation of R-Ras (Mochizuki et al., 2000). Therough eye phenotype caused by constitutive over-activation ofDC3G in D. melanogasterin vivo can be suppressed by

reducing the gene dosage of Ras1, rolled (MAPK) and Rap1(Ishimaru et al., 1999). Taken together, these reports suggestan activation of both, the Ras/Rap/MAPK-pathway and theRas/Jun kinase-pathway, by C3G.

C3G was initially isolated as a protein binding to the amino-terminal SH3 domain of the adaptor protein, chicken retroviruskinase (Crk) (Knudsen et al., 1994; Tanaka et al., 1994). C3Gcontains a carboxy-terminal CDC25-type catalytic domainstimulating the guanine nucleotide exchange on Ras-familymembers (Tanaka et al., 1994). Centrally located are fourproline-rich domains that confer binding to the amino-terminalSH3 domain of Crk and the adaptor protein Grb2 (Kirsch etal., 1998; Tanaka et al., 1994). Amino-terminal to these is oneproline-rich region that confers binding to the SH3 domain ofthe focal adhesion molecule p130Cas(Kirsch et al., 1998).

Crk and p130Cas have been shown to be integral parts offocal adhesions. They form a complex with the docking proteinpaxillin and link filamentous actin of the cytoskeleton tointegrin receptors (reviewed by Turner, 2000; Schaller, 2001).Crk, CrkL and Grb2 have been shown to enhance C3G activity(Ichiba et al., 1997). Crk does so by recruiting C3G to thecytoplasmic membrane. The SH2 domain of Crk that confersbinding to paxillin or the adaptor protein Shc is essential formembrane recruitment of C3G (Ichiba et al., 1997). Paxillin,in turn, binds to integrins via src and focal adhesion kinase(reviewed by Schaller, 2001).

A recently reported mutation of the C3G(Grf2) gene in miceresulted in early post-implantation lethality thereby excludingthe study of the role of C3G later in development (Ohba etal., 2001). We have generated a mouse strain carrying a

355Development 130, 355-367 © 2003 The Company of Biologists Ltddoi:10.1242/dev.00217

The Ras signalling pathway has major roles in normal cellfunction and oncogenesis. C3G is a guanine nucleotideexchange factor for members of the Ras family of GTPases.We generated a mouse strain with a hypomorphic C3Gallele. C3Ggt/gt mutant embryos died of vascular defectsaround E11.5 due to haemorrhage and vascular integritydefects. Vascular supporting cells did not developappropriately. C3G-deficient fibroblasts responded toPDGF-BB abnormally, exhibited cell adhesion defects andlacked paxillin and integrin-β1-positive cell adhesions.

In contrast, integrin-β3-positive cell adhesions formednormally. These results show that C3G is required for (1)vascular myogenesis, (2) the formation of paxillin- andintegrin β1-positive, but not integrin β3-positive, celladhesions and (3) normal response to PDGF, necessary forvascular myogenesis.

Key words: C3G, Ras signalling, Vascular development, Focaladhesions, Integrins, PDGF

SUMMARY

The guanine nucleotide exchange factor C3G is necessary for the formation

of focal adhesions and vascular maturation

Anne K. Voss 1, Peter Gruss 2 and Tim Thomas 1

1Development and Neurobiology, Walter and Eliza Hall Institute of Medical Research, Parkville, Victoria 3050, Australia2Max-Planck-Institute of Biophysical Chemistry, Goettingen, Germany*Author for correspondence (e-mail: [email protected])

Accepted 11 October 2002

356

hypomorphic C3G allele, C3Ggt. Here we reveal threepreviously unreported functions of C3G. Firstly, C3G isrequired for the formation or stabilisation of integrin β1- andpaxillin-positive focal adhesions. Secondly, C3G is necessaryfor a normal response to PDGF. Thirdly, C3G is crucial forvascular maturation. The observed defects in cell adhesion andPDGF-response may explain the blood vessel developmentdefect.

MATERIALS AND METHODS

Generation of the mutant C3Ggt allele, identification andcharacterisation of the trapped locusThe promoter-less reporter construct pGT1.8geo[kindly provided byW. Skarnes (Skarnes et al., 1995)] was electroporated into the parentalmurine embryonic stem (ES) cell line MPI-II (Voss et al., 1997) asdescribed previously (Voss et al., 1998b).

Total RNA was isolated from ES cells heterozygous for theinsertion of pGT1.8geointo the murine genome (clone F82). 5′ rapidamplification of cDNA ends was performed as described previously(Voss et al., 1998b) using oligonucleotides complementary tothe lacZ gene of pGT1.8geoas gene-specific primers. RACEproducts were cloned into pGemTand sequenced. The gene-specificsequence obtained was used to screen databases. To identify thesite of integration within the C3G locus a series of probes wereused for Southern analysis including several probes cloned by PCRfrom intron 1. These probes were generated through the use ofthe Celera Discovery System and Celera Genomics’ associateddatabases.

To test for the presence of C3GmRNA 3′ of the gene trap insertionsite, RT-PCR was performed as described previously (Voss et al.,1998b) on total RNA isolated from homozygous embryos usingoligo(dTTP) for the reverse transcriptase reaction and using 5′ TGTTTT CCG CCT TGT TAT GTT CCT 3′ as the forward and 5′ AGTGGC AGC GGC AGA CCT CAG 3′ as the reverse primer. Thisreaction generated a product complementary to bases 268-647 of amouse EST (GenBank accession no. BG916265) spanning exons 21-24 of the C3G locus. To test for the presence of full-length proteincoding mRNA in homozygous embryos, RT-PCR representing allcoding exons was performed using primers in exon 21 (5′ TGT TTTCCG CCT TGT TAT GTT CCT 3′) and exon 1 (5′ GCCCGG-AAATGTCCGGCAAGATCG 3′).

Generation of the C3Ggt mutant mouse line andgenotyping of mice and embryosChimeras were produced as describes previously (Voss et al., 1998b)and mated to wild-type females of the mouse strains C57Bl/6 andRandom Swiss. In timed pregnant female mice, noon of the day onwhich the vaginal plug was observed was termed day 0.5 of gestation(E0.5). Genotyping was carried out using DNA isolated from tailbiopsies or embryonic membranes by C3G gene-specific Southernanalysis using an [α-32P]dATP-labelled probe near the insertion siteof pGT1.8geoin intron 1 (probe 1, Fig. 1A). Probe 1 comprised 203bp sequence located 22 kb 3′ of exon 1. It was PCR generated using5′ GGA GCA GGC AGG ACT CCT CAC T 3′ and 5′ TCT CCC ATCTAA GAG GCT CTG G 3′ as primers.

Whole embryo and histological analysisEmbryos were dissected from the uteri, placed in phosphate-bufferedsaline (PBS), viewed and photographed using a low-magnificationstereomicroscope (Zeiss) and a digital camera (Axiocam, Zeiss).Embryonic membranes were used for genotyping by C3G-specificSouthern analysis. For histological analysis, embryos were fixed in4% paraformaldehyde, dehydrated, infiltrated and embedded inparaffin. 8 µm serial sections were cut, deparaffinised and stained with

Haematoxylin and Eosin using standard techniques. Slides wereviewed and photographed using a compound microscope (Zeiss) anda digital camera. Photographs of mutant and control embryos werecompared in detail.

Northern analysis and in situ hybridisationA lacZ- or C3G-specific [α-32P]dATP-labelled probe 3′ of theinsertion site (probe 2, Fig. 1A) was used for northern analysis of totalRNA isolated from embryos, adult tissues and cell lines and, in vitrotranscribed and labelled with 35S-labelled UTPαS, for in situhybridisation as described previously (Voss et al., 2000). Probe 2 wasPCR generated using 5′ CGC CCC TGC TTC TCA ATG GTT AGCC 3′ and 5′ CCA TCC AAG AAG GGA AAG CCA GCC G 3′ andwas complementary to 431 bases spanning exons 2-5 of the C3Glocus. Probe 2 was cloned into pBluescript KS+ and sequenced.

Western analysisLysates of primary murine embryonic fibroblasts (MEFs) isolatedfrom E10.5 embryos of heterozygous intercrosses were separated by8% to 16% SDS-PAGE, electrotransferred to membrane (ImmobilonP), blocked in 1% BSA, 0.1% Tween 20 PBS, incubated with primaryrabbit polyclonal antibody (1:200 or 1:1000, Santa Cruz), washed,incubated with secondary horseradish peroxidase-conjugatedantibody (1:3000 to 1:20000, Biorad), washed and detected bychemiluminescence (Pierce). Densitometry was carried out using acomputing densitometer (Molecular Dynamics).

Immunohistochemistry and immunocytochemistryHorseradish peroxidase-immunohistochemistry was performed asdescribed previously (Thomas et al., 2000). In addition, horseradishperoxidase-immunohistochemistry was performed on cryosections forthe detection of platelet and endothelial cell adhesion molecule 1(PECAM 1). Fluorescence immunohistochemistry was carried outusing standard techniques. In brief, after treatment with 0.25%gelatine and 0.1% Triton X-100 in PBS paraformaldehyde-fixed oracetone-fixed cryosections (PECAM 1 and smooth muscleα-actin(SMαA) staining) or deparaffinised sections (SMαA and all otherantigens) were treated with intervening PBS washes with firstantibody solution, second fluorescent antibody solution and thencounter-stained with 0.1 µg/ml bis-benzimide in 0.25% gelatine inPBS (Molecular Probes). The sections were then mounted withMowiol, viewed and photographed using a fluorescence microscope(Zeiss) and a digital camera. Immunocytochemistry was carried outas described previously (Voss et al., 2000).

The first antibodies and dilutions used were mouse anti-smoothmuscle α-actin (SMαA; Sigma, 1:400), rat anti-mouse CD31(PECAM 1; Pharmingen, 1:200 for cryosections and 1:20 for wholemounts), rabbit anti-nidogen (kindly provided by M. Dziadek,1:200), biotinylated hamster anti-rat CD29 (Integrin β1 chain,Pharmingen, 1:250), mouse anti-chick paxillin (BD Transduction,1:250), rabbit anti-human integrin β3 (Chemicon, 1:20), mouse anti-chicken vinculin (Sigma, 1:50). Secondary antibodies andstreptavidin used were biotinylated goat anti-rat (Vector, 1:150),biotinylated horse anti-mouse and rabbit (Vector, 1:50), Alexa Fluor546 goat anti-mouse (Molecular Probes, 1:1000), Alexa Fluor 488goat anti-mouse (Molecular Probes, 1:1000), TRITC goat anti-rabbit(Jackson, 1:200), TRITC-streptavidin (Southern Biotechnology,1:50).

Isolation of primary murine embryonic fibroblasts and cellculture, PDGF treatment, cell adhesion and cell migrationassays and statisticsPrimary murine embryonic fibroblasts (MEFs) were isolatedessentially as described previously (Voss and Thomas, 2001)omitting the glass bead dissociation step. Briefly, E10.5 embryoswere transferred individually into 250 µl of trypsin/EDTA solution,incubated for 5 minutes at 37°C followed by mechanical

A. K. Voss, P. Gruss and T. Thomas

357C3G in cell adhesions and vascular myogenesis

dissociation by pipetting and plated into 6-well tissue cultureplates in Dulbecco’s minimum essential medium with 10% foetalbovine serum (DMEM 10% FBS). The cells were used at passage 3and 4. Ninety-six-well plates or cover slip inserts for 24-well plateswere coated for 1 hour with 10 µg/ml human plasma fibronectin(Roche), 10 µg/ml gelatine (Sigma) or 10 µg/ml natural mouselaminin (Life Technology). Cells were plated at subconfluent density(20,000 cells per well, 24-well plate) with or without 0.1% or 10%FBS.

For PDGF treatment, cells were serum starved for 16 hours andthen treated with or without 2 ng/ml or 10 ng/ml platelet-derivedgrowth factor-AA (PDGF-AA) or PDGF-BB (both recombinanthuman; Pepro Tech). Numbers of actin rings were counted pertreatment in more than 500 cells per cultures isolated from 3 C3Ggt/gt

homozygous and 3 wild-type embryos in duplicates and compared byχ2 test. Results are given as mean ± standard error of the mean withP values.

For adhesion assays 7,000 and 14,000 cells were plated into 96-well plates. After 1 hour or 2 hours, non-adherent cells were washedoff twice with PBS and cells were fixed with 3% paraformaldehyde,2% sucrose at room temperature for 5 minutes. Adherent cellswere stained with 0.5% Crystal Violet in 20% methanol. Plateswere washed in H2O and air-dried. Stained adherent cells weresolubilised in 0.1 M citric acid in 50% ethanol, pH 4.2. The opticalabsorbance was determined at 570 nm using an ELISA reader(SpectraFluor Plus, Tecan). Results were compared by two-factoranalysis of variance with genotype and animal as the two factors.Results are given as mean ± standard error of the mean with Pvalues.

Cell migration assays were performed as described previously(Lallemand et al., 1998). In brief, a rectangular wound was createdin monolayers of cells on the day they became confluent using a 20-200 µl micropipette tip. Dislodged cells were washed off andmedium was replaced. Images of the monolayer wounds weretaken at the time of wounding (0 hours) and 7 hours later. The areafree of cells was measured using image analysis software (NIHImage 1.62). The area covered by cells within 7 hours afterwounding was calculated. Results were compared by two-factoranalysis of variance with genotype and experiment as the twofactors. Results are given as means ± standard error of the mean withP values.

RESULTS

Generation and characterisation of the C3Ggt mutantalleleWe generated a mutant C3G allele, C3Ggt, in a gene trapscreen in murine embryonic stem cells (Voss et al., 1998b). 5′RACE of RNA isolated from clone F82 revealed the fusion ofcoding sequences of the gene trap construct pGT1.8geo(Skarnes et al., 1995) to exon 1 (Fig. 1A) of the murine C3Glocus (base 297 of the sequence with GenBank accession no.NM_504050) (Zhai et al., 2001). The protein deduced fromthis fusion mRNA contained the first 19 amino acids of C3Gfused to β-galactosidase and neomycin phosphotransferase(Fig. 1A). This fusion protein lacked the carboxy-terminalCDC25-type catalytic domain, the 4 central Crk SH3-bindingdomains and the central p130CasSH3-binding domain. Thesedomains have been shown to be essential for C3G function(Kirsch et al., 1998). No functional domains have beenidentified within the first 19 amino acids of C3G (Ichiba et al.,1999). Therefore no abnormal function is to be expected fromthis fusion protein.

The gene trap insertion was located in the 59 kb long intron1, approximately 16 kb 3′ of exon 1 (Fig. 1A). Southernanalysis of EcoRV-digested wild-type, heterozygous andhomozygous DNA showed that the insertion of the gene trapconstruct caused a shift in the size of the third EcoRV fragmentof intron 1 (14830 bases to 26997 bases from exon 1, Fig. 1B).Southern analysis of the 5′ and 3′ adjacent EcoRV fragmentsof intron 1 revealed no polymorphism in heterozygous orhomozygous DNA, confirming the absence of rearrangementsof the C3G locus outside of the gene trap insertion site. Lossof sequences 5′ or 3′ of the gene trap insertion was ruled outby 5′RACE (see above) and RT-PCR (below).

Northern analysis of total RNA isolated from E10.5embryos wild type, heterozygous or homozygous for theC3Ggt mutant allele showed C3G/β-gal/neofusion mRNA ofthe expected size of about 4.7 kb and no detectable mRNA 3′

Fig. 1. The C3G locus and the mutated allele C3Ggt.(A) The 124 kb C3G locus comprises 24 exons. Thestart of translation is encoded by exon 1. The gene trapinsertion occurred in intron 1. Intronic probe 1 is usedfor Southern analysis of genomic DNA (probe 1). Themajor product of the native locus is a 1095 aa protein

with a carboxy-terminal CDC25-like catalytic domain (black box) and 4 internal Crk-SH3 domain-binding sites (black stripes) and one p130cas-SH3 binding site (grey box). Probe 2 used for northern analysis and in situ hybridisation of paraffin sections of embryos is indicated (probe 2).The gene trap disrupts the coding region after the first 19 codons (black arrowhead). The mutated allele, C3Ggt, codes for a fusion between thefirst 19 aa of C3G, β-galactosidase (β-gal) and neomycin phosphotransferase (neoR). (B) Southern analysis of genomic DNA isolated fromembryos of heterozygous intercrosses hybridised with probe 1 as indicated in A. Bands of 12 kb and 16 kb for the wild-type and the C3Ggt

mutant allele, respectively. Homozygous mutant, mt; heterozygous, ht; wild type, wt.

358

of the gene trap insertion site (Fig. 2A). Likewise radioactivein situ hybridisation on homozygous and wild-type E11.5embryos did not reveal mRNA 3′ of the gene trap insertion(compare Fig. 3K and L with I and J). However, limitingdilution RT-PCR revealed C3G mRNA 3′ of the gene trapinsertion in homozygous mutant embryos at less than 1% ofwild-type levels (Fig. 2B). This 3′ mRNA was generatedpresumably by splicing around the gene trap insertion inintron 1. RT-PCR using primers to exons 21 and 1 revealedthe presence of C3G mRNA containing all coding exons inhomozygous mutant embryos (Fig. 2C). We have documentedthis form of splicing for one of our other gene trap mousestrains previously (Voss et al., 1998a). Finally, westernanalysis of total primary embryonic fibroblast lysates showedtwo prominent C3G bands in wild-type and heterozygouslysates. Instead of these, two weak bands were visible inhomozygous lysate (Fig. 2D). Densitometrically theseamounted to less than 5% of the normal amount of C3Gprotein. In conclusion, the C3Ggt allele generates less than 1%

mRNA containing all coding exons and no more than 5% C3Gprotein and, therefore, is a hypomorphic rather than a nullallele.

Expression of C3G in mid-gestation embryos andadult tissuesIt was previously reported that the C3G gene is expressedubiquitously in Drosophila melanogaster(Ishimaru et al.,1999) and in human foetal and adult tissues (Tanaka et al.,1994). Northern analysis of adult mouse tissues showed thatC3G is expressed at low levels ubiquitously, with higher levelsof expression in testes (not shown). The activity pattern of theβ-galactosidase reporter inserted into the C3G locus (Fig. 3A-D) and radioactive in situ hybridisation of sections of E11.5and E12.5 embryos confirmed this result (Fig. 3E-L). Notably,reporter activity and expression of the endogenous locus wereobserved in all cells including vascular endothelial cells andblood vessel surrounding cell, both in large (Fig. 3I) and smallblood vessels (Fig. 3D,J).


Fig. 2. The mRNA and proteinproducts of the wild-type andthe mutant C3Galleles.(A) Northern analysis of totalRNA isolated from E10.5embryos of C3Ggt/+

heterozygous intercrosses. 10µg of total RNA were loadedper lane. The genotype of theembryos is indicated above, aswild type (wt), heterozygousmutant (ht) and homozygousmutant (mt) for the gene trapmutation in the C3G locus.(Upper panel) Hybridisationwith the C3Gprobe 2 asindicated in Fig. 1A. (Middlepanel) hybridisation with alacZ-specific probe. (Lowerpanel) ethidium bromidestaining of the 18S rRNA. Notethe absence of detectable C3GmRNA in homozygous mutantembryos by this method. ThelacZprobe detected a fusionmRNA of the expected size of4.7 kb (169 bases 5′ UTR andcoding sequence 5′ of the genetrap insertion of the C3G locusand 4.42 kb mRNA product ofpGT1.8geopluspolyadenylation tail). (B) RT-PCR of total RNA isolated from macroscopically normal E10.5 embryos of heterozygous intercrosses. Oligo-dTTP was used to generate cDNA. Lanes 1, 2, 3, 4 and 5: cDNA of wild-type embryos used diluted 1:10 (1), 1:50 (2), 1:100 (3), 1:500 (4),1:1000 (5). Lanes 6, 7, 8, 9 and 10: cDNA of homozygous C3Ggt/gt embryo was used diluted 1:10 (6), 1:50 (7), 1:100 (8), 1:500 (9) 1:1000(10). Lane 11: control without cDNA template. M=100 bp ladder, intense band=500 bp. Note that even the 1:1000 diluted wild-type RNAyielded a prominent RT-PCR product, whereas the amount of product yielded from 1:10 and 1:50 dilution of the homozygous RNA was small.(C) PCR of undiluted homozygous mutant (mt) and wild-type cDNA (wt) as in B, but using primers spanning exons 1-21 representing allcoding exons of the C3G locus. Note the presence of a small amount of product from homozygous template indicating the generation of normalprotein coding mRNA from the homozygous mutant allele. Lanes 1 and 3 are controls. M=1 kb ladder; intense band=5 kb. (D) Western analysisof cell lysates of primary embryonic fibroblasts isolated from E10.5 homozygous, heterozygous and wild-type littermates. 40 µg of lysate wereloaded per lane. Genotypes are wild type (wt), homozygous mutant (mt) and heterozygous (ht). C3G protein bands were detected with an anti-C3G antibody in the expected position (compare to Posern et al., 2000). Note the two prominent C3G bands in wild-type and heterozygous celllysates. The two weak bands visible in homozygous lysate are likely to be residual C3G protein. They comprise no more than 5% of the normallevel of C3G protein as determined by densitometry.


The C3Ggt/g t mutant phenotype in vivoWe did not recover any homozygous C3Ggt/gt mutant offspringfrom heterozygous inter-crosses at weaning. Among 361embryos (36 litters) recovered between E9.5 and E14.5, 26%were homozygous, 49% heterozygous for the C3Ggt allele and25% wild-type at the C3G locus (Table 1 and Fig. 1B). Ninepercent of the implanted embryos were resorbed at an earlystage and could not be genotyped. Given the Mendeliandistribution of the C3Ggt allele, these early resorptions arelikely to involve all three genotypes. The majority of thehomozygotes recovered at E9.5 were phenotypically normal(Table 2). At E10.5, one third of the homozygotes showedphenotypic abnormalities or were dead. At E11.5 four fifthswere abnormal or dead. At E13.5 and 14.5 all homozygoteswere either dead or highly abnormal.

The majority of the C3Ggt/gt mutant embryos died betweenE10.5 and E11.5 exhibiting haemorrhage into the lumen of the

neural tube (Fig. 4A-C). Mutant embryos dying between E13.5and E14.5 showed massive subcutaneous oedema orhaemorrhagic oedema (Fig. 4D-F). Most likely hypovolaemiccardiovascular failure was the cause of death. The vascularsystem defects are the subject of this report. Apart from thedefects observed in the vascular system the homozygous

Fig. 3.Expression pattern of the C3G gene. (A-D) β-galactosidase reporter activity in whole heterozygous (A) and wild-type E11.5 embryo(B), and in a 16 µm paraffin sagittal section of an E11.5 heterozygote embryo (C) and in a small cephalic blood vessel at E12.5 (D).(E-L) Radioactive in situ hybridisation using an antisense C3Griboprobe (E,F, probe 2, Fig. 1A) and a sense control probe (G,H) on paraffinsections of E12.5 wild-type embryos and the antisense probe on sections of an E11.5 wild type (I,J) or homozygous mutant embryo (K,L).Ubiquitous expression of the C3G locus is shown by the reporter activity and by in situ hybridisation. Note specifically presence of reporteractivity and endogenous gene activity in all cells including blood vessel endothelial cells and blood vessel surrounding cells in both small (D,J)and large blood vessels (I). Also note the absence of silver grains in the sections of homozygous embryos (K,L). Arrows point to blood vesselsurrounding cells, arrowheads to endothelial cells (D,I,J). Bar represents 70 µm in A,B,C,E,F,G,H and 10 µm in I,J,K,L.

Table 1. Distribution of the C3Ggt allele among offspring ofheterozygous intercrosses

Embryonic stage +/+ gt/+ gt/gt

E9.5 10 30 14E10.5 33 54 29E11.5 20 40 24E12.5 11 14 10E13.5 8 21 6E14.5 6 19 12

Total 88 (25%) 178 (49%) 95 (26%)

360

embryos also showed defects in the nervous system (A. K. V.and T. T., unpublished).

Histological analysis of serial sections of 19 homozygousembryos and 19 controls at E9.5, E10.5 and E11.5 revealedthat haemorrhage initiated from small cephalic blood vesselsmost often in the vicinity of the hindbrain neural epithelium(Fig. 4G-I), occasionally near the olfactory epithelium. Smallblood vessels burst leading to an accumulation of blood in theinterstitial space until the neural epithelium gave way andruptured. The blood then drained into the lumen of the neuraltube, probably causing hypovolaemia leading to death.

Mutant C3G small cephalic blood vessels lack bloodvessel supporting cellsBlood vessel maturation occurs in mice between E10.5 andE11.5 (Takahashi et al., 1996). Blood vessel maturationinvolves the recruitment of supporting cells of mesenchymaland other origin that differentiate into pericytes and smoothmuscle cells (reviewed by Carmeliet, 2000). This process lendsmechanical stability to blood vessels.

In order to test if the fragility in C3Ggt/gt blood vessels wasdue to a lack of blood vessel maturation we examined the

expression of smooth muscle α-actin (SMαA) in 8 pairs ofC3G homozygous mutant embryos and littermate controls atE10.5 and E11.5. We found few or no SMαA-positive cellsaround small cephalic blood vessels in homozygous mutantembryos (compare Fig. 5A with B and C with D). Those fewSMαA-positive cells that were found did not form a tight ringaround the blood vessels as seen in the controls. Rather theywere loosely associated with the blood vessels and with theirenvironment. They had a rounded appearance with minimalcontacts to neighbouring cells. Larger blood vessels, e.g. thedorsal aorta or the basilar artery, did acquire SMαA-positivecells. However, the coating of the larger blood vessels was lesscomplete in C3G mutant embryos than in controls (basilarartery in Fig. 5G compared to H). Significantly, the dorsal aortawas deficient in SMαA-staining cells in its dorsal aspectswhich are known to develop blood vessel supporting cells laterthan the ventral aspects (Takahashi et al., 1996). The mutantvascular endothelial cells expressed platelet and endothelialcell adhesion molecule 1 (PECAM 1) normally and secretednidogen normally (Fig. 5E,F,I,J,K,L). Structure and arborationof blood vessels appeared normal in embryos stained forPECAM 1 as whole mounts (not shown). These data suggestedthat the primary defect in C3G mutant vasculature waslocalised to vascular supporting cells. Vascular endothelialcells, in contrast, appeared to be functioning normally.

C3G mutant primary murine embryonic fibroblastsexhibit an abnormal response to PDGF-BBThe initial stimulus for recruitment of vascular supporting cellshas been proposed to be an increase in shear force experiencedby blood vessel endothelial cells due to a gradual increase inblood pressure and flow as the embryos grow and their heartfunction becomes more efficient. It has been hypothesised that,in response to the increase in shear force, the endothelial cellsrelease PDGF, which signals to perivascular cells (Risau,1997). These then migrate to and along the blood vessels and


Table 2. Change in phenotype of homozygous C3Ggt/gtmutant embryos with developmental age

Phenotype Phenotypic Embryonic normal abnormalities Deadage gt/gt gt/gt gt/gt

E9.5 13 1 –E10.5 18 8 3E11.5 5 14 5E12.5 1 5 4E13.5 – 3* 3E14.5 – 3* 9

*Phenotypic abnormalities incompatible with further survival.

Fig. 4.C3Ggt/gt mutantphenotype. Homozygousmutant embryos at E11.5(A-C), E13.5 (D,E), E14.5(F, dead), and (G-I)Haematoxylin and Eosin-stained paraffin sections atE11.5. Note the formation ofhaemorrhages in the majorityof cases (E10.5-E11.5), whichinitiated near the hindbrainneuroepithelium (arrows inA,B,G). These haemorrhagesenlarged (H) and finally theneuroepithelium ruptured(arrow in I). In other cases,embryos exhibited massivesubcutaneous oedema andhaemorrhagic oedema atE13.5 (arrows in D,E). Barrepresents 77 µm in A-C, 150µm in D,E, 190 µm in F, 106µm in G, 66 µm in H and 80µm in I.


differentiate into supporting cells. It appears that PDGF-BB isrequired for recruitment of pericytes to capillaries, but is notrequired for recruitment of PDGFRβ-positive cells to largeblood vessels (Leveen et al., 1994; Lindahl et al., 1997).

In order to study in more detail the possible phenotypicaberration of mesenchymal cells, one source of blood vesselsupporting cells, we isolated primary murine embryonicfibroblasts (MEFs) from 3 C3Ggt/gt and 3 wild-type littermatecontrol E10.5 embryos. We exposed them to PDGF, which isone of the signals that mesenchymal cells surrounding bloodvessels would receive from endothelial cells in vivo.Immortilised fibroblast cell lines have been reported torespond to PDGF-BB with a prominent reorganisation of actinfilaments into actin rings. The mutant MEFs showed a muchhigher frequency of actin ring formation than control cells inresponse to PDGF-BB, but not to PDGF-AA (compare Fig.6A,B with C and D). Mutant MEFs showed not only one actinring more frequently, they also often formed multiple rings percell (Fig. 6B) or actin rings spanning the whole perimeter ofthe cell (Fig. 6A). In contrast, the majority of control MEFsdid not form actin rings. In 2% of these cells one small actin

ring was seen (Fig. 6C). After exposure to 10 ng/ml PDGF-BB, actin rings occurred with a frequency of 0.33 per cell inmutant MEFs versus 0.02 in control MEFs (P

362

SMαA-positive cells appeared to be only loosely attached totheir environment (see above). To examine the formation ofcell adhesions in C3G mutant cells we stained mutant andcontrol MEFs for paxillin and vinculin. The number and thesize of paxillin-positive focal adhesions were greatly reducedin C3G mutant MEFs (Fig. 6E compared with F). Likewise,the number and size of vinculin-positive focal adhesions wasreduced (Fig. 6G compared with H). The fact that the mutantcells generated a few small paxillin-positive focal points maybe explained by the presence of residual wild-type C3Ggenerated from the mutant allele. Alternatively, C3G may beimportant in the stabilisation of paxillin-positive focaladhesions rather than the initial establishment. Paxillin doesnot bind directly to C3G. Rather they both bind to the adaptorprotein Crk. Paxillin binds to the SH2 domain and C3G to theamino-terminal SH3 domain of Crk (Senechal et al., 1998).Therefore, the paucity of paxillin-positive focal adhesions inC3Ggt/gt mutant MEFs suggests that the C3G molecule may beessential for the assembly of the protein complex containing,

among other proteins, C3G, Crk and paxillin. Alternatively,signalling through C3G and activation of downstream cascadesmay be important for the stabilisation of focal adhesions.

C3G is necessary for cell adhesion to laminin andgelatine and regulation of cell migrationTo study the effects of the C3Ggt mutation on cell adhesion,homozygous mutant and control MEFs were plated ontolaminin, gelatine or fibronectin in the absence of serum or ontogelatine or fibronectin in the presence of 10% serum. Theplating efficiency of C3Ggt/gt mutant was significantly reducedon laminin and gelatine with and without serum (Fig. 7A). Incontrast, cell adhesion to fibronectin was not affected by theC3G mutation. Over a period of 2 days the majority of themutant cells grown on laminin or gelatine in the absence ofserum rounded up and lifted off the plate. However, the mutantcells grown on fibronectin remained attached, although theydid not spread in a normal manner (without serum shown inFig. 7B). These data suggested that C3G was required for

binding to specific components of theextracellular matrix and for cell spreading ingeneral.

In monolayer wounding assays of isolates ofprimary fibroblasts from 3 C3G mutant E10.5embryos and 3 littermate control embryos weobserved a significant increase in cell migrationactivity in the mutant cells (Fig. 7C,D). Anincrease in cell motility had been observedpreviously in C3G-deficient fibroblasts (Ohbaet al., 2001). The requirement of C3G for celladhesion and the regulation of cell migrationmay not be unrelated events.

C3G is required for integrin β1 clusterformationIn order to explore the effects of the C3Ggt

mutation on integrin expression or


Fig. 6.PDGF response and paxillin and vinculindistribution in C3Gmutant MEFs. FITC-phalloidinstaining for filamentous actin of homozygous C3Gmutant MEFs (A,B) or wild-type MEFs (C), platedonto gelatine-coated coverslips in 24-well plates inDME plus 10% FBS, serum-starved for 16 hoursand treated for 10 minutes with 10 ng/ml PDGF-BB. Note the formation of rings of filamentousactin. Actin rings were significantly more numerous(B) and larger (A) in C3Gmutant MEFs ascompared to wild-type MEFs (C). (D) Graphicrepresentation of the PDGF-BB response of C3Gmutant (gt/gt) vs. wild-type MEFs. For a


distribution we stained mutant and control MEFs forintegrin β1 and paxillin (Figs 8 and 9) or integrin β3 andpaxillin (Fig. 9). Control cells showed a punctate stainingfor integrin β1 (Fig. 8B,H; Fig. 9F) and β3 (Fig. 9B,D).Both integrin β1- and β3-positive foci coincided withpaxillin staining (Fig. 8F,L; Fig. 9K; and data not shown).The integrin β1 foci were large and sparse, whereas theintegrin β3 foci were more numerous and generally smaller.C3Ggt/gt mutant cells exhibited a marked reduction in thenumber of integrin β1 foci (Fig. 8A compared with B, andG with H). In contrast integrin β3 foci appeared to beunaffected by the C3Ggt mutation (Fig. 9A-D). Integrin β1and paxillin double staining visualised the highly abnormalmorphology of C3Ggt/gt mutant cells in the absence ofserum, particularly on laminin (Fig. 8E) and gelatine (notshown) and, to a lesser extend, on fibronectin (Fig. 8K).Integrin β1 immunostaining was markedly abnormal in themutant cells. Instead of the punctate staining distributedevenly across the cell as seen in the control cells, integrin

β1 appeared to form abnormal aggregates usually near thenucleus. Abnormal aggregates of integrin β1 were mostpronounced on laminin (Fig. 8A), but also obvious ongelatine (not shown) and fibronectin (Fig. 8G).

To enable us to examine C3Ggt/gt mutant cells in a healthierstate we grew cells for 2 days on gelatine in the presence of10% FBS before starving them in 0.1% serum overnight. Inaddition, we grew cells in 10% serum on tissue culture plasticbefore plating them for 40 minutes onto fibronectin in thepresence of 10% serum. In both cases the cells attached, spreadmore efficiently and exhibited a healthier morphology thanwithout serum. However, integrin β1 immunostaining wasabnormal even under these conditions. Instead of the evenlydistributed punctate staining seen in the control cells (Fig. 9F),mutant cells exhibited a few or no integrin β1-positive foci(Fig. 9E). The few integrin β1-positive foci sometimescoincided with paxillin staining in the C3Gmutant cells (Fig.9J compared with K). These data suggested that C3G wasinstrumental in the assembly or stabilisation of integrin β1

Fig. 7.C3Gmutant MEFs showedimpaired cell adhesion and spreadingand enhanced cell migration.(A,B) Cells were plated in serum-freemedium onto substrates as indicated.(A) One hour after plating, non-adherent cells were removed bywashing. Attached cells were stainedand counted by photometricabsorption. C3Gmutant MEFs showeda 66% reduction in adhesion to laminin(*, P

364

clusters. Although adhesion to laminin was most severelyaffected, adhesion and spreading on other substrates was alsoabnormal. The observed abnormalities indicated that C3G isimportant as an integral part of the complex formation aroundintegrin β1, but not integrin β3. Alternatively, signallingthrough C3G may be important for the stabilisation of integrinβ1 clusters.

DISCUSSION

We have shown that C3G was essential for fibroblast celladhesion, for the formation of paxillin- and integrin β1-, butnot integrin β3-positive focal adhesions, for a normal responseto PDGF and for blood vessel maturation. Migration ordifferentiation (or both) of the perivascular cells was disturbedby the C3Gmutation. This led to haemorrhages in the majorityof the cases (80%). In the remaining 20% it led to disruptedvascular integrity with resulting oedema. The difference in theC3G mutant phenotype observed by us and by Ohba and co-workers is probably attributable to the ameliorating effect ofthe less than 1% residual normal mRNA observed in ourC3Ggt/gt embryos. Expression of normal mRNA at only 5% ofthe normal level has been shown previously to lead to a

disproportionally large correction of a mutant phenotype(Dorin et al., 1996).

Our hypothesis is that in C3G mutants PDGF signals fromblood vessel endothelial cells did not elicit a normal responsein perivascular cells. PDGF-B mutant embryos have beenreported to fail to develop normal numbers of vascular smoothmuscle cells in microvasculature (Lindahl et al., 1997), whereC3Ggt/gt mutants also show defects in smooth muscle cells. Incontrast, larger blood vessels develop smooth muscle cells atleast to some extent in PDGF-B, PDGF-Rβ and C3G mutantembryos (Leveen et al., 1994; Lindahl et al., 1997; Soriano,1994) (and our data). However, the C3G mutant phenotypemanifests itself earlier and shows a wider variety of defectsthan mutations affecting the PDGF signalling pathway only.This is presumably due to the fact that C3G is required forother signalling pathways, such as integrin signalling.

Integrins mediate interactions between cells andextracellular matrix components such as laminin andfibronectin. Integrins are heterodimers consisting of an α anda β subunit. There are a large number of α and β subunits, manyof which bind multiple substrates (Hynes, 1992). Cellspreading requires the cytoplasmic domain of the β subunits(Berrier et al., 2000; Ylanne et al., 1993) and involvesactivation of the Ras signalling pathway (Berrier et al., 2000).


Fig. 8. Integrin β1 and paxillin distribution in C3Gmutant MEFs cultured without serum. Integrin β1 (A,B,G,H) and paxillin (C,D,I,J)immunostaining of MEFs plated on laminin (A-F) or fibronectin (G-L). (E,F,K,L) Merged images with nuclear counterstain bis-benzimide.Note the punctate staining of integrin β1 in wild-type MEFs on both substrates (B,H), which was lost in MEFs lacking normal C3Gexpression(A,G). In wild-type MEFs integrin β1 focal points were also paxillin-positive. Arrowheads in B,D,F and H,J,L point to prominent examples. Incontrast, C3Gmutant MEFs on laminin and fibronectin exhibited an aggregation of integrin β1 near the nucleus (A,G). Paxillin was poorlydistributed in C3Gmutant MEFs on laminin (C,D) and more normally distributed in mutant cells on fibronectin (I,J). A few foci of integrin β1and paxillin co-localisation were observed in C3Gmutant MEFs (A,C,E, arrowhead). However, paxillin- and integrin β1-positive foci were byfar more numerous in the controls on both laminin (F) and on fibronectin (L). Bar represents 14 µm in all panels.


In this context the requirement of C3Gspecifically for integrin β1 focal pointswas particularly interesting. In contrastto integrin β1, integrin β3 foci formedwithout or with very little C3G.Although integrin β1 is normallyinvolved, adhesion to fibronectin canbe mediated by integrin β3. In contrast,adhesion to laminin and collagen relieson integrin β1 for the β subunit of theintegrin heterodimers (Hynes, 1992).Accordingly, C3Gmutant cells, whichform integrin β3-positive celladhesions, but are deficient in integrinβ1-positive cell adhesions, adhere tofibronectin, but only poorly to lamininor gelatine.

3T3 cell adhesion to fibronectin,laminin, collagen I and vitronectinresults in CrkL-mediated C3Gphosphorylation and, presumably,activation (de Jong et al., 1998).Adhesion of other cell types is at leastpartly mediated by C3G (Arai et al.,2001; Arai et al., 1999). While thesestudies used cell adhesion toextracellular matrix components astheir end point, we showed here thatC3G was crucial for the assembly orstabilisation of integrin-positive focal adhesions. Moreover, weshowed that lack of C3G affected specifically integrin β1, andnot integrin β3. Interestingly, the reported phenotype of C3Gnull mutant mice (Ohba et al., 2001) is similar to that ofintegrin β1 null mutants (Fassler and Meyer, 1995; Stephenset al., 1995). Taking our data into consideration it is possiblethat the peri-implantation lethality of C3Gnull mutants may in

part be caused by a deficiency in integrin β1-mediated celladhesion.

Focal adhesions are large extracellular matrix-inducedclusters of integrins linked to the ends of large actin bundles.One of the proteins making the link is paxillin (Turner, 2000;Schaller, 2001). C3G deficiency caused a greatly reducednumber of much smaller paxillin-positive focal adhesions. The

Fig. 9. Integrin β3 and β1 distribution inC3Gmutant MEFs cultured with serum.(A-D) Integrin β3 distribution wasunaffected by the C3Gmutation in theabsence (A,B) or the presence (C,D) ofserum. Integrin β3 foci were morenumerous and smaller than integrinβ1-positive foci (D,F). (E-K) We investigatedintegrin β1 and paxillin distribution inMEFs plated in the presence of serum,because the morphology of C3GmutantMEFs was highly abnormal without serumon all substrates (see Fig. 8). AlthoughC3Gmutant MEFs appeared healthy whenplated onto gelatine-coated cover slips inthe presence of serum, integrin β1 fociwere similarly reduced in number(compare E and F) as in culturs withoutserum (compare Fig. 8A with B and 8Gwith H). Only very few integrin β1-positive foci coincided with paxillinstaining (J, yellow) as compared tocontrols (K). Arrowheads point toexamples. Bar represents 14.5 µm in allpanels.

366

reduction in focal adhesions is most likely the cause of reducedcell attachment properties and failure of cell spreading.Previously, it had been shown that C3G binds to Crk and it wasknown that Crk in turn binds to paxillin (Knudsen et al., 1994;Salgia et al., 1995; Schaller, 2001; Tanaka et al., 1994). Weshowed here for the first time that C3G is crucial for theformation and/or re-enforcement of focal adhesions. Lack offocal adhesions has been observed in fibroblasts expressing anegative regulator of Src. Consistent with our finding, over-expression of C3G reverses this phenotype (Li, 2002). Thistogether with our data suggests that signalling through C3Grather than the C3G molecule itself is important for theformation and/or stabilisation of focal adhesions.

The loss-of-function phenotype observed in our study showsconsiderable overlap with mutant phenotypes caused by loss offunction of other members of the same signalling pathway.Besides activation through Crk-C3G, members of the Rasfamily of GTPases can be activated through thephosphotyrosine docking molecule ShcA, the adaptor proteinGrb2 and the guanine nucleotide exchange factors Sos1 andSos2. Embryos lacking ShcA succumb to cardiovasculardefects between E10.5 and E11.5 (Lai and Pawson, 2000).Similar to C3G mutants, in ShcAmutant embryos the maturevasculature is not established efficiently. Reduced staining forsmooth muscle α-actin and impaired cell-cell and cell-extracellular matrix adhesions were observed in the ShcAmutant embryos (Lai and Pawson, 2000). The loss-of-functionphenotype of Sos1was reported as cardiovascular defects,haemorrhage and death between E11.5 and E12.5 (Wang et al.,1997) or as death due to placental defects between E9.5 andE11.5 (Qian et al., 2000). Sos2mutant mice are viable (Estebanet al., 2000). Similarities between the loss-of-functionphenotypes of the Crk-C3G and the Shc-Sos aspects of thesignalling pathway were somewhat unexpected. These twoaspects of the Ras-signalling pathway had been observed to actantagonistically in vitro (Okada et al., 1998). In contrast, theobserved similarities in the loss-of-function phenotypes wouldsuggest synergistic action, at least with respect to blood vesselmaturation.

Taken together our data show that C3G is essential fornormal PDGF signalling and for normal response to theextracellular matrix environment. C3G is critical for theprocess of vascular myogenesis. These results place C3G in acentral role for cell-cell and cell-matrix interactions duringblood vessel maturation.

We would like to thank W. Skarnes for providing the pGT1.8geogene trap construct and M. Dziadek for the anti-nidogen antisera. Wegratefully appreciate the excellent technical assistance of C. Cowled,S. Mihajlovic, E. Loza and A. Conrad. We thank V. Foletta and H.Cooper for fruitful discussions. This project was funded by the Walterand Eliza Hall Institute, the Max Planck Society and Amgen Inc.

REFERENCES

Alsayed, Y., Uddin, S., Ahmad, S., Majchrzak, B., Druker, B. J., Fish, E.N. and Platanias, L. C. (2000). IFN-gamma activates the C3G/Rap1signaling pathway. J. Immunol.164, 1800-1806.

Arai, A., Nosaka, Y., Kanda, E., Yamamoto, K., Miyasaka, N. and Miura,O. (2001). Rap1 is activated by erythropoietin or interleukin-3 and isinvolved in regulation of beta1 integrin-mediated hematopoietic celladhesion. J. Biol. Chem.276, 10453-10462.

Arai, A., Nosaka, Y., Kohsaka, H., Miyasaka, N. and Miura, O.(1999).CrkL activates integrin-mediated hematopoietic cell adhesion through theguanine nucleotide exchange factor C3G. Blood93, 3713-3722.

Berrier, A. L., Mastrangelo, A. M., Downward, J., Ginsberg, M. andLaFlamme, S. E. (2000). Activated R-ras, Rac1, PI 3-kinase andPKCepsilon can each restore cell spreading inhibited by isolated integrinbeta1 cytoplasmic domains. J. Cell Biol.151, 1549-1560.

Carmeliet, P. (2000). Mechanisms of angiogenesis and arteriogenesis. Nat.Med.6, 389-395.

de Jong, R., van Wijk, A., Heisterkamp, N. and Groffen, J.(1998). C3G istyrosine-phosphorylated after integrin-mediated cell adhesion in normal butnot in Bcr/Abl expressing cells. Oncogene17, 2805-2810.

Dorin, J. R., Farley, R., Webb, S., Smith, S. N., Farini, E., Delaney, S. J.,Wainwright, B. J., Alton, E. W. and Porteous, D. J. (1996). Ademonstration using mouse models that successful gene therapy forcystic fibrosis requires only partial gene correction. Gene Ther.3, 797-801.

Esteban, L. M., Fernandez-Medarde, A., Lopez, E., Yienger, K., Guerrero,C., Ward, J. M., Tessarollo, L. and Santos, E.(2000). Ras-guaninenucleotide exchange factor sos2 is dispensable for mouse growth anddevelopment. Mol. Cell. Biol.20, 6410-6413.

Fassler, R. and Meyer, M.(1995). Consequences of lack of beta 1 integringene expression in mice. Genes Dev.9, 1896-1908.

Hedberg, K. M., Bengtsson, T., Safiejko-Mroczka, B., Bell, P. B. andLindroth, M. (1993). PDGF and neomycin induce similar changes in theactin cytoskeleton in human fibroblasts. Cell Motil. Cytoskeleton24, 139-149.

Hynes, R. O.(1992). Integrins: versatility, modulation, and signaling in celladhesion. Cell 69, 11-25.

Ichiba, T., Hashimoto, Y., Nakaya, M., Kuraishi, Y., Tanaka, S., Kurata,T., Mochizuki, N. and Matsuda, M. (1999). Activation of C3G guaninenucleotide exchange factor for Rap1 by phosphorylation of tyrosine 504. J.Biol. Chem.274, 14376-14381.

Ichiba, T., Kuraishi, Y., Sakai, O., Nagata, S., Groffen, J., Kurata, T.,Hattori, S. and Matsuda, M. (1997). Enhancement of guanine-nucleotideexchange activity of C3G for Rap1 by the expression of Crk, CrkL, andGrb2. J. Biol. Chem.272, 22215-22220.

Ishimaru, S., Williams, R., Clark, E., Hanafusa, H. and Gaul, U.(1999).Activation of the Drosophila C3G leads to cell fate changes andoverproliferation during development, mediated by the RAS-MAPKpathway and RAP1. EMBO J.18, 145-155.

Kirsch, K. H., Georgescu, M. M. and Hanafusa, H.(1998). Direct bindingof p130(Cas) to the guanine nucleotide exchange factor C3G. J. Biol. Chem.273, 25673-25679.

Knudsen, B. S., Feller, S. M. and Hanafusa, H.(1994). Four proline-richsequences of the guanine-nucleotide exchange factor C3G bind with uniquespecificity to the first Src homology 3 domain of Crk. J. Biol. Chem.269,32781-32787.

Lai, K. M. and Pawson, T. (2000). The ShcA phosphotyrosine dockingprotein sensitizes cardiovascular signaling in the mouse embryo. Genes Dev.14, 1132-1145.

Lallemand, D., Ham, J., Garbay, S., Bakiri, L., Traincard, F., Jeannequin,O., Pfarr, C. M. and Yaniv, M. (1998). Stress-activated protein kinases arenegatively regulated by cell density. EMBO Journal.17, 5615-5626.

Leveen, P., Pekny, M., Gebre-Medhin, S., Swolin, B., Larsson, E. andBetsholtz, C. (1994). Mice deficient for PDGF B show renal,cardiovascular, and hematological abnormalities. Genes Dev.8, 1875-187.

Li, L., Okura, M. and Imamoto, A. (2002). Focal adhesions require catalyticactivity of SRC family kinases to mediate integrin-matrix adhesion. Mol.Cell. Biol. 22, 1203-1217.

Lindahl, P., Johansson, B. R., Leveen, P. and Betsholtz, C.(1997). Pericyteloss and microaneurysm formation in PDGF-B-deficient mice. Science277,242-245.

Mochizuki, N., Ohba, Y., Kobayashi, S., Otsuka, N., Graybiel, A. M.,Tanaka, S. and Matsuda, M.(2000). Crk activation of JNK via C3G andR-Ras. J. Biol. Chem.275, 12667-12671.

Nosaka, Y., Arai, A., Miyasaka, N. and Miura, O.(1999). CrkL mediatesRas-dependent activation of the Raf/ERK pathway through the guaninenucleotide exchange factor C3G in hematopoietic cells stimulated witherythropoietin or interleukin-3. J. Biol. Chem.274, 30154-30162.

Ohba, Y., Ikuta, K., Ogura, A., Matsuda, J., Mochizuki, N., Nagashima,K., Kurokawa, K., Mayer, B. J., Maki, K., Miyazaki, J. et al. (2001).Requirement for C3G-dependent Rap1 activation for cell adhesion andembryogenesis. EMBO J.20, 3333-3341.



Okada, S., Matsuda, M., Anafi, M., Pawson, T. and Pessin, J. E.(1998).Insulin regulates the dynamic balance between Ras and Rap1 signaling bycoordinating the assembly states of the Grb2-SOS and CrkII-C3Gcomplexes. EMBO J.17, 2554-2565.

Okada, S. and Pessin, J. E.(1997). Insulin and epidermal growth factorstimulate a conformational change in Rap1 and dissociation of the CrkII-C3G complex. J. Biol. Chem.272, 28179-28182.

Posern, G., Rapp, U. R. and Feller, S. M.(2000). The Crk signaling pathwaycontributes to the bombesin-induced activation of the small GTPase Rap1in Swiss 3T3 cells. Oncogene19, 6361-6368.

Qian, X., Esteban, L., Vass, W. C., Upadhyaya, C., Papageorge, A. G.,Yienger, K., Ward, J. M., Lowy, D. R. and Santos, E.(2000). The Sos1and Sos2 Ras-specific exchange factors: differences in placental expressionand signaling properties. EMBO J.19, 642-654.

Reedquist, K. A., Fukazawa, T., Panchamoorthy, G., Langdon, W. Y.,Shoelson, S. E., Druker, B. J. and Band, H.(1996). Stimulation throughthe T cell receptor induces Cbl association with Crk proteins and the guaninenucleotide exchange protein C3G. J. Biol. Chem.271, 8435-8442.

Risau, W. (1997). Mechanisms of angiogenesis. Nature386, 671-674.Salgia, R., Uemura, N., Okuda, K., Li, J. L., Pisick, E., Sattler, M., de Jong,

R., Druker, B., Heisterkamp, N. and Chen, L. B.(1995). CRKL linksp210BCR/ABL with paxillin in chronic myelogenous leukemia cells. J.Biol. Chem.270, 29145-29150.

Schaller, M. D. (2001). Paxillin: a focal adhesion-associated adaptor protein.Oncogene20, 6459-6472.

Senechal, K., Heaney, C., Druker, B. and Sawyers, C. L.(1998). Structuralrequirements for function of the Crkl adapter protein in fibroblasts andhematopoietic cells. Mol. Cell. Biol.18, 5082-5090.

Skarnes, W. C., Moss, J. E., Hurtley, S. M. and Beddington, R. S.(1995).Capturing genes encoding membrane and secreted proteins important fordevelopment. Proc. Natl. Acad. Sci. USA92, 6592-6596.

Smit, L., van der Horst, G. and Borst, J. (1996). Sos, Vav, and C3Gparticipate in B cell receptor-induced signaling pathways and differentiallyassociate with Shc-Grb2, Crk, and Crk-L adaptors. J. Biol. Chem.271, 8564-8569.

Soriano, P. (1994). Abnormal kidney development and hematologicaldisorders in PDGF beta-receptor mutant mice. Genes Dev.8, 1888-1896.

Stephens, L. E., Sutherland, A. E., Klimanskaya, I. V., Andrieux, A.,Meneses, J., Pedersen, R. A. and Damsky, C. H.(1995). Deletion of beta1 integrins in mice results in inner cell mass failure and peri-implantationlethality. Genes Dev.9, 1883-1895.

Takahashi, Y., Imanaka, T. and Takano, T.(1996). Spatial and temporalpattern of smooth muscle cell differentiation during development of the

vascular system in the mouse embryo. Anat. Embryol. (Berl).194, 515-526.

Tanaka, S., Morishita, T., Hashimoto, Y., Hattori, S., Nakamura, S.,Shibuya, M., Matuoka, K., Takenawa, T., Kurata, T. and Nagashima,K. (1994). C3G, a guanine nucleotide-releasing protein expressedubiquitously, binds to the Src homology 3 domains of CRK and GRB2/ASHproteins. Proc. Natl. Acad. Sci. USA91, 3443-3447.

Tanaka, S., Ouchi, T. and Hanafusa, H.(1997). Downstream of Crk adaptorsignaling pathway: activation of Jun kinase by v-Crk through the guaninenucleotide exchange protein C3G. Proc. Natl. Acad. Sci. USA94, 2356-2361.

Thomas, T., Voss, A. K., Chowdhury, K. and Gruss, P.(2000). Querkopf,a MYST family histone acetyltransferase, is required for normal cerebralcortex development. Development127, 2537-2548.

Turner, C. E. (2000). Paxillin and focal adhesion signalling. Nature Cell Biol.2, E231-236.

Uemura, N. and Griffin, J. D. (1999). The adapter protein Crkl links Cbl toC3G after integrin ligation and enhances cell migration. J. Biol. Chem.274,37525-37532.

Voss, A. K. and Thomas, T.(2001). Identification of novel genes by gene trapmutagenesis. Methods Mol. Biol.175, 377-396.

Voss, A. K., Thomas, T. and Gruss, P.(1997). Germ line chimeras fromfemale ES cells. Exp. Cell Res.230, 45-49.

Voss, A. K., Thomas, T. and Gruss, P.(1998a). Compensation for a gene trapmutation in the microtubule associated protein 4 locus by alternativepolyadenylation and alternative splicing. Dev. Dyn.212, 258-266.

Voss, A. K., Thomas, T. and Gruss, P.(1998b). Efficiency assessment of thegene trap approach. Dev. Dyn.212, 171-180.

Voss, A. K., Thomas, T., Petrou, P., Anastassiadis, K., Scholer, H.and Gruss, P. (2000). Taube nuss is a novel gene essential for thesurvival of pluripotent cells of early mouse embryos. Development127,5449-5461.

Wang, D. Z., Hammond, V. E., Abud, H. E., Bertoncello, I., McAvoy, J. W.and Bowtell, D. D.(1997). Mutation in Sos1 dominantly enhances a weakallele of the EGFR, demonstrating a requirement for Sos1 in EGFRsignaling and development. Genes Dev.11, 309-320.

Ylanne, J., Chen, Y., O’Toole, T. E., Loftus, J. C., Takada, Y. and Ginsberg,M. H. (1993). Distinct functions of integrin alpha and beta subunitcytoplasmic domains in cell spreading and formation of focal adhesions. J.Cell Biol. 122, 223-233.

Zhai, B., Huo, H. and Liao, K. (2001). C3G, a guanine nucleotide exchangefactor bound to adapter molecule c-Crk, has two alternative splicing forms.Biochem. Biophys. Res. Commun.286, 61-66.

The guanine nucleotide exchange factor C3G is necessary for the … · hypomorphic C3G allele,...

Documents

Transcript of The guanine nucleotide exchange factor C3G is necessary for the … · hypomorphic C3G allele,...