Jour

nal o

f Cel

l Sci

ence

SHORT REPORT

CIIA negatively regulates the Ras–Erk1/2 signaling pathwaythrough inhibiting the Ras-specific GEF activity of SOS1

Hyun Sub Hwang*, Sang Gil Hwang*, Kyoung-Wan Yoon, Je-Hyun Yoon, Kyung-Hye Rohand Eui-Ju Choi`

ABSTRACT

Son of sevenless 1 (SOS1) is a Ras-specific guanine-nucleotide-

exchange factor (GEF) that mediates intracellular signaling

processes induced by receptor tyrosine kinases. In this study, we

show that CIIA (also known as VPS28) physically associates with

SOS1 and thereby inhibits the GEF activity of SOS1 on Ras, which

prevents the epidermal growth factor (EGF)-induced activation of

the Ras–Erk1/2 pathway. Furthermore, CIIA inhibited cyclin D1

expression, as well as DNA, synthesis in response to EGF.

Intriguingly, CIIA failed to inhibit the Ras-specific GEF activity of

Noonan-syndrome-associated SOS1 mutants (M269R, R552G,

W729L and E846K). Taken together, our results suggest that CIIA

functions as a negative modulator of the SOS1–Ras signaling

events initiated by peptide growth factors including EGF.

KEY WORDS: CIIA, Ras, SOS1, VPS28

INTRODUCTIONSmall GTPases play a pivotal role in intracellular signaling

processes initiated by ligand-bound receptor tyrosine kinases

(RTKs), which mediate diverse cellular activities, including cell

proliferation and differentiation (Scita et al., 2000). Activities of

small GTPases are regulated by cycling between inactive GDP-

and active GTP-bound states, and this regulation is achieved by

guanine-nucleotide-exchange factors (GEFs), GTPase-activating

proteins (GAPs), and guanine-nucleotide-dissociation inhibitors

(GDIs). Son of sevenless 1 (SOS1) is a dual GEF for the Ras and

Rac1 GTPases (Nimnual et al., 1998). SOS1 consists of two

tandem histone fold (HF) domain, a diffuse B cell lymphoma

homology (DH) domain, and a pleckstrin homology (PH)

domain in the N-terminal region, a Ras-binding region (REM

domain) and a Ras catalytic domain (Cdc25 domain) in the

central region, and a proline-rich domain in the C-terminal

region (Bar-Sagi, 1994; Corbalan-Garcia et al., 1998). The

Cdc25 domain is responsible for the Ras-specific GEF activity

of SOS1, whereas the DH domain is necessary for the Rac1-

specific GEF activity of the protein (Bar-Sagi, 1994; Das et al.,

2000; Scita et al., 2001). The REM domain is important for an

allosteric interaction between SOS1 and Ras-GTP, which

induces the activation of the Cdc25 domain (Margarit et al.,

2003). The proline-rich C-terminal region of SOS1 interacts with

growth factor receptor bound protein 2 (Grb2) (Bar-Sagi, 1994).

The Grb2–SOS1 complex mediates the stimulation of the Ras–

Erk1/2 pathway by ligand-activated RTKs (Lowenstein et al.,

1992).

CIIA was initially discovered as an anti-apoptotic protein that

turned out to be identical to VPS28 (Cho et al., 2003; Kim et al.,

2010). Aside from these initial findings, we have recently shown

that CIIA physically associates with SOS1 and promotes the

SOS1-mediated activation of Rac1 (Hwang et al., 2011). We now

report that CIIA, by binding SOS1, inhibits the GEF activity of

SOS1 on Ras. Our results suggest that CIIA functions as a

negative switch of the SOS1–Ras signaling events initiated by

RTK-coupled peptide growth factors.

RESULTS AND DISCUSSIONCIIA inhibits the activation of the SOS1–Ras–Erk1/2 signalingevents induced by peptide growth factorsAs we reported previously (Hwang et al., 2011), co-

immunoprecipitation analysis revealed that CIIA physically

associates with SOS1 in NIH3T3 cells (Fig. 1A; supplementary

material Fig. S1A). This interaction was enhanced after the cells

were exposed to peptide growth factors, including epidermal

growth factor (EGF), platelet-derived growth factor (PDGF) and

fibroblast growth factor (FGF).

Next, given that many peptide growth factors stimulate the

tyrosine kinase activities of their specific receptors and thereby

facilitate the SOS1–Ras signaling processes, we examined the

possible effects of CIIA on Ras and Erk1/2 activation initiated by

peptide growth factors. After stably transfecting NIH3T3 cells or

PC12 cells with either an empty plasmid vector or a vector for

hemagglutinin (HA)-tagged CIIA or FLAG-tagged CIIA, we

treated those cells with the indicated peptide growth factors and

then examined the activities of Ras and Erk1/2 in the cells. Ras

activation was monitored by binding of active Ras (Ras-GTP) to a

GST-fused Raf fragment containing the Ras-GTP-binding

domain (GST–RBD), and Erk1/2 activation was detected by

immunoblotting with antibodies to phosphorylated Erk1/2.

Ectopic CIIA dramatically suppressed the activation of Ras and

Erk1/2 induced by EGF, PDGF, FGF or nerve growth factor

(NGF) (Fig. 1B–E; supplementary material Fig. S1B–E). Next,

in order to examine the effect of endogenous CIIA on Ras and

Erk1/2 activities, we constructed HeLa cells expressing either

small interfering RNA (siRNA) against GFP (siGFP) or CIIA

siRNA (siCIIA). siRNA-mediated depletion of CIIA expression

potentiated the EGF-induced activation of both Ras and Erk1/2

activities, and this effect was reversed by ectopic expression of

Myc-tagged CIIA* (which contains three silent point mutations

within the region targeted by the CIIA siRNA) (Fig. 1F;

supplementary material Fig. S1F). Moreover, siRNA-mediated

Laboratory of Cell Death and Human Diseases, Department of Life Sciences,Korea University, Seoul 136-701, South Korea.*These authors contributed equally to this work

`Author for correspondence (ejchoi@korea.ac.kr)

Received 8 August 2013; Accepted 24 January 2014

� 2014. Published by The Company of Biologists Ltd | Journal of Cell Science (2014) 127, 1640–1646 doi:10.1242/jcs.139931

1640

Jour

nal o

f Cel

l Sci

ence

Fig. 1. See next page for legend.

SHORT REPORT Journal of Cell Science (2014) 127, 1640–1646 doi:10.1242/jcs.139931

1641

Jour

nal o

f Cel

l Sci

ence

depletion of SOS1 (siSOS1) abolished the potentiating effect ofsiCIIA on EGF-induced Ras activation (Fig. 1G; supplementary

material Fig. S1G). Collectively, these results suggest that CIIAinteracts with SOS1 and thereby inhibits the activation of the Rasand Erk1/2 pathway.

CIIA blocks Ras activation by inhibiting the Ras-GEF activityof SOS1We previously observed that CIIA binds to the central region ofSOS1 (SOS1-CEN; amino acid residues 551–1050) (Hwang et al.,2011), which is involved in Ras activation (Chardin et al., 1993).

We, therefore, examined the effect of siCIIA on the interactionbetween SOS1 and Ras in HeLa cells. siRNA-mediated depletionof CIIA enhanced the interaction between SOS1 and Ras (Fig. 2A;supplementary material Fig. S2A). Furthermore, overexpressed

CIIA inhibited the interaction between ectopic Ras and SOS1-CENin HEK293T cells (Fig. 2B; supplementary material Fig. S2B), aswell as the EGF-induced interaction between endogenous Ras and

SOS1 in MDCK cells (supplementary material Fig. S2C). Next,given that SOS1-CEN possesses the Ras-specific GEF activity(Corbalan-Garcia et al., 1998), we examined the effect of CIIA on

Ras activation induced by SOS1-CEN. In a Ras-GEF assay, thefluorescence intensity increased when SOS1-CEN was addedto GST–Ras, and this increase was reduced by CIIA (Fig. 2C),

indicating that CIIA inhibited the GEF activity of SOS1-CEN onRas. Taken together, these results suggest that CIIA inhibitsthe SOS1–Ras interaction and the GEF activity of SOS1 on Ras,thereby suppressing SOS1-mediated Ras activation. By contrast,

CIIA did not affect the interaction between ectopic SOS1 and Grb2in HEK293T cells (Fig. 2D; supplementary material Fig. S2D) orthe interaction between endogenous SOS1 and Grb2 in MDCK

cells (supplementary material Fig. S2C). siRNA-mediated depletionof CIIA also did not affect either the binding between Grb2 andSOS1 under basal conditions, or the EGF-induced dissociation of

SOS1 and Grb2 (Fig. 2E; supplementary material Fig. S2E).

CIIA inhibits EGF-induced cyclin D1 expression and DNAsynthesisThe peptide-growth-factor-stimulated Erk1/2 pathway mediatescyclin D1 induction and DNA synthesis (Cheng et al., 1998).

Indeed, EGF treatment increased the abundance of cyclin D1transcript (Fig. 3A) and protein (Fig. 3B) in siGFP-expressing

HeLa cells. This effect of EGF was inhibited by U0126(Fig. 3A,B), which is a MEK1 inhibitor that can block EGF-induced Erk1/2 activation (supplementary material Fig. S3).Thus, our results suggest that EGF induces cyclin D1 expression

through stimulation of the Erk1/2 pathway. Furthermore, siRNA-mediated depletion of CIIA potentiated the EGF-induced cyclinD1 expression in siCIIA-expressing HeLa cells (Fig. 3A,B).

Next, we examined the effect of CIIA on EGF-induced DNAsynthesis in cultured cells. DNA synthesis analysis using a 59-bromo-29deoxyuridine (BrdU) incorporation assay revealed that

EGF induced DNA synthesis in siGFP-expressing HeLa cells,and that this effect was inhibited by PD98059 (Fig. 3C),which is another MEK1 inhibitor that blocks EGF-induced

Erk1/2 activation (supplementary material Fig. S3). siRNA-mediated depletion of CIIA expression enhanced EGF-inducedDNA synthesis in HeLa cells (Fig. 3C). Consistently, ectopicexpression of CIIA in MDCK cells abrogated EGF-induced DNA

synthesis (Fig. 3D). These results suggest that CIIA blocks cyclinD1 expression, as well as DNA synthesis, induced by EGF-stimulated Erk1/2 pathway.

CIIA fails to inhibit Ras and Erk1/2 activation mediated byNoonan-syndrome-associated SOS1 mutantsNext, we examined whether CIIA could regulate the function of theNoonan-syndrome-associated mutants of SOS1. We chose fourNoonan-syndrome-associated SOS1 mutants (M269R, R552G,

W729L and E846K), whose mutations are located within the DH,helical linker (HL), REM and Cdc25 domains, respectively. TheseSOS1 mutants appear to lose the autoinhibitory regulation of theRas-GEF activity (Robert et al., 2007; Sondermann et al., 2004;

Tartaglia et al., 2007). We transfected 293T cells with a vectorencoding the indicated SOS1 variant and then examined thebinding of CIIA to each of the SOS1 variants. CIIA was able to

bind M269R, R552G, W729L and E846K mutants of SOS1 as wellas wild-type SOS1 in the cells (Fig. 4A; supplementary materialFig. S4C). Next, we examined whether CIIA could affect the

binding of the SOS1 variants to Ras. CIIA failed to inhibit thebinding between Ras and each of the four mutants of SOS1,whereas it blocked the binding between Ras and wild-type SOS1(Fig. 4B; supplementary material Fig. S4D). Furthermore, CIIA

did not affect the EGF-induced activation of Ras and Erk1/2mediated by the SOS1 W729L (Fig. 4C), E846K (Fig. 4D),M269R (supplementary material Fig. S4E) or R552G mutants

(supplementary material Fig. S4F). The GEF assay data alsoshowed that CIIA did not inhibit the Ras-GEF activity of SOS-CEN (W729L) (Fig. 4E), unlike its inhibition on the Ras-GEF

activity of SOS-CEN (Fig. 2C). In addition, CIIA failed to inhibitEGF-induced DNA synthesis mediated by SOS1 W729L or SOS1E846K (Fig. 4F). Collectively, these results suggest that the

Noonan-syndrome-associated SOS1 mutants are resistant to theinhibitory action of CIIA on the Ras-GEF activity, although themolecular mechanism of this is not clear yet. Given that theSOS1 mutants used in the present study are known to lose the

autoinhibitory regulation of the Ras-GEF activity (Robert et al.,2007), it would be intriguing to investigate whether CIIA promotesthe autoinhibitory regulation of the SOS1 Ras-GEF activity.

Moreover, our findings suggest that malfunction of CIIA couldcause misregulation of the SOS1–Ras signaling pathway, therebyleading to abnormality in various cellular events, including cell

proliferation.

Fig. 1. CIIA inhibits the activation of Ras and Erk1/2 induced by peptidegrowth factors. (A) NIH3T3 cells were incubated in the absence orpresence of EGF (100 ng/ml), PDGF (20 ng/ml) or FGF (10 ng/ml) for 5 min.Cell lysates were subjected to immunoprecipitation (IP) with rabbit anti-CIIAantibody or rabbit preimmune IgG, and the resulting precipitates wereimmunoblotted (IB) with anti-SOS1 antibody. (B–E) NIH 3T3/neo and NIH3T3/HA-CIIA cells (B–D) or PC12/control and PC12/Flag-CIIA cells (E) wereserum-starved for 16 h and then were left untreated or treated with EGF(100 ng/ml) (B), FGF (10 ng/ml) (C) or PDGF (20 ng/ml) (D) for 5 min, or withNGF (100 ng/ml) for 10 min (E). Cell lysates were subjected to pull-down(PD) with glutathione–agarose beads coupled to GST–RBD. Bead-boundproteins were immunoblotted with anti-Ras antibody. Cell lysates were alsoimmunoblotted with antibodies to phosphorylated Erk1/2 (indicated by the Pwithin a circle), Erk1/2, Ras or HA. (F,G) HeLa cells expressing siGFP, siCIIAor siCIIA plus CIIA*-Myc (F), or siGFP, siSOS1 or siCIIA plus siSOS(G), were serum-starved for 16 h, then incubated for 5 min without or withEGF (100 ng/ml). Cell lysates were assayed for Ras activity as in B. Celllysates were also immunoblotted with the indicated antibodies. Ras pull-down and Erk1/2 phosphorylation data in B–G were quantified and themeans6s.d. from two or three independent experiments are shown insupplementary material Fig. S1B–G, respectively. The intensities of thebands were determined by densitometry and expressed as the fold increaseas shown underneath the blots.

SHORT REPORT Journal of Cell Science (2014) 127, 1640–1646 doi:10.1242/jcs.139931

1642

Jour

nal o

f Cel

l Sci

ence

Fig. 2. CIIA inhibits the Ras-GEF activity of SOS1. (A) HeLa cells expressing siGFP (control) or siCIIA were serum-starved for 16 h, then incubated for 5 minwithout or with EGF (100 ng/ml). Cell lysates were immunoprecipitated with anti-Ras antibody or preimmune IgG and the resulting precipitates wereimmunoblotted with anti-SOS1 antibody. (B) 293T cells were transfected for 48 h with vectors encoding SOS1-CEN–Myc, HA–Ras and Flag–CIIA, as indicated.Cell lysates were immunoprecipitated (IP) with anti-HA antibody, and the resulting precipitates were immunoblotted (IB) with anti-Myc antibody. (C) The Ras-specific GEF activity of SOS1-CEN (CEN) was determined in vitro in the absence or presence of CIIA by fluorescence spectroscopy with a GEF exchange assaykit. Left: relative fluorescence intensity data from one representative experiment. Right: means6s.d. of three independent experiments. Results are expressedas the N9- methylanthraniloyl-GTP (mant-GTP) incorporation into GST–Ras after 600 s relative to time 0 after adding SOS1-CEN. *P,0.01. (D) 293T cells weretransfected for 48 h with expression vectors encoding HA–SOS1, Flag–Grb2, and CIIA–Myc, as indicated. Cell lysates were immunoprecipitated with anti-Flagantibody, and the resulting precipitates were immunoblotted with anti-HA antibody. (E) siGFP- or siCIIA-expressing or HeLa cells were serum-starved for 16 hand then left untreated or treated with 100 ng/ml EGF for 5 min. Cell lysates were immunoprecipitated with anti-Grb2 antibody, and the resulting precipitateswere analyzed by immunoblotting with anti-SOS1 antibody.

SHORT REPORT Journal of Cell Science (2014) 127, 1640–1646 doi:10.1242/jcs.139931

1643

Jour

nal o

f Cel

l Sci

ence

MATERIALS AND METHODSCell culture, transfection and antibodies293T, MDCK, NIH3T3 and HeLa cells were maintained under a humidified

atmosphere of 5% CO2 at 37 C̊ in Dulbecco’s modified Eagle’s medium

(DMEM; Hyclone, South Logan, UT) supplemented with 10% fetal bovine

serum (FBS; Hyclone). PC12 cells were cultured in DMEM containing 10%

FBS and 5% horse serum (Invitrogen) on culture dishes coated with poly-D-

lysine (Sigma). MDCK and PC12 cells were stably transfected with either

pcDNA3-puro/Flag-CIIA or an empty vector, and were selected with

puromycin. NIH3T3 cells were stably transfected with either pcDNA3/HA-

CIIA or an empty vector, and selected with G418. HeLa cells were stably

transfected with pSUPER.retro/siGFP or pSUPER.retro/siCIIA, and were

selected with puromycin. Rabbit polyclonal antibodies to CIIA, to SOS1 and

to Grb2 were from Santa Cruz Biotechnology (Santa Cruz, CA). Mouse

monoclonal antibodies to Ras and to cyclin D were purchased from BD

Biosciences (Franklin Lakes, NJ) and from Santa Cruz Biotechnology,

respectively. Mouse monoclonal antibodies to Flag and to HA were obtained

from Sigma-Aldrich (St Louis, MO). Rabbit polyclonal antibodies to Erk1/2

and to phosphorylated Erk1/2 were from Cell Signaling (Beverly, MA).

Co-immunoprecipitationCells were lysed in NETN lysis buffer and subjected to co-

immunprecipitation as described previously (Hwang et al., 2011).

In vitro small-GTP-binding protein activity assayCell lysates were subjected to centrifugation at 12,000 g at 4 C̊, and the

supernatants were subjected to pull-down with GST-Raf (RBD)-coupled

glutathione–agarose beads. Ras-GTP in the pull-down precipitates was

examined by immunoblotting analysis with anti-Ras antibody.

In vitro GEF assayIn vitro GEF activity of SOS1 variants was examined with a GEF

exchange assay biochem kit (Cytoskeleton Inc.), as described previously

(Hwang et al., 2011).

BrdU incorporation assayCells were incubated with 10 mM BrdU for 6 h, and analyzed for BrdU

incorporation with a cell proliferation ELISA kit (Roche Applied

Science, Indianapolis, IN). Data are the mean6s.d. of quadruplicate

samples from four representative experiments.

Semi-quantitative RT-PCRTotal RNA was extracted from cultured cells using RNeasy Mini kit

(Qiagen, Valencia, CA). RT-PCR analysis was performed with primers

to cyclin D1 or to glyceraldehyde-3-phosphate dehydrogenase as a

control.

AcknowledgementsWe thank Pier P. Di Fiore (University of Milan, Milan, Italy), Kozo Kaibuchi(University of Nagoya, Nagoya, Japan), and Christian Herrmann (Ruhr UniversityBochum, Bochum, Germany) for providing SOS1, Ras and GST–RBD cDNAclones, respectively.

Competing interestsThe authors declare no competing interests.

Fig. 3. CIIA inhibits EGF-dependent cyclin D1 expression and DNA synthesis. (A,B) siGFP- or siCIIA-expressing or HeLa cells were serum-starved for16 h, and incubated for 1 h without or with 10 mM U0126. Then, the cells were left untreated or treated with 100 ng/ml EGF for 6 h. (A) Total RNA wasisolated from the cells and subjected to semi-quantitative RT-PCR using primers for cyclin D1 or GAPDH. (B) Cell lysates were immunoblotted withantibodies against cyclin D1 or CDK4. (C,D) Control or Flag–CIIA expressing MDCK cells (C) or siGFP- or siCIIA-expressing or HeLa cells (D) wereserum-starved for 16 h and incubated in the absence or presence of 100 ng/ml EGF and 50 mM PD98059 for 6 h. The cells were further incubated for 4 hwith 10 mM of BrdU and then lyzed. Cell lysates were assayed for BrdU incorporation with an ELISA kit. Results are mean6s.d. of quadruplicate samples fromfour representative experiments.

SHORT REPORT Journal of Cell Science (2014) 127, 1640–1646 doi:10.1242/jcs.139931

1644

Jour

nal o

f Cel

l Sci

ence

Fig. 4. CIIA does not inhibit Ras and Erk1/2activation induced by Noonan-syndrome-associated SOS1 mutants. (A) 293T cells weretransfected for 48 h with vectors encoding HA-taggedwild-type (WT) SOS1, SOS1 (W729L) or SOS1(E846K). Cell lysates were immunoprecipitated (IP)with an anti-CIIA antibody and the resultingprecipitates were immunoblotted (IB) with an anti-HAantibody. (B) 293T cells were transfected for 48 h withvectors encoding the indicated proteins. Cell lysateswere immunoprecipitated with anti-Flag antibody andthe resulting precipitates were immunoblotted withanti-HA antibody. (C,D) 293Tcells were transfected for48 h with indicated combinations of vectors encodingFlag–CIIA, HA–SOS1 and HA–SOS1 (W729L) (C) orHA-SOS1 (E846K) (D). Then, the cells were serum-starved for 16 h, and incubated with 100 ng/ml EGFfor indicated times. Cell lysates were examined forRas activity by GST pull-down assay. The cell lysateswere also immunoblotted with antibodies againstphosphorylated Erk1/2 (indicated by the P within acircle), Erk1/2, Ras, Flag or HA. The intensities of thebands were determined by densitometry andexpressed as the fold increase as shown underneaththe blots. (E) The GEF activity of SOS1-CEN (W729L)on Ras was determined in vitro in the absence orpresence of CIIA by fluorescence spectroscopy with aGEF exchange assay kit. Left: relative fluorescenceintensity data from one representative experiment.Right: data are the means6s.d. of three independentexperiments. Results are expressed as the N9-methylanthraniloyl-GTP (mant-GTP) incorporation intoGST-Ras after 600 s relative to time 0 after addingSOS1-CEN (W729L). *P,0.01, N.S., not significant.(F) Hela cells were transfected for 30 h with plasmidvectors encoding the indicated proteins. Cells werethen serum-starved for 16 h, and incubated for 6 hwithout or with 100 ng/ml EGF. Then, the cells weretreated with 10 uM BrdU for 4 h and assayed for BrdUincorporation. Data are the means6s.d. of threeindependent experiments.

SHORT REPORT Journal of Cell Science (2014) 127, 1640–1646 doi:10.1242/jcs.139931

1645

Jour

nal o

f Cel

l Sci

ence

Author contributionsH.S.H., S.G.H. and E.-J.C. designed experiments, analyzed data, and wrote thepaper; H.S.H., S.G.H., K.-W. Y., J.-H. Y., and K.-H.R. performed research.

FundingThis work was supported by the National Research Foundation [grant numbers2011-0030141, 2009-0081488, 2010-0001197 to E.-J.C.] funded by the Ministry ofScience, ICT and Future Planning of Korea.

Supplementary materialSupplementary material available online athttp://jcs.biologists.org/lookup/suppl/doi:10.1242/jcs.139931/-/DC1

ReferencesBar-Sagi, D. (1994). The Sos (Son of sevenless) protein. Trends Endocrinol.Metab. 5, 165-169.

Chardin, P., Camonis, J. H., Gale, N. W., van Aelst, L., Schlessinger, J., Wigler,M. H. and Bar-Sagi, D. (1993). Human Sos1: a guanine nucleotide exchangefactor for Ras that binds to GRB2. Science 260, 1338-1343.

Cheng, M., Sexl, V., Sherr, C. J. and Roussel, M. F. (1998). Assembly of cyclinD-dependent kinase and titration of p27Kip1 regulated by mitogen-activatedprotein kinase kinase (MEK1). Proc. Natl. Acad. Sci. USA 95, 1091-1096.

Cho, S. G., Kim, J. W., Lee, Y. H., Hwang, H. S., Kim, M. S., Ryoo, K., Kim, M. J.,Noh, K. T., Kim, E. K., Cho, J. H. et al. (2003). Identification of a novel antiapoptoticprotein that antagonizes ASK1 and CAD activities. J. Cell Biol. 163, 71-81.

Corbalan-Garcia, S., Margarit, S. M., Galron, D., Yang, S. S. and Bar-Sagi, D.(1998). Regulation of Sos activity by intramolecular interactions. Mol. Cell. Biol.18, 880-886.

Das, B., Shu, X., Day, G. J., Han, J., Krishna, U. M., Falck, J. R. and Broek, D.(2000). Control of intramolecular interactions between the pleckstrin homologyand Dbl homology domains of Vav and Sos1 regulates Rac binding. J. Biol.Chem. 275, 15074-15081.

Hwang, H. S., Hwang, S. G., Cho, J. H., Chae, J. S., Yoon, K. W., Cho, S. G. andChoi, E. J. (2011). CIIA functions as a molecular switch for the Rac1-specificGEF activity of SOS1. J. Cell Biol. 195, 377-386.

Kim, K. J., Yu, J. W., Hwang, H. S. and Choi, E. J. (2010). CIIA is a novelregulator of detachment-induced cell death. Cancer Res. 70, 6352-6358.

Lowenstein, E. J., Daly, R. J., Batzer, A. G., Li, W., Margolis, B., Lammers, R.,Ullrich, A., Skolnik, E. Y., Bar-Sagi, D. and Schlessinger, J. (1992). The SH2and SH3 domain-containing protein GRB2 links receptor tyrosine kinases to rassignaling. Cell 70, 431-442.

Margarit, S. M., Sondermann, H., Hall, B. E., Nagar, B., Hoelz, A., Pirruccello,M., Bar-Sagi, D. and Kuriyan, J. (2003). Structural evidence for feedbackactivation by Ras.GTP of the Ras-specific nucleotide exchange factor SOS. Cell112, 685-695.

Nimnual, A. S., Yatsula, B. A. and Bar-Sagi, D. (1998). Coupling of Ras and Racguanosine triphosphatases through theRas exchanger Sos.Science 279, 560-563.

Roberts, A. E., Araki, T., Swanson, K. D., Montgomery, K. T., Schiripo, T. A.,Joshi, V. A., Li, L., Yassin, Y., Tamburino, A. M., Neel, B. G. et al. (2007).Germline gain-of-function mutations in SOS1 cause Noonan syndrome. Nat.Genet. 39, 70-74.

Scita, G., Tenca, P., Frittoli, E., Tocchetti, A., Innocenti, M., Giardina, G. and DiFiore, P. P. (2000). Signaling from Ras to Rac and beyond: not just a matter ofGEFs. EMBO J. 19, 2393-2398.

Scita, G., Tenca, P., Areces, L. B., Tocchetti, A., Frittoli, E., Giardina, G.,Ponzanelli, I., Sini, P., Innocenti, M. and Di Fiore, P. P. (2001). An effectorregion in Eps8 is responsible for the activation of the Rac-specific GEF activityof Sos-1 and for the proper localization of the Rac-based actin-polymerizingmachine. J. Cell Biol. 154, 1031-1044.

Sondermann, H., Soisson, S. M., Boykevisch, S., Yang, S. S., Bar-Sagi, D. andKuriyan, J. (2004). Structural analysis of autoinhibition in the Ras activator Sonof sevenless. Cell 119, 393-405.

Tartaglia, M., Pennacchio, L. A., Zhao, C., Yadav, K. K., Fodale, V., Sarkozy, A.,Pandit, B., Oishi, K., Martinelli, S., Schackwitz, W. et al. (2007). Gain-of-function SOS1 mutations cause a distinctive form of Noonan syndrome. Nat.Genet. 39, 75-79.

SHORT REPORT Journal of Cell Science (2014) 127, 1640–1646 doi:10.1242/jcs.139931

1646