RS

TDF

R

tcihtabpfhdBA

aksisttBbpht

1rtP

aS

m

3

Biochemical and Biophysical Research Communications 271, 626–629 (2000)

doi:10.1006/bbrc.2000.2671, available online at http://www.idealibrary.com on

0CA

ho Small G-Protein-Dependent Binding of mDia to anrc Homology 3 Domain-Containing IRSp53/BAIAP21

akeshi Fujiwara,2 Akiko Mammoto, Yongman Kim, and Yoshimi Takai3

epartment of Molecular Biology and Biochemistry, Osaka University Graduate School of Medicine/aculty of Medicine, Suita 565-0871, Japan

eceived April 1, 2000

be involved in cytokinesis and cell polarity (reviewed inR

sRRb1ialnbiAmimbfi

tihaa5t2b(

M

rmLpta

mDia1 is a downstream effector of Rho small G pro-ein that is implicated in stress fiber formation andytokinesis. We isolated an mDia1-binding protein anddentified it to be IRSp53/BAIAP2. IRSp53 and BAIAP2ave independently been isolated as a 58/53-kDa pro-ein tyrosine phosphorylated in response to insulinnd a BAI1-binding protein, respectively. BAI1 is arain-specific seven-span transmembrane protein ca-able of inhibiting angiogenesis. The proline-richormin homology 1 domain of mDia1 bound the Srcomology 3 domain of IRSp53/BAIAP2 in a GTP-Rho-ependent manner. The results suggest that IRSp53/AIAP2 is a downstream effector of mDia1. © 2000

cademic Press

Rho small G proteins, including RhoA, -B, and -C,re implicated in actin stress fiber formation and cyto-inesis (reviewed in Refs. 1, 2). Many potential down-tream effectors of Rho have been identified (reviewedn Refs. 3, 4), and mDia has been shown to regulatetress fiber formation and cytokinesis (5–7). mDia haswo isoforms, mDia1 and -2, and is a mammalian coun-erpart of the budding yeast Saccharomyces cerevisiaeni1p and Bnr1p and Drosophila diaphanous whichelong to the FH family (5, 8–11). The FH familyroteins are defined by the presence of two forminomology domains, the proline-rich FH1 domain, andhe FH2 domain, and have genetically been shown to

Abbreviations used: FH, formin homology; EF1a, elongation factora; SH, Src homology; aa, amino acid(s); PCR, polymerase chaineaction; DBDLexA, DNA-binding domain of LexA; ADGAL4, transcrip-ional activation domain of GAL4; MBP, maltose binding protein;VDF, polyvinylidene fluoride; PBS, phosphate buffered saline.

1 This work was supported by grants-in-aid for Scientific Researchnd for Cancer Research from the Ministry of Education, Science,ports, and Culture, Japan (1999).

2 Current address: Vanderbilt University Medical Center, Depart-ent of Cell Biology, Nashville, TN 37232-2175.3 To whom correspondence should be addressed. Fax: 181-6-6879-

419. E-mail: ytakai@molbio.med.osaka-u.ac.jp.

626006-291X/00 $35.00opyright © 2000 by Academic Pressll rights of reproduction in any form reserved.

ef. 12).Our series of experiments in Saccharomyces cerevi-

iae indicate that Bni1p is a downstream effector ofho1p whereas Bnr1p is a downstream effector ofho4p (9, 10). Bni1p directly binds at least three actin-inding proteins: profilin, EF1a, and Aip3p/Bud6p (10,3, 14). Profilin is an actin monomer-binding proteinmplicated in actin polymerization (13); EF1a is anctin filament (F-actin)-binding protein which stimu-ates bundling of F-actin (15–17); and Aip3p/Bud6p is aovel actin-binding protein (13, 18). Bni1p furthermoreinds Spa2p that is required for the membrane local-zation of Bni1p (19). Bnr1p binds at least profilin andip3p/Bud6p (10, 20). Bnr1p, but not Bni1p, further-ore binds Hof1p which contains one SH3 domain and

s implicated in the regulation of cytokinesis (21–23).Dia1 has been shown to bind profilin, but no mDia1-

inding proteins other than profilin have been identi-ed.We have isolated here an mDia1-binding protein by

he yeast two-hybrid screening method and identifiedt to be a mouse counterpart of hamster IRSp53 anduman BAIAP2-a and -b, that contain one SH3 domainnd are homologous to yeast Hof1p (21–25). IRSp53nd BAIAP2 have independently been isolated as a8/53-kDa protein tyrosine phosphorylated in responseo insulin and a BAI1-binding protein, respectively (24,5). BAI1 is a brain-specific seven-span transmem-rane protein that is capable of inhibiting angiogenesis26).

ATERIALS AND METHODS

Cloning of a mouse IRSp53/BAIAP2. A reporter strain, L40 car-ying pBTM116-HA-mDia1 (aa 1-1117), was transformed with aouse embryonic fibroblast cDNA library made in pACTII (Clontechaboratories, Inc., California). The yeast two-hybrid screening andlasmid recovery were performed as described (19, 27). For quanti-ative assay for b-galactosidase activity, the activity was measuredccording to the ONPG assay method (28).

LLhmtlf

taDs2

smpdmbwtps

1pmuM1tISw01l

(Aia

R

pt1stccqehg3Tm

b6rbtBt(ptI

tBR

Vol. 271, No. 3, 2000 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Strains and media. Yeast strains L40 (MATa trp1 leu2 his3YS2<lexA-HIS3 URA3<lexA-lacZ) and TAT7 (MATa trp1 leu2 his3YS2<lexA-HIS3 ura3<lexA-lacZ) were used for the yeast two-ybrid studies (10, 29). Yeast strains were usually grown in richedium (YPDAU) and yeast transformants were grown in a selec-

ion medium (SD) (19). Yeast transformations were performed by theithium acetate methods (30). Escherichia coli strain DH5a was usedor construction and propagation of plasmids.

Molecular biological techniques. Standard molecular biologicalechniques were used for construction of plasmids, DNA sequencing,nd PCR (31). DNA sequences were determined using ALFexpressNA sequencer (Amersham Pharmacia Biotech, Inc., Buckingham-

hire, UK) and PCRs were performed using GeneAmp PCR System400 (Perkin-Elmer, Norwalk, CT).

Plasmid construction. Yeast two-hybrid plasmids were con-tructed by inserting various mDia1 and IRSp53/BAIAP2 DNA frag-ents generated by PCR into pBTM116-HA (10) encoding DBDLexA or

ACTII-HK (32) encoding ADGAL4. Yeast two-hybrid plasmids of aominant active mutant of RhoA, RhoA (G14V), a dominant negativeutant of RhoA, RhoA(G17A), and wild-type RhoA were constructed

y inserting cDNA fragments into pTHF1 encoding ADGAL4. pTHF1as constructed by disrupting the LEU2 gene in pGAD424 (29) with

he URA3 gene. For His6-mDia1 (aa 1-1117), we first constructedlasmid pRSET-mDia1 (aa 1-1255) and subsequently digested andelf-ligated to remove the mDia1 (aa 1118-1255).

In vitro binding of MBP-IRSp53/BAIAP2 with His6-mDia1 (aa-1117). Recombinant full-length IRSp53/BAIAP2 (aa 1-522) wasurified from overexpressing Escherichia coli DH5a containing plas-id pMALC2-IRSp53/BAIAP2 (aa 1-522) as an MBP fusion proteinsing an amylose resin column (New England BioLabs, Inc., Beverly,A) as described (33). The [35S]methionine-labeled His6-mDia1 (aa

-1117) was generated using the TNT T7 quick coupled transcrip-ion/translation system (Promega Corporation). Three pmol of MBP-RSp53/BAIAP2 (aa 1-522) and 40 pmol of MBP were subjected toDS–PAGE and transferred to a PVDF membrane. The membraneas washed with PBS and sequentially blocked with PBS containing.1% (W/V) Tween 20 and 5% (W/V) defatted powder milk at 4°C for2 h. The membrane was then incubated with [35S]methionine-abeled His6-mDia1 (aa 1-1117) (;1 pmol) in PBS containing 0.1%

FIG. 1. Binding of mDia1 to IRSp53/BAIAP2 and their binding rruncated forms of mDia1 was transformed into strain L40 expressininding of mDia1 to ADGAL4-IRSp53/BAIAP2 (aa 1-522). (B) BindhoA-binding domain.

627

W/V) Tween 20 and 5% (W/V) defatted powder milk at 4°C for 15 h.fter the incubation, the membrane was washed with PBS contain-

ng 0.1% (W/V) Tween 20 at 4°C, followed by autoradiography usingn image analyzer (Fujix BAS-2000).

ESULTS

Isolation of IRSp53/BAIAP2 as an mDia1-bindingrotein. We searched an mDia1-binding protein byhe yeast two-hybrid screening using mDia1 (aa-1117) as a bait. Among 1.2 3 106 transformantscreened, 15 positive clones (His1 and lacZ1) were iden-ified. Among these clones, 11 clones were found toonfer both the His1 and lacZ1 phenotypes on L40ontaining pBTM116-HA-mDia1 (aa 1-1117). DNA se-uencing of the insert DNAs revealed that 8 clonesncoded a protein homologous to hamster IRSp53 anduman BAIAP2-a and -b. The isolated portion of theene product corresponded to aa 340-522, 339-521, and39-520 of IRSp53, BAIAP2-a, and -b, respectively.herefore, we used IRSp53 instead of the isolatedouse gene for further analysis.The binding of mDia1 to IRSp53/BAIAP2 was tested

y the yeast two-hybrid method. Deletion analysis (aa22-683) indicated that the FH1 domain-containingegion of mDia1 was necessary and sufficient for theinding to full-length IRSp53/BAIAP2 (Fig. 1A), andhat the SH3 domain-containing region of IRSp53/AIAP2 (aa 339-522) bound the FH1 domain-con-

aining region of mDia1 (aa 552-820) (Fig. 1B). mDia1aa 552-820) also bound the isolated mouse generoduct (data not shown). These results indicate thathe FH1 domain of mDia1 binds the SH3 domain ofRSp53/BAIAP2.

ons. A two-hybrid plasmid encoding DBDLexA fused to full-length orDGAL4 fused to full-length or truncated forms of IRSp53/BAIAP2. (A)of IRSp53/BAIAP2 to DBDLexA-mDia1 (aa 552-820). RhoA B.D.,

egig Aing

aMSb[m1m

Bthaa(ttbBm3bBobsbd

DISCUSSION

ScialfRsHiiyitftBc

liFtbcnmmbI

BA(RPbm

Iab[IB

Vol. 271, No. 3, 2000 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Direct binding of mDia1 to IRSp53/BAIAP2. To ex-mine whether mDia1 directly binds IRSp53/BAIAP2,BP-IRSp53/BAIAP2 (aa 1-522) was subjected to

DS–PAGE, followed by transfer to a PVDF mem-rane, and the membrane was incubated with

35S]methionine-labeled His6-mDia1 (aa 1-1117). His6-Dia1 (aa 1-1117) bound MBP-IRSp53/BAIAP2 (aa

-522), but not MBP (Fig. 2). This result indicates thatDia1 directly binds IRSp53/BAIAP2.

GTP-RhoA-dependent binding of mDia1 to IRSp53/AIAP2. Although the FH1 domain of mDia1 bound

he SH3 domain of IRSp53/BAIAP2 by the yeast two-ybrid method, full-length mDia1 showed little inter-ction with full-length IRSp53/BAIAP2 (see Figs. 1And 1B). Since mDia1 is a downstream effector of RhoA5), the requirement of the activated form of RhoA inhis protein interaction was examined by the yeastwo-hybrid method. No interaction was still observedetween full-length mDia1 and full-length IRSp53/AIAP2 even after the addition of a dominant negativeutant of RhoA, RhoA (G17A), or wild-type RhoA (Fig.

). In contrast, marked interaction was observedetween full-length mDia1 and full-length IRSp53/AIAP2 after the addition of a dominant active mutantf RhoA, RhoA(G14V). No interaction was observedetween RhoA(G14V) and IRSp53/BAIAP2 (data nothown). These results indicate that full-length mDia1inds full-length IRSp53/BAIAP2 in a GTP-RhoA-ependent manner.

FIG. 2. Direct binding of mDia1 to IRSp53/BAIAP2. MBP-RSp53/BAIAP2 (aa 1-522) and MBP were subjected to SDS–PAGEnd transferred to a PVDF membrane. The membrane was incu-ated with [35S]methionine-labeled His6-mDia1 (aa 1-1117). The

35S]methionine-labeled His6-mDia1 (aa 1-1117) bound to MBP-RSp53/BAIAP2 (aa 1-522) was detected. Lane 1, MBP-IRSp53/AIAP2 (aa 1-522); lane 2, MBP.

628

We have shown here that mDia1 directly binds anH3 domain-containing protein, IRSp53/BAIAP2. Be-ause mDia1 is a downstream effector of RhoA which ismplicated in stress fiber formation and cytokinesisnd this binding is dependent on GTP-RhoA (5–7), it isikely that IRSp53/BAIAP2 may be a downstream ef-ector of mDia1 that mediates at least a part of thehoA-mDia1 signaling pathway. We have previouslyhown in the budding yeast that Bnr1p directly bindsof1p in a GTP-Rho4p-dependent manner (21). Hof1p

s a protein which contains one SH3 domain and ismplicated in cytokinesis (21–23). FH proteins in theeast and fruit flies have genetically been shown to benvolved in cytokinesis (12). mDia1 has been showno localize in a ring-like structure at the cleavageurrow in mitotic fibroblast cells (5). Taken together,hese findings suggest that the RhoA-mDia1-IRSp53/AIAP2 system plays an important role at least inytokinesis.It has recently been shown that mDia1 intramolecu-

arly interacts at its N- and C-terminal regions, mak-ng it an inactive conformation (6). Our results that theH1 domain of mDia1 binds IRSp53/BAIAP2 even inhe absence of GTP-RhoA and that full-length mDia1inds it only in a GTP-RhoA-dependent manner areonsistent with this earlier observation (6). It may beoted that GTP-RhoA does not bind to full-lengthDia1 at least as estimated by the yeast two-hybridethod (Fig. 3). Another factor may furthermore

e involved in the formation of GTP-RhoA-mDia1-RSp53/BAIAP2 ternary complex. It has been shown

FIG. 3. GTP-RhoA-dependent binding of mDia1 to IRSp53/AIAP2. Cells of TAT7 expressing DBDLexA-mDia1 (aa 1-1255) andDGAL4-IRSp53/BAIAP2 (aa 1-522) were transformed with pTHF1

vector), pTHF1-RhoA(G14V), pTHF1-RhoA(G17A), or pTHF1-hoA. Cells of TAT7 expressing DBDLexA-mDia1 (aa 1-1255) andACTII-HK (vector) were transformed with pTHF1-RhoA(G14V).-Galactosidase activity was measured by the qualitative assayethod.

that Src tyrosine kinase, an SH3 domain-containingplHtbtc

A

R

1

1

11

1

1

16. Yang, F., Demma, M., Warren, V., Dharmawardhane, S., and

1

1

1

2

2

22

2

2

2

2

22

3

3

3

3

3

3

Vol. 271, No. 3, 2000 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

rotein, co-immunoprecipitates with mDia1 and co-ocalizes with mDia1 at the cleavage furrow of dividingeLa cells (34). Src has been shown to be implicated in

he completion of mitosis (reviewed in Ref. 35). Src maye a possible candidate molecule to regulate the forma-ion of the GTP-RhoA-mDia1-IRSp53/BAIAP2 ternaryomplex.

CKNOWLEDGMENT

We thank Richard A. Roth for IRSp53 cDNA-containing plasmids.

EFERENCES

1. Hall, A. (1994) Annu. Rev. Cell Biol. 10, 31–54.2. Takai, Y., Sasaki, T., Tanaka, K., and Nakanishi, H. (1995)

Trends Biochem. Sci. 20, 227–231.3. Tapon, N., and Hall, A. (1997) Curr. Opin. Cell Biol. 9, 86–92.4. Hall, A. (1998) Science 279, 509–514.5. Watanabe, N., Madaule, P., Reid, T., Ishizaki, T., Watanabe, G.,

Kakizuka, A., Saito, Y., Nakao, K., Jockusch, B. M., and Naru-miya, S. (1997) EMBO J. 16, 3044–3056.

6. Watanabe, N., Kato, T., Fujita, A., Ishizaki, T., and Narumiya, S.(1999) Nat. Cell Biol. 1, 136–143.

7. Nakano, K., Takaishi, K., Kodama, A., Mammoto, A., Shiozaki, H.,Monden, M., and Takai, Y. (1999) Mol. Biol. Cell 10, 2481–2491.

8. Castrillon, D. H., and Wasserman, S. A. (1994) Development 120,3367–3377.

9. Kohno, H., Tanaka, K., Mino, A., Umikawa, M., Imamura, H.,Fujiwara, T., Fujita, Y., Hotta, K., Qadota, H., Watanabe, T.,Ohya, Y., and Takai, Y. (1996) EMBO J. 15, 6060–6068.

0. Imamura, H., Tanaka, K., Hihara, T., Umikawa, M., Kamei, T.,Takahashi, K., Sasaki, T., and Takai, Y. (1997) EMBO J. 16,2745–2755.

1. Alberts, A. S., Bouquin, N., Johnston, L. H., and Treisman, R.(1998) J. Biol. Chem. 273, 8616–8622.

2. Wasserman, S. A. (1998) Trends Cell Biol. 8, 111–115.3. Evangelista, M., Blundell, K., Longtine, M. S., Chow, C. J.,

Adames, N., Pringle, J. R., Peter, M., and Boone, C. (1997)Science 276, 118–122.

4. Umikawa, M., Tanaka, K., Kamei, T., Shimizu, K., Imamura, H.,Sasaki, T., and Takai, Y. (1998) Oncogene 16, 2011–2016.

5. Demma, M., Warren, V., Hock, R., Dharmawardhane, S., andCondeelis, J. (1990) J. Biol. Chem. 256, 2286–2291.

629

Condeelis, J. (1990) Nature 347, 494–496.7. Murray, J. W., Edmonds, B. T., Liu, G., and Condeelis, J. (1996)

J. Cell Biol. 135, 1309–1321.8. Amberg, D. C., Zahner, J. E., Mulholland, J. W., Pringle, J. R.,

and Botstein, D. (1997) Mol. Biol. Cell 8, 729–753.9. Fujiwara, T., Tanaka, K., Mino, A., Kikyo, M., Takahashi,

K., Shimizu, K., and Takai, Y. (1998) Mol. Biol. Cell 9, 1221–1233.

0. Kikyo, M., Tanaka, K., Kamei, T., Ozaki, K., Fujiwara, T., Inoue,E., Takita, Y., Ohya, Y., and Takai, Y. (1999) Oncogene 18,7046–7054.

1. Kamei, T., Tanaka, K., Hihara, T., Umikawa, M., Imamura, H.,Kikyo, M., Ozaki, K., and Takai, Y. (1998) J. Biol. Chem. 273,28341–28345.

2. Lippincott, J., and Li, R. (1998) J. Cell Biol. 143, 1947–1960.3. Vallen, E. A., Caviston, J., and Bi, E. (2000) Mol. Biol. Cell 11,

593–611.4. Yeh, T. C., Ogawa, W., Danielsen, A. G., and Roth, R. A. (1996)

J. Biol. Chem. 271, 2921–2928.5. Oda, K., Shiratsuchi, T., Nishimori, H., Inazawa, J., Yoshikawa,

H., Taketani, Y., Nakamura, Y., and Tokino, T. (1999) Cytogenet.Cell Genet. 84, 75–82.

6. Nishimori, H., Shiratsuchi, T., Urano, T., Kimura, Y., Kiyono, K.,Tatsumi, K., Yoshida, S., Ono, M., Kuwano, M., Nakamura, Y.,and Tokino, T. (1997) Oncogene 15, 2145–2150.

7. Vojtek, A. B., Hollenberg, S. M., and Cooper, J. A. (1993) Cell 74,205–214.

8. Guarente, L. (1983) Methods Enzymol. 101, 181–191.9. Nonaka, H., Tanaka, K., Hirano, H., Fujiwara, T., Kohno, H.,

Umikawa, M., Mino, A., and Takai, Y. (1995) EMBO J. 14,5931–5938.

0. Gietz, D., Jean, A. S., Woods, R. A., and Schiestl, R. H. (1992)Nucleic Acids Res. 20, 1425.

1. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) in Molecu-lar Cloning: A Laboratory Manual, 2nd ed., Cold Spring HarborLaboratory Press, Cold Spring Harbor, NY.

2. Ozaki, K., Tanaka, K., Imamura, H., Hihara, T., Kameyama, T.,Nonaka, H., Hirano, H., Matsuura, Y., and Takai, Y. (1996)EMBO J. 15, 2196–2207.

3. di Guan, C., Li, P., Riggs, P. D., and Inouye, H. (1988) Gene 67,21–30.

4. Tominaga, T., Sahai, E., Chardin, P., McCormick, F., Court-neidge, S. A., and Alberts, A. S. (2000) Mol. Cell 5, 13–25.

5. Thomas, S. M., and Brugge, J. S. (1997) Annu. Rev. Cell Dev.Biol. 13, 513–609.