Structural basis for mutual relief of the Rac guaninenucleotide exchange factor DOCK2 and its partnerELMO1 from their autoinhibited formsKyoko Hanawa-Suetsugua, Mutsuko Kukimoto-Niinoa,b, Chiemi Mishima-Tsumagaria, Ryogo Akasakaa, Noboru Ohsawaa,Shun-ichi Sekinea,c, Takuhiro Itoa,c, Naoya Tochioa, Seizo Koshibaa, Takanori Kigawaa,d, Takaho Teradaa,b,Mikako Shirouzua, Akihiko Nishikimib,e,f, Takehito Urunoe, Tomoya Katakaig, Tatsuo Kinashig, Daisuke Kohdah,Yoshinori Fukuib,e,f,1, and Shigeyuki Yokoyamaa,b,c,1

aRIKEN Systems and Structural Biology Center, 1-7-22 Suehiro, Tsurumi, Yokohama 230-0045, Japan; bJapan Science and Technology Agency, CoreResearch for Evolutional Science and Technology, Tokyo 102-0075, Japan; cDepartment of Biophysics and Biochemistry, Graduate School of Science,University of Tokyo, 7-3-1 Hongo, Bunkyo, Tokyo 113-0033, Japan; dTokyo Institute of Technology, 4259 Nagatsuta, Midori, Yokohama 226-8502, Japan;eDivision of Immunogenetics, Department of Immunobiology and Neuroscience, Medical Institute of Bioregulation, Kyushu University, Fukuoka 812-8582,Japan; fResearch Center for Advanced Immunology, Kyushu University, Fukuoka 812-8582, Japan; gDepartment of Molecular Genetics, Institute ofBiomedical Science, Kansai Medical University, Osaka 570-8506, Japan; and hDivision of Structural Biology, Medical Institute of Bioregulation, KyushuUniversity, Fukuoka 812-8582, Japan

Edited by John Kuriyan, University of California, Berkeley, CA, and approved December 31, 2011 (received for review September 6, 2011)

DOCK2, a hematopoietic cell-specific, atypical guanine nucleotideexchange factor, controls lymphocyte migration through ras-relatedC3 botulinum toxin substrate (Rac) activation. Dedicator of cytokin-esis 2–engulfment and cell motility protein 1 (DOCK2•ELMO1) com-plex formation is required for DOCK2-mediated Rac signaling. In thisstudy, we identified the N-terminal 177-residue fragment and theC-terminal 196-residue fragment of human DOCK2 and ELMO1, re-spectively, as the mutual binding regions, and solved the crystalstructure of their complex at 2.1-Å resolution. The C-terminal Pro-rich tail of ELMO1 winds around the Src-homology 3 domain ofDOCK2, and an intermolecular five-helix bundle is formed. Overall,the entire regions of both DOCK2 and ELMO1 assemble to create arigid structure, which is required for the DOCK2•ELMO1 binding, asrevealed by mutagenesis. Intriguingly, the DOCK2•ELMO1 interfacehydrophobically buries a residue which, when mutated, reportedlyrelieves DOCK180 from autoinhibition. We demonstrated that theELMO-interacting region and the DOCK-homology region 2 guaninenucleotide exchange factor domain of DOCK2 associate with eachother for the autoinhibition, and that the assembly with ELMO1weakens the interaction, relieving DOCK2 from the autoinhibition.The interactions between the N- and C-terminal regions of ELMO1reportedly cause its autoinhibition, and binding with a DOCKprotein relieves the autoinhibition for ras homolog gene family,member G binding and membrane localization. In fact, the DOCK2•-ELMO1 interface also buries the ELMO1 residues required for theautoinhibition within the hydrophobic core of the helix bundle.Therefore, the present complex structure reveals the structural basisby which DOCK2 and ELMO1 mutually relieve their autoinhibitionfor the activation of Rac1 for lymphocyte chemotaxis.

X-ray crystallography ∣ NMR ∣ protein complex ∣ CDM proteins ∣immunology

Dedicator of cytokinesis 2 (DOCK2), a large protein with amolecular mass of 213 kDa, is specifically expressed in he-

matopoietic cells (1). DOCK2 plays a critical role in lymphocytemigration and activation by regulating the actin cytoskeletonthrough ras-related C3 botulinum toxin substrate (Rac) activa-tion (2, 3), and the deletion of DOCK2 enables long-term cardiacallograft survival (4). DOCK2 also controls various immunolo-gical functions, including helper Tcell differentiation (5), neutro-phil chemotaxis (6, 7), and type I interferon induction in plasma-cytoid dendritic cells (8). DOCK2 belongs to the CDM family(Caenorhabditis elegans CED-5, mammalian DOCK180, andDrosophila melanogaster Myoblast city) of the evolutionally con-served, atypical guanine nucleotide exchange factors (GEFs)

for the Rho-family GTPases. There are 11 mammalian members(DOCK180, DOCK2-11) of the CDM family. The CDM proteinsshare the DOCK-homology regions (DHR)-1 and DHR-2 (alsoknown as Docker domains), and lack the Dbl homology (DH)and pleckstrin homology (PH) domains typically present in themammalian Rho-family GEFs (6, 9, 10). DOCK2 mediates theRac GEF reaction by means of the DHR-2 domain (6, 10, 11).DOCK2 also interacts with phosphatidylinositol 3,4,5-tripho-sphate [PtdInsð3;4;5ÞP3] through the DHR-1 domain (7, 12), totarget the GEF activity to the plasma membrane. Recently, thecrystal structure of the DHR-1 domain of DOCK180 was deter-mined (13), as well as those of the DHR-2 domains of DOCK2and DOCK9 bound with their substrates, Rac1 and cell divisioncycle 42 (Cdc42), respectively (14, 15).

DOCK2, like DOCK180 and DOCK3-5, has a Src-homology3 (SH3) domain at its N terminus. The N-terminal 502-residue re-gion, including the SH3 domain, of DOCK2 interacts with engulf-ment and cell motility protein 1 (ELMO1), a mammalianhomologue of C. elegans CED-12 (16). ELMO1 contains the N-terminal ras homolog gene family, member G (RhoG)-binding re-gion, the ELMO domain, the PH domain, and the C-terminal se-quence with three PxxP motifs. The C-terminal region of ELMO1,including the Pro-rich sequence, binds the SH3-containing regionof DOCK2, which is required for DOCK2 to activate Rac in vivo(16). Similarly, DOCK180 forms a ternary complex with ELMO1and Rac (17, 18), which is essential for the in vivo catalytic activity(17, 19). ELMO1 binds to the N-terminal region (about 200 resi-dues), including the SH3 domain, of DOCK180 (17, 19). However,an ELMO1mutant lacking the Pro-rich sequence retains the abilityto bind DOCK180 (20), suggesting that the region flanking the Pro-rich sequence of ELMO1 also participates in DOCK180 binding.Furthermore, the DOCK180•ELMO1 complex was proposed tofunction as a bipartite GEF for Rac, in which the DHR-2 domain

Author contributions: M.S., Y.F., and S.Y. designed research; K.H.-S., M.K.-N., C.M.-T., R.A.,N.O., S.-i.S., T.I., N.T., S.K., T. Kigawa, T.T., A.N., T.U., D.K., and S.Y. performed research;T. Katakai and T. Kinashi contributed new reagents/analytic tools; K.H.-S., M.K.-N., Y.F.,and S.Y. analyzed data; and K.H.-S., M.K.-N., Y.F., and S.Y. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

Freely available online through the PNAS open access option.

Data deposition: The atomic coordinates have been deposited in the Protein Data Bank,www.pdb.org (PDB ID codes 2RQR, 3A98, 3B13).1To whom correspondence may be addressed. E-mail: yokoyama@biochem.s.u-tokyo.ac.jpor fukui@bioreg.kyushu-u.ac.jp.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1113512109/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1113512109 PNAS ∣ February 28, 2012 ∣ vol. 109 ∣ no. 9 ∣ 3305–3310

BIOCH

EMISTR

Y

Dow

nloa

ded

by g

uest

on

July

5, 2

020

of DOCK180 and the PH domain of ELMO1 cooperatively func-tion (20). The crystal structure of the ELMO1 PH domain has re-cently been determined, and the unusual α-helical N-terminalextension of the PH domain in ELMO1 was found to be essentialfor DOCK180 binding and DOCK180-mediated Rac signal-ing (21).

In the absence of ELMO1, the N-terminal region, includingthe SH3 domain, of DOCK180 interacts with the DHR-2 domain,which may facilitate the autoinhibition of the DOCK180 GEFactivity (22). ELMO1 binding to the SH3 domain disrupts theSH3•DHR-2 interaction, and allows Rac binding to the DHR-2domain for the GEF activity (22). This autoinhibition ofDOCK180 is relieved by the mutation of Ile31 in the SH3 do-main, even though the location of this residue is distant from thePxxP-binding pocket. ELMO1 by itself is also reportedly auto-inhibited because of the interaction of the ELMO inhibitory do-main (EID), lying between the C-terminal PxxP motif and the PHdomain, with the ELMO autoregulatory domain (EAD) residingbetween the N-terminal RhoG-binding domain and the ELMOdomain. Actually, EAD binding to DOCK180, as well as themutations of Met692 and Glu693 in the EAD, disrupts this intra-molecular inhibitory interaction, resulting in RhoG binding andmembrane localization (23). Consequently, DOCK2 and ELMO1mutually relieve their autoinhibitions by an unknown interactionmode involving the DOCK2 N-terminal and ELMO1 C-terminalregions, a scenario that is far more complicated than simpleSH3•PxxP binding.

In the present study, we identified the regions required for theinteraction between DOCK2 and ELMO1, and determined theircomplex structure. The SH3 domain of DOCK2 binds to theC-terminal sequence of ELMO1 much more extensively thanother SH3 domains, which only bind the PxxP motif. Moreover,the α-helical region following the SH3 domain forms an intermo-lecular five-helix bundle with both the N- and C-terminal exten-sions of the PH domain of ELMO1. Thus, the entire region isassembled into a rigid structural unit. The residues required forthe respective autoinhibitions (i.e., Ile31 of DOCK2 and Met692of ELMO1) are buried in the hydrophobic core of the intermo-lecular assembly, thus revealing the structural basis of the mutualautoinhibition relief. Further mutagenesis demonstrated that thisunique assembly mode is important for ELMO1 to relieve theautoinhibition of DOCK2, so it can participate in Rac activationin lymphocyte chemotaxis.

ResultsNMR Analysis of the DOCK2 SH3 Domain Interaction with ELMO1 Pep-tides.The C terminus of ELMO1 contains six Pro residues locatedclose together (residues 707, 710, 711, 712, 714, and 717), con-stituting three potential PxxP motifs, which are the potential SH3domain binding sites. To define the interaction between theDOCK2 SH3 domain and the ELMO1 Pro-rich sequence, we firstgenerated various fusion constructs for cell-free protein synthesis(24–26) and obtained one sample suitable for structure determi-nation by NMR spectroscopy (Fig. S1). The fusion constructincludes a 26-residue ELMO1 peptide (residues 697–722) fusedto the N terminus of the DOCK2 SH3 domain (residues 8–70)(Fig. 1A and Fig. S2A). For comparison, we also analyzed theDOCK2 SH3 domain alone and compared its chemical shifts forthe backbone 15N and 1HN atoms with those in the fusion sample.The residues located in the arginine and threonine pair (RT)loop, the n-Src loop, β4, and the helix-like loop between β4and β5 exhibited large chemical shift changes (Fig. S2B).

We determined the solution structure of the DOCK2 SH3–ELMO1 peptide fusion complex (Table S1). The rmsd from themean structure was 0.78 Å for the backbone atoms and 1.19 Å forall heavy (non-hydrogen) atoms in the well-ordered region (resi-dues 704–722 for the ELMO1 peptide and residues 9–68 for theDOCK2 SH3 domain) (Fig. S2C). The topology of the DOCK2

SH3 domain resembles those of other SH3 domains, and consistsof five β-strands packed in two β-sheets, which form a β-barrel-like structure (Fig. S2D). The ELMO1 peptide binds to theDOCK2 SH3 domain as a polyproline II helix in a class II orien-tation (27, 28). Tyr44 of DOCK2 is buried in the center of theinterface and hydrophobically interacts with Pro714 and Pro717of ELMO1. These analyses revealed that, among the three can-didates, P714-x-x-P717 serves as the PxxP motif. On the otherhand, the N-terminal seven residues of the ELMO1 peptide(residues 697–703) were poorly defined, probably because of highmobility, although they were necessary for the structure determi-nation (Figs. S1 and S2C). This observation prompted us to exam-ine whether any additional interactions, besides the SH3 domain–PxxP motif interaction, also occur between DOCK2 and ELMO1.

Crystal Structure of the Complex of the DOCK2 and ELMO1 InteractingDomains. First, we determined the ELMO1-interacting region ofDOCK2 and the DOCK2-interacting region of ELMO1 that aresuitable for crystallographic studies (see SI Text). We generatedlonger fragments of DOCK2 (residues 1–177; the SH3 domainand the flanking region) and ELMO1 (residues 532–727; includ-ing the PH domain and the Pro-rich region) by cell-free proteinsynthesis (Fig. S3). The DOCK2 fragment precipitated duringthe synthesis, but it was prepared as a soluble protein complexby cell-free coexpression with the ELMO1 fragment. The purifiedcomplex between the DOCK2 and ELMOl fragments (hereafterreferred to as the DOCK2•ELMOl complex) was stable andmonomeric with 1∶1 stoichiometry, as determined by analyticalultracentrifugation (Fig. S4A).

We determined the crystal structure of the DOCK2•ELMOlcomplex (Table S2). The asymmetric unit contained two virtuallyidentical DOCK2•ELMOl complexes. Overall, DOCK2 andELMO1 form a large, rigid complex involving the entire regionsof the two protein fragments. The DOCK2 structure is composedof the N-terminal SH3 domain (residues 1–69) and a three-helixbundle domain (residues 81–167; Dα1–3). The electron density

Fig. 1. Mutually interactive regions of DOCK2 and ELMO1. (A) Domain or-ganizations of DOCK2 and ELMO1. Known protein interaction and functionalregions are indicated. Blue and red bars indicate the DOCK2 and ELMO1regions included in the NMR construct. Green and yellow bars indicate theregions used for the crystal structure determination. ELMO1-interacting re-gion (EIR) and DOCK2-interacting region (DIR) indicate the ELMO1-interact-ing region and the DOCK2-interacting region, respectively. (B) Overviews ofthe crystal structure of the DOCK2•ELMOl complex. Ribbon representationsof the DOCK2•ELMOl complex. DOCK2 is colored green and ELMO1 is yellow.The six proline residues in the ELMO1 Pro-rich tail are colored red. The twoviews are related by a 90° rotation about the vertical axis.

3306 ∣ www.pnas.org/cgi/doi/10.1073/pnas.1113512109 Hanawa-Suetsugu et al.

Dow

nloa

ded

by g

uest

on

July

5, 2

020

of the 11-residue region connecting the two domains was notobserved. The ELMO1 structure is composed of the PH domain(residues 560–674), its N- and C-terminal extended helices Eα1and Eα3 (residues 532–558 and 681–697, respectively), and the C-terminal Pro-rich tail (residues 700–727) (Fig. 1B and Fig. S3).The DOCK2 SH3 domain and the ELMO1 Pro-rich tail interactextensively with each other. Furthermore, Dα1–3 of DOCK2 andEα1 and Eα3 of ELMO1 form an intermolecular five-helix bun-dle (Fig. 1B). A surface area of approximately 2;370 Å2 is buriedin the interface between DOCK2 and ELMO1.

The SH3 Domain of DOCK2 and the Pro-Rich Tail of ELMO1.The overallfold of the DOCK2 SH3 domain and the binding manner of thePxxP-motif (Pro714–Pro717) are consistent with those definedby our NMR analysis. However, the SH3 domain reinforces theinteraction with the regions flanking the PxxP motif of ELMO1(Fig. 2A). The overall intermolecular binding mode of the DOCK2SH3 domain with ELMOl, burying approximately 1;450 Å2, is byfar the most extensive among the SH3 domain complexes charac-terized to date (SI Text and Fig. S5). The interactions occur in threeregions. First, the 28-residue Pro-rich tail (Asp700–Asn727) windsaround the SH3 domain, burying approximately 1;310 Å2. TheDOCK2 SH3 domain forms extensive hydrophobic interactionsand several hydrophilic interactions with the ELMO1 tail. At thecenter of the interface, the side-chain imino group of Trp44 ofDOCK2 hydrogen bonds with the main-chain carbonyl group ofLys715, within the PxxP motif of ELMO1. The DOCK2 SH3 do-main forms six hydrogen bonds that interact with the C-terminalhalf (Asn719–Asn727), following the PxxP motif, of the ELMO1

tail. In contrast, the DOCK2 SH3 domain binds the N-terminalhalf (Asp700–Ile713), mostly through hydrophobic interactions.

Second, the DOCK2 SH3 domain also interacts with Eα1 andEα3 of ELMO1, through its RT loop (Fig. 2B). Among the RT-loop residues, Tyr17 stacks its side chain between those of Pro532of Eα1 and Pro710 in the ELMO1 tail, whereas Ile31 and Gly32hydrophobically interact with Ile533 of Eα1, Leu698 of Eα3, andLeu701 and Ile706 of the tail. In addition, Asn18 hydrogen bondswith Pro532 and Elu535 of Eα1, whereas Gln30 hydrogen bondswith Arg697 of Eα3. Within the SH3 domain, Ile31, correspond-ing to Ile32, which is reportedly required for the autoinhibition ofDOCK180 (21), is located at the center of the interactions(Fig. 2B).

Third, DOCK2 interacts with the boundary between Eα3 andthe ELMO1 tail. Lys67, at the C terminus of the SH3 domain,hydrogen bonds with Leu699 of Eα3 and Asp700 in the tail.Arg128 and Ser129 of Dα1 form hydrogen bonds with Leu699of Eα3 and Glu702 in the tail. The residues involved in these in-teractions in the three regions are well conserved in the CDMfamily (Fig. S3).

Five-Helix Bundle. The intermolecular five-helix bundle consistsof Dα1–3 of DOCK2 and Eα1 and Eα3 of ELMO1 (Fig. 3 andFig. S4 B and C). In the DOCK2 three-helix bundle domain, Dα2is arranged in an antiparallel manner relative to Dα1 and Dα3.Almost all of the interactions between the three helices are hy-drophobic, involving a large number of aliphatic side chains of Ile,Val, and Leu residues and those of Trp98 and Met120 (Fig. S4B).Similarly, the ELMO1 Eα1 and Eα3 helices are arranged in anantiparallel manner and form multiple hydrophobic interactionsinvolving the side chains of Ile and Leu residues (Fig. S4C). Onthe other hand, Dα1 and Dα2 of DOCK2 and Eα1 and Eα3 ofELMO1 form the interface between the DOCK2 and ELMO1helix bundles, burying approximately 930 Å2 (Fig. 3). The inter-face involves not only a number of hydrophobic interactions, butalso several hydrophilic interactions. The residues Leu95, Trp96,Trp102, Val107, Phe114, Met121, and Met125 of DOCK2 hydro-phobically interact with Ile544, Leu547, Ile548, Leu681, Leu689,Met692, Leu696, and Leu699 of ELMO1. Met692 of ELMO1,which is required for ELMO autoinhibition (23), is buried atthe center of the five-helix bundle. In addition, the Glu693 sidechain of ELMO1 hydrogen bonds with the Lys103 side chain ofDOCK2. Similarly, the Tyr106 side chain of DOCK2 hydrogenbonds with the Asp685 side chain of ELMO1, and the Arg128

Fig. 2. DOCK2•ELMO1 interactions around the SH3 domain. (A) Interfacebetween the DOCK2 SH3 domain and the ELMO1 Pro-rich tail. The six prolineresidues (Pro707, Pro710, Pro711, Pro712, Pro714, and Pro717) in the ELMO1Pro-rich tail are colored red, and the labels of the two Pro residues formingthe PxxPmotif are underlined. Hydrogen bonds are indicated by dashed lines.The DOCK2 and ELMO1 residues are labeled in black and red, respectively.(B) Interface between the DOCK2•ELMO1 five-helix bundle and the DOCK2SH3 domain. The DOCK2 and ELMO1 residues are labeled in black and red,respectively.

Fig. 3. DOCK2•ELMO1 five-helix bundle formation. Interface within theDOCK2•ELMO1 five-helix bundle, showing the interactions between theDOCK2 and ELMO1 helices. Hydrogen bonds are indicated by dashed lines.

Hanawa-Suetsugu et al. PNAS ∣ February 28, 2012 ∣ vol. 109 ∣ no. 9 ∣ 3307

BIOCH

EMISTR

Y

Dow

nloa

ded

by g

uest

on

July

5, 2

020

side chain of DOCK2 hydrogen bonds with the Leu699 main-chain carbonyl group of ELMO1.

The hydrophobic residues within the three helices of DOCK2are highly conserved among the CDM family members (Fig. S3A).Actually, themutations of such hydrophobic residues, L96D/W99Eand I132D/L133D in DOCK180 (L95/W98 and L131/L132, re-spectively, in DOCK2), reportedly disrupt the ability to bindELMO1 (21). Therefore, DOCK180 forms a three-helix bundlesimilar to that of DOCK2, and uses it to interact with ELMO1. Onthe other hand, nearly all of the hydrophobic interactions betweenthe two helices are conserved in the ELMO family (Fig. S3B). TheDOCK2 residues on the interface with ELMO1 are well conservedwith those of DOCK180, whereas three residues, Trp96, Met121,Met125, are not conserved (Fig. S3A). As for ELMO1, the presentinteraction mode with DOCK2 (Fig. 3) agrees well with the resultsof a previous mutational study, which showed that Leu547 andIle548 of ELMO1 are required for DOCK180 binding (21). There-fore, DOCK180 is also likely to form a five-helix bundle, asobserved in the present DOCK2•ELMO1 complex.

In addition, the C terminus of Eα1 of ELMO1 forms hydro-phobic interactions with the Eα2 helix within the PH domain(Fig. S4D). These hydrophobic residues are strictly conservedin the ELMO family (Fig. S3B). The Eα1•Eα2 interactions fixthe arrangement of the PH domain relative to the five-helixbundle (Fig. 1B). The core structure of the PH domain ofELMO1 in the DOCK2•ELMOl complex is essentially the sameas that of the ELMO1 fragment consisting of the PH domain andthe Eα1 helix (21). However, Eα1 significantly bends, at the cen-tral Pro542, and interacts with Eα3 to form the five-helix bundlewith DOCK2, whereas Eα1 was straight in the previous structure,due to crystal packing (21) (Fig. S4E).

Characterization of DOCK2 and DOCK180Mutants.On the basis of theDOCK2•ELMO1 interactions described above, we mutatedseveral ELMO1-interacting residues in the full-length DOCK2protein. First, six putative hydrogen-bonding residues in the SH3domain (Fig. 2 A and B) were changed to Ala (Fig. 4A). Althoughthe Q30A and S62A mutations had no effect, the W4A, E39A,W44A, and R46A mutations significantly reduced the ELMO1binding. In particular, the Ala mutation of Trp44, which is locatedin the center of the SH3•PxxP interface (Fig. 2A), had the mostprofound effect on ELMO1 binding (Fig. 4A). Therefore, thehydrogen bonds of Trp44 with the PxxP motif, and those of Trp4,Glu39, and Arg46 with the C-terminal region downstream of thePxxP motif, are important for ELMO1 binding. On the other hand,the point mutations of Trp102 and Tyr106 of DOCK2 (Fig. 3) toGlu (Y106E and W102E) substantially reduced ELMO1 binding(Fig. 4B). Similarly, the R128A mutation nearly completely abol-ished the ELMO1 binding, whereas the S129A mutation did notaffect it (Fig. 4A). Arg128 is located at the C terminus of the Dα2helix, in the vicinity of the SH3 domain, and interacts with themainchain at the border between the Eα3 helix and the Pro-rich tail ofELMO1 (Figs. 2B and 3). These results indicate that both bindinginterfaces, the SH3 domain and the helix bundle, are important forthe DOCK2•ELMO1 complex formation.

In contrast to the marked reduction in ELMO1 binding by theDOCK2 W44A mutation (Fig. 4A), the point mutations of thecorresponding Trp residue of DOCK180 (W45A and W45K) re-portedly had minimal effects on ELMO1 binding (21, 22). Con-sequently, we confirmed the observed difference by mutating thecorresponding Trp residues to Lys in DOCK2 and DOCK180.The W44K mutation almost completely abolished the ability ofDOCK2 to bind to ELMO1, whereas the DOCK180 W45K mu-tation retained it (Fig. 4C). Therefore, these mutagenesis resultsindicated that DOCK2 and DOCK180 differ in their require-ments for the SH3•PxxP interaction in ELMO1 binding. Thisfinding led us to examine the functional significance of theSH3•PxxP interaction in DOCK2-mediated chemotaxis. For this

purpose, we stably expressed wild-type DOCK2 and the W44Amutant in BW5147α−β− cells, which lack the endogenous expres-sion of DOCK2 (2). As compared with the nontransfected cells,the expression of wild-type DOCK2 in BW5147α−β− cells signif-icantly increased the migration speed (Fig. 4D). However, the ex-pression of the W44A mutant failed to restore the motility in two

Fig. 4. The SH3–PxxP interaction is essential for ELMO1-binding in DOCK2.(A–C) Interaction of the ELMO1 binding residues with DOCK2. HEK293T celllysates expressing HA-tagged DOCK2 (WTand mutants) were incubated withGST-tagged ELMO1. The DOCK2-HA in the pull-downs and total cell lysateswas detected by immunoblotting with an anti-HA antibody. The results werequantified by densitometry and are expressed as the ratio of ELMO1 boundto total DOCK2 after normalization to the WT value, with an arbitrary valueof 1. The results indicate mean� SD of three independent experiments.(D) Chemotactic response of BW5147α−β− cells expressing DOCK2 (WTor mu-tant). BW5147α−β− cells (−) stably expressing HA-tagged DOCK2 (clones 4B3and 4C2) or HA-tagged DOCK2W44A (clones 2B9 and 3G10) were allowed tomigrate on stromal cells prepared from mouse lymph nodes. The migrationspeed was calculated by analyzing at least 148 cells per clone. **P < 0.01. (E)The effect of the W44A mutation on Rac activation. BW5147α−β− cells (−)expressing wild-type DOCK2 (clone 4C2) or the W44A mutant (clone 3G10)were stimulated with CXCL12 (400 ng∕mL) before the assay. Data are repre-sentative of three independent experiments. (F) The effect of the W44A mu-tation on ELMO1-mediated relief of DOCK2 autoinhibition. The GFP-taggedDOCK2 N-terminal fragment (Met1–Glu200), with or without the W44Amutation, and the FLAG-tagged DOCK2 C-terminal fragment (Ser1195–Met1828) were expressed in HEK293T cells in the presence or absence of theHA-tagged ELMO1. TCL, total cell lysate. The association between the N- andC-terminal fragments was analyzed by reciprocal immunoprecipitations withthe relevant antibodies. Data are representative of two independent experi-ments.

3308 ∣ www.pnas.org/cgi/doi/10.1073/pnas.1113512109 Hanawa-Suetsugu et al.

Dow

nloa

ded

by g

uest

on

July

5, 2

020

independent clones, although its expression level was comparableto that of the wild-type DOCK2 (Fig. 4D). Consistent with thisfinding, chemokine-induced Rac activation was readily detectedin BW5147α−β− cells expressing the wild-type DOCK2, but notthe W44A mutant (Fig. 4E). Collectively, these results indicatethat the SH3•PxxP interaction is required for DOCK2-mediatedRac activation.

The SH3 domain of DOCK180 has been implicated in its auto-inhibition by binding to the catalytic DHR-2 domain and thusblocking Rac access to it (22). To explore the mechanism by whichthe SH3•PxxP interaction regulates the DOCK2-mediated Racactivation, we transiently expressed the DOCK2 N-terminalfragment (Met1–Glu200), with or without the W44A mutation,and the C-terminal fragment (Ser1195–Met1828), including theDHR-2 domain, in HEK293Tcells. Regardless of the presence ofthe W44A mutation, the N-terminal fragment of DOCK2 asso-ciated with the C-terminal fragment, suggesting that a similarautoinhibitory mechanism occurs in DOCK2 (Fig. 4F). However,upon the coexpression of ELMO1, this association was disruptedin the case of the wild-type DOCK2, but not the W44A mutant(Fig. 4F). These results suggest that the SH3•PxxP interaction isrequired to relieve the autoinhibition of DOCK2 and to allowRac access to the catalytic DHR-2 domain.

DiscussionDOCK2 is a key regulator of the immune system. In this study, wedetermined the crystal structure of the complex of the DOCK2and ELMO1 interacting regions, which revealed a large, rigid as-sembly. Three α-helices of DOCK2 and two α-helices of ELMO1form a helix bundle. The SH3 domain of DOCK2 interacts withthe long Pro-rich tail and the two α-helices of ELMO1 in an unu-sually extensive manner. Together, these structural elements, aswell as the PH domain of ELMO1, accrete to form a rigid assem-bly. A mutational analysis demonstrated that the entire assemblyof these two regions is required for the full-length DOCK2 andELMO1 proteins to form a complex and activate Rac in vivo.

The SH3 domain of DOCK180 reportedly interacts with thecatalytic DHR-2 domain, which may sterically block Rac accessto the DHR-2 domain and thereby cause the autoinhibition ofDOCK180 (22). Actually, in the present study, we observed thebinding of the N-terminal ELMO-interacting region with the C-terminal DHR-2 containing region of DOCK2 and its reductionby the assembly with ELMO (Fig. 4F). Therefore, a similar me-chanism underlies the autoinhibition of the DOCK2 GEF activ-ity. In the case of DOCK180, the mutation of the conserved Ileresidue within the SH3 domain (I32K) disrupts the SH3•DHR-2interaction, and relieves DOCK180 of the autoinhibition (22).This Ile residue corresponds to Ile31 in DOCK2, within the RTloop of the SH3 domain. Ile31 is not close to the PxxP-bindingsite, but is buried at the center of the hydrophobic interface withthe Eα1 and Eα3 helices and the following peptide segment ofELMO1 (Fig. 2B).

On the other hand, the ELMO1 protein also has an autoinhi-bitory mechanism (23). The EID binds to the EAD in the auto-inhibited form, and EAD binding with DOCK180 relievesELMO1 of its autoinhibition. Therefore, DOCK2/DOCK180 andELMO1 mutually relieve their autoinhibited forms (Fig. 5A). Thedouble mutation of Met692 and Glu693 to Ala within the EADreportedly disrupts the interaction between the EID and the EAD.Correspondingly, Met692 is buried at the center of the hydro-phobic core of the intermolecular five-helix bundle, whereas theGlu693 side chain interacts with the Lys103 side chain of DOCK2,as revealed by the present structure. As described above, thehydrophobic core of the five-helix bundle also buries Ile31 ofDOCK2, which is the putative key residue for the DOCK2 auto-inhibition. Therefore, the tight assembly between the ELMO1-interacting region of DOCK2 and the DOCK2-interacting regionof ELMO1 deprives these regions of their respective intramolecu-

lar autoinhibitory interactions. Thus, the structural basis by whichDOCK2 and ELMO1 mutually relieve their autoinhibited formshas now been revealed (Fig. 5A).

The PH domain of ELMO1 is firmly fixed, relative to the five-helix bundle, in the rigid DOCK2•ELMO1 complex. TheELMO1 PH domain reportedly cooperates with DOCK180 toactivate Rac (20). The mutations of the PH domain residues,such as G559A and W665A, in ELMO1 abrogate the formationof the functional ternary complex with DOCK180 and nucleo-tide-free Rac, although they do not affect the DOCK180 binding(20). In the present DOCK2•ELMO1 structure, Gly559 andTrp665 of ELMO1 are involved in intramolecular interactionsbetween the Eα1 and Eα2 helices (Fig. S4D). Therefore, theirmutations are expected to impair the proper positioning of theELMO1 PH domain relative to the other portions of theDOCK•ELMO complex. Thus, the fixed arrangement of the

Fig. 5. Hypothetical model of the DOCK2•ELMO1•Rac1 ternary complex. (A)The schematic model of the mutual relief of DOCK2 and ELMO1 from theirautoinhibited forms for Rac activation. EIR, ELMO1-interacting region; DIR,DOCK2-interacting region. (B) Side view, showing the membrane attachmentof the DOCK2•ELMO1•Rac1 ternary complex. The membrane is shown as alight yellow bar. The membrane attachment site of Rac1 at the C terminus isdepicted by a red dashed line. (C) Top view from the membrane side. Thepresent structure of the DOCK2•ELMO1 complex (green and yellow) is over-laid onto the structures of the DOCK180 DHR-1 domain (orange, PDB code3L4C) (13), the dimeric DOCK2 DHR-2 domain•Rac1 complex (blue and red,PDB code 3B13), and importin subunit β-1 (ARM, armadillo-like repeat) (gray,PDB code 2QNA) (30). The two views are related by a 90° rotation about thehorizontal axis.

Hanawa-Suetsugu et al. PNAS ∣ February 28, 2012 ∣ vol. 109 ∣ no. 9 ∣ 3309

BIOCH

EMISTR

Y

Dow

nloa

ded

by g

uest

on

July

5, 2

020

ELMO1 PH domain may be important for functional ternarycomplex formation with DOCK and Rac.

In the typical mammalian RhoGEFs, such as Trio and Dbs, theDH and PH domains act as a “cassette” and coordinately bindtheir cognate Rho GTPases (Rac1 and Cdc42) for efficient acti-vation (29). By analogy to the DH–PH domains, the PH domainof ELMO1 may facilitate Rac1 binding by the DHR-2 domain. Astructural model of the DOCK180•Rac complex has been pro-posed (13), and thus we integrated the present DOCK2•ELMO1structure into a similar model of the DOCK2•Rac1 complex(Fig. 5 B and C, and Table S3). In this preliminary model, thetwo DOCK2•ELMO1 complexes might interact with two Rac1molecules on the flat surface of the plasma membrane. ThePH domain of ELMO1 may be arranged similarly to that inthe Rac1•Trio DH–PH cassette (Protein Data Bank code 2NZ8).

The ELMO1-interacting residues of DOCK2 are also highlyconserved in DOCK180. In fact, the present structure indicatesthat their overall interaction modes may be similar to each other.Nevertheless, we found an interesting difference betweenDOCK2 and DOCK180. The mutation of Trp44, within the SH3domain of DOCK2, abolished the ELMO1 binding ability, butthe corresponding mutation of Trp45 of DOCK180 did not affectit (Fig. 4C). Phe63, within the SH3 domain of DOCK2, formshydrophobic interactions with the ELMO1 PxxP-tail, but it is re-placed with Tyr in DOCK180. Similarly, in the binding interfacein the five-helix bundle, Trp96, Met121, and Met125 of DOCK2are replaced with Arg, Ile, and Ile, respectively, in DOCK180(Fig. S3A). These amino acid replacements may alter the ELMO-binding affinity between DOCK2 and DOCK180. Alternatively,an additional region of DOCK180, including Gly171, whose mu-tation reduces ELMO1 binding (22), may contribute to the tigh-ter binding of ELMOl to DOCK180, as compared to DOCK2.In the current DOCK2•ELMOl complex structure, the corre-sponding Gly residue of DOCK2 (Gly172) is located within a dis-

ordered C-terminal region (residues 168–177), and thus is notinvolved in ELMO1 binding.

Our observation that DOCK2•ELMO1 complex formationis essential for the DOCK2 function makes it an attractive ther-apeutic target for immunologic disorders, such as autoimmunediseases and graft rejection, caused by tissue infiltration by lym-phocytes. Although the mechanism by which DOCK180 main-tains the SH3 domain-independent, tight binding of ELMO1remains elusive, the observed difference raises the possibility thatELMO1 binding might be specifically disrupted by small com-pounds that bind to the interface in DOCK2, but not DOCK180.In this context, our structure will provide important informationfor the development of an inhibitor that selectively disruptsDOCK2•ELMO1 complex formation.

Materials and MethodsThe protein samples for structure determination were produced by the cell-free synthesis method. The solution structure of the DOCK2 SH3–ELMO1 fu-sion protein [Protein Data Bank (PDB) code 2RQR] was determined by NMR.The crystal structures of the DOCK2•ELMOl (PDB code 3A98) and DOCK2-DHR2•Rac1 (PDB code 3B13) complexes were determined by the single-wa-velength anomalous dispersion and molecular replacement methods, respec-tively. Detailed experimental procedures are provided in the SI Text.

ACKNOWLEDGMENTS.We thank Y. Tomo, M. Aoki, E. Seki, M. Ikari, K. Hanada,T. Matsuda, N. Matsuda, Y. Motoda, Y. Fujikura, T. Harada, S. Watanabe, andT. Nagira for preparing proteins for NMR experiments. We also thankH. Niwa, K. Katsura, S. Tojo, M. Ueno, K. Honda, M. Goto, M. Inoue, M. Aoki,A. Sakamoto, M. Toyama, Y. Terazawa, T. Fujimoto, M. Wakiyama, A. Shima-da, and Y. Fujii for technical assistance. We are grateful to Dr. Osamu Ohara(Kazusa DNA Research Institute, Japan) for the DOCK2 clone (KIAA0209). Thiswork was supported by the RIKEN Structural Genomics/Proteomics Initiative,the National Project on Protein Structural and Functional Analyses, and bythe Targeted Proteins Research Program, the Ministry of Education, Culture,Sports, Science, and Technology of Japan.

1. Nishihara H, et al. (1999) Non-adherent cell-specific expression of DOCK2, amember ofthe human CDM-family proteins. Biochim Biophys Acta 1452:179–187.

2. Fukui Y, et al. (2001) Haematopoietic cell-specific CDM family protein DOCK2 is essen-tial for lymphocyte migration. Nature 412:826–831.

3. Sanui T, et al. (2003) DOCK2 is essential for antigen-induced translocation of TCR andlipid rafts, but not PKC-theta and LFA-1, in T cells. Immunity 19:119–129.

4. Jiang H, et al. (2005) Deletion of DOCK2, a regulator of the actin cytoskeleton in lym-phocytes, suppresses cardiac allograft rejection. J Exp Med 202:1121–1130.

5. Tanaka Y, et al. (2007) T helper type 2 differentiation and intracellular trafficking ofthe interleukin 4 receptor-alpha subunit controlled by the Rac activator Dock2. NatImmunol 8:1067–1075.

6. Kunisaki Y, et al. (2006) DOCK2 is a Rac activator that regulates motility and polarityduring neutrophil chemotaxis. J Cell Biol 174:647–652.

7. Nishikimi A, et al. (2009) Sequential regulation of DOCK2 dynamics by two phospho-lipids during neutrophil chemotaxis. Science 324:384–387.

8. Gotoh K, et al. (2010) Selective control of type I IFN induction by the Rac activatorDOCK2 during TLR-mediated plasmacytoid dendritic cell activation. J Exp Med207:721–730.

9. Côté JF, Vuori K (2007) GEF what? Dock180 and related proteins help Rac to polarizecells in new ways. Trends Cell Biol 17:383–393.

10. Meller N, Irani-Tehrani M, Kiosses WB, Del Pozo MA, Schwartz MA (2002) Zizimin1, anovel Cdc42 activator, reveals a new GEF domain for Rho proteins. Nat Cell Biol4:639–647.

11. Miyamoto Y, Yamauchi J, Sanbe A, Tanoue A (2007) Dock6, a Dock-C subfamily gua-nine nucleotide exchanger, has the dual specificity for Rac1 and Cdc42 and regulatesneurite outgrowth. Exp Cell Res 313:791–804.

12. Côté JF, Motoyama AB, Bush JA, Vuori K (2005) A novel and evolutionarily conservedPtdIns(3,4,5)P3-binding domain is necessary for DOCK180 signalling. Nat Cell Biol7:797–807.

13. Premkumar L, et al. (2010) Structural basis of membrane targeting by the Dock180family of Rho family guanine exchange factors (Rho-GEFs). J Biol Chem285:13211–13222.

14. Kulkarni K, Yang J, Zhang Z, Barford D (2011) Multiple factors confer specific Cdc42and Rac activation by Dock exchange factors. J Biol Chem 286:25341–25351.

15. Yang J, Zhang Z, Roe SM, Marshall CJ, Barford D (2009) Activation of Rho GTPases byDOCK exchange factors is mediated by a nucleotide sensor. Science 325:1398–1402.

16. Sanui T, et al. (2003) DOCK2 regulates Rac activation and cytoskeletal reorganizationthrough interaction with ELMO1. Blood 102:2948–2950.

17. Brugnera E, et al. (2002) Unconventional Rac-GEF activity is mediated through theDock180-ELMO complex. Nat Cell Biol 4:574–582.

18. Gumienny TL, et al. (2001) CED-12/ELMO, a novel member of the CrkII/Dock180/Racpathway, is required for phagocytosis and cell migration. Cell 107:27–41.

19. Grimsley CM, et al. (2004) Dock180 and ELMO1 proteins cooperate to promote evo-lutionarily conserved Rac-dependent cell migration. J Biol Chem 279:6087–6097.

20. Lu M, et al. (2004) PH domain of ELMO functions in trans to regulate Rac activation viaDock180. Nat Struct Mol Biol 11:756–762.

21. Komander D, et al. (2008) An alpha-helical extension of the ELMO1 pleckstrin homol-ogy domain mediates direct interaction to DOCK180 and is critical in Rac signaling.Mol Biol Cell 19:4837–4851.

22. Lu M, et al. (2005) A Steric-inhibition model for regulation of nucleotide exchange viathe Dock180 family of GEFs. Curr Biol 15:371–377.

23. Patel M, et al. (2010) An evolutionarily conserved autoinhibitory molecular switch inELMO proteins regulates Rac signaling. Curr Biol 20:2021–2027.

24. Kigawa T, Matsuda T, Yabuki T, Yokoyama S (2007) Cell-free Protein Synthesis, edsAlexander S Spirin and James R Swartz (Wiley, New York), pp 83–97.

25. Kigawa T, et al. (2004) Preparation of Escherichia coli cell extract for highly productivecell-free protein expression. J Struct Funct Genomics 5:63–68.

26. Matsuda T, et al. (2007) Improving cell-free protein synthesis for stable-isotope label-ing. J Biomol NMR 37:225–229.

27. Cesareni G, Panni S, Nardelli G, Castagnoli L (2002) Can we infer peptide recognitionspecificity mediated by SH3 domains? FEBS Lett 513:38–44.

28. Mayer BJ (2001) SH3 domains: Complexity in moderation. J Cell Sci 114:1253–1263.29. Chhatriwala MK, Betts L, Worthylake DK, Sondek J (2007) The DH and PH domains of

Trio coordinately engage Rho GTPases for their efficient activation. J Mol Biol368:1307–1320.

30. Wohlwend D, Strasser A, Dickmanns A, Ficner R (2007) Structural basis for RanGTP in-dependent entry of spliceosomal U snRNPs into the nucleus. J Mol Biol 374:1129–1138.

3310 ∣ www.pnas.org/cgi/doi/10.1073/pnas.1113512109 Hanawa-Suetsugu et al.

Dow

nloa

ded

by g

uest

on

July

5, 2

020