GTC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Genes to Cells (2007) 12, 1255-1265. doi:10.1111/j.1365-2443.2007.01132.x
© 2007 Blackwell Publishing or its licensors
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yamashita, M.
Right arrow Articles by Nureki, O.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Yamashita, M.
Right arrow Articles by Nureki, O.

Crystal structure of human DAAM1 formin homology 2 domain

 Masami Yamashita1,a, Tomohito Higashi2,a, Shiro Suetsugu3,4, Yusuke Sato1, Tomoyuki Ikeda2, Ryutaro Shirakawa2, Toru Kita2, Tadaomi Takenawa5, Hisanori Horiuchi2, Shuya Fukai1,6,b,* and Osamu Nureki1,7,8,*

1 Department of Biological Information, Graduate School of Bioscience and Biotechnology, Tokyo Institute of Technology, 4259 Nagatsuta-cho, Midori-ku, Yokohama-shi, Kanagawa 226-8501, Japan
2 Department of Cardiovascular Medicine, Graduate School of Medicine, Kyoto University, 54 Shogoinkawara-machi, Sakyo-ku, Kyoto 606-8507, Japan
3 Institute of Molecular and Cellular Biosciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-0032, Japan
4 Precursory Research for Embryonic Science and Technology (PRESTO), Japan Science and Technology Agency (JST), Honcho, Kawaguchi-shi, Saitama 332-0012, Japan
5 Graduate School of Medicine, Kobe University, 7-5-1, Kusunoki-cho, Chuo-ku, Kobe-shi, Hyogo 650-0017, Japan
6 Center for Biological Resources and Informatics, Tokyo Institute of Technology, 4259 Nagatsuta-cho, Midori-ku, Yokohama-shi, Kanagawa 226-8501, Japan
7 RIKEN Genomics Sciences Center, 1-7-22 Suehiro-cho, Tsurumi, Yokohama 230-0045, Japan
8 Solution Oriented Research for Science and Technology (SORST), Japan Science and Technology Agency (JST), Honcho, Kawaguchi-shi, Saitama 332-0012, Japan


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Experimental procedures
 Note added in proof
 References
 
Reorganization of the actin filament is an essential process for cell motility, cell–cell attachment and intracellular transport. Formin proteins promote nucleation and elongation of the actin filament, and thus are key regulators for this process. The formin homology 2 (FH2) domain forms a head-to-tail ring-shaped dimer, and processively moves towards the barbed end. Dishevelled-associated activator of morphogenesis (DAAM) is a Rho-regulated formin implicated in neuronal development. Here, we present the crystal structure of human DAAM1 FH2 dimer at 2.8 Å resolution. This is the first dimeric structure of the mammalian formin. The core structure of human DAAM1 is similar to those of mouse mDia1 and yeast Bni1p, whereas the orientations of the FH2 dimeric rings are different between human DAAM1 and yeast Bni1p, despite their similar dimer interactions. This difference supports the previous prediction that the dimer architecture of the formin is highly flexible in the actin-free state. The results of the actin assembly assays using the DAAM1 mutants demonstrated that the length of the linker connecting the N-terminal domain and the core region is crucial for the activity.


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Experimental procedures
 Note added in proof
 References
 
Formin protein family regulates nucleation and elongation of the actin filament (Evangelista et al. 1997; Ishizaki et al. 2001; Pruyne et al. 2002; Sagot et al. 2002; Dong et al. 2003; Kovar et al. 2003; Li & Higgs 2003; Higashida et al. 2004). The formin family is conserved beyond species from yeast to mammals (Higgs & Peterson 2005). In general, the Diaphanous-related formins are direct effector molecules of the small GTPase, Rho (Kohno et al. 1996; Evangelista et al. 1997; Watanabe et al. 1997; Yayoshi-Yamamoto et al. 2000). In the resting stage, the Diaphanous-related formins remain inactive due to the intramolecular interaction between the N-terminal Diaphanous-inhibitory domain (DID) and the C-terminal Diaphanous-autoregulatory domain (DAD) (Watanabe et al. 1999; Alberts 2001). All the formin members have common domains referred to as formin-homology-1 (FH1) and FH2 domains between DID and DAD (Higgs 2005). Upon binding with the GTP-bound Rho, the interaction between DID and DAD is disrupted, which exposes FH1–FH2 domains, leading to the initiation of the actin polymerization (Lammers et al. 2005; Otomo et al. 2005a; Rose et al. 2005). Proline-rich FH1 can bind profilin and enhance the actin filament elongation (Kovar et al. 2003, 2006; Kovar & Pollard 2004; Romero et al. 2004), while FH2 directly binds to the actin filament at its barbed end. FH2 catalyzes the nucleation and elongation of the actin filament, preventing capping proteins from binding to the barbed end (Kovar et al. 2003; Zigmond et al. 2003; Harris et al. 2004; Higashida et al. 2004; Kovar & Pollard 2004; Moseley et al. 2004; Romero et al. 2004).

The structure of the FH2 domain of Bni1p, a yeast formin, has been reported (Xu et al. 2004; Otomo et al. 2005b). The FH2 domain is divided into five subdomains termed as lasso, linker, knob, coiled-coil and post. The structure of Bni1p FH2 was interpreted as a dimer based on the crystallographic symmetry. The dimer formation is mediated by unique interactions of "lasso" with "post" of the partner FH2, exhibiting a closed ring structure. FH2 incapable of dimer formation lost the actin polymerization activity, indicating that the dimer formation is essential for the activity (Shimada et al. 2004; Xu et al. 2004). The complex structure of the actin and Bni1p FH2 has been recently determined (Otomo et al. 2005b). The structure exhibited that the post/lasso and knob regions are involved in the contact with the actin monomer. A structural unit including these two contact sites is termed as a bridge: a single bridge interacts with one actin monomer. In the crystal, the actin monomers are related by 21 helical symmetry (Holmes et al. 1990), where Bni1p FH2 formed a helical oligomer around the actin molecules. Thus, Rosen and his colleagues constructed the dimeric FH2 model from the Bni1p FH2 oligomeric structure by swapping the alternate lasso–post connection, and unraveled the details of the "stair-stepping" model for the actin polymerization: the "stair-stepping" of each bridge on the barbed end maintains a space between the actin filament and FH2 to accommodate the next incoming actin monomer, and thereby substantiate the processive movement of FH2. However, the stair-stepping along the actin filament intrinsically causes the helical tension. Therefore, the screw model to relax the helical tension has also been optionally proposed (Shemesh et al. 2005).

Formins are widely diverged in between non-metazoan and metazoan. For mammalian formins, only the core FH2 structure, lacking lasso and linker subdomains, has been reported for mDia1 (Shimada et al. 2004). The core FH2 of mDia1 exhibited three helical domains similar to Bni1p. Dishevelled-associated activator of morphogenesis-1 (DAAM1) is a Rho-regulated formin (Habas et al. 2001; Nakaya et al. 2004; Higgs & Peterson 2005), which was identified as a binding protein of the Wnt receptor-associated protein, dishevelled. Interestingly, DAAM1 was reported to bind the GTP-bound RhoA and increase its GTP-bound form population in cells (Habas et al. 2001). Over-expression and knockdown of cellular DAAM1 altered the features of actin filaments. In development, DAAM1 and its homologue DAAM2 are expressed complementarily in different regions in brain (Nakaya et al. 2004). These data suggest important roles of DAAM1 for development and other cellular functions. However, the physiological and biochemical functions of DAAM1 as a formin remain to be elucidated.

In this study, we determined the crystal structure of the homodimeric FH2 domain of human DAAM1 at 2.8 Å resolution. The homodimeric FH2 domains of DAAM1 indeed form a contracted closed ring of a rectangular shape, in contrast to the parallelogram-shaped ring of the Bni1p FH2 domain (Xu et al. 2004): in the dimeric ring, the orientation of the coiled-coil structures of the DAAM1 FH2 domain is largely distinct from that of the Bni1p FH2 domain. This supports the previous prediction that the dimeric structure of the formin FH2 is flexible. Docking analysis using the actin-bound Bni1p FH2 indicates that the present ring structure of DAAM1 FH2 has to be expanded for the actin elongation. The functional importance of the linker length was revealed by pyrene-labeled actin assembly assay using mutants with various linker lengths.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Experimental procedures
 Note added in proof
 References
 
Overall structure of human DAAM1 FH2

Based on the primary sequence, DAAM proteins apparently belong to the formin family. However, their biochemical properties have not been well characterized so far. First, we analyzed the actin assembly activity of human DAAM1 by using pyrene-labeled actin. As shown in Fig. 1A, the human DAAM1 FH1–FH2–DAD assembled the actin molecules in a concentration-dependent manner. To determine the crystal structure, the FH2 region of human DAAM1 (residues 593–1022, referred to as "DAAM1 FH2" hereafter) was overproduced in Escherichia coli as a GST-fusion protein. The GST-fused DAAM1 FH2 was purified by Glutathione Sepharose (GE Healthcare, USA), and then, the GST-tag was cleaved by thrombin. Subsequently, DAAM1 FH2 was purified by cation-exchange chromatography and gel filtration chromatography. The native crystal belongs to the space group of P1 with cell constants, a = 69.2 Å, b = 91.9 Å, c = 97.7 Å, {alpha} = 98.1°, ß = 90.3°, {gamma} = 104.8°. The phase was determined by multiple-wavelength anomalous dispersion (MAD) method using the selenomethionine-labeled protein. The asymmetric unit contains two copies of the FH2 dimer. Here, each molecule is termed as chains A, B, C and D, corresponding to chain IDs in the deposited PDB file (PDB ID 2Z6E). The final model contains residues 601–655 and 680–1012 for the chain A, residues 594–659 and 682–1011 for the chain B, residues 601–657 and 681–1002 for the chain C, and residues 601–657 and 680–1012 for the chain D. The values of Rwork and Rfree for the final model (30–2.8 Å) were 0.225 and 0.288, respectively (Table 1).


Figure 1
View larger version (52K):
[in this window]
[in a new window]

  		Figure 1  Human DAAM1 function and structure. (A) Concentration-dependent actin assembly of the ring-shaped FH2 domain of human DAAM1. Actin assembly was measured by incubation of the pyrene-labeled actin in the presence of various concentrations of DAAM1 FH1-FH2-DAD. (B) Human DAAM1 FH2 dimer structure. The dimeric ring structure of human DAAM1 FH2 is shown as a ribbon model. The lasso (residues 593–650), linker (residues 651–689), knob (residues 690–768), coiled-coil (residues 769–807 and 888–961) and post (residues 808–887, 962–1012) regions are colored blue, orange, green, pink and yellow, respectively. Structurally disordered region in the linker is shown as a dotted orange line. The ring shape of human DAAM1 FH2 is clearly different from those of yeast Bni1p FH2 molecules in the actin-free and actin-bound states (Xu et al. 2004; Otomo et al. 2005b).

 

View this table:
[in this window]
[in a new window]

  		Table 1  Data collection, phasing and refinement statistics
 
One DAAM1 FH2 monomer interacts with the other monomer in a head-to-tail manner to form a closed rectangular ring (Fig. 1B), in contrast to the parallelogram-shaped ring of yeast Bni1p FH2 dimer in the actin-free state (Xu et al. 2004). The diameter of the DAAM1 FH2 dimer is smaller than that of yeast Bni1p FH2 dimer in the actin-bound state (Otomo et al. 2005b). As well as the other FH2 structures, DAAM1 FH2 can be divided into five functional regions called "lasso (residues 593–650, colored blue)," "linker (residues 651–686, colored orange)," "knob (residues 687–768, colored green)," "coiled-coil (residues 769–807 and 888–961, colored pink)" and "post (residues 808–887 and 962–1012, colored yellow)," as shown in Fig. 1B. In the present structure, most part of the linker region was structurally disordered, except for the proximity of the ß-sheet-like structure comprising residues 651–653 and 684–686. Even in the chain B, which are ordered better than any other chains, we could not completely build the model for the linker region.

Lasso–post interactions of the DAAM1 FH2 dimer

The head-to-tail dimer of DAAM1 FH2 is formed by the intermolecular interactions between the lasso region of one monomer and the post region of the other monomer. The lasso–post interactions are almost the same in four DAAM1 FH2 molecules in the asymmetric unit. Thus, we describe here the interactions between the post region of the chain A and the lasso region of the chain B. Chain IDs are attached to the residue numbers as suffixes for clarity (e.g. Trp615B). The lasso–post interactions in human DAAM1 FH2 dimer are mostly hydrophobic as well as in yeast Bni1p FH2 dimer (Fig. 2A–D, right). Three hydrophobic pockets are formed around the central {alpha}-helix (residues 808–823) of the post region (Fig. 2A, left). The pocket formed by Leu814A, Leu818A and Phe835A surrounds the side chain of Trp628B (Fig. 2B, left), whereas that formed by Phe820A, Val817A, Met824A, Phe835A, Ile843A, Leu856A and Leu860A surrounds the side chains of Phe613B and Trp615B (Fig. 2C, left). These two pockets are formed by the post region of the chain A. On the other hand, Phe637B, Leu642B, Pro604B and Pro606B, which are in the lasso region of the chain B (Fig. 2D, left), surrounds the side chain of Tyr823A. Note that Trp615B, Trp628B and Tyr823A are conserved or replaced with the functionally-equivalent amino acid among the FH2 domains (Fig. 3A).


Figure 2
View larger version (63K):
[in this window]
[in a new window]

  		Figure 2  The lasso–post interactions of human DAAM1 FH2 dimer. (A–D) The post region of the chain A and the lasso region of the chain B are colored blue and pink, respectively. The amino acid residues involved in the lasso–post interactions are shown as sticks. The hydrogen bonds are shown as dotted yellow lines.

 

Figure 3
View larger version (39K):
[in this window]
[in a new window]

  		Figure 3  Amino acid sequence alignment of the FH2 domains among human DAAM1, mouse mDia1 and yeast Bni1p, and the actin assembly assay using DAAM1 FH2 mutated in the dimer interface. (A) Colored grey were regions whose structures were not reported. Color bars above the sequence alignment show each subdomain, where coloring scheme is the same as Fig. 1C. Human DAAM1 residues are numbered every 10 amino acid residues. The amino acid residues involved in the lasso–post interactions are colored orange and indicated by asterisks (*). (B) Assembly rates of the human DAAM1 FH1-FH2-DAD mutants, W615A and W628A. Actin assembly was measured by incubation of the pyrene-labeled actin in the presence of the DAAM1 FH1-FH2-DAD mutants at 50 nM.

 
In addition to the hydrophobic interactions, eight hydrogen bonds are formed between the lasso and post regions. Around the Phe613B/Trp615B-bound pocket, the O{delta} atom of Asn825A hydrogen bonds with the main-chain NH of Phe613B (Fig. 2C, left). Asn825A is in the most conserved sequence motif (-G-N-Y/F-M-N-) in the post region (Fig. 3A), and its counterpart of yeast Bni1p, Asn1580, makes bipartite hydrogen bonds with the main-chain NH and CO of Leu1361 (Fig. 2C, right). In yeast Bni1p FH2, Gln1360 also form bipartite hydrogen bonds with the main-chain NH and CO of Gly1578 and Asn1580, respectively (Fig. 2C, right), but this hydrogen bonding network was not formed in human DAAM1 FH2. Besides these hydrogen bonds, the main-chain CO of Arg829A hydrogen bonds with the N{varepsilon} atom of Trp615B (Fig. 2C, right). The N{xi} atom of Lys617B hydrogen bonds with the main-chain CO of Gly830A. In the Bni1p FH2, the hydrogen bonding network is formed only around the Leu1361B/Trp1363B-bound pocket (corresponding to the Phe613B/Trp615B-bound pocket of DAAM1 FH2) (Fig. 2C, right). On the other hand, the hydrogen bonding network is also formed around the Tyr823A-bound pocket in DAAM1 FH2 (corresponding to the Phe1578-bound pocket in Bni1p FH2) (Fig. 2D). The O{eta} atoms of Tyr823A and Tyr859A hydrogen bond with the O{varepsilon} atom of Glu646B and the O{delta} atom of Asp644B, respectively. The O{gamma} atom of Ser849A simultaneously hydrogen bonds with the main-chain CO of Glu647B and the main-chain NH of Ser650B.

The dimer interface of the rectangular human DAAM1 FH2 ring resembles that of the parallelogram-shaped yeast Bni1p FH2 ring. In addition, the replacement of the conserved Trp residue, Trp615 or Trp628, by an Ala residue reduced or lost the actin assembly activity, respectively (Fig. 3B). This result shows that the conserved dimer interface architecture is important for the activity. Therefore, the architecture of the dimer interface may be generalized to all, or most of, formin FH2 homodimeric structures.

Different orientation of the FH2 dimer ring between yeast Bni1p and human DAAM1

The core structure of the human DAAM1 FH2 monomer (including the knob, coiled-coil and post regions) can be well superposed on that of yeast Bni1p (2.9 Å rmsd over 282 C{alpha} atoms) in the actin-free state (Fig. 4A). In addition, the lasso–post interactions are well conserved as described above. Nevertheless, the orientations of the FH2 dimer rings are quite different between yeast Bni1p and human DAAM1 (Fig. 4A). The direction of the pseudo twofold axis of the DAAM1 FH2 dimer is perpendicular to that of the yeast Bni1p dimer, due to the structural difference in the linker region and the different lasso orientation relative to the FH2 core (Fig. 4B). In yeast actin-free Bni1p FH2, the linker region consists of a single {alpha}-helix with short loops attached with both ends. This structure is expanded upon binding with an actin molecule (Fig. 1B) (Otomo et al. 2005b). Since the linker region is substantially elastic, the orientation of the lasso region relative to the FH2 core is not restricted. In contrast, in the DAAM1 FH2 dimer, the ß-sheet-like structure, comprising residues 651–653 and 684–686, fixes the orientation of the lasso region, although the other part of the linker region was structurally disordered (Fig. 4C). This ß-sheet-like structure seems to function as a fastener to stick together the N- and C-terminal ends of the linker region.


Figure 4
View larger version (49K):
[in this window]
[in a new window]

  		Figure 4  Dimeric ring structure. (A) Comparison of the FH2 dimer ring orientation between human DAAM1 FH2 and yeast Bni1p FH2. DAAM1 FH2 and Bni1p FH2 are superposed on the FH2 core (including the knob, coiled-coil and post regions). DAAM1 and Bni1p FH2 are colored blue and pink, respectively. The pseudo twofold axis of the DAAM1 ring and twofold axis of the Bni1p ring are shown as blue and pink solid lines, respectively. (B) Comparison of the orientation of the lasso region relative to the FH2 core among human DAAM1 FH2, yeast actin-free Bni1p FH2, and yeast actin-bound Bni1p FH2. Human DAAM1 FH2, yeast actin-free Bni1p FH2 and yeast actin-bound Bni1p FH2 are colored blue, pink and green, respectively. Structurally disordered region in the DAAM1 FH2 linker is shown as a dotted blue line. (C) 2Fo–Fc electron density map around the ß-sheet-like structure of the chain B in the linker region (contoured at 1.1{sigma} level).

 
Functional importance of the linker length

The core structure of the human DAAM1 FH2 monomer can be well superposed on that of the yeast actin-bound Bni1p (data not shown) (2.7 Å rmsd over 282 C{alpha} atoms). Based on this superposition, we generated the docking model of human DAAM1 FH2 and the actin filament. First, we placed the DAAM1 FH2 dimer so that its monomer could be superposed on the actin-bound Bni1p FH2 (i.e. a DAAM1 FH2 dimer–actin monomer complex was modeled). Then, the ideal actin filament modeled by Holmes and Kabsch (Holmes et al. 1990) was placed so that its actin monomer located on the barbed end could be superposed on the FH2-bound actin monomer (i.e. the actin monomer in the DAAM1 FH2 dimer–actin monomer complex was replaced by the modeled ideal actin filament) (Fig. 5A). Binding of the DAAM1 FH2 to the actin filament forms an intermolecular cleft for the incorporation of the next incoming actin monomer, although the size is not enough. Therefore, the DAAM1 FH2 ring must be expanded so as to accept the incoming actin monomer by approximately 25 Å (Fig. 5A). This mechanism is reasonable because the linker region was extended in the actin-bound form of the Bni1p FH2 dimer, as compared to its actin-free form (Otomo et al. 2005b). Upon this conformational change in the DAAM1 FH2 dimer, the ß-sheet-like structure in the linker region should be unfolded to release the lasso region from the FH2 core. According to the mechanism explained above, the linker region should play a critical role in the actin assembly. We next examined the functional role of the linker region. We produced various FH1-FH2-DAD proteins with mutations in the linker region (Fig. 5B), and analyzed their actin assembly activities (Fig. 5C). The {Delta}17 mutant (21-amino-acid linker inserted with the additional -Gly-Ser- residues) had a similar actin assembly activity to the wild-type, whereas the {Delta}23 mutant lost the activity. These results indicate that -Asp657-Phe-Phe659- and/or -Ser677-Ser-Lys679-, are/is required for the activity, since these two sequences are present in the {Delta}17 mutant but not in the {Delta}23 mutant. On the other hand, the {Delta}20N and {Delta}20C mutants, which are equivalent with the (-Asp657-Phe-Phe659-)-lacking and (-Ser677-Ser-Lys679-)-lacking {Delta}17 mutants, respectively, retained similar activities to the wild-type. These results mean that either -Asp657-Phe-Phe659- or -Ser677-Ser-Lys679- is required, and that the sequence is irrelevant with the activity. The {Delta}11, {Delta}17, {Delta}20N and {Delta}20C mutants (29, 23, 20 and 20-amino-acid linkers are contained, respectively) had similar activities to the wild-type, while the {Delta}23 and {Delta}29 mutants (17- and 11-amino-acid linkers inserted with the additional -Gly-Ser- residues, respectively) completely lost the activity. Therefore, the linker length over 21 amino acids is essential for the actin assembly by human DAAM1 FH1-FH2-DAD. This is reasonable; because the 22-amino-acid linker would have a 28-Å length in minimum if it folds into an {alpha} helix, while a 59-Å length in maximum if it folds into an extended ß strand, which corresponds well to the difference in the circumferential length between the DAAM1 FH2 narrow ring and the Bni1p FH2 wide ring. The {Delta}17d mutant retained the activity, although its activity is weaker than that of the wild-type. This could be due to the mutation in the ß-sheet-like structure in the linker region.


Figure 5
View larger version (29K):
[in this window]
[in a new window]

  		Figure 5  Docking analysis and the actin assembly assay using mutants with various linker lengths. (A) Human DAAM1 FH2 in the complex with the ideal actin filament model (docking model; in this study). The structures of FH2 and the actin monomer are shown as ribbon and surface representations, respectively. One monomer of FH2 is colored green, and the other monomer is colored yellow. The actin monomer in one side is colored pink, and that in the other side is colored blue. The incoming actin monomer in the barbed end is shown as a transparent surface. The contracted and expanded states are shown in the top and bottom panels, respectively. (B) Amino acid sequences around the linker region of the human DAAM1 mutants used for the actin assembly assay. (C) Assembly rates of the human DAAM1 FH1-FH2-DAD mutants. Actin assembly was measured by incubation of the pyrene-labeled actin in the presence of the DAAM1 FH1-FH2-DAD mutants at 50 nM. The actin assembly rates were calculated as slopes of curves at 50% assembly, and shown as means ± SD in five independent experiments.

 

   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Experimental procedures
 Note added in proof
 References
 
Functional implications on the dimeric state of the present FH2 structure

The present structure is the first native dimeric form of the mammalian formin FH2, and shows that the architecture of the dimer interface is conserved between yeast and mammalian formins. On the other hand, the relative orientation of the dimer is different between the human DAAM1 and yeast Bni1p. In contrast to Bni1p, the dimer conformation of the DAAM1 FH2 was stabilized by the ß-sheet-like structure, which is formed by the N- and C-terminal ends of the linker region. This stabilization implies that the contracted state of the DAAM1 FH2 might be more favorable than the expanded state, suggesting a regulatory role of the linker region. Actually, the linker length, but not specific amino acids in the region, affected the actin assembly activity, at least in human DAAM1, as shown in this study. Also, the linker sequence and length are divergent in the formin family, and the difference in the actin assembly activity might reflect a physiological role of each formin protein. Although we do not know the actual conformation of FH2 bound to the barbed end of the elongating actin filament, our docking model with the ideal actin filament implies that expansion and contraction of the FH2 dimeric ring could occur on the barbed end. The contracted state of FH2 on the barbed end might contribute to effective blocking of the capping protein and/or relaxation of the helical tension caused by the stair-stepping movement. Further structural study would be required to capture the conformation of the formin during the actin assembly.


   Experimental procedures
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
Note added in proof
 References
 
Materials

Complementary DNA encoding human DAAM1 was kindly provided as KIAA0666 by Kazusa DNA Institute (Japan). The FH1–FH2–DAD region of the DAAM1 and its various mutants, and FH2 domain were generated by PCR-based mutagenesis. Deleted sequences of linker region mutants were replaced by Gly-Ser linker. All of the sequences of the PCR products were confirmed by sequencing using a 3100 Genetic Analyzer (GE Healthcare). All of these cDNAs were subcloned into pGEX-2T (GE Healthcare) and these proteins were produced as glutathione-S-transferase (GST)-fused proteins in Escherichia coli strain BL21 and purified on with Glutathione Sepharose according to the manufacture's instruction (GE Healthcare). All of the purified recombinant proteins were extensively dialyzed against 50 mM Hepes–K buffer (pH 7.4) containing 78 mM KCl, 4 mM MgCl2, 0.2 mM CaCl2, 2 mM EGTA and 1 mM dithiothreitol (DTT), and stored at 4 °C until use. The protein concentrations were determined by the Bradford method (Bio-Rad, Tokyo, Japan) or from the intensities of the bands in Coomassie Blue-stained SDS-PAGE gels using bovine serum albumin as a standard.

Purification, crystallization and structure determination

Escherichia coli strain RosettaTM (DE3) cells (Invitrogen, USA) were transformed with the expression vector, and cultured in LB media containing 100 mg/L ampicillin. Expression was induced by addition of 0.1 mM isopropyl-ß-D-thiogalactopyranoside (IPTG) when the culture reached an OD660 of approximately 0.5. After 16 h induction at 20 °C, the cells were collected by centrifugation at 8000 g for 15 min, and suspended in phosphate buffered saline (PBS) containing 1 mM DTT and 1 mM phenylmethylsulfonyl fluoride (PMSF). The cells were disrupted by sonication. Cell debris and larger particles were removed by centrifugation at 12 000 g for 40 min. The supernatant was applied onto Glutathione Sepharose FF column (GE Healthcare), pre-equilibrated with PBS containing 1 mM DTT. The GST tag was cleaved by addition of thrombin (Haematologic Technology, USA) on the column, and the protein was eluted with 2 column volumes of 50 mM Tris–HCl buffer (pH 6.8) containing 50 mM NaCl and 1 mM DTT. The elution was loaded onto MonoS column (GE Healthcare), pre-equilibrated with 50 mM Tris–HCl buffer (pH 6.8) containing 1 mM DTT (buffer A). The protein was eluted with a linear gradient of NaCl from 50 to 500 mM in buffer A. The fractions containing the FH2 domain was loaded onto Superdex 200 16/60 (prep grade) column (GE Healthcare), pre-equilibrated with 10 mM Tris–HCl buffer (pH 7.4) containing 50 mM NaCl and 5 mM ß-mercaptoethanol. The purified protein was concentrated to 8 mg/mL by using Amicon Ultra 15 (Millipore, USA). The SeMet-labeled FH2 was overproduced in E. coli B834 (DE3) cells (Invitrogen), and cultured in CoreTM media (Wako, Japan) containing 30 µg/mL L-selenomethionine. The protein was purified in the same way as the native FH2. The purified FH2 solution was concentrated up to 8 mg/mL by using Amicon Ultra 15 (Millipore), following the manufacturer's instruction. Best crystals grew with the reservoir solution of 50 mM cacodylate–Na buffer (pH 6.5) containing 10% PEG4000. The native crystal, which was used for the final structure refinement, belongs to the space group P1 with the unit cell parameters, a = 69.2 Å, b = 91.9 Å, c = 97.7 Å, {alpha} = 98.1°, ß = 90.3°, {gamma} = 104.8°. Diffraction data of the native and selenomethionine-labeled protein crystals were collected at the beamline BL41XU in SPring 8 (Hyogo, Japan), and processed with the program HKL2000 (Otwinowski & Minor 1997) and the CCP4 program suite (CCP4 1994). Phases were determined by a multi-wavelength anomalous dispersion (MAD) method using the selenium edge. 18 of 40 possible selenium sites were identified with the program SNB (Weeks & Miller 1999). Refinement of the selenium sites and phase calculation were carried out using the MAD dataset up to 3.0 Å with the program SHARP (Fortelle & Bricogne 1997). The calculated phases were improved by solvent-flattening, histogram matching and non-crystallographic symmetry-related averaging with the program DM (CCP4 1994). The atomic model was built with the program O (Jones et al. 1991). The model was refined against the native data set up to 2.8 Å resolution by using the program CNS (Brunger et al. 1998). Data collection, phasing and refinement statistics are shown in Table 1. All molecular graphics are prepared with the program PYMOL (DeLano Scientific, USA; <http://www.pymol.org>).

In vitro actin polymerization assay

Rabbit skeletal muscle actin was purified from acetone powder (Sigma, USA) and labeled with pyrenemethyliodoacetamide (Invitrogen) (Higgs et al. 1999), according to the manufacturer's instructions. Monomeric actin was prepared from the purified actin by gel filtration chromatography using a Superdex 200 column (GE Healthcare), pre-equilibrated in 2 mM Tris buffer (pH 8.0) containing 0.5 mM ATP, 0.2 mM CaCl2 and 0.2 mM dithiothreitol. Actin assembly was induced by incubation of 2 µM pyrene-labeled actin in the absence or presence of FH1-FH2-DAD domains of DAAM1, and evaluated by measuring fluorescence intensity over time at an excitation of 365 nm and emission of 407 nm in a fluorescence spectrophotometer (F-2500, HITACHI, Japan) at 25 °C. Rates of the actin assembly were calculated from the slopes of the assembly curves at 50% polymerization.


   Note added in proof
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 Note added in proof
References
 
The structure of the FH2 domain (residues 596–1078) of human DAAM1 has been reported by the Eck group online since 5 April 2007 in the Journal of Molecular Biology. In their structure, the two monomers are related by the crystallographic twofold axis, whereas our dimer structure is related by the pseudo twofold axis in the asymmetric unit. The dimer orientation in their structure was stabilized in the ß-strand-like structure, as observed in our structure. They showed that this ß-strand-like structure locked the dimer conformation by a mutational analysis, and explained the weak actin assembly activity of human DAAM1. Despite of the difference in the crystal form, the ß-strand-like structure was observed in both structures, and the dimer conformation is the almost same. This indicates that the present dimer conformation is preferable in the actin-free state of DAAM1 FH2, which may represent the auto-inhibition state, as suggested by the Eck group.


   Acknowledgements
 
We thank Drs M. Kawamoto and N. Shimizu (JASRI), and Ms T. Matsubara for their help in data collection at SPring 8, and for excellent technical assistance, respectively. We also thank Kazusa DNA Institute for providing the KIAA plasmid. This work was supported by a grant from the Ministry of Education, Culture, Sports, Science and Technology to S.F., H.H., and O.N., a SORST program grant from JST to O.N. This work was also supported in part by grants from Takeda Science Foundation to H.H. and Japan Heart Foundation Pfizer Japan Inc. Grant for Research on Cardiovascular Disease to T.H. T.H. and R.S. were supported by JSPS Research Fellowships for Young Scientists. Coordinates and structure factors are deposited in the Protein Data Bank under accession code 2Z6E.


   Footnotes
 
Communicated by: Toshio Hakoshima

aThese authors contributed equally to this work. Back

bPresent address: Structural Biology Laboratory, Life Science Division, Synchrotron Radiation Research Organization, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-0032, Japan. Back

* Correspondence: E-mail: nureki{at}bio.titech.ac.jp


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 Note added in proof
 References
 
Alberts, A.S. (2001) Identification of a carboxyl-terminal diaphanous-related formin homology protein autoregulatory domain. J. Biol. Chem. 276, 2824–2830.[Abstract/Free Full Text]

Brunger, A.T., Adams, P.D., Clore, G.M., DeLano, W.L., Gros, P., Grosse-Kunstleve, R.W., Jiang, J.S., Kuszewski, J., Nilges, M., Pannu, N.S., Read, R.J., Rice, L.M., Simonson, T. & Warren, G.L. (1998) Crystallography and NMR system: a new software suite for macromolecular structure determination. Acta Crystallogr. D 54, 905–921.[CrossRef][Medline]

CCP4 (1994) The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D 50, 760–763.[CrossRef][Medline]

Dong, Y., Pruyne, D. & Bretscher, A. (2003) Formin-dependent actin assembly is regulated by distinct modes of Rho signaling in yeast. J. Cell Biol. 161, 1081–1092.[Abstract/Free Full Text]

Evangelista, M., Blundell, K., Longtine, M.S., Chow, C.J., Adames, N., Pringle, J.R., Peter, M. & Boone, C. (1997) Bni1p, a yeast formin linking cdc42p and the actin cytoskeleton during polarized morphogenesis. Science 276, 118–122.[Abstract/Free Full Text]

Fortelle, E.D.L. & Bricogne, G. (1997) Maximum-likelihood heavy-atom parameter refinement for multiple isomorphous replacement and multiwavelength anomalous diffraction methods. Methods Enzymol. 276, 472–494.

Habas, R., Kato, Y. & He, X. (2001) Wnt/Frizzled activation of Rho regulates vertebrate gastrulation and requires a novel Formin homology protein Daam1. Cell 107, 843–854.[CrossRef][Medline]

Harris, E.S., Li, F. & Higgs, H.N. (2004) The mouse formin, FRL{alpha}, slows actin filament barbed end elongation, competes with capping protein, accelerates polymerization from monomers, and severs filaments. J. Biol. Chem. 279, 20076–20087.[Abstract/Free Full Text]

Higashida, C., Miyoshi, T., Fujita, A., Oceguera-Yanez, F., Monypenny, J., Andou, Y., Narumiya, S. & Watanabe, N. (2004) Actin polymerization-driven molecular movement of mDia1 in living cells. Science 303, 2007–2010.[Abstract/Free Full Text]

Higgs, H.N. (2005) Formin proteins: a domain-based approach. Trends Biochem. Sci. 30, 342–353.[CrossRef][Medline]

Higgs, H.N. & Peterson, K.J. (2005) Phylogenetic analysis of the formin homology 2 domain. Mol. Biol. Cell 16, 1–13.[Abstract/Free Full Text]

Higgs, H.N., Blanchoin, L. & Pollard, T.D. (1999) Influence of the C terminus of Wiskott–Aldrich syndrome protein (WASp) and the Arp2/3 complex on actin polymerization. Biochemistry 38, 15212–15222.[CrossRef][Medline]

Holmes, K.C., Popp, D., Gebhard, W. & Kabsch, W. (1990) Atomic model of the actin filament. Nature 347, 44–49.[CrossRef][Medline]

Ishizaki, T., Morishima, Y., Okamoto, M., Furuyashiki, T., Kato, T. & Narumiya, S. (2001) Coordination of microtubules and the actin cytoskeleton by the Rho effector mDia1. Nat. Cell Biol. 3, 8–14.[CrossRef][Medline]

Jones, T.A., Zou, J.Y., Cowan, S.W. & Kjeldgaard (1991) Improved methods for binding protein models in electron density maps and the location of errors in these models. Acta Crystallogr. A 47, 110–119.[CrossRef]

Kohno, H., Tanaka, K., Mino, A., Umikawa, M., Imamura, H., Fujiwara, T., Fujita, Y., Hotta, K., Qadota, H., Watanabe, T., Ohya, Y. & Takai, Y. (1996) Bni1p implicated in cytoskeletal control is a putative target of Rho1p small GTP binding protein in Saccharomyces cerevisiae. EMBO J. 15, 6060–6068.[Medline]

Kovar, D.R. & Pollard, T.D. (2004) Insertional assembly of actin filament barbed ends in association with formins produces piconewton forces. Proc. Natl. Acad. Sci. USA 101, 14725–14730.[Abstract/Free Full Text]

Kovar, D.R., Harris, E.S., Mahaffy, R., Higgs, H.N. & Pollard, T.D. (2006) Control of the assembly of ATP- and ADP-actin by formins and profilin. Cell 124, 423–435.[CrossRef][Medline]

Kovar, D.R., Kuhn, J.R., Tichy, A.L. & Pollard, T.D. (2003) The fission yeast cytokinesis formin Cdc12p is a barbed end actin filament capping protein gated by profilin. J. Cell Biol. 161, 875–887.[Abstract/Free Full Text]

Lammers, M., Rose, R., Scrima, A. & Wittinghofer, A. (2005) The regulation of mDia1 by autoinhibition and its release by Rho•GTP. EMBO J. 24, 4176–4187.[CrossRef][Medline]

Li, F. & Higgs, H.N. (2003) The mouse Formin mDia1 is a potent actin nucleation factor regulated by autoinhibition. Curr Biol. 13, 1335–1340.[CrossRef][Medline]

Moseley, J.B., Sagot, I., Manning, A.L., Xu, Y., Eck, M.J., Pellman, D. & Goode, B.L. (2004) A conserved mechanism for Bni1- and mDia1-induced actin assembly and dual regulation of Bni1 by Bud6 and profilin. Mol. Biol. Cell 15, 896–907.[Abstract/Free Full Text]

Nakaya, M.A., Habas, R., Biris, K., Dunty, W.C. Jr., Kato, Y., He, X. & Yamaguchi, T.P. (2004) Identification and comparative expression analyses of Daam genes in mouse and Xenopus. Gene Expr. Patterns 5, 97–105.[CrossRef][Medline]

Otomo, T., Otomo, C., Tomchick, D.R., Machius, M. & Rosen, M.K. (2005a) Structural basis of Rho GTPase-mediated activation of the formin mDia1. Mol. Cell 18, 273–281.[CrossRef][Medline]

Otomo, T., Tomchick, D.R., Otomo, C., Panchal, S.C., Machius, M. & Rosen, M.K. (2005b) Structural basis of actin filament nucleation and processive capping by a formin homology 2 domain. Nature 433, 488–494.[CrossRef][Medline]

Otwinowski, Z. & Minor, W. (1997) Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307–326.

Pruyne, D., Evangelista, M., Yang, C., Bi, E., Zigmond, S., Bretscher, A. & Boone, C. (2002) Role of formins in actin assembly: nucleation and barbed-end association. Science 297, 612–615.[Abstract/Free Full Text]

Romero, S., Le Clainche, C., Didry, D., Egile, C., Pantaloni, D. & Carlier, M.F. (2004) Formin is a processive motor that requires profilin to accelerate actin assembly and associated ATP hydrolysis. Cell 119, 419–429.[CrossRef][Medline]

Rose, R., Weyand, M., Lammers, M., Ishizaki, T., Ahmadian, M.R. & Wittinghofer, A. (2005) Structural and mechanistic insights into the interaction between Rho and mammalian Dia. Nature 435, 513–518.[CrossRef][Medline]

Sagot, I., Rodal, A.A., Moseley, J., Goode, B.L. & Pellman, D. (2002) An actin nucleation mechanism mediated by Bni1 and profilin. Nat. Cell Biol. 4, 626–631.[Medline]

Shemesh, T., Otomo, T., Rosen, M.K., Bershadsky, A.D. & Kozlov, M.M. (2005) A novel mechanism of actin filament processive capping by formin: solution of the rotation paradox. J. Cell Biol. 170, 889–893.[Abstract/Free Full Text]

Shimada, A., Nyitrai, M., Vetter, I.R., Kuhlmann, D., Bugyi, B., Narumiya, S., Geeves, M.A. & Wittinghofer, A. (2004) The core FH2 domain of diaphanous-related formins is an elongated actin binding protein that inhibits polymerization. Mol. Cell 13, 511–522.[CrossRef][Medline]

Watanabe, N., Kato, T., Fujita, A., Ishizaki, T. & Narumiya, S. (1999) Cooperation between mDia1 and ROCK in Rho-induced actin reorganization. Nat. Cell Biol. 1, 136–143.[CrossRef][Medline]

Watanabe, N., Madaule, P., Reid, T., Ishizaki, T., Watanabe, G., Kakizuka, A., Saito, Y., Nakao, K., Jockusch, B.M. & Narumiya, S. (1997) p140mDia, a mammalian homolog of Drosophila diaphanous, is a target protein for Rho small GTPase and is a ligand for profilin. EMBO J. 16, 3044–3056.[CrossRef][Medline]

Weeks, C.M. & Miller, R. (1999) The design and implementation of SnB v2.0. J. Appl. Cryst. 32, 120–124.[CrossRef]

Xu, Y., Moseley, J.B., Sagot, I., Poy, F., Pellman, D., Goode, B.L. & Eck, M.J. (2004) Crystal structures of a formin homology-2 domain reveal a tethered dimer architecture. Cell 116, 711–723.[CrossRef][Medline]

Yayoshi-Yamamoto, S., Taniuchi, I. & Watanabe, T. (2000) FRL, a novel formin-related protein, binds to Rac and regulates cell motility and survival of macrophages. Mol. Cell. Biol. 20, 6872–6881.[Abstract/Free Full Text]

Zigmond, S.H., Evangelista, M., Boone, C., Yang, C., Dar, A.C., Sicheri, F., Forkey, J. & Pring, M. (2003) Formin leaky cap allows elongation in the presence of tight capping proteins. Curr. Biol. 13, 1820–1823.[CrossRef][Medline]

Received: 13 June 2007
Accepted: 6 August 2007




This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
T. Higashi, T. Ikeda, R. Shirakawa, H. Kondo, M. Kawato, M. Horiguchi, T. Okuda, K. Okawa, S. Fukai, O. Nureki, et al.
Biochemical Characterization of the Rho GTPase-regulated Actin Assembly by Diaphanous-related Formins, mDia1 and Daam1, in Platelets
J. Biol. Chem., March 28, 2008; 283(13): 8746 - 8755.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yamashita, M.
Right arrow Articles by Nureki, O.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Yamashita, M.
Right arrow Articles by Nureki, O.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS