HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 45, 43089-43095, November 8, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/45/43089    most recent M208225200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Paavilainen, V. O. Articles by Lappalainen, P. Search for Related Content PubMed PubMed Citation Articles by Paavilainen, V. O. Articles by Lappalainen, P. Social Bookmarking                      What's this?

Structural Conservation between the Actin Monomer-binding Sites of Twinfilin and Actin-depolymerizing Factor (ADF)/Cofilin^*

Ville O. Paavilainen^§, Michael C. Merckel^¶, Sandra Falck, Pauli J. Ojala, Ehmke Pohl, Matthias Wilmanns, and Pekka Lappalainen^**

From the Programs in  Cellular Biotechnology, and ^¶ Structural Biology and Biophysics, Institute of Biotechnology, P.O. Box 56, University of Helsinki, 00014 Helsinki, Finland and the  European Molecular Biology Laboratory (EMBL)-Hamburg Outstation, Notkestr. 85, 22603 Hamburg, Germany

Received for publication, August 12, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Twinfilin is an evolutionarily conserved actin monomer-binding protein that regulates cytoskeletal dynamics in organisms from yeast to mammals. It is composed of two actin-depolymerization factor homology (ADF-H) domains that show ~20% sequence identity to ADF/cofilin proteins. In contrast to ADF/cofilins, which bind both G-actin and F-actin and promote filament depolymerization, twinfilin interacts only with G-actin. To elucidate the molecular mechanisms of twinfilin-actin monomer interaction, we determined the crystal structure of the N-terminal ADF-H domain of twinfilin and mapped its actin-binding site by site-directed mutagenesis. This domain has similar overall structure to ADF/cofilins, and the regions important for actin monomer binding in ADF/cofilins are especially well conserved in twinfilin. Mutagenesis studies show that the N-terminal ADF-H domain of twinfilin and ADF/cofilins also interact with actin monomers through similar interfaces, although the binding surface is slightly extended in twinfilin. In contrast, the regions important for actin-filament interactions in ADF/cofilins are structurally different in twinfilin. This explains the differences in actin-interactions (monomer versus filament binding) between twinfilin and ADF/cofilins. Taken together, our data show that the ADF-H domain is a structurally conserved actin-binding motif and that relatively small structural differences at the actin interfaces of this domain are responsible for the functional variation between the different classes of ADF-H domain proteins.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The actin cytoskeleton is essential for diverse cellular processes such as morphogenesis, motility, endocytosis, and cell division. The structure and dynamics of actin filaments are regulated by a large number of actin-binding proteins (1). Despite the large variation in the biochemical activities of these proteins, many of them interact with actin through a relatively small number of actin-binding motifs. The actin-depolymerization factor homology (ADF-H)¹ domain is an ~150-amino acid motif that is present in three phylogenetically distinct classes of actin-binding proteins: ADF/cofilins, Abp1/drebrins, and twinfilins. Although all these proteins appear to use the ADF-H domain for their interactions with actin, they are biochemically distinct and play different roles in actin dynamics (2).

ADF/cofilins are small actin-binding proteins composed of a single ADF-H domain. They bind both actin monomers and filaments and promote rapid filament turnover in cells by depolymerizing/fragmenting actin filaments. ADF/cofilins bind ADP-actin with higher affinity than ATP-actin and inhibit the spontaneous nucleotide exchange on actin monomers (3, 4).

Abp1/drebrins are relatively large proteins composed of an N-terminal ADF-H domain followed by a variable region and a C-terminal SH3 domain. Unlike ADF/cofilins, Abp1/drebrins interact only with actin filaments and do not promote filament depolymerization or fragmentation. Abp1/drebrins appear to be involved in endocytosis as well as in promoting neuronal plasticity in animals (5, 6). Interestingly, at least yeast Abp1 also activates the Arp2/3 complex and may therefore function as a link between endocytosis and actin polymerization (47).

Twinfilins are actin-binding proteins that are composed of two ADF-H domains. Mutations in budding yeast and Drosophila twinfilin genes result in defects in actin-dependent cell biological and developmental processes, indicating that twinfilin is intimately involved in the regulation of actin dynamics in these organisms (7, 8). Unlike ADF/cofilins and Abp1/drebrins, which interact with actin filaments, twinfilins bind only actin monomers (9). Twinfilins bind ADP-G-actin with ~10-fold higher affinity than ATP-G-actin and prevent the nucleotide exchange on actin monomers (10). Analysis of yeast twinfilin suggested that this protein contributes to actin dynamics by localizing actin monomers, in their "inactive" ADP-form, to the sites of rapid actin assembly in cells. The localization of twinfilin at these regions is regulated by direct interactions between twinfilin and capping protein (11). Although twinfilin is composed of two ADF-H domains, it appears to form a 1:1 complex with actin monomers (7, 12). Analysis of mouse twinfilin demonstrated that the two ADF-H domains are biochemically different from each other. The C-terminal domain forms a stable (k_off = 1.8 s¹), high affinity (K_d = 0.05 µM) complex with ADP-actin, whereas the N-terminal domain forms a more transient (k_off = 20 s¹), lower affinity (K_d = 0.7 µM) complex with ADP-G-actin. Kinetic analysis further suggests that the actin monomer first associates with the N-terminal ADF-H domain and is then delivered to the C-terminal domain through a conformational change within the twinfilin molecule (10).

The atomic structures of four ADF/cofilin proteins have been determined by x-ray crystallography and NMR (13-16). To elucidate structural differences underlying the biochemical differences between the three ADF-H domain protein families, it will be essential to gain structural information also from the two other members of this family: Abp1/drebrin and twinfilin. Here, we present the crystal structure of the N-terminal ADF-H domain of mouse twinfilin. Furthermore, we mapped the actin monomer-binding site of this domain by site-directed mutagenesis. These data show that ADF/cofilins and twinfilin interact with G-actin through structurally similar interfaces but that the regions involved in F-actin binding in ADF/cofilin are not structurally conserved in twinfilin.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Site-directed Mutagenesis, Protein Expression, and Purification-- The recombinant proteins containing the N-terminal ADF-H domain of mouse twinfilin (residues 1-142 and 1-174) were expressed in Escherichia coli BL21(DE3) cells as glutathione S-transferase (GST) fusions from the vector pGAT2 (17). The proteins were cleaved from GST by thrombin digestion and purified by gel filtration chromatography as described previously (10). The site-directed mutations were introduced to the Twf_1-174 construct using the PCR-based overlap extension method described by Higuchi et al. (18). To express an M1-T5 deletion of the domain, a fragment lacking the codons for the 5 N-terminal amino acids was amplified by PCR with oligonucleotides that create NcoI and HindIII sites at the 5' and 3' ends of the product, respectively. The fragment was digested with NcoI and HindIII and ligated into the pGAT2 vector, and the protein was expressed and purified as described above. All DNA constructs were sequenced to verify the correct sequence. Rabbit muscle actin was purified from acetone powder (19) and labeled with 7-chloro-4-nitrobenz-2-oxa-1,3-diazole (NBD-Cl) as described in Detmers et al. (20) with modifications by Weeds et al. (21). Protein concentrations were determined with a Hewlett-Packard 8452A diode array spectrophotometer by using calculated extinction co-efficients: ₂₈₀ = 26,1 mM¹ cm¹ (for the N-terminal ADF-H domain of twinfilin) and _290-400 = 26,6 mM¹ cm¹ (for actin).

Crystallization and Data Collection-- The twinfilin_1-142 construct was crystallized by mixing the protein solution (1 µl of 10 mg/ml in 50 mM NaCl, 10 mM Tris, pH 7.5) with the reservoir solution (1.6 M sodium citrate, 3% polyethylene glycol 400, 0.1 M Na-Hepes, pH 7.5) and equilibrating against the reservoir buffer by the hanging drop vapor diffusion method at +4 °C. Regular crystals of up to 0.3 mm appeared after 3-5 days, belonged to the space group P2₁2₁2₁, and had cell constants a = 47.54 Å, b = 74.49Å, c = 76.95 Å,  =  =  = 90^o. V_m calculations indicated that there were most likely two monomers in the asymmetric unit with a solvent content of 44%. The gold-derivative crystals were prepared by soaking native crystals in well buffer supplemented with 0.1 M cyano-aurate for 24-48 h at +4 °C. The crystals were mounted in cryoloops directly from mother liquor and flash-frozen in liquid nitrogen. All data were collected at the European Molecular Biology Laboratory (EMBL) Hamburg outstation on the wiggler beamline BW7B equipped with a Mar345 imaging plate scanner (22). Images were processed with programs DENZO and SCALEPACK (23).

Structure Solution and Refinement-- The structure was solved by single anomalous dispersion phasing. The gold positions were identified using the program SHELXS (24) and refined with SOLVE (25). The program SHARP (26) was used to extend the phases and calculate the initial experimental map at a resolution of 2.5 Å. These maps were further enhanced by density modification. An initial model was carefully built into the experimental map using the program O (27). The model was then refined using the maximum likelihood target function with the phase probability distribution in CNS 1.1 (28). These coordinates were subsequently used as a molecular replacement search model for the 1.6-Å native data set. The refinement using data from 20 to 1.6 Å was completed by iterative cycles of model building and simulated annealing and conjugate gradient least squares coordinate and restrained B-factor refinement in CNS. During all steps, 9% of the data was used to calculate the free R-factor. The final model was analyzed with Procheck (29) and CNS. The coordinates of the model were deposited in the Protein Data Bank (PDB code 1M4J).

Actin Monomer Binding Assay-- The change in the fluorescence of NBD-labeled ADP-G-actin was used to monitor the binding of wild-type and mutant twinfilin domains to actin monomers as described (10). ADP-actin was prepared by incubating NBD-actin with hexokinase-agarose beads (Sigma) and 1 mM glucose overnight as described (30). Different concentrations (0-150 µM) of wild-type and mutant twinfilin domains were mixed with 0.2 µM ADP-G-actin (in 5 mM Tris, pH 7.5, 0.08 mM CaCl₂, 0.2 mM dithiothreitol, 0.2 mM ADP, 1 mg/ml bovine serum albumin, 0.1 M KCl, 1 mM MgCl₂). The normalized decrease of fluorescence, shown in Equation 1,

(Eq. 1)

was measured with a BioLogic MOS250 fluorometer at each concentration of wild-type and mutant twinfilin ADF-H domains with an excitation at 482 nm and emission at 535 nm. The data were analyzed and fitted using Equation 2

(Eq. 2)

where Equation 3 provides

(Eq. 3)

and Equation 4 provides

(Eq. 4)

Urea Denaturation Assays-- Wild-type and mutant Twf_1-174 proteins were used at a final concentration of 2 µM in 10 mM Tris, pH 7.5, 50 mM NaCl. The proteins were diluted into 0-7 M buffered urea and incubated at room temperature for 60 min. Fluorescence measurements were carried out at an excitation of 280 nm, and emission was monitored at 355 nm. The fluorescence change versus urea concentration was then plotted, and the midpoint of unfolding was determined.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Overall Structure and Comparison with Known ADF/Cofilin Structures-- The structure of the N-terminal ADF-H domain of mouse twinfilin (Twf_1-142) was determined at 1.6-Å resolution (Table I). The experimental single anomalous dispersion map showed two molecules in the asymmetric unit and allowed for a continuous chain trace, excluding the six N-terminal and three C-terminal residues, which were disordered. The N-terminal ADF-H domain of twinfilin consists of a five-stranded mixed -sheet in which the four N-terminal strands are antiparallel and the two C-terminal strands run parallel to each other. These strands are surrounded by a pair of -helices on both sides of the sheet (Fig. 1A).

View larger version (65K): [in this window] [in a new window]   Fig. 1.   Structure of the N-terminal ADF-H domain of mouse twinfilin. A, a schematic ribbon diagram of the N-terminal ADF-H domain of twinfilin, Twf1-142. The structure is color-ramped from blue (N terminus) via green to red (C terminus). B, a representative section of the 2Fo  Fc 1.6-Å electron density map contoured at 1 centered around residue Asp-74. C, C superimposition of Twf1-142 (red) and yeast cofilin (blue). Twf1-142 is in the same orientation as in panel A. The strands 3 and 4 (red arrow) and the C-terminal helix 4 (blue arrow) are oriented differently in Twf1-142 and ADF/cofilins. The superposition was produced with DALI (41).

Although twinfilin and ADF/cofilin display different biochemical activities and the ADF-H domains of twinfilin are only ~20% identical to ADF/cofilins at the amino acid level, Twf_1-142 superimposes well with the ADF/cofilin structures. An average root mean square deviation for all C atoms between Twf_1-142 and yeast cofilin is 1.9 Å (Fig. 1C). The orientation of helices 1, 2, and 3 as well as strands 1, 2, and 5 are especially well conserved between twinfilin and ADF/cofilin, whereas the orientations of helix 4 and strands 3 and 4 deviate between ADF/cofilin and Twf_1-142. The long -sheet composed of strands 3 and 4 protrudes away from the main protein body in all ADF/cofilin structures determined so far (13-16), whereas in Twf_1-142, these strands and the loop connecting them are tilted toward the C-terminal end of helix 3 (Fig. 1C, red arrow). Furthermore, the C-terminal helix, which is oriented roughly parallel to the strand 5 in ADF/cofilins, is tilted ~25^o in the N-terminal ADF-H domain of twinfilin (Fig. 1C, blue arrow). These structural variations are well defined in the electron density maps, are accompanied by several contacts to the main protein body, and are important for the integrity and activity of twinfilin as discussed below.

A structure-based sequence alignment between Twf_1-142 and yeast cofilin shows that the structurally most highly conserved regions between these molecules (1, 2, 3, 1, 2, 4, 5) also display the most obvious sequence homology to each other (Fig. 2). The hydrophobic core, including twinfilin residues Ala-9, Phe-17, Ala-20, Leu-28, Ile-30, Ile-32, Leu-37, Phe-56, Leu-60, Trp-83, Ile-86, Leu-108, Val-121, Val-129, Tyr-134, is well conserved between ADF/cofilins and Twf_1-142. Furthermore, Trp-88 makes similar hydrogen bonding to Leu-58, Pro-59, and Pro-92 as the corresponding residues in ADF/cofilin molecules. It is also important to note that the regions important for actin monomer binding in ADF/cofilins, especially the long helix 3, are well conserved in the N-terminal ADF-H domain of twinfilin. Also, the Tyr-68 and Tyr-101 residues, which have been suggested to stabilize and orient this long actin-binding helix in ADF/cofilins (31), are stacked against each other in a similar orientation to the corresponding invariable tyrosines in ADF/cofilins. In contrast, the C-terminal helix and the region surrounding the loop between strands 3 and 4 show a lower degree of sequence and structural conservation between ADF/cofilins and Twf_1-142. Interestingly, these structurally less conserved regions have been shown to play an important role in actin filament binding in ADF/cofilin proteins (32, 33).

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Structural sequence alignment of yeast cofilin and the N-terminal ADF-H domain of twinfilin. The secondary structure elements of cofilin and twinfilin are indicated above and below the sequences, respectively. The residues that have been shown to be important for actin monomer and actin filament interactions in ADF/cofilins are indicated with asterisks and hash marks, respectively, above the cofilin sequence (32-34, 42). The positions of the residues mutated in this study are indicated below the twinfilin sequence.

The Actin-binding Site of Twinfilin-- The N-terminal ADF-H domain of twinfilin is structurally similar to ADF/cofilins (Fig. 1C), and recent studies showed that these two proteins compete with each other in actin monomer binding (10). This suggests that they interact with actin through at least partially overlapping interfaces. However, it is important to note that whereas ADF/cofilins bind both F- and G-actin, twinfilin interacts only with G-actin.

To map the actin monomer-binding site on the surface of the N-terminal ADF-H domain of twinfilin, we introduced six mutations on the putative actin-binding surface of twinfilin. The mutagenesis was carried out for Twf_1-174 construct (containing twinfilin residues 1-174) instead of Twf_1-142 because Twf_1-174 promotes a stronger change in the fluorescence upon binding to NBD-G-actin than Twf_1-142. However, the extra residues in Twf_1-174 do not contribute to actin binding because both of these constructs interact with actin monomers with nearly identical affinities (10). Five of the mutations in this study were alanine substitutions of surface residues at corresponding regions that have been shown to be important for actin monomer and/or actin filament binding in ADF/cofilin. The sixth mutation was the deletion of the 5 N-terminal amino acids that were disordered in our crystal structure. The N-terminal region of ADF/cofilins has been shown to play an essential role in interactions with actin (32, 34, 35). The residues important for actin binding in ADF/cofilin as well as the twinfilin mutations generated in this study are shown in Fig. 2.

The mutant proteins were expressed in E. coli as GST-fusion proteins, enriched with glutathione-agarose beads, separated from GST by thrombin digestion, and purified by gel filtration chromatography. All Twf_1-174 mutant proteins were fully soluble and monomeric according to their elution positions and profiles in gel filtration chromatography. To determine the effects of these mutations for the overall stability of the molecule, a fluorescence-monitored urea denaturation assay was carried out for each protein (Fig. 3). The behavior of the wild-type ADF-H domain and all mutant proteins is consistent with a simple two-state unfolding transition. The wild-type Twf_1-174 and most of the mutants show a midpoint for the transition at ~4 M urea, indicating that these mutations do not significantly affect the stability of the protein. However, the mutant K135A,K136A showed a midpoint for the transition at 3.1 M urea and was therefore significantly less stable than the wild-type protein. However, these mutations do not affect the overall structure of the protein because the K135A,K136A mutant protein still binds to actin monomers with almost identical affinity to the wild-type Twf_1-174 (Table II).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   The stability of the recombinant proteins measured by fluorescence monitored urea denaturation assay. The arbitrary fluorescence units are shown on the y axis, and the urea concentration is shown on the x axis. The wild-type Twf1-174 and five of the mutant proteins unfold at ~4 M urea, whereas the K135A,K136A mutant unfolds at 3.1 M urea.

The fluorescence of NBD-G-actin is modulated upon binding to twinfilin and its individual ADF-H domains, thereby providing a means of determining the affinity of these proteins for actin monomers (10). Here, we determined the affinities of wild-type and mutant Twf_1-174 proteins for G-actin under physiological conditions (0.1 M KCl, 1 mM MgCl₂). In these assays, we used ADP-actin because twinfilin has been shown to bind ADP-G-actin with ~10-fold higher affinity than ATP-G-actin (10). The wild-type Twf_1-174 and five of the mutant proteins (M1-T5; Q76A,Q79A; K109A,K110A; E127A,D128A; K135A,K136A) resulted in 10-20% quenching of the NBD-fluorescence upon binding to actin monomers. However, whereas the quenching was almost saturated with 2 µM K135A,K136A mutant protein (Fig. 4), much larger concentrations of the other four mutants were required for maximal fluorescence quenching (Fig. 4, insets). The affinity of the K135A,K136A mutant for ADP-G-actin (K_D = 0.8 µM) is very similar to the one determined previously for the wild-type Twf_1-174 for ADP-G-actin (K_D = 0.7 µM, ref. 10), suggesting that these mutations do not interfere with the actin monomer binding activity. In contrast, the N-terminal deletion as well as three of the point mutation proteins (Q76A,Q79A; K109A,K110A; E127A,D128A) show over 5-fold decrease (K_D = 4.5-7.3 µM) in the affinity for actin monomers. The R96A,K98A mutant does not result in a detectable quenching of NBD-actin fluorescence, suggesting that this mutant protein does not have significant affinity for ADP-G-actin (Fig. 4). It is also formally possible that the R96A,K98A mutant binds to G-actin but would be unable to quench NBD-fluorescence upon binding. However, previous mutagenesis studies demonstrated that corresponding residues in yeast twinfilin play an important role in actin monomer binding/sequestering activity (11). Therefore, these results suggest that five of the mutations introduced in this study (M1-T5; Q76A,Q79A; R96A,K98A; K109A,K110A; E127A,D128A) disrupt the interaction between twinfilin and actin monomer, and only the mutant K135A,K136A does not interfere with actin binding (Table II).

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Binding of Twf1-174 mutant proteins to ADP-G-actin. The decrease in the fluorescence of NBD-labeled Mg-ADP-G-actin was measured at different concentrations of Twf1-174 mutant proteins. The experiment was carried out with 0.2 µM actin under physiological ionic conditions. Symbols indicate data, and solid lines indicate fitted binding curves for a complex with 1:1 stoichiometry. It is important to note that the K135A, K136A mutant protein binds ADP-G-actin with similar affinity (KD = 0.8 µM) as that of wild-type Twf1-142 (KD = 0.7 µM, see also Ref. 10). In contrast, five other mutants show significantly decreased affinity for actin monomers. The insets show the quenching of NBD-G-actin fluorescence at higher concentrations of Twf1-174 mutant proteins.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The analysis presented here demonstrates that ADF-H domains form a structurally conserved family of protein domains: the overall fold of the N-terminal ADF-H domain of twinfilin is similar to the known ADF/cofilin structures, and these proteins also show structural similarity to the repeated segments of the gelsolin family of actin filament severing/capping proteins (Fig. 5). Although the main part of the Twf_1-142 structure is very similar to ADF/cofilins, significant structural differences are seen at certain regions of these two biochemically distinct proteins (Fig. 1C). Similarly, structural differences have been shown to exist between the repeated segments of gelsolin, each of which display discrete biochemical activities (36).

View larger version (55K): [in this window] [in a new window]   Fig. 5.   Comparison of the actin-binding sites of the N-terminal ADF-H domain of twinfilin, cofilin, and gelsolin. A, ribbon diagrams of mouse Twf1-142, yeast cofilin and human gelsolin segment-1. The side chains of the residues important for actin monomer binding in these proteins are indicated by red. The side chains of the cofilin residues important for F-actin binding are indicated by blue, and the twinfilin residues mutated in this study that do not contribute to actin binding are indicated by green. The gelsolin segment-1 residues important for G-actin interaction are taken from the segment-1 actin monomer co-crystal structure (43), and the cofilin residues important for G- and F-actin interactions are taken from (32-34, 42, 44, 45). The twinfilin residues mutated in this study are indicated by letters and numbers. B, electrostatic surface potential of the actin-binding sites of Twf1-142, yeast cofilin, and gelsolin segment-1 displayed at ± 10 kT/e. The orientation of the proteins is identical to panel A, and the actin monomer-binding surfaces are circled by orange dashed lines. Regions of positive and negative potential are blue and red, respectively. The surface potentials of the actin monomer-binding sites of Twf1-142 and yeast cofilin are similar to each other, whereas this site on gelsolin segment-1 is more strongly and uniformly negatively charged. This figure was prepared with GRASP (46).

Analysis of the six site-directed Twf_1-174 mutants demonstrates that the N-terminal ADF-H domain of twinfilin utilizes a similar interface for interactions with actin monomers as ADF/cofilins (Fig. 5A). Because the electrostatic surface potentials on the actin monomer-binding surfaces of the N-terminal ADF-H domain of twinfilin and yeast cofilin are very similar to each other (Fig. 5B), we propose that these proteins also bind to the same site on the actin monomer. This is supported by the observation that twinfilin and ADF/cofilins compete with each other in binding to actin monomers (10). However, it is important to note that the actin monomer-binding site of twinfilin is somewhat extended as compared with the one of ADF/cofilin because additionally, in Twf_1-174, the residues Gln-76 and/or Gln-79, located in the loop between strands 3 and 4, are important for G-actin binding. Mutations at the corresponding region of ADF/cofilins do not result in defects in actin monomer binding (32). This region also maps outside the actin monomer-binding site of yeast cofilin determined using synchroton x-ray footprinting (34). Molecular dynamics simulation suggested that in ADF/cofilin-actin monomer complex, this region points away from the binding site on actin monomer toward the next subunit in the actin filament (37). However, in Twf_1-142, this loop is bent toward the C-terminal end of the helix 3. Assuming that the N-terminal ADF-H domain of twinfilin binds to an actin monomer in a similar orientation as modeled for cofilin by Wriggers et al. (37), the strands 3 and 4 in Twf_1-142 are bent to place residues Gln-76 and Gln-79 close to the N and C termini of actin in subdomain 1. Therefore, it is possible that the small differences in the actin monomer interfaces between twinfilin and ADF/cofilins account for differences in the actin monomer interactions between these two proteins.

ADF/cofilins interact with both actin monomers and filaments, whereas all twinfilins characterized so far bind only to actin monomers (3, 9). The residues specific for F-actin binding in ADF/cofilins have been mapped to the strand 4 and to the C-terminal helix (32, 33). Interestingly, the most pronounced structural differences between ADF/cofilin and Twf_1-142 are found in these two regions: the loop connecting the strands 3 and 4 in Twf_1-142 is bent toward the C-terminal end of helix 3, and the C-terminal helix in Twf_1-142 is tilted ~25^o as compared with the one in ADF/cofilins (Fig. 1C). When ADF/cofilin molecule is replaced by Twf_1-142 in the currently available ADF/cofilin-F-actin models (33, 38), the strands 3 and 4 of Twf_1-142 point toward the actin filament, causing a steric hindrance. Thus, our data provide a structural explanation for the lack of actin filament binding activity in twinfilin.

Taken together, these data suggest that the ADF-H domains form a structurally conserved actin-binding motif. Because the most ancient member of this family, ADF/cofilin, binds actin filaments, it is possible that during evolution, the actin filament binding activity was lost in twinfilin as a result of structural changes in the actin filament-binding site of this domain. This led to an evolution of a new family of actin monomer-binding proteins, which enabled more efficient and complex regulation of actin dynamics in cells. The ADF-H domain appears to be a specific ADP-actin-binding motif because at least ADF/cofilins and twinfilins bind ADP-actin with a significantly higher affinity than ATP-actin (10, 11, 39, 40). Thus, the actin monomer binding properties of twinfilin are different from other actin monomer binding/sequestering proteins such as thymosin-4 and profilin, which bind ATP-G-actin with higher affinity than ADP-G-actin (1). Therefore, the ancient duplication of the ADF-H domain and subsequent modifications in the structure and actin binding properties of this domain were critical for the appearance of a specific ADP-actin monomer-binding/sequestering protein, twinfilin.

	FOOTNOTES

^* This work was supported by grants from the Academy of Finland, Biocentrum Helsinki, and the European Molecular Biology (EMBO) Young Investigator Program (to P. L.). Access to the beam lines at European Molecular Biology Laboratory (EMBL)/Deutsches Elektronen Synchrotron (DESY), Hamburg, was supported by the contract HPRI-1999-00017 by the European Commission.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The atomic coordinates and the structure factors (code 1M4J) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).

^§ Supported by a fellowship from Helsinki Graduate School in Biosciences.

^** To whom correspondence should be addressed. Tel.: 358-9-19159499; Fax: 358-9-19159366; E-mail: pekka.lappalainen@helsinki.fi.

Published, JBC Papers in Press, August 30, 2002, DOI 10.1074/jbc.M208225200

	ABBREVIATIONS

The abbreviations used are: ADF, actin-depolymerizing factor; ADF-H, ADF homology; GST, glutathione S-transferase; NBD, 7-chloro-4-nitrobenz-2-oxa-1,3-diazole; Twf, twinfilin.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				V. O. Paavilainen, E. Oksanen, A. Goldman, and P. Lappalainen Structure of the actin-depolymerizing factor homology domain in complex with actin J. Cell Biol., July 14, 2008; 182(1): 51 - 59. [Abstract] [Full Text] [PDF]

				V. O. Paavilainen, M. Hellman, E. Helfer, M. Bovellan, A. Annila, M.-F. Carlier, P. Permi, and P. Lappalainen Structural basis and evolutionary origin of actin filament capping by twinfilin PNAS, February 27, 2007; 104(9): 3113 - 3118. [Abstract] [Full Text] [PDF]

				J. B. Moseley, K. Okada, H. I. Balcer, D. R. Kovar, T. D. Pollard, and B. L. Goode Twinfilin is an actin-filament-severing protein and promotes rapid turnover of actin structures in vivo J. Cell Sci., April 15, 2006; 119(8): 1547 - 1557. [Abstract] [Full Text] [PDF]

				D. Korkin, F. P. Davis, and A. Sali Localization of protein-binding sites within families of proteins Protein Sci., September 1, 2005; 14(9): 2350 - 2360. [Abstract] [Full Text] [PDF]

				H. Schuler, A.-K. Mueller, and K. Matuschewski A Plasmodium Actin-depolymerizing Factor That Binds Exclusively to Actin Monomers Mol. Biol. Cell, September 1, 2005; 16(9): 4013 - 4023. [Abstract] [Full Text] [PDF]

				O. Quintero-Monzon, A. A. Rodal, B. Strokopytov, S. C. Almo, and B. L. Goode Structural and Functional Dissection of the Abp1 ADFH Actin-binding Domain Reveals Versatile In Vivo Adapter Functions Mol. Biol. Cell, July 1, 2005; 16(7): 3128 - 3139. [Abstract] [Full Text] [PDF]

				X. Li, X. Liu, Z. Lou, X. Duan, H. Wu, Y. Liu, and Z. Rao Crystal structure of human coactosin-like protein at 1.9 A resolution Protein Sci., November 1, 2004; 13(11): 2845 - 2851. [Abstract] [Full Text] [PDF]

				B. J. Pope, K. M. Zierler-Gould, R. Kuhne, A. G. Weeds, and L. J. Ball Solution Structure of Human Cofilin: ACTIN BINDING, pH SENSITIVITY, AND RELATIONSHIP TO ACTIN-DEPOLYMERIZING FACTOR J. Biol. Chem., February 6, 2004; 279(6): 4840 - 4848. [Abstract] [Full Text] [PDF]

				M. K. Vartiainen, E. M. Sarkkinen, T. Matilainen, M. Salminen, and P. Lappalainen Mammals Have Two Twinfilin Isoforms Whose Subcellular Localizations and Tissue Distributions Are Differentially Regulated J. Biol. Chem., September 5, 2003; 278(36): 34347 - 34355. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/45/43089    most recent M208225200v1