Arabidopsis AtADF1 is Functionally Affected by Mutations on Actin Binding Sites

Chun-Hai Dong,Wei-Ping Tang and Jia-Yao Liu

College of Life Sciences,Qingdao Agricultural University,Qingdao 266109,China

Introduction

Actin exists either in the monomeric(G-actin)or the polymeric form(F-actin).In response to endogenous cues and external signals,the polymerization of G-actin and the depolymerization of F-actin occur dynamically.Actin turnover in vivo is 100–200 fold faster than it is in vitro.This enhanced turnover is due primarily to the activity of proteins belonging to the actin depolymerizing factor(ADF)/cofilin family(Bamburg 1999).

The ADF/cofilin family is highly conserved,and family members have been shown to be essential for the survival of eukaryotes.In Saccharomyces cerevisiae,disruption of the cofilin gene was shown to be lethal because mutant cells were unable to divide(Iida et al.1993;Moon et al.1993).In Drosophila melanogaster,a cofilin mutant called twinstar was found to be defective in centrosome migration and cytokinesis,and the mutant embryo was arrested at the larval stage(Gunsalus et al.1995).The essential function of ADF/cofilin was also reported in Dictyostelium discoideum(Aizawa et al.1995)and Caenorhabditis elegans(Ono et al.1999).Arabidopsis thaliana L.contains a large ADF family(Dong et al.2001a;Maciver and Hussey 2002;Ruzicka et al.2007),and moderate changes of ADF levels in transgenic Arabidopsis significantly affect actin organization and F-actin-dependent cellular activities(Dong et al.2001b).Recently,it was reported that increased actin filament bundling occurred when ADF2 was decreased in inducible RNA interference Arabidopsis lines(Clément et al.2009),and similar enhanced F-actin bundling was found in the ADF4 knockout Arabidopsis mutant(Henty et al.2011).

Actin depolymerizing factor/cofilin proteins bind to both G-and F-actin.Under physiological conditions,ADF/cofilin has a higher affinity for ADP-bound than ATP-bound actin filaments(Maciver et al.1991;Carlier et al.1997;Blanchoin and Pollard 1999).This difference in affinity plays a major role in differentiating the pool of newly polymerized ATP-actin filaments that are resistant to ADF-inducing depolymerization from the pool of aging ADP-actin filaments that can be rapidly depolymerized by ADF(Pollard 2001).The binding of ADF/cofilin to F-actin causes a twist in the filament and lowers the persistence length(McGough et al.1997;McCullough et al.2008).This leads to destabilization of the lateral contacts of actin monomers within the filaments(McGough and Chiu 1999)and renders the filament more susceptible to breakage events at boundaries between bare segments and ADF/cofilindecorated segments(De La Cruz 2009;Suarez et al.2011),resulting in an increase of actin filament barbed ends and filament disassembly(Andrianantoandro and Pollard 2006;Reymann et al.2011).The mechanism by which ADF induces actin depolymerization requires both an increase in the rate of depolymerization at the pointed ends,and the severing of actin filaments at the barbed ends.

Molecular structures of a number of ADF/cofilin proteins from different organisms have been determined(Hatanaka et al.1996;Fedorov et al.1997;Leonard et al.1997;Bowman et al.2000),and these proteins share a similar topology,all being built around a four strand mixed β-sheet surrounded by two αhelices on each face.Because of the difficulty in crystallization of the ADF/cofilin-actin complex,confirmation of a detailed interface of the ADF/cofilin in G-and F-actin binding sites has been controversial.Analysis using radiolytic oxidative protein footprinting and mass spectrometry to probe the conformation of G-actin in the G-actin/cofilin binary complex suggested that cofilin binds to the cleft between subdomains 1 and 2 in actin(Kamal et al.2007).However,based on the structural homology between cofilin and the gelsolin segment 1(GS1)(Wriggers et al.1998),and the competitive binding of cofilin against GS1(Mannherz et al.2007),cofilin has been suggested to bind to G-actin at the cleft between subdomains 1 and 3.This model is supported by a recent finding from analysis of the crystal structure of the mouse twinfilin’s C-terminal ADFH domain in complex with an actin monomer(Paavilainen et al.2008),indicating that ADF-H domains bind between actin subdomains 1 and 3 with an insertion of the long α-helix into the hydrophobic cleft of actin.The mechanism by which the ADF/cofilin proteins dynamically interact with actin is poorly understood.

Some studies,based mostly on the ADF/cofilin residue mutagenesis,have been reported using mutants derived from either amino acid substitutions or residue deletions.Peptide mapping identified the ADF/cofilin N-terminal Ser-3(i.e.Ser-6 in plant ADF)as the phosphorylation site(Agnew et al.1995;Moriyama et al.1996;Smertenko et al.1998).In vitro experiments showed that the Arabidopsis AtADF1(S6D)with residue substitution of the Ser-6 with aspartic acid,mimicking phosphorylated AtADF1,displayed significantly decreased affinity for G-and F-actin(Ressad et al.1998).In addition,a study of yeast cofilin mutagenesis suggested that the basic residues located at the N-terminus of α-helix 3 are essential for both G-and F-actin binding(Lappalainen et al.1997;Ojala et al.2001).Similar studies of yeast and human cofilin(Lappalainen et al.1997;Pope et al.2000)and of maize ADF3(Jiang et al.1997)suggest that basic residues located at β-strand 4,5,and αhelix 4 are important for interactions with F-actin.However,the actin binding sites of Arabidopsis AtADF1 on the interface of the ADF-actin complex is not clear,and in vivo F-actin binding assay is missing.

In this study,we took the Arabidopsis AtADF1 as an example to analyze actin binding activities by application of different approaches,including a computational structural prediction of the actin-AtADF1 complex,an in vitro G-and F-actin binding assay,and an in vivo F-actin binding and depolymerizing analysis.Our results showed that the Arabidopsis AtADF1 α-helix 3 is in close proximity and forms a binding surface that interacts with actin subdomains 1 and 3,and the residue mutations with chargedto-alanine substitutions on the α-helix 3(R98/A,K100/A)disrupt both G-and F-actin binding.In addition,our study suggests that the AtADF1 α-helix 4 and β-strand 5 are spatially close but opposite to the N-terminus and α-helix 3,and probably form the actin binding site important for F-actin binding.Through use of an in vivo F-actin binding assay and transgenic Arabidopsis plants,our study demonstrates the crucial importance of the basic residues of β-strand 5 and α-helix 4 in F-actin binding and depolymerizing activities.This study thus advances our understanding of the molecular interaction of the Arabidopsis ADF and cellular actin structures.

Results

Generation of AtADF1 actin binding site mutants based on structural analysis of the AtADF1-actin complex

Analysis of the crystal structure of the mouse twinfilin’s C-terminal ADF-H domain in complex with G-actin,showing that ADF-H domains bind between actin subdomains 1 and 3 with an insertion of the long α-helix into the hydrophobic cleft of actin(Paavilainen et al.2008),prompted us to study the detailed interface of the Arabidopsis AtADF1-actin complex.We used the molecular simulation approach as described previously by Wriggers et al.(1998),in whose research it was suggested that yeast cofilin interacts with actin in a manner similar to the GS1(Hatanaka et al.1996;Wriggers et al.1998)to replace the GS1 in the complex structure of the actin-GS1(McLaughlin et al.1993)with the Arabidopsis AtADF1(Bowman et al.2000),generating a putative AtADF1-actin complex structure.Our study showed a very similar folding of the Arabidopsis AtADF1 to GS1,and the AtADF1-actin complex produced steric clashes of the AtADF1 N-terminus with actin.The AtADF1 N-terminus contains a phosphorylation site at Ser-6,and it is known that a mutation at this site(AtADF1S6D)significantly affects both G-and F-actin binding(Ressad et al.1998).In addition,inspection of the structural model revealed that the α-helix 3(Lys96–Leu114)is close to actin subdomains 1 and 3,and is within a distance suitable for protein-protein interaction.The N-terminus and α-helix 3 are in close proximity in the 3-D structure,and form a binding surface that interacts with subdomains 1 and 3 of G-actin.Moreover,the computational prediction from the AtADF1-actin complex showed that α-helix 4 and β-strand 5 are spatially close but opposite to the N-terminus and α-helix 3,and probably form another actin binding site important for F-actin binding.

Based on the predicted AtADF1-actin complex structure,we decided to use site-directed mutagenesis to create two mutants with either charged-to-alanine substitutions on α-helix 3(R98/A,K100/A),named AtADF1(α3),or with amino acid substitutions on strand β5(K82/A)and α-helix 4(R135/A,R137/A),named AtADF1(β5α4)(Figure 1).Nucleotide mutations were confirmed by sequencing both strands of the generated cDNA clones.

Figure 1.Amino acid sequences of the wild-type AtADF1(WT)and mutants AtADF1(α3)and AtADF1(β5α4).Amino acid sequences of AtADF1(WT)and mutants AtADF1(α3)and AtADF1(β5α4)are shown in the figure.The positions of secondary structure elements based on the Arabidopsis AtADF1 crystal structure(Bowman et al.2000)are shown above the sequences.The mutated residues corresponding to those in yeast cofilin(Lappalainen et al.1997)are highlighted.The asterisk(?)marks the position of the phosphorylation site at the N-terminus.Numbers indicate the positions of the amino acid residues.

Highly-conserved residues Arg98 and Lys100 in AtADF1 α-helix 3 are essential for both actin monomer binding and actin filament binding

We first determined the effects of residue mutations with charged-to-alanine substitutions on α-helix 3(R98/A,K100/A)on the overall stability of the mutant proteins.The wild-type AtADF1(WT)and the mutant AtADF1(α3)were expressed in Escherichia coli as the glutathione-S-transferase fusion protein.The expression level of AtADF1(α3)was comparable to that of AtADF1(WT).Fusion proteins were purified on a glutathione column(see Materials and Methods).To test the effect of the residue mutations(R98/A,K100/A)on the stability of mutant proteins,we performed a urea denaturation assay as compared to the AtADF1(WT).The transition state for the mutant AtADF1(α3)was at approximately 4.2 mol/L urea,similar to that of the AtADF1(WT),indicating that the purified mutant proteins were properly folded.

It is known that the ADF/cofilin proteins decrease the rate of nucleotide dissociation from actin monomers.To characterize the binding of ADF proteins to actin monomers,we measured the rate of nucleotide dissociation from Mg-εADP-actin in the presence of AtADF1(WT)and mutant AtADF1(α3).A saturating amount of AtADF1(WT)blocks Mg-εADP dissociation from muscle actin(Figure 2A).A saturating amount of AtADF1(α3)mutant also inhibits the rate of nucleotide dissociation from actin(Figure 2A,closed circles),but with a lower efficiency when compared to the AtADF1(WT),suggesting that the mutations in AtADF1(α3)decrease the affinity of the ADF for actin monomer binding.

We characterized the interaction of AtADF1(α3)with actin filaments using a pyrene fluorescence assay(Carlier et al.1997;Blanchoin and Pollard 1999).Pyrene-labeled actin filaments were approximately 20 fold more fluorescent than actin monomers.The pyrene fluorescence was quenched by the binding of ADF to pyrene-labeled actin filaments.It is possible for the decrease in pyrene fluorescence due to the depolymerization of actin filaments by AtADF1 to affect the binding experiment.To avoid this potential problem,we incubated AtADF1 and pyrene-labeled actin filaments for only 60 s before measurements.Under these conditions,the variation of the fluorescence decrease after binding of AtADF1 to pyrene-labeled actin filaments produced a saturation curve that fits with equation 1.The affinity of AtADF1 for actin filaments is Kd=1± 0.4μM Figure 2B,closed squares;Ressad et al.1998).Under the same conditions,the AtADF1(α3)mutant binds actin filaments with an affinity at least 20 fold lower(Figure 2B,closed circles).

Figure 2.In vitro actin binding assays of the wild-type AtADF1(WT)and mutants AtADF1(α3)and AtADF1(β5α4).(A)Effect of AtADF1(WT)and AtADF1(α3)on the rate of nucleotide dissociation from actin.Time course of fluorescence change after mixing muscle Mg-εADP-actin with ATP.Open circles,6 μM MgεADP-actin with 1000 μM ATP.Closed squares,6 μM Mg-εADP-actin plus 6 μM AtADF1(WT).Closed circles,6 μM Mg-εADP-actin plus 6 μM mutant AtADF1(α3).Solid lines are the best fits to single exponentials.(B)Fluorescence assay for the interaction of AtADF1 with pyrenelabeled actin filaments.Dependence of the decrease of pyrenyl fluorescence after mixing partially-labeled actin filaments with ADF for 60 s was plotted as a function of AtADF1 concentration.Closed squares,0.5μM pyrenyl-labeled actin filaments(10%labeled)plus AtADF1(WT).Closed circles,0.5μM pyrenyl-labeled actin filaments(10%labeled)plus the AtADF1(α3)mutant.Solid lines are the best fits of equation 1 to the data.(C)Depolymerization of actin filaments assayed by pelleting.The percent of actin in the pellets(black)and the supernatants are shown:1,actin alone;2,actin plus 0.36 μM AtADF1(α3);3,actin plus 0.36 μM AtADF1(β5α4);4,actin plus 0.66 μM AtADF1(WT).

Residue mutations on β-strand 5 and α-helix 4 affect F-actin binding and depolymerizing activities

The effects of specific mutations with amino acid substitutions on strand β5(K82/A)and α-helix 4(R135/A,R137/A)on the overall stability of the mutant AtADF1(β5α4)proteins were examined as described above.The AtADF1(β5α4)mutants were expressed in E.coli as glutathione-S-transferase fusion protein.Unfortunately,the mutant AtADF1(β5α4)was expressed at lower levels,and the solubility of AtADF1(β5α4)was lower than that of the AtADF1(WT),resulting in a very low yield of purified protein.The urea denaturation examination indicated that the transition state for the mutant was at approximately 4.2 M urea,similar to that of the AtADF1(WT),suggesting that the purified mutant proteins were properly folded.

The poor solubility and low yield of the mutant AtADF1(β5α4)made it impossible to examine actin binding activity using the nucleotide dissociation assay or the pyrene fluorescence assay described for AtADF1(α3).We decided to test for interactions between the mutant AtADF1(β5α4)and actin filaments through use of a pelleting assay as shown in Figure 2(C).As a control,the AtADF1(WT)converted most actin filaments into a nonpelletable species after 30 min of incubation in the presence of AtADF1(WT)(0.66μM).Because of the poor solubility and low yield of the mutant AtADF1(β5α4),its pelleting assay together with that of the AtADF1(α3)was performed at a lower concentration.In the presence of 0.36 μM AtADF1(α3),no change in the amount of pelletable actin was observed.All AtADF1(α3)remained in the supernatant,in good agreement with the lower affinity of this mutant for actin filaments(Figure 2B).In the presence of 0.36 μM AtADF1(β5α4),approximately 25%of the actin filaments did not pellet,indicating that the mutant is able to depolymerize actin filaments.As none of the AtADF1(β5α4)mutant was found in the pellet,our results suggest that AtADF1(β5α4)has a low affinity for actin filaments.

In vivo F-actin binding and depolymerizing assay of AtADF1(α3)and AtADF1(β5α4)in onion(Allium cepa)peel epidermal cells

Previously,we examined in vivo F-actin binding of the AtADF1(WT)by transiently expressing the fluorescent GFPAtADF1 fusion proteins in living onion peel epidermal cells(Dong et al.2001a).Using the same approach,we fused the open reading frames(ORFs)of AtADF1(α3)and AtADF1(β5α4)with GFP at its C-terminus,and the fusion genes were placed downstream of a 35S promoter in pGFP(GA)5 II(Kost et al.1998).Nucleotide sequences were confirmed by double-strand DNA sequencing.The resulting constructs were transiently expressed in onion peel epidermal cells mediated by particle bombardment gene transfer.Confocal microscopy observations revealed that both mutants AtADF1(α3)and AtADF1(β5α4)failed to bind to filamentous actin structures in living plant cells compared to the wild-type GFP-AtADF1 fusion protein(Figure 3).

To investigate whether overexpression of the AtADF1 mutant proteins affects actin organization in living plant cells,we employed a double-labeling approach using the following GFP variants as tags:CFP(ex:458 nm,em:>475 nm)and YFP(ex:514 nm,em:>530 nm).The yellow fluorescent YFP-mTn fusion was used to tag actin filaments,and the blue fluorescent CFP-ADF was used to label AtADF1(WT)and its mutants.We replaced the GFP-mTn fusion gene with YFP-mTn(Kost et al.1998).Gene fusions were confirmed by sequencing both strands of the generated cDNA clones.Confocal microscopy showed that both YFP-mTn and CFP-mTn fusion proteins labeled actin filaments as well as GFP-mTn(Kost et al.1998;Figure 4).By co-transformation of YFP-mTn and CFP-AtADF1,we double-labeled actin filaments observed from the YFP channel,and ADF from the CFP channel in the same cell.When CFP-AtADF1(WT)was co-expressed with YFP-mTn,a breakdown of thick actin cables was clearly observed(Figure 4E,F).In contrast,normal actin organization was observed when CFP-AtADF1(α3)was co-expressed with YFP-mTn(Figure 4G).When CFP-AtADF1(β5α4)was co-expressed with YFP-mTn,a significant partial depolymerization of thick actin cables was observed(Figure 4I).A summary of the examined cells can be found in Table 1.Some of the cells showed no clear fluorescence signal,probably because little protein was expressed or actin structures were not labeled well.

In vivo analysis of AtADF1(α3)and AtADF1(β5α4)in transgenic Arabidopsis

We also generated transgenic Arabidopsis plants overexpressing AtADF1(α3)and AtADF1(β5α4)to study in vivo functions of the mutant proteins in Arabidopsis seedling growth.We transformed wild-type Arabidopsis with the constructs containing 35S-AtADF1(α3)and 35S-AtADF1(β5α4)in a binary vector.A total of six homozygous lines carrying the 35S-AtADF1(α3)transgene and nine lines carrying the 35S-AtADF1(β5α4)mutant were obtained.Of these,three independent lines from each construct and the wild-type AtADF1(WT)transgenic lines that express comparable ADF protein levels were chosen for further studies(Figure 5).

Figure 3.In vivo F-actin binding of GFP-AtADF1(α3)and GFP-AtADF1(β5α4)fusion proteins in onion(Allium cepa)peel epidermal cells.(A)A living cell showing native actin organization as visualized by the GFP-mTn fusion protein(Kost et al.1998).(B)A living cell showing depolymerized filamentous actin structure after expression of the wild-type GFP-AtADF1(WT)fusion protein.(C)A living cell showing diffuse localization of the GFP-AtADF1(α3)fusion protein.(D)A living cell showing diffuse localization of the GFP-AtADF1(β5α4)fusion proteins.Bars=100μm.

Figure 4.F-actin organization of onion(Allium cepa)peel epidermal cells co-expressing YFP-mTn and CFP-AtADF1 fusion proteins.(A)A living cell showing native actin organization after expression of YFP-mTn only.(B)Image of the same cell as in(A).(C)Image of the same cell as in(D).(D)A living cell showing native actin organization after expression of CFP-mTn only.(E)A living cell showing disrupted actin organization after co-expression of YFP-mTn and wild-type CFP-AtADF1(WT).(F)Image of the same cell as in(E)showing in vivo F-actin binding of CFP-AtADF1(WT)after co-expression of YFP-mTn and CFPAtADF1(WT).(G)A living cell showing normal actin organization after co-expression of YFP-mTn and CFP-AtADF1(α3).(H)Image of the same cell as in(G)showing subcellular distribution of CFP-AtADF1(α3)after co-expression of YFP-mTn and CFPAtADF1(α3).(I)A living cell showing partially-depolymerized actin filaments after co-expression of YFP-mTn and CFP-AtADF1(β5α4).(J)Image of the same cell as in(I)showing diffuse distribution of CFP-AtADF1(β5α4)after co-expression of YFP-mTn and CFP-AtADF1(β5α4).(A),(C),(E),(G)and(I)Images from the YFP channel(ex:514 nm;em:>530 nm).(B),(D),(F),(H)and(J)Images from the CFP channel(ex:458 nm;em:>475 nm).Bars=100μm.

Table 1.Summary of onion(Allium cepa)peel epidermal cells co-expressing YFP-mTn and CFP-AtADF1(WT),CFP-AtADF1(α3),or CFP-AtADF1(β5α4)

Morphological analysis of the transgenic Arabidopsis lines revealed that none of the AtADF1(α3)lines with mutant protein levels at 25–40 fold over the endogenous AtADF1 displayed an abnormal phenotype compared to the AtADF1(WT)transgenic lines(Figure 6C).The transgenic lines overexpressing AtADF1(β5α4)by 15–25 fold showed a significant phenotype when grown in darkness.The length of seedling hypocotyls was decreased by 30–35%,although the other aspects of plant growth and development appeared normal(Figure 6D).

Discussion

The lack of atomic-resolution structural data on actin/cofilin complexes limits our understanding of the functional mechanisms of ADF/cofilin action.The interface of G-actin/cofilin and F-actin/cofilin complexes has been studied by mutagenesis(Lappalainen et al.1997),radiolytic footprinting(Guan et al.2002),and nuclear magnetic resonance(NMR)(Pope et al.2004).Based on the structural homology between cofilin and GS1(Wriggers et al.1998)and the competitive binding of cofilin against GS1(Mannherz et al.2007),cofilin was suggested to bind to G-actin at the cleft between subdomains 1 and 3.This model is supported by an analysis of the crystal structure of the mouse twinfilin’s C-terminal ADF-H domain in complex with an actin monomer(Paavilainen et al.2008).In this study,our computational imaging also showed that the N-terminus and αhelix 3 of the Arabidopsis AtADF1 are physically close to actin subdomains 1 and 3,supporting the view that residues from these elements may form an essential binding site for G-actin.This notion is supported by the results of Ressad et al.(1998)who demonstrated an essential role of the phosphorylation site(serine 6)in the N-terminus of the Arabidopsis AtADF1 for its binding to G-actin.In addition,this study further confirmed the essential activity of the N-terminal basic residues Arg98 and Lys100 in α-helix 3 for G-actin binding,as shown in Figure 2 and Figure 3,by using both in vitro and in vivo assays.

F-actin binding involves interactions between ADF/cofilin and two actin subunits.Compared to the elements essential for G-actin binding,a more extended surface of ADF/cofilin is likely required for F-actin binding.Based on residue mutagenesis studies of yeast cofilin(Lappalainen et al.1997)and human cofilin(Pope et al.2000),it was reported that β-strand 5 and α-helix 4 were required for F-actin binding,while the mutations of tyrosine residues(Tyr67 and Tyr70)on β-strand 4 in the hydrophobic core of the maize ZmADF3 did not affect the affinity of ADF3 for G-actin,but abolished its F-actin binding(Jiang et al.1997).The models for the F-actin binding sites have been lacking in definitive structural support until recently,when the crystal structure of the mouse twinfilin’s C-terminal ADF-H domain in complex with an actin monomer was obtained(Paavilainen et al.2008),and the detail interface analysis indicated that the ADF/cofilin residues R80,K82,E134,R135,and R138,which were previously shown to be important for F-actin binding by mutagenesis(Lappalainen et al.1997),are located at the interface.In this study,our computational imaging showed that the residues from β-strand 5 and α-helix 4 of the Arabidopsis AtADF1 are spatially close but opposite to the N-terminus and α-helix 3,similarly forming the F-actin binding site for an adjacent monomer in the actin filament.Mutations of the available basic residues at β-strand 5 and α-helix 4 significantly affect F-actin binding activity,providing direct evidence for its importance in the binding of AtADF1 to actin filaments(Figures 2C,3).

In this study,we used the charged-to-alanine mutation approach to neutralize charged amino acids in the generation of AtADF1(α3)and AtADF1(β5α4)mutants.In contrast,Van Troys et al.(2000)used charge reversal mutations instead(e.g.lysine to glutamate),although it was suggested that charge reversal mutations affect protein stability and structure.To examine the effects of basic residues-to-alanine mutations on the overall stability of AtADF1(α3)and AtADF1(β5α4)proteins,in vitro urea denaturation experiments were performed,and the results indicate that the mutant proteins are folded properly.However,we found that AtADF1(β5α4),which has mutations in four residues,displayed a decreased solubility.

Figure 5.Analysis of ADF levels in transgenic Arabidopsis seedlings.(A)AtADF1(WT)lines.(B)AtADF1(α3)lines.(C)AtADF1(β5α4)lines.A transgenic line carrying a 35S-GFP-mTn transgene wild-type(WT)was retransformed with constructs containing 35S-AtADF1,35S-AtADF1(α3)and 35S-AtADF(β5α4).Three independent lines of each double transgenic construct were selected for western blot analysis using an anti-AtADF1 polyclonal antibody(Dong et al.2001b).The expression levels of GFP-mTn detected by a GFP antibody were used as a loading reference.

Figure 6.Phenotypes of AtADF1(α3)and AtADF1(β5α4)transgenic plants.(A)Non-transgenic wild-type(WT)control.(B)AtADF1(WT).(C)AtADF1(α3).(D)AtADF1(β5α4).Bars=10 mm.

Characterization of the two ADF1 mutants,AtADF1(α3)and AtADF1(β5α4),revealed their different proprieties both in vitro and in vivo.Both mutants have a lower affinity for actin filaments than the AtADF1(WT),suggesting that both(α3)and(β5α4)regions of AtADF1 are important for the binding of actin filaments.However,AtADF1(β5α4)efficiently depolymerizes actin filaments both in vitro(Figure 2)and in vivo(Figure 4),but AtADF1(α3)does not act as a depolymerizing factor.One possible explanation for this difference is that the affinity of AtADF1(α3)for actin monomers is lower than that of the AtADF1(WT).Unfortunately,the very limited yield of pure AtADF1(β5α4)proteins owing to its low solubility prevented us from clearly characterizing its affinity for actin monomers.Because AtADF1(β5α4)depolymerizes actin filaments both in vitro and in vivo with no evidence of a direct interaction with actin filaments,it is likely that this mutant may depolymerize actin filaments simply by sequestering actin monomers.The result suggests that a strong affinity for actin filaments does not seem to be essential for the Arabidopsis AtADF1 to function in the control of actin filament depolymerization.However,if the affinity for both actin monomers and actin filaments is altered as is the case with AtADF1(α3),the AtADF1 mutant derivative becomes non-functional.In conclusion,comparative studies of these two mutants suggest that if the affinity of AtADF1 for actin filaments is lowered,a strong affinity for actin monomers is essential to maintain the in vivo function of AtADF1.

Materials and Methods

Site-directed mutagenesis

Mutations of the Arabidopsis thaliana L.AtADF1 were generated according to Horton et al.(1989).Fragments from the AtADF1 cDNA were generated in separate polymerase chain reactions(PCRs).The primers containing mutated base(s)were designed,and the PCR product ends contained complementary sequences.When these PCR products were mixed,denatured,and reannealed,the strands having the matching sequences overlapped and acted as primers for each other.Extension of this overlap by PCR produced a recombinant molecule that contained the mutated base(s).Based on this technique,mutants of the Arabidopsis AtADF1(α3)and AtADF1(β5α4)were generated

Biochemical assay of mutant proteins

Reagents and materials were purchased from Sigma-Aldrich(St Louis,MO,USA)(dithiothreitol(DTT),ethylenediamine tetraacetic acid(EDTA),N-Lauroylsarcosine,Tris,sodium azide,hexokinase,ATP,and ADP),Molecular Probes(Eugene,OR,USA)(1,N6-ethenoadenosine-5′-diphosphate(ε-ADP)),Roche(Basel,Switzerland)(protease inhibitors),Amersham Pharmacia Biotech(Piscataway,NJ,USA)(Glutathione Sepharose 4B),and Bio-Rad(Hercules,CA,USA)(DEAE 5 FPLC column),respectively.

Wild-type and mutant AtADF1 were expressed in E.coli BL21(DE3)p LysS as glutathione-S-transferase(GST)protein fusions using pGEX-4T-2.Cells were grown in 1,000 mL of LB(Lysogeny Broth)medium to an optical density of 0.8 at 595 nm,and induced with isopropyl-thio-β-D-galactoside(IPTG,1 mM)for 4 h.After centrifugation(3,500 g,20 min),the cell pellet was resuspended in 35 mL of lysis buffer(20 mM Tris pH=7.5,500 mM NaCl,1 mM EDTA,10%glycerol,1%triton(4%for AtADF1(β5α4)mutant),0.5%N-Lauroylsarcosine,and two tablets of protease inhibitor).The lysate was centrifuged at 100,000 g for 30 min,and the supernatant was loaded onto a 4 mL glutathione column.The beads were washed sequentially with 2 volumes of TBSE buffer(20 mM Tris HCl pH 8;100 mM NaCl,1 mM EDTA),4 volumes of TBSE plus 250 mM NaCl and 2 volumes of TBSE.GST-AtADF1 bound to the beads was cleaved by thrombin(25 U/mL)for 2 h.Cleaved AtADF1 was eluted with 8 mL of TBSE buffer.The fractions containing AtADF1 proteins were dialyzed overnight against buffer DEAE 5(10 mM Tris-Cl pH 8,50 mM NaCl,5 mM DTT,1 mM EDTA,0.1 mM phenylmethylsulfonyl fluoride(PMSF),1 mM NaN3).AtADF1 proteins were then loaded on a DEAE 5 FPLC(fast protein liquid chromatography)column and eluted with a linear 0–500 mM gradient of NaCl.The peak fraction containing AtADF1 was pooled,dialyzed overnight in storage buffer(10 mM Tris-Cl pH 7,1 mM DTT,1 mM EDTA,1 mM NaN3),and concentrated using a Centriprep YM-3 Amicon(Millipore,USA).

Actin was purified from rabbit skeletal muscle acetone powder according to Spudich and Watt(1971),and monomeric Ca-ATP-actin was purified on a Sephacryl S-300 chromatography at 4°C in G(Gactin)buffer(5 mM Tris-Cl,pH 8,0.2 mM ATP,0.1 mM CaCl2,0.5 mM DTT).Actin was labeled with pyrene iodoacetamide according to Kouyama and Mihashi(1981)modified by Pollard et al.(1984).Mg-ATP-actin was prepared by addition of 0.2 mM ethyleneglycoltetraacetic(EGTA)and 11 fold molar excess of MgCl2over actin,and used within hours.Polymerization was initiated by the addition of 1/9(v/v)10 X KME(500 mM KCl,MgCl2,10 mM EGTA,100 mM Tris-Cl,pH 7).

For the pyrene fluorescence assay,we used the change in fluorescence with excitation at 366 nm and emission at 387 nm to follow the interaction of the AtADF1(WT)and AtADF1(α3)with pyrenyl-labeled actin filaments(Carlier et al.1997).Data were collected using a FluoroMax spectrofluorometer(Spex,Metuchen,NJ,USA).

The nucleotide dissociation assay was performed according to Blanchoin and Pollard(1998).Interaction of the AtADF1(WT)and mutant AtADF1(α3)with actin monomers was followed using the ability of AtADF1 to inhibit the rate of nucleotide dissociation from actin.Mg-ε-ADP-actin was prepared according to Blanchoin and Pollard(1998).The dissociation of ε-ADP from actin by a large excess of ATP was observed in fluorescence with excitation at 360 nm and emission at 410 nm.

For the pelleting assay,the AtADF1 proteins and actin filaments were incubated together for 30 min and centrifuged for 20 min to pellet actin filaments.The supernatant and pellet were collected,and proteins were separated by sodium dodecylsulfate polyacrylamide gel electrophoresis.Coomassie blue-stained gels were digitized with a GS-800 calibrated densitometer(Umax;Bio-Rad)and the density of gel bands was quantified in the two fractions using quantityOne(Bio-Rad).

Kinetics of dissociation of the nucleotide was fit to exponentials with the Kaleidagraph software(Synergy Software,Reading,PA,USA).Binding experiments were fitted with equation 1:

where F is the observed fluorescence,Ff is the fluorescence of actin filaments,Fb is the fluorescence of actin filaments bound to AtADF1,(A)is the total concentration of actin filaments,(ADF)is the total concentration of ADF,and Kd is the dissociation equilibrium constant of the complex.

Plant materials and growth conditions

Transgenic lines mediated by vacuum infiltration transformation(Bechtold et al.1993)were selected on Murashige and Skoog(MS)plates containing kanamycin(50μg/mL).Seeds were sterilized in a solution containing 25%(v/v)bleach and 0.01%(v/v)triton X-100 for 15 min with gentle agitation.After washing with sterile water threefold,the seeds were resuspended in 0.2%agar and plated on MS medium(MS salts,8%agar,and 0%or 3%sucrose).Seeded plates were placed at 4°C for 2 d to improve germination.Germinated seedlings and plants were grown under white fluorescence light(45μEm-2s-1,16:8 h light:dark cycle).Seedlings of 15–20 d old were transferred to soil and grown in a greenhouse.

Plant cell transfection by biolistic bombardment

Coding sequences of AtADF1 mutants were PCR amplified from the cDNA clones and cloned in frame at the 3′-end of the GFP cDNA into Pgfp(GA)5 II(Kost et al.1998).Expression of the resulting GFP-ADF fusion proteins were examined in onion peel epidermal cells mediated by particle bombardment gene transfer as described by Dong et al.(2001a).Confocal images were obtained using a water-immersion lens(Zeiss,Uberkuchen,Germany).

Immunoblot analysis

Protein extracts were obtained from 10 d old seedlings with phosphate-buffered saline(PBS)buffer containing 1 mM PMSF.Protein concentration in the supernatant was determined by the Bradford method(1976).For western blot analysis,protein bands were blotted onto a membrane(Millipore)according to the manufacturer’s instructions.The membrane was blocked with 3%non-fat milk in PBS buffer containing 0.05%Tween-20 for 1 h at room temperature before incubation with affinity-purified anti-ADF immunoglobulin(Ig)G(Dong et al.2001b).Horseradish peroxidase-linked antirabbit IgG(ECL;Amersham)was used as the secondary antibody.The reaction was detected according to the ECL(Amersham Pharmacia Biotech).

Acknowledgements

We would like to thank Dr.Nam-Hai Chua(The Rockefeller University)for critical reading of the manuscript,Dr.Swaminathan Kunchithapadam(National University of Singapore)for help in computational imaging,Dr.Laurent Blanchoin and Dr.Solenne Dufour(Laboratoire de Physiologie Cellulaire Végétale,CEA/CNRS/UJF,France)for the in vitro actin binding assay.This work was supported by the Shandong Taishan Scholar program,the Shandong Natural Science Foundation(ZR2012CM022,ZRB019E7),and the Laboratory of Biotechnology of Qingdao Agricultural University.

Agnew BJ,Minamide LS,Bamburg JR(1995)Reactivation of phosphorylated actin depolymerizing factor and identification of the regulatory site.J.Biol.Chem.270,17582–17587.

Aizawa H,Sutoh K,Tsubuki S,Kawashima S,Ishii A,Yahara I(1995)Identification,characterization,and intracellular distribution of cofilin in Dictyostelium discoideum.J.Biol.Chem.270,10923–10932.

Andrianantoandro E,Pollard TD(2006)Mechanism of actin filament turnover by severing and nucleation at different concentrations of ADF/cofilin.Mol.Cell 24,13–23.

Bamburg JR(1999)Proteins of the ADF/cofilin family:Essential regulators of actin dynamics.Annu.Rev.Cell Biol.15,185–230.

Bechtold N,Ellis J,Peletier G(1993)In planta Agrobacteriummediated gene transfer by infiltration of adult Arabidopsis thaliana plants.C.R.Acad.Sci.Paris Life Sci.316,1194–1199.

Blanchoin L,Pollard TD(1998)Interaction of actin monomers with Acanthamoeba actophorin(ADF/cofilin)and profilin.J.Biol.Chem.273,25106–25111.

Blanchoin L,Pollard TD(1999)Mechanism of interaction of Acanthamoeba actophorin(ADF/Cofilin)with actin filaments.J.Biol.Chem.274,15538–15546.

Bowman GD,Nodelman IM,Hong Y,Chua NH,Lindberg U,Schutt CE(2000)A comparative structural analysis of the ADF/cofilin family.Proteins 41,374–384.

Bradford MM(1976)A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proteindye binding.Anal.Biochem.72,248–254.

Carlier MF,Santolini J,Lanrent V,Melki R,Didry D,Hong Y,Xia GX,Chua NH,Pantolni D(1997)Actin depolymerizing factor(ADF/cofilin)enhances the rate of filament turnover:Implication in actin-based motility.J.Cell Biol.136,1307–1322.

Clément M,Ketelaar T,Rodiuc N,Banora MY,Smertenko A,Engler G,Abad P,Hussey PJ,de Almeida Engler J(2009)Actindepolymerizing factor2-mediated actin dynamics are essential for root-knot nematode infection of Arabidopsis.Plant Cell 21,2963–2979.

De La Cruz EM(2009)How cofilin severs an actin filament.Biophys.Rev.1,51–59.

Dong CH,Kost B,Xia G,Chua NH(2001a)Molecular identification and characterization of the Arabidopsis AtADF1,AtADF5 and AtADF6 genes.Plant Mol.Biol.45,517–527.

Dong CH,Xia G-X,Hong Y,Ramachandran S,Kost B,Chua NH(2001b)ADF proteins are involved in the control of flowering and regulate F-actin organization,cell expansion,and organ growth in Arabidopsis.Plant Cell 13,1333–1346.

Fedorov AA,Lappalainen P,Fedorov EV,Drubin DG,Almo SC(1997)Structure determination of yeast cofilin.Nat.Struct.Biol.4,366–369.

Guan JQ,Vorobiev S,Almo SC,Chance MR(2002)Mapping the G-actin binding surface of cofi lin using synchrotron protein footprinting.Biochemistry 41,5765–5775.

Gunsalus KC,Bonaccorsi S,Williams E,Verni F,Gatti M,Goldberg ML(1995)Mutations in twinstar,a Drosophila gene encoding a cofilin/ADF homologue,result in defects in centrosome migration and cytokinesis.J.Cell Biol.131,1243–1259.

Hatanaka H,Ogura K,Moriyama K,Ichikawa S,Yahara I,Inagaki F(1996)Tertiary structure of destrin and structural similarity between two actin-regulating protein families.Cell 85,1047–1055.

Henty JL,Bledsoe SW,Khurana P,Meagher RB,Day B,Blanchoin L,Staiger CJ(2011)Arabidopsis actin depolymerizing factor4 modulates the stochastic dynamic behavior of actin filaments in the cortical array of epidermal cells.Plant Cell 23,3711–3726.

Horton RM,Hunt HD,Ho SN,Pullen JK,Pease LR(1989)Engineering hybrid genes without the use of restriction enzymes:Gene splicing by overlap extension.Gene 77,61–68.

Iida K,Moriyama K,Matsumoto S,Kawasaki H,Nishida E,Yahara I(1993)Isolation of a yeast essential gene,COF1,that encodes a homologue of mammalian cofilin,a low-Mr actin-binding and depolymerizing protein.Gene 124,115–120.

Jiang CJ,Weeds AG,Khan S,Hussey PJ(1997)F-actin and G-actin binding are uncoupled by mutation of conserved tyrosine residues in maize actin depolymerizing factor(ZmADF).Proc.Natl.Acad.Sci.USA.94,9973–9978.

Kamal JK,Benchaar SA,Takamoto K,Reisler E,Chance MR(2007)Three-dimensional structure of cofilin bound to monomeric actin derived by structural mass spectrometry data.Proc.Natl.Acad.Sci.USA.104,7910–7915.

Kost B,Spielhofer P,Chua NH(1998)A GFP-mouse talin fusion protein labels plant actin filaments in vivo and visualizes the actin.Plant J.16,393–401.

Kouyama T,Mihashi K (1981)Fluorimetry study of N-(1-pyrenyl)iodoacetamide-labeled F-actin.Local structural change of actin protomer both on polymerization and on binding of heavy meromyosin.Eur.J.Biochem.114,33–38.

Lappalainen P,Fedorov EV,Fedorov AA,Almo SC,Drubin DG(1997)Essential functions and actin-binding surfaces of yeast cofilin revealed by systematic mutagenesis.EMBO J.16,5520–5530.

Leonard SA,Gittis AG,Petrella EC,Pollard TD,Lattman EE(1997)Crystal structure of the actin-binding protein actophorin from Acanthamoeba.Nat.Struct.Biol.4,369–373.

Maciver SK,Hussey PJ(2002)The ADF/cofilin family:Actinremodeling proteins.Genome Biol.3,3007.1–3007.12.

Maciver SK,Zot HG,Pollard TD(1991)Characterization of actin filament severing by actophorin from Acanthamoeba castellanii.J.Cell Biol.115,1611–1620.

Mannherz HG,Ballweber E,Galla M,Villard S,Granier C,Steegborn C,Schmidtmann A,Jaquet K,Pope B,Weeds AG(2007)Mapping the ADF/Cofilin binding site on monomeric actin by competitive cross-linking and peptide array:Evidence for a second binding site on monomeric actin.J.Mol.Biol.366,745–755.

McGough A,Pope B,Chiu W,Weeds A(1997)Cofilin changes the twist of F-actin:Implications for actin filament dynamics and cellular function.J.Cell Biol.138,771–781.

McCullough BR,Blanchoin L,Martiel JL,De la Cruz EM(2008)Cofilin increases the bending flexibility of actin filaments:Implications for severing and cell mechanics.J.Mol.Biol.381,550–558.

McGough A,Chiu W(1999)ADF/cofilin weakens lateral contacts in the actin filament.J.Mol.Biol.291,513–519.

McLaughlin PJ,Gooch JT,Mannherz HG,Weeds AG(1993)Structure of gelsolin segment 1-actin complex and the mechanism of filament severing.Nature 364,685–692.

Moon A,Janmey PA,Louie KA,Drubin DG(1993)Cofilin is an essential component of the yeast cortical cytoskeleton.J.Cell Biol.120,421–435.

Moriyama K,Iida K,Yaharra I(1996)Phosphorylation of Ser-3 of cofilin regulates its essential function on actin.Genes Cells 1,73–86.

Ojala PJ,Paavilainen V,Lappalainen P(2001)Identification of yeast cofilin residues specific for actin monomer and PIP2 binding.Biochemistry 40,15562–15569.

Ono S,Baillie DL,Benian GM(1999)UNC-60B,an ADF/cofilin family protein,is required for proper assembly of actin into myofibrils in Caenorhabditis elegans body wall muscle.J.Cell Biol.145,491–502.

Paavilainen VO,Oksanen E,Goldman A,Lappalainen P(2008)Structure of the actin-depolymerizing factor homology domain in complex with actin.J.Cell Biol.182,51–59.

Pollard TD(2001)Life scientists and politics in the United States.Nat.Rev.Mol.Cell Biol.2,929–931.

Pollard TD,Selden SC,Maupin P(1984)Interaction of actin filaments with microtubules.J.Cell Biol.99,33–37.

Pope BJ,Gonsior SM,Yeoh S,McGough A,Weeds AG(2000)Uncoupling actin filament fragmentation by cofilin from increased subunit turnover.J.Mol.Biol.298,649–661.

Pope BJ,Zierler-Gould KM,Kuhne R,Weeds AG,Ball LJ(2004)Solution structure of human cofilin:Actin binding,pH sensitivity,and relationship to actin-depolymerizing factor.J.Biol.Chem.279,4840–4848.

Ressad F,Didry D,Xia GX,Hong Y,Chua NH,Pantaloni D,Carlier MF(1998)Kinetic analysis of the interaction of actin-depolymerizing factor(ADF)/cofilin with G-and F-actins–comparison of plant and human ADFs and effect of phosphorylation.J.Biol.Chem.273,20894–20902.

Reymann AC,Suarez C,Guérin C,Martiel JL,Staiger CJ,Blanchoin L,Boujemaa-Paterski R(2011)Turnover of branched actin filament networks by stochastic fragmentation with ADF/cofilin.Mol.Biol.Cell 22,2541–2550.

Ruzicka DR,Kandasamy MK,McKinney EC,Burgos-Rivera B,Meagher RB(2007)The ancient subclasses of Arabidopsis ACTIN DEPOLYMERIZING FACTOR genes exhibit novel and differential expression.Plant J.52,460–472.

Smertenko AP,Jiang CJ,Simmons NJ,Weeds AG,Davies DR,Hussey PJ(1998)Ser6 in the maize actin-depolymerizing factor,ZmADF3,is phosphorylated by a calcium-stimulated protein kinase and is essential for the control of functional activity.Plant J.14,187–193.

Spudich JA,Watt S(1971)The regulation of rabbit skeletal muscle contraction.I.Biochemical studies of the interaction of the tropomyosin-troponin complex with actin and the proteolytic fragments of myosin.J.Biol.Chem.246,4866–4871.

Suarez C,Roland J,Boujemaa-Paterski R,Kang H,McCullough BR,Reymann AC,Guérin C,Martiel JL,De la Cruz EM,Blanchoin L(2011)Cofilin tunes the nucleotide state of actin filaments and severs at bare and decorated segment boundaries.Curr.Biol.21,862–868.

Van Troys M,Dewitte D,Verschelde JL,Goethals M,Vandekerckhove J,Ampe C(2000)The competitive interaction of actin and PIP2 with actophorin is based on overlapping target sites:Design of a gain-of-function mutant.Biochemistry 39,12181–12189.

Wriggers W,Tang JX,Azuma T,Marks PW,Janmey PA(1998)Cofilin and gelsolin segment-1:Molecular dynamics simulation and biochemical analysis predict a similar actin binding mode.J.Mol.Biol.282,921–932.

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Figure S1.Structural modeling of predicted actin-AtADF1 complex.

(A)Superposition of AtADF1(red)and gelsolin segment-1(gold).

(B)Putative AtADF1-actin complex.

Actin subdomains 1,2,3,4 and the α-helix 3,β-strand 5,and α-helix 4 of AtADF1 are indicated.

The coordinates of AtADF1 residues were provided by Bowman et al.(2000).The coordinates of actin as well as the crystal structure of the actin-gelsolin segment-1 complex were supplied by McLaughlin et al.(1993).