The actin bundling activity of actin bundling protein 34 is inhibited by calcium binding to the EF2

The actin bundling activity of actin bundling protein 34 is inhibited by calcium binding to the EF2

Biochemical and Biophysical Research Communications xxx (2018) 1e5 Contents lists available at ScienceDirect Biochemical and Biophysical Research Co...

2MB Sizes 1 Downloads 37 Views

Biochemical and Biophysical Research Communications xxx (2018) 1e5

Contents lists available at ScienceDirect

Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc

The actin bundling activity of actin bundling protein 34 is inhibited by calcium binding to the EF2 Jeong Min Chung, Han-ul Kim, Gwang Joong Kim, Dooil Jeoung, Hyun Suk Jung* Department of Biochemistry, College of Natural Sciences, Kangwon National University, 1 Kangwondaehak-gil, Chuncheon-si, Gangwon-do, 24341, Republic of Korea

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 July 2018 Accepted 23 July 2018 Available online xxx

Actin bundling protein 34 (ABP34) is the one of 11 actin-crosslinking proteins identified in Dictyostelium discoideum, a novel model organism for the study of actin-associated neurodegenerative disorders such as Alzheimer's disease and Huntington's disease. ABP34 localizes at the leading and trailing edges of locomotory cells, i.e., at the cell cortex, filopodia, and pseudopodia. Functionally, it serves to stabilize membrane-associated actin at sites of cellecell contact. In addition, this small crosslinking protein is involved in actin bundle formation, and its bundling activity is regulated by the concentration of calcium ion. Several studies have sought to determine the mechanism underlying the calcium-regulated actin bundling activity of ABP34, but it remains unclear. Using several mutational and structural analyses, we revealed that calcium binding to the EF2 motif disrupts the inter-domain interaction between the N- and C-domains, thereby inhibiting the actin bundling activity of ABP34. This finding provides clues about the pathogenesis of neurodegenerative disorders related to actin bundling. © 2018 Elsevier Inc. All rights reserved.

Keywords: Actin bundling protein Actin-crosslinking protein Protein structure Actin-binding protein EF-Hand

1. Introduction Actin bundling protein 34 (ABP34) is a monomeric 34 kDa protein (295 amino acids) in Dictyostelium discoideum. This small actincrosslinking protein plays a crucial role in formation of isotropic Factin networks and anisotropic bundles of filaments in the cytoplasm [1]. The crystal structure of ABP34 reveals that it has a “bent arm” shape with a two-domain structure consisting of the N-terminal calcium-binding domain and a C-terminal actin-binding domain, linked by six amino acid loops [2]. The N-terminal domain consists of eight alpha-helices (a1 to a8), five 310-helices (h1 to h5), and two beta-sheets (b1 to b2), whereas the C-terminal domain consists of four alpha-helices (a9 to a12) and one 310-helix (h6) [2]. Unlike other actin-binding proteins, which are pH-

Abbreviations: aa, amino acids; ABP, Actin-binding Protein; ABS, Actin-binding Site; CCD, Charge Coupled Device; DTT, Dithiothreitol; EGTA, Ethylene Glycol Tetraacetic Acid; IPTG, Isopropyl-b-D-Thiogalactopyranoside; LaB6, Lanthanum hexaboride; MOPs, 3-morpholinopropane-1-sulfonic acid; OD, Optical Density; LB, Luria Broth; PCR, Polymerase Chain Reaction; TEM, Transmission Electron Microscopy; LaB6, Lanthanum hexaboride; WT, Wild-Type. * Corresponding author. Department of Biochemistry, College of Natural Sciences, Kangwon National University, 1, Kangwondaehak-gil, Chuncheon-si, Gangwon-do, 24341, Republic of Korea. E-mail address: [email protected] (H.S. Jung).

sensitive, ABP34 is not affected by pH within physiological ranges. Instead, it is regulated by micromolar concentrations of calcium [3,4]. ABP34 has two putative calcium-binding EF-hand motifs, EF1 and EF2 [5]. Its bundling activity is activated at low calcium concentrations (1  108 M) and deactivated at high calcium levels (1  106 M) [3,6]. Biochemical studies of various truncated mutants have suggested that ABP34 has three actin-binding sites: one strong site (amino acids [aa] 193e254) and two weaker sites (aa 1e123 and 279e295) [7]. Moreover, the intramolecular interaction between interaction zone 1 (IZ-1, aa 71e123) and 2 (IZ-2, aa 193e254) plays a key role in regulating actin binding by maintaining the N-terminus (aa 1e76) in close proximity to the strong actin-binding site [8]. Functional and structural analyses have been performed in several previous studies, but the structure of the ABP34/F-actin complex has not yet been solved, and the regulatory mechanism of ABP34 is not fully understood. In this study, through several mutational and structural approaches, we obtained insights into the calcium-dependent regulation of actin bundle formation by ABP34.

https://doi.org/10.1016/j.bbrc.2018.07.122 0006-291X/© 2018 Elsevier Inc. All rights reserved.

Please cite this article in press as: J.M. Chung, et al., The actin bundling activity of actin bundling protein 34 is inhibited by calcium binding to the EF2, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.07.122

2

J.M. Chung et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e5

2. Materials and methods 2.1. Purification of recombinant proteins The full-length ABP34 gene (GenBank accession no. U32112) was PCR-amplified from synthesized D. discoideum DNA, and the amplicon was inserted downstream of the T7 promoter of the expression vector pET-28a(þ) (Novagen, U.S). This plasmid was used as a template for construction of ABP34 mutants. The mutants were PCR-amplified using primers containing either a stop codon in the middle of the amplified DNA fragment or a codon encoding alanine amino acid. The resultant constructs in pET-28a(þ) were transformed into Escherichia coli DH5a (DE3) cells, and stable transformants were selected on LB plates supplemented with 50 mg/ml kanamycin. From selected transformants, plasmid DNA was isolated and transformed into E. coli BL21(DE3). The transformants were grown in LB medium containing 50 mg/ml kanamycin at 37  C with shaking at 180 rpm until the optical density at 600 nm (OD600) reached 0.5e0.6. Protein expression was induced by the addition of IPTG to a final concentration of 1 mM. After 4 h, the cells were centrifuged at 3000  g for 10 min, washed once with low calcium solution containing 50 mM Na acetate, 2 mM MgCl2, 1 mM ethylene glycol tetraacetic acid (EGTA), 20 mM 3morpholinopropane-1-sulfonic acid (MOPs) at pH 7.0, and centrifuged again at 3000  g for 10 min. The pellets were resuspended in ice-cold lysis buffer. For insoluble mutants, 0.5% sarkosyl detergent was added to improve their solubility (details described in our previous study [9]). The resuspended pellets were sonicated for 10 min at 34% amplitude with 2 s intervals between each sonication pulse (Sonics, U.S) The lysates were centrifuged at 11,000  g for 10 min, and the supernatant was collected and loaded onto a prepacked 5 ml Ni-NTA column (GE Healthcare, U.K.) preequilibrated with lysis buffer. Wild-type (WT) ABP34 and mutant proteins were eluted stepwise with 25 ml of a 5e250 mM imidazole gradient. Fractions containing the ABP34 proteins were analyzed on 12.5% or 15% SDS-PAGE gels. 2.2. Size-exclusion chromatography (SEC) € The AKTA pure chromatography system and a Superdex 200 10/ 300 GL (GE Healthcare, USA) column were used for automated sizeexclusion chromatography (SEC). The Superdex 200 10/300 GL column was washed and equilibrated with one column volume each of MilliQ water and low calcium solution prior to sample injection. The concentrated protein samples were centrifuged at 13,000  g at 4  C for 15 min on an Avanti J25 preparative centrifuge (Beckman Coulter, USA), and the soluble supernatant was carefully injected into the system using a 1 ml syringe with a needle. SEC was performed at 0.5 ml/min, and 1 ml fractions were collected during the separation process as the OD280 was continuously monitored. Fractions corresponding to peaks were analyzed by 12.5% SDSPAGE. 2.3. F-actin bundling assay (low-speed co-sedimentation) The F-actin bundling assay was performed as described by Fechheimer and Taylor [4] and Fechheimer [3]. In the low-speed centrifugation assay, skeletal striated muscle G-actin was induced to polymerize at room temperature in F-actin solution (50 mM Na acetate, 2 mM MgCl2, 1 mM EGTA, 1 mM ATP, 1 mM Dithiothreitol (DTT), 1 mM NaN3, 20 mM MOPs, pH 7.0). To sustain the polymerized form, F-actin was mixed with 2% Alexa Fluor 488ephalloidin, which binds to actin filaments much more tightly than to actin monomers and prevents its depolymerization. Recombinant proteins (WT and mutant constructs) were subsequently incubated

with F-actin for 60 min at room temperature with the low and high (the low calcium buffer including 6 mM CaCl2) calcium solution, then centrifuged for 30 min at 10,000  g in a MICRO 17 TR tabletop centrifuge (Hanil, South Korea). Both supernatants and pellets were dissolved in equivalent volumes of SDS-PAGE sample buffer. Equal volumes of the total, supernatant solution, and pellet samples were analyzed on 12% SDS-PAGE gels. Protein bands were visualized by Coomassie brilliant blue staining. 2.4. Transmission electron microscopy Actin bundling activity was visualized by negative staining electron microscopy. Phalloidin-treated F-actin was mixed with purified recombinant proteins in the F-actin solution and incubated at RT for 60 min. The mixtures were diluted to a final concentration of 100 nM with the low and high calcium solution. A 5 ml of the mixture sample was applied to a glow-discharged carbon-coated grid and negatively stained with 3e4 drops of 1% aqueous uranyl acetate. The prepared grids were examined on a Tecnai 10 transmission electron microscope (FEI, USA) equipped with a lanthanum hexaboride (LaB6) cathode operating at 100 kV. Images were collected with a US1000 CCD camera (Gatan, USA) at a magnification of 0.32 nm/pixel. 3. Results 3.1. Actin bundling function of ABP34 ABP34 protein consists of 295 amino acids with two putative EFhand motifs (aa 82e112 and 135e163) in the N-domain and three distinct actin-binding sites in aa 1e124, 193e254, and 279e295 (Fig. 1A) [8]. To visualize the actin bundling activity of WT ABP34, negative staining electron microscopy was carried out (Fig. 1B). Because previous studies suggested that ABP34 is a calciumsensitive actin bundling protein [8,10], we performed the actin bundling assay under two different conditions, i.e., low and high concentrations of CaCl2 using low and high calcium solution, respectively. As shown in Fig. 1B, WT ABP34 protein induced formation of actin bundles at low calcium concentration (middle row in Fig. 1B). However, when the concentration of CaCl2 was elevated to 5 mM, the bundling activity of ABP34 was diminished (bottom row in Fig. 1B). Taken together, these findings indicate that ABP34 can induce actin bundling at low calcium concentration, but that its activity is inhibited at higher concentrations. 3.2. Calcium-dependent actin bundling mechanism of ABP34 As described above, calcium ion is a key player in the regulation of the actin bundling mechanism of ABP34. The ABP34 protein contains EF-hand motifs, a type of helix-loop-helix motif, which can be classified as functional or non-functional [8,10]. As shown in sequence homologues in some species of amoeba (Fig. 2A) [2], the functional EF-hand (EF2) has a 12-residue Ca2þ-binding loop that starts with an aspartate and ends with glutamate (144DVNFDGRVSFLE155) [2]. Within the loop, calcium ion interacts with oxygen ligands in conjunction with the ligands that participate in calcium coordination, Asp144, Asn146, and Asp148 and the main-chain carbonyl groups of Arg150 and Glu155 (Fig. 2B). To investigate the calcium-dependent actin bundling mechanism of ABP34, we produced variants of the protein by replacing several residues of EF2 with alanine (DEF2 mutants; D144A, R150A, and E155A) (Fig. 2B), and then examined actin bundling activity by high speed co-sedimentation assay and negative staining EM analysis (Fig. 3). The EF2 mutants (Fig. 3A) show similar bundling function to that of the WT ABP34 at low calcium concentration

Please cite this article in press as: J.M. Chung, et al., The actin bundling activity of actin bundling protein 34 is inhibited by calcium binding to the EF2, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.07.122

J.M. Chung et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e5

3

Fig. 2. Sequence analysis of ABP34 with various amoebal species and its EF2 motif. (A) Sequence alignment of ABP34 with various amoebal species; Dictyostelium discoideum (D. discoideum), Dictyostelium fasciculatum (D. fasciculamtum), Dictyostelium purpureum (D. purpureum), Physarum polycephalum (P. polycephalum), Polysphondyium pallidum (P. Pallidum), Acanthamoeba castellanii (A. castellanii), and Entamoeba histolytica (E. histolytica) [2]. The 12 calcium binding residues of EF2 (144DVNFDGRVSFLE155) are highlighted by black box. Three actin binding sites are underlined by purple line (ABS1, ABS2 and ABS3). (B) Cartoon diagram of the EF2. The Ca2þ-coordinating ligands are denoted as green sticks. Arrows point to the mutated residues (conversion to alanine) in the 12 calcium binding residues of EF2. The Ca2þ is represented as black sphere.

Fig. 1. ABP34 and its actin bundling activity. (A) Schematic diagrams and structure of ABP34 in surface (left) and cartoon representation (right) (PDB code: AX3N) [2]. Two putative EF-hands are colored in red (EF1) and blue (EF2). The two putative actinbinding sites are indicated by dashed black circles and the strong actin-binding site (ABS2) is represented by a dashed purple circle. (B) Electron micrographs taken from: F-actin alone (top row); mixture of WT ABP34 and F-actin in low calcium concentration (middle row); mixture of WT ABP34 and F-actin in high calcium solution (bottom row). Note that molecular appearances of actin-bundling were found in the mixture in low calcium solution. Scale bar represents 100 nm.

(Fig. 3B and C). However, the EF2 mutants did not exhibit calciumsensitive function at high calcium concentration, as expected from the fact that they could not bind Ca2þ (bottom row in Fig. 3C, c.f. mutants to WT). These results indicate that the actin bundling activity of ABP34 is activated when calcium ion is not bound to the protein via the EF2. In addition, when calcium ion is present in EF2, the bundling activity of ABP34 is inhibited. The amino acid sequence of ABP34 from Dictyostelium revealed homologues in some species of amoeba, but no sequence homology to other actin-binding proteins [2,11]. Sequence analysis indicated that ABP34 has conserved actin-binding residues (aa 216e244) in actin-binding site 2 (ABS2), suggesting that ABS2 is a strong actinbinding site (Fig. 2A) [2]. To further investigate the mechanisms involved in the EF2/calcium-regulated actin bundling activity of ABP34, we constructed several C-terminally truncated mutants: aa 1e279, 1e254, 1e193, 1e114, and investigated their bundling activity by negative staining EM analysis (Fig. 4A). At low calcium concentration, the 1e279 mutant (which lacks ABS3) exhibited actin bundling activity. Moreover, the bundling activity was inhibited when the calcium concentration was elevated, indicating

Fig. 3. Actin bundling activity of EF2 mutant proteins. (A) Expression test for DEF2 mutants. The bands representing the mutant proteins are highlighted by a red box (Left). (B) Actin bundling assay (co-sedimentation assay) for DEF2 mutants. Actin and ABP34 proteins are indicated by arrowheads. S indicates supernatant, while p represents pellet. (c) Negatively stained fields showing the actin bundling activity of WT and DEF2 mutants at low and high calcium concentrations. Scale bar represents 100 nm.

that ABS3 is not important for either actin binding or calciumregulated actin bundling activity. On the other hand, the truncated mutants 1e254, 1e193, and 1e114 did not activate actin bundling at either low or high calcium concentration (Fig. 4A), thus truncated mutants, 1e193, 1e165, and 1e114 exhibited calciumindependent bundling activity because their strong actin-binding site 2 (ABS2) was deleted. However, the bundling activity of truncated mutant 1e254 was inhibited at both low and high calcium concentration, even though

Please cite this article in press as: J.M. Chung, et al., The actin bundling activity of actin bundling protein 34 is inhibited by calcium binding to the EF2, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.07.122

4

J.M. Chung et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e5

the binding of calcium to the EF2 may rely on this conformational shift between the N-domain and C-domain. 3.3. Inter-domain interaction in actin bundling mechanism The overall structure of ABP34 resembles the shape of a bent arm due to inter-domain interactions between the N- and Cdomain [2]. These inter-domain interactions are of two types, electrostatic and hydrophobic interactions. The inter-domain contact is mainly sustained by hydrophobic interactions between a3 and h1 helices in the N-domain and the a12 helix in the C-domain (Fig. 4B). The hydrophobic core is formed by tryptophan 54 (W54) in the h1 helix, tyrosine 65 (Y65) and tryptophan 69 (W69) in the a3 helix, and tryptophan 277 and 278 (W277 and W278) in the a12 helix. Additional, electrostatic inter-domain interactions are formed between the residues in the domain interface. To illustrate the importance of the inter-domain interaction in the calcium-dependent actin bundling activity of ABP34, we mutated the residues participating in the hydrophobic interactions to alanine (hereafter referred to as Dhydrophobic interaction; W54A, W69A and W278A; Fig. 4B). All of the mutants were expressed and purified from the supernatants (Fig. 4C). Purified Dhydrophobic interaction mutants revealed that all of the mutants had impaired bundling activity at both low and high calcium concentrations (Fig. 4D). This is consistent with the results observed from the bundling assay of the a12-deleted mutant (1e254) (Fig. 4A). It is interesting to note that the inhibitory effect of disrupting the inter-domain interaction on the actin bundling activity of ABP34 was comparable to that of calcium regulation of the EF2 motif. In other words, binding of calcium to the EF2 motif may induces a conformational shift between the N-domain and Cdomain, thereby inhibiting the actin bundling activity of ABP34 protein. 4. Discussion

Fig. 4. Actin bundling assay for C-terminally truncated and Dhydrophobic interaction mutants of ABP34. (A) Negative staining EM micrographs showing bundling activity of N-terminally truncated mutants at low (top) and high calcium concentration (bottom). Cartoon structures above each micrograph show the deleted region in each mutant. Deletion of the a12 helix in the C-domain is indicated by a dashed red circle. Scale bar represents 100 nm. (B) Diagrams showing side chains involved in hydrophobic interactions. (C) Expression results of Dhydrophobic interaction mutants with sarkosyl-detegent treatment. Migration band relevant to ABP34 mutants are indicated by arrowheads: S, Supernatant; M, Size Marker; P, Pellet. (D) Negatively stained fields showing general appearances of actin bundling activity of the mutants at low and high CaCl2 concentrations. Scale bar indicates 100 nm.

it contained ABS2 and both EF-hand motifs. In the truncated mutant, only a12 helix (Fig. 4B; 271SSAGAIWWMNRDLEEKKKRY290) in the C-domain was deleted (indicated by dashed red circle on Fig. 4A). This result indicates that the deletion of only the a12 helix in the C-domain had a similar effect on actin bundling activity as binding of calcium ion by EF2 (i.e., under high calcium concentration). A previous study suggested that the a12 helix in the Cdomain contributes to inter-domain interactions in combination with the h1 and a3 helices of the N-domain (Fig. 4B), resulting in the formation of a bent overall structure [2]. Thus, disruption of protein stability by removing the key helix involved in the interdomain interaction can cause the inhibition of actin bundling. In addition, the a12 helix not only stabilizes the protein structure, but is also important in the bundling activity of the protein. The mechanism underlying the inhibition of actin bundling of ABP34 by

ABP34 is a calcium concentration-dependent actin bundling protein in D. discoideum. In this study, the actin bundling properties of recombinant WT ABP34 and various mutant proteins were determined by co-sedimentation assay and TEM. ABP34 protein activates the actin bundling in the absence of calcium, or in the presence of a low concentration of calcium (1  108 M), but its bundling activity is inhibited when the calcium concentration is elevated (1  104 M) [3,4,12]. Previous studies suggested that the N-domain of ABP34 contributes to calcium binding and contains an EF-hand motif, a conserved structure found in many calciumbinding proteins [8,13]. ABP34 protein has two putative EF-hand motifs, the non-functional EF1 and the functional EF2 [2,14]. The latter consists of a 12-residue calcium-binding loop, which starts with an Asp and ends with Glu (144DVNFDGRVSFLE155) [2]. In almost all cases, the EF-hand occupies in pairs, which creates a distance of 11 Å between the two calcium ions [15]. ABP34 protein contains an EF-hand pair formed by a four-helix bundle (a3, a4, a6, and a7), connected by a hydrophobic interaction. Further stabilization of the EF-pair is provided by two short anti-parallel b-sheets (b1 and b2) [2]. However, the intervening loop of EF1 has lost the proper composition of calcium-coordinating ligands. Consequently, EF1 of ABP34 is not a functional calcium-binding motif, in accordance with the previous findings [2]. As shown in Fig. 3, the actin bundling activity of DEF2 mutant proteins was similar to that of WT ABP34 at low calcium concentrations. By contrast, in DEF2 mutant proteins, the calcium-sensitive inhibition of actin bundling activity was not observed at high calcium concentration. This is consistent

Please cite this article in press as: J.M. Chung, et al., The actin bundling activity of actin bundling protein 34 is inhibited by calcium binding to the EF2, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.07.122

J.M. Chung et al. / Biochemical and Biophysical Research Communications xxx (2018) 1e5

with a previous study suggesting that the introduction of mutations in both EF-hands abolishes the actin bundling inhibitory effect at elevated calcium concentrations [10]. These results indicate that EF2 of ABP 34 plays a crucial role in calcium ion binding, thereby inhibiting the actin bundling activity of the protein. Disruption of the regulatory effect of calcium ion on the actin bundling activity of ABP34 was also observed in the 1e254 mutant of ABP34, which lacks the a12 helix (Fig. 4A). Although the truncated mutant had both functional calcium-binding motif (EF-hands) and strong actinbinding site (ABS2), the mutant did not exhibit actin bundling activity at any calcium concentration. It implies that the a12 helix may be involved in the conformational changes induced by the calcium binding on the EF2 motif and this may provide a plausible explanation for the effects of EF2-calcium binding on the WT ABP34 protein. To understand this outcome, it is necessary to explain the inter-domain interaction between the N-domain and C-domain. The overall structure of ABP34 is stabilized by two different types of inter-domain interaction, electrostatic and hydrophobic. The hydrophobic interaction is the main force that maintains the interdomain contacts of ABP34. The a12 helix in the C-domain is involved in the hydrophobic interaction between the N- and Cdomains. Consequently, disruption of this inter-domain interaction inhibits the actin bundling activity of ABP34. This conclusion was supported by the results of actin bundling assays using mutants in the residues that participate in the hydrophobic interaction (Fig. 4B and D). The hydrophobic force is primarily the result of interactions between W54 in h1 and Y65 and W69 in a3 with W277 and W278 in a12 [2]. When the interaction was diminished by introducing mutations in these residues, actin bundling activity was also inhibited at both low and high calcium concentrations (Fig. 4D). The residues in the EF-hand pair of ABP34 are involved in the N-domain and directly contribute to the interaction with the C-domain [8]. Moreover, the study suggested that the intramolecular interaction between the domains maintains the N-terminal region in close proximity to ABS2 to regulate the interaction of ABP34 with F-actin [8]. Taken together, these findings indicate that the presence of calcium ion in EF2 affects the stability and conformation of the EFhand pair, leading to the disruption of the interaction between the N- and C-domains, eventually resulting in the inhibition of actin bundling activity of ABP34. A previous study suggested that ABP34 inhibits depolymerization at both ends of actin filaments, even under conditions in which it cannot induce actin bundle formation [16]. This implies that ABP34 affects actin filament dynamics within cells, possibly by causing local accumulation of actin filaments by inhibiting actin depolymerization. Actin accumulation is often observed in the brains of patients who suffer from neurodegenerative disorders such as Alzheimer's and Parkinson's diseases [17,18]. Moreover, calcium dysregulation has been consistently implicated in Alzheimer's disease [19,20]. Thus, the findings of this study provide clues about the pathogenesis of actin filament-associated neurodegenerative diseases that are regulated by calcium concentration. Conflicts of interest No potential conflict of interest relevant to this article was reported.

5

Acknowledgements Authors thank Dr. S Lee for providing plasmid with assisting data analysis. This study was supported by the Basic Science Research Program through the National Research Foundation of Korea funded by the Ministry of Science, ICT & Future Planning (NRF-2015R1C1A1A01053611 and 2018R1D1A1B07045580 to H. S Jung, and NRF-2017M3A9G7072417 to D. Jeoung).

References [1] M. Fechheimer, R. Furukawa, A 27,000-D core of the Dictyostelium 34,000-D protein retains Ca(2þ)-regulated actin cross-linking but lacks bundling activity, J. Cell Biol. 120 (1993) 1169e1176. [2] M.K. Kim, J.H. Kim, J.S. Kim, S.O. Kang, Structure of the 34 kDa F-actinbundling protein ABP34 from Dictyostelium discoideum, Acta Crystallogr D Biol Crystallogr 71 (2015) 1835e1849. [3] M. Fechheimer, The Dictyostelium discoideum 30,000-dalton protein is an actin filament-bundling protein that is selectively present in filopodia, J. Cell Biol. 104 (1987) 1539e1551. [4] M. Fechheimer, D.L. Taylor, Isolation and characterization of a 30,000-dalton calcium-sensitive actin cross-linking protein from Dictyostelium discoideum, J. Biol. Chem. 259 (1984) 4514e4520. [5] M. Fechheimer, D. Murdock, M. Carney, C.V. Glover, Isolation and sequencing of cDNA clones encoding the Dictyostelium discoideum 30,000-dalton actinbundling protein, J. Biol. Chem. 266 (1991) 2883e2889. [6] R.W. Lim, M. Fechheimer, Overexpression, purification, and characterization of recombinant Dictyostelium discoideum calcium-regulated 34,000-dalton Factin bundling protein from Escherichia coli, Protein Expr. Purif. 9 (1997) 182e190. [7] R.W. Lim, R. Furukawa, S. Eagle, R.C. Cartwright, M. Fechheimer, Three distinct F-actin binding sites in the Dictyostelium discoideum 34,000 dalton actin bundling protein, Biochemistry 38 (1999) 800e812. [8] R.W. Lim, R. Furukawa, M. Fechheimer, Evidence of intramolecular regulation of the Dictyostelium discoideum 34 000 Da F-actin-bundling protein, Biochemistry 38 (1999) 16323e16332. [9] J.M. Chung, S. Lee, H.S. Jung, Effective non-denaturing purification method for improving the solubility of recombinant actin-binding proteins produced by bacterial expression, Protein Expr. Purif. 133 (2017) 193e198. [10] R. Furukawa, A. Maselli, S.A. Thomson, R.W. Lim, J.V. Stokes, M. Fechheimer, Calcium regulation of actin crosslinking is important for function of the actin cytoskeleton in dictyostelium, J. Cell Sci. 116 (2003) 187e196. [11] R. Furukawa, M. Fechheimer, Differential localization of alpha-actinin and the 30 kD actin-bundling protein in the cleavage furrow, phagocytic cup, and contractile vacuole of Dictyostelium discoideum, Cell Motil Cytoskeleton 29 (1994) 46e56. [12] M. Fechheimer, R. Furukawa, Preparation of 30,000-Da actin cross-linking protein from Dictyostelium discoideum, Meth. Enzymol. 196 (1991) 84e91. [13] H. Kawasaki, R.H. Kretsinger, Calcium-binding proteins. 1: EF-hands, Protein Profile 1 (1994) 343e517. [14] S. Vijay-Kumar, W.J. Cook, Structure of a sarcoplasmic calcium-binding protein from Nereis diversicolor refined at 2.0 A resolution, J. Mol. Biol. 224 (1992) 413e426. [15] R.R. Biekofsky, J. Feeney, Cooperative cyclic interactions involved in metal binding to pairs of sites in EF-hand proteins, FEBS Lett. 439 (1998) 101e106. [16] S.H. Zigmond, R. Furukawa, M. Fechheimer, Inhibition of actin filament depolymerization by the Dictyostelium 30,000-D actin-bundling protein, J. Cell Biol. 119 (1992) 559e567. [17] J.F. Reyes, K. Stone, J. Ramos, A. Maselli, Formation of Hirano bodies after inducible expression of a modified form of an actin-cross-linking protein, Eukaryot. Cell 8 (2009) 852e857. [18] M. Furgerson, J.K. Clark, J.D. Crystal, J.J. Wagner, M. Fechheimer, R. Furukawa, Hirano body expression impairs spatial working memory in a novel mouse model, Acta neuropathologica communications 2 (2014) 131. [19] N.K. Woods, J. Padmanabhan, Neuronal calcium signaling and Alzheimer's disease, Adv. Exp. Med. Biol. 740 (2012) 1193e1217. [20] M.P. Mattson, R.E. Rydel, I. Lieberburg, V.L. Smith-Swintosky, Altered calcium signaling and neuronal injury: stroke and Alzheimer's disease as examples, Ann. N. Y. Acad. Sci. 679 (1993) 1e21.

Please cite this article in press as: J.M. Chung, et al., The actin bundling activity of actin bundling protein 34 is inhibited by calcium binding to the EF2, Biochemical and Biophysical Research Communications (2018), https://doi.org/10.1016/j.bbrc.2018.07.122