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

J. Biol. Chem., Vol. 277, Issue 46, 44140-44146, November 15, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/46/44140    most recent M202596200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sasagawa, K. Articles by Hatano, M. Search for Related Content PubMed PubMed Citation Articles by Sasagawa, K. Articles by Hatano, M. Social Bookmarking                      What's this?

Identification of Nd1, a Novel Murine Kelch Family Protein, Involved in Stabilization of Actin Filaments^*

Kazushi Sasagawa^§, Yuji Matsudo, Myenmo Kang, Lisa Fujimura, Yoshinori Iitsuka, Seiji Okada, Takenori Ochiai^§, Takeshi Tokuhisa, and Masahiko Hatano^¶

From the  Department of Developmental Genetics (H2) and the ^§ Department of Academic Surgery (M9), Graduate School of Medicine, Chiba University, Chiba 260-8670, Japan

Received for publication, March 18, 2002, and in revised form, August 21, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We isolated Nd1, a novel kelch family gene that encodes two forms of proteins, Nd1-L and Nd1-S. Nd1-L contains a BTB/POZ domain in its N terminus and six kelch repeats in the C terminus. Nd1-S has the BTB/POZ domain but lacks the six kelch repeats. Nd1-L but not Nd1-S mRNA is detected ubiquitously in normal mouse tissues. Nd1-L and Nd1-S proteins can form a dimer through the BTB/POZ domain. Nd1-L colocalizes with actin filaments detected using a confocal microscope, and its kelch repeats bind to them in vitro. Overexpression of Nd1-L in NIH3T3 cells delayed cell growth by affecting the transition of cytokinesis. Furthermore, the overexpression prevented NIH3T3 cells from cell death induced by actin destabilization but not by microtubule dysfunction. These data suggest that Nd1-L functions as a stabilizer of actin filaments as an actin-binding protein and may play a role in the dynamic organization of the actin cytoskeleton.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Dynamics of the actin cytoskeleton regulate a wide range of structures and functions of eukaryotic cells. Purified actin exists as a monomer and spontaneously assembles into actin filaments in the presence of salt and ATP. In the cortex of cells, actin molecules continually polymerize and depolymerize to generate cell surface protrusions such as lamellipodia and microspikes. Polymerization is regulated by extracellular signals binding to cell surface receptors that act through G proteins and the small GTPases Rac and Rho (1). Thus, the actin-based cytoskeleton is responsible for a wide range of cellular functions such as the generation and maintenance of cell polarity and cellular motility. These properties of actin filaments depend on a large retinue of actin-binding proteins that bind to actin filaments and modulate their properties and functions. Actin-binding proteins cross-link actin filaments into loose gels, bind them into stiff bundles, attach them to the plasma membrane, or forcibly move them relative to one another. Sets of actin-binding proteins are thought to act cooperatively in generating events on the cell surface, including cytokinesis, phagocytosis, and cell locomotion.

The kelch family protein, defined by an ~50 amino acid motif repeat (2, 3), is one of the actin-binding proteins. The kelch motif was originally discovered as a 6-fold tandem element in the sequence of the Drosophila kelch ORFI protein. The kelch repeats motif appears to function as a novel actin binding domain (4). More than 20 proteins containing kelch repeats have been identified in a diverse set of organisms, including virus, yeast, Caenorhabditis elegans, Drosophila, and mammals (5). The predicted structure of -sheet repeat motifs may have functional significance in binding actin, protein folding, or protein-protein interactions. In addition to the kelch repeats motif, most kelch family proteins also possess a 120-amino acid motif referred to as the BTB/POZ domain at the extreme N terminus of the protein (6, 7). The BTB/POZ domain was originally identified in a group of transcription factors such as Bcl6 (8-10), PLZF (11), Drosophila Tramtrack (12), and bric-a brac proteins (13). It has been reported that BTB/POZ domains mediate homo- and heterodimerization in vitro and the formation of multimeric complexes in vivo (6, 7, 14).

In Drosophila, the kelch protein is a component of ring canal and cross-links actin bundles. Loss of the kelch protein causes disorganization of ring canal formation resulting in infertility (15). Disruption of kelch-containing protein, tea1, in yeast is lethal with a defect in cell division (16). While growing numbers of kelch family member proteins have been identified in mammals the physiological roles of kelch family proteins are not fully understood. We identified a novel murine kelch family gene, Nd1. The long form of Nd1 (Nd1-L) contains a BTB/POZ domain in its N terminus and six kelch repeats in the C terminus. Nd1-L protein localizes in the cytoskeleton and associates with actin filaments. The overexpression of Nd1-L in NIH3T3 cells prevented cell death induced by treatment with cytochalasin D. The role of Nd1 in the actin cytoskeletal organization is discussed.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cloning of the Nd1 cDNA-- Representational difference analysis (RDA)¹ for cDNA was done as described (17, 18). Briefly, cDNA from enteric neurons was digested with Dpnll and ligated to the R-Bgl-12/24 adapters. Subtractive hybridization was done using enteric neuron cDNA from Ncx-deficient mice (19) as testers and cDNA from wild type mice as drivers. After three rounds of subtractive hybridization, cDNA fragments were subcloned into a pGEM-T vector (Promega, Madison, WI) and sequenced using an automatic DNA sequencer (ALF Express; Amersham Biosciences). One novel cDNA fragment designated Nd1 was chosen for further characterization. This fragment was used as a probe to screen a murine heart cDNA library (Stratagene, La Jolla, CA). The 5'- and the 3'- most sides of the cDNA were cloned with poly(A)+ mRNAs from the mouse heart with the 5'-RACE and 3'-RACE system for rapid amplification of cDNA ends (Invitrogen), respectively. Those cDNA fragments were reconstructed into pGEM-4Z (Promega) to obtain a full-length of the open reading frame.

Northern Blot Analysis-- Total RNAs were extracted from adult mouse tissues using the TRIzol reagent (Invitrogen). A 625-bp EcoRI-XbaI DNA fragment (+30 to +655) of the Nd1 cDNA, including a BTB/POZ domain, was used as a POZ probe. The fragment was subcloned into pGEM-4Z and labeled with the digoxigenin DNA labeling mixture (Roche Molecular Biochemicals) by PCR with primers T7 and SP6. The filter was hybridized overnight with the digoxigenin-labeled probe at 55 °C in the presence of 50% formamide then washed at 58 °C in 0.1× SSC, 0.1% SDS.

Construction of FLAG or HA Epitope-tagged Nd1 Expression Plasmids-- Flag-Nd1 and HA-Nd1 expression plasmids (pCR-2FLAG-Nd1-L, pCR-2FLAG-Nd1-S, and pCR-2HA-Nd1-L) were constructed by PCR amplifying Nd1-L or Nd1-S cDNA fragments containing an open reading frame and ligating them to the EcoRI sites of pCR-2FLAG and pCR-2HA (20). To construct cadmium-inducible expression plasmids (pSMT-2FLAG-Nd1-L and pSMT-2FLAG-Nd1-S), FLAG-tagged Nd1-L and Nd1-S fragments were isolated from the pCR-2FLAG-Nd1-L and pCR-2FLAG-Nd1-S, respectively, then subcloned into sheep metallothionein (SMT) promoter expression plasmids (21). Primer sequences used for various fragments are as follows: Nd1 5' primer; 5'-GCCGGAATTCATGATTCCCAATGGAT-3', Nd1 3' primer; 5'-CGCCGAATTCTTAAAACTGGAAAATC-3'.

Glutathione S-Transferase (GST) Fusion Protein-- A HindIII fragment encoding the Nd1-L kelch domain was subcloned into pGEX-4T-2 in frame with the GST reading frame to generate the pGEX-Nd1-kelch. The plasmid was transformed into Escherichia coli JM109 cells. Following 3 h of incubation with 0.1 mM isopropyl--D-thiogalactopyranoside, the bacteria were collected by centrifugation, washed twice with cold PBS, resuspended in PBS, and lysed by sonication. The GST fusion protein was isolated from the supernatant with glutathione-Sepharose beads (Amersham Biosciences), as described in the manufacturer's protocol.

Cell Culture and Transfection-- 293 cells were maintained in Dulbecco's modified Eagle's medium (Sigma) supplemented with 10% fetal bovine serum (Sigma) at 37 °C. NIH3T3 cells were cultured in RPMI 1640 (Sigma) supplemented with 10% fetal bovine serum. For transient transfection, 293 cells (2-5 × 10⁶) were transfected with 2 µg of pCR-2FLAG-Nd1-L or pCR-2FLAG-Nd1-S expression plasmids using LipofectAMINE (Invitrogen) and harvested after 18-24 h. To establish stable transfectants, NIH3T3 cells were transfected with 10 µg of pSMT-2FLAG-Nd1-L or pSMT-2FLAG-Nd1-S along with 1 µg of pSV2Neo using electroporation and selected with 400 µg/ml G418 (Invitrogen). For induction of the SMT promoter, cells were cultured in the presence of 5 µM CdCl₂. In some experiments, 1 µM of cytochalasin D (Sigma), 2 µM taxol (Sigma) or 0.8 µg/ml nocodazole (Sigma) was added to the culture.

Immunoprecipitation and Immunoblot Analysis-- 293 cells transiently transfected with FLAG-tagged Nd1-L or FLAG-tagged TIAP (22) were lysed at 4 °C in 1.0 ml of RIPA buffer (25 mM Tris-HCl, pH 7.5, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 150 mM NaCl, 1 mM phenylmethylsulfonyl fluoride, 10 µl/ml leupeptin, and 1 mM Na₃VO₄) (23). Lysates were clarified by centrifugation, and the supernatants were shaken at 4 °C for 2 h with a rabbit anti-actin antibody (Sigma) or with a mouse anti-FLAG M2 antibody (Sigma). Immunoprecipitates were obtained with 30 µl of protein G-Sepharose 4FF (Amersham Biosciences) for 1 h. The samples were centrifuged for 1 min at 4 °C, the pellets were washed four times with 1 ml of RIPA buffer without sodium deoxycholate and SDS then resolved in 20 µl of 2 × SDS sample buffer, boiled for 5 min, and fractionated by 10% SDS-PAGE. SDS-PAGE followed by electroblotting onto Immobilon-P filters (Millipore, Bedford, MA) using the minigel system (Bio-Rad Laboratories, Hercules, CA). The filters were blocked overnight with Block Ace (Yukijirushi, Sapporo, Japan) at 4 °C. Blots were washed and incubated in PBS-T (0.1% Tween 20 in PBS) with a mouse anti-FLAG M2 monoclonal antibody or an affinity-purified rabbit anti-actin antibody for 1 h then washed and incubated with horseradish peroxidase-conjugated anti-mouse or anti-rabbit IgG (Amersham Biosciences) for 1 h, respectively. Immunoreactive bands were visualized using ECL detection reagents (Amersham Biosciences).

To examine the dimerization through the BTB/POZ domain, 293 cells were transiently transfected with HA-tagged Nd1-L in combination with FLAG-tagged Nd1-S or FLAG-tagged TIAP. These cells were lysed in RIPA buffer and immunoprecipitated with an anti-FLAG M2 antibody. Immunoprecipitates were blotted with an anti-HA antibody (Roche Molecular Biochemicals), and immunoreactive bands were visualized using ECL reagents.

Actin Binding Assay-- F-actin was preassembled from purified rabbit skeletal muscle actin (2 µM) in G-buffer (5 mM Tris-HCl, pH 8.0, 0.1 mM ATP, 0.2 mM CaCl₂, 0.02% (v/v) NaN₃) by the addition of 0.02 volumes of 50× initiation buffer (100 mM MgCl₂, 50 mM ATP, 2.5 M KCl) (Cytoskeleton Inc., Denver, CO) and incubated for 30 min at 25 °C. For co-sedimentation, mixtures of the GST-Nd1-kelch and F-actin in F-buffer (20 mM Tris-HCl, pH 8.0, 100 mM KCl, 1 mM MgCl₂, 0.1 mM ATP, 0.1 mM CaCl₂, 0.5 mM dithiothreitol, 0.01% (v/v) NaN₃) were centrifuged (100,000 × g, 25 min, 4 °C), and the pellets were analyzed using Coomassie Blue-stained SDS-PAGE.

Immunohistochemistry-- Subcellular localization of the Nd1 protein was examined in NIH3T3 cells or 293 cells transfected with FLAG-tagged Nd1-L. Cells were cultured in chamber slides, fixed for 30 min in cold acetone then blocked with 5% bovine serum albumin. Transfectants were stained with a mouse anti-FLAG M2 monoclonal antibody (1:200 dilution in PBS containing 0.1% Triton X-100) followed by FITC-conjugated goat anti-mouse IgG (1:200) (Cappel, Aurora, OH) then rhodamine-conjugated phalloidin (Sigma). Immunostained cells were visualized on a Zeiss Axioskop20 or a Zeiss CLSM410 confocal laser scanning microscope equipped with 63×/NA 1.4 objective lens with an argon/krypton laser (Carl Zeiss Co., Ltd, Gottingen, Germany). Images were acquired and analyzed using a Scanalytics Cellscan system with exhaustive photon reassignment deconvolution image enhancement (24).

Cell Cycle Analysis-- One million cells were fixed with 70% ethanol for 18 h and incubated in 1 ml of propidium iodide staining solution (0.05 mg/ml propidium iodide, 0.02 mg/ml ribonuclease A, 1% glucose in PBS) for 30 min. Fluorescence from propidium iodide-nuclear DNA complexes was analyzed using a FACSCalibur (Becton Dickinson, San Jose, CA). Proportions of cells in G₁, S, and G₂/M phase of the cell cycle were analyzed using ModFit LT software (Verity Software House, Inc., Topsham, ME).

Statistical Analysis-- Data were analyzed using a single-tailed Student's t test. Data are given as mean ± S.D.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Molecular Cloning of the Nd1 cDNA-- During a process of a RDA method using enteric neuron cDNAs from wild type and Ncx-deficient mice (19, 25), we obtained several cDNA fragments. One clone of 103 bp, representing a novel gene, was selected and further characterized. The 103-bp cDNA fragment was used as a probe to screen a murine heart cDNA library. Several overlapping cDNA clones were sequenced and aligned. Two related cDNAs were obtained. One of them contained a long open reading frame of 1926 nucleotides that encoded the entire protein of 642 amino acids with a predicted molecular mass of 71.9 kDa (Nd1-L; Ncx downstream gene 1 long form) (Fig. 1A). Another cDNA contained an open reading frame of 663 nucleotides that encoded a 221-amino acid protein with a predicted molecular mass of 24.7 kDa (Nd1-S; Nd1 short form) (Fig. 1B). Nucleotide sequences of Nd1-L and Nd1-S are identical from 360 to +661. From nucleotide +662 the sequences diversify and an open reading frame continues up to +1936 in Nd1-L, whereas a stop codon appears at +664 in Nd1-S. A part of the 3'-untranslated sequence of Nd1-S (+1623 to 1727, Fig. 1B) was identified in the open reading frame of the Nd1-L sequence (+661 to 765). Sequence analysis of cDNA and genomic DNA of the Nd1-L and Nd1-S revealed that these two forms of mRNA were created by alternative splicing (26).

View larger version (74K): [in this window] [in a new window]   Fig. 1.   Nucleotide and amino acid sequences of mouse Nd1 cDNA. A, sequence of the Nd1-L cDNA. The open reading frame extended from nucleotide 1 to 1926 and encoded a protein of 642 amino acids. The BTB/POZ domain is underlined. Hatched boxes indicate the six kelch repeats motif. The polyadenylation signal sequence between nucleotides 2798 and 2804 is underlined. B, sequence of the Nd1-S cDNA. The open reading frame extended from nucleotide 1 to 663 and encoded a protein of 221 amino acids. The single dashed line nucleotides (1623-1727) were the same as nucleotides 661-765 of the Nd1-L cDNA. The double dashed line nucleotides (715-818) were a cDNA fragment probe isolated from RDA.

The cDNA and the deduced amino acid sequence of Nd1 were compared with sequences in the GenBank^TM and Swiss Prot databases using BLASTN and BLAST. Nd1-L contains two major domains, a BTB/POZ domain in its N terminus and six repeats of the kelch domain in its C terminus. In contrast, Nd1-S lacks the kelch repeats and has only the BTB/POZ domain. The BTB/POZ domain has 30-37% identity to that of other kelch family proteins such as mouse Keap1 (27), human ENC-1 (28), human Mayven (29), and Drosophila kelch protein (2) (Fig. 2A). It has also 30-38% identity to that of BTB/POZ-zinc finger transcription factors such as Bcl6 (8-10), BAZF (30), and PLZF (11). The kelch repeats motif of Nd1-L has 29-38% identity to other kelch member proteins (Fig. 2B).

View larger version (58K): [in this window] [in a new window]   Fig. 2.   Sequence alignment of kelch family proteins. A, The BTB/POZ domains of six kelch family proteins, Nd1, ENC-1, Mayven, Keap1, IPP, and Drosophila kelch. Shaded boxes indicate the consensus sequence of BTB/POZ proteins (6). B, Kelch repeats domains of six kelch family proteins. The six kelch repeats domain (Krep1-6) of Nd1-L was compared with that of other kelch family proteins. Shaded boxes indicate identical residues.

Expression of Nd1-- Expression of Nd1 mRNA was examined in normal mouse tissues and various cell lines using Northern blot analysis with the BTB/POZ domain probe. A 3.2-kb band, which corresponds to the Nd1-L mRNA, was detected in all tissues examined and most abundantly in the heart, kidney, and small and large intestine (Fig. 3A). The expression was also observed in murine lymphoid (WEHI-231), macrophage (RAW-264), neuroblastoma (C1300), and fibroblast (L and NIH3T3) cell lines (Fig. 3B), thereby indicating that the expression of Nd1-L is ubiquitous. A 2.4-kb band, which corresponds to Nd1-S mRNA, was also detected in the heart, kidney, testis, ovary, and small and large intestine but not in the skeletal muscle, liver, and lung. In cell lines, expression of Nd1-S was detected in WEHI-231, C1300, and weakly in L and NIH3T3 cells but not in RAW-264 cells (Fig. 3B). Thus, expression of Nd1-S is restricted compared with that of Nd1-L.

View larger version (48K): [in this window] [in a new window]   Fig. 3.   Northern blot analysis of Nd1-L and Nd1-S mRNA expression. A, the expression was analyzed in total RNA (20 µg) from indicated tissues of adult mice. 3.2-kb Nd1-L and 2.4-kb Nd1-S mRNAs are indicated by arrows. The lower part of the panel shows 28 S and 18 S rRNA from the ethidium bromide-stained gel used as the loading control. B, the expression was analyzed in total RNA (20 µg) from various cell lines. Expression of Nd1-L and Nd1-S detected with the BTB/POZ probe are indicated by arrows. The lower part of the panel shows the loading control detected with the G3PDH probe.

Colocalization of Nd1-L with Actin Cytoskeleton-- Drosophila kelch protein forms a homodimer through its BTB/POZ domain and associates with actin through kelch repeats (5). First, we examined the dimerization of Nd1 proteins through the BTB/POZ domain using Nd1-L and Nd1-S. HA-tagged Nd1-L expression plasmids were cotransfected with either FLAG-tagged Nd1-S or FLAG-tagged TIAP (22) into 293 cells. Immunoprecipitation was done using an anti-FLAG antibody followed by immunoblotting with an anti-HA antibody. HA-tagged Nd1-L was specifically coimmunoprecipitated with FLAG-tagged Nd1-S but not with FLAG-tagged TIAP in 293 cells (Fig. 4A). Yeast two-hybrid analysis also confirmed the dimerization of Nd1 proteins through the BTB/POZ domain (data not shown). Second, we examined the association of Nd1-L with the actin cytoskeleton. We transiently transfected FLAG-tagged Nd1-L expression plasmids or FLAG-tagged TIAP plasmids into 293 cells. When immunoprecipitation was done using an anti-actin antibody followed by immunoblotting with an anti-FLAG antibody, FLAG-tagged Nd1-L but not FLAG-tagged TIAP was detected (Fig. 4B, upper panel). Furthermore, 42 kDa actin was also coimmunoprecipitated with FLAG-tagged Nd1-L but not with FLAG-tagged TIAP (Fig. 4B, middle panel). To confirm a direct interaction between actin and the kelch repeats of Nd1-L, a co-sedimentation assay was done. Following incubation with F-actin, GST-Nd1-kelch fusion protein partitioned predominantly to the pellet fraction with F-actin, whereas GST protein or GST-Nd1-kelch fusion protein without F-actin remained in the soluble fraction (Fig. 4C). Thus, Nd1-L associates with F-actin through kelch repeats.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Molecules associated with Nd1-L. A, in vivo association between Nd1-L and Nd1-S. 293 cells were transfected with FLAG- or HA-tagged Nd1 expression vectors. Lysates from transfected 293 cells were immunoprecipitated with an anti-FLAG antibody and blotted with an anti-HA antibody. HA-tagged Nd1-L (~80 kDa) was coimmunoprecipitated with FLAG-tagged Nd1-S. 293 cells transfected with HA-tagged Nd1-L and FLAG-tagged Nd1-S (lane 1), HA-tagged Nd1-L and FLAG-tagged TIAP (lane 2), and HA-tagged Nd1-L alone (lane 3). B, in vivo association of Nd1 with actin. 293 cells were transfected with FLAG-tagged Nd1-L expression vectors. Upper panel, lysates from transfected 293 cells were immunoprecipitated with an anti-actin antibody and blotted using an anti-FLAG antibody. FLAG-tagged Nd1-L (~80 kDa) was detected in lane 1 (arrow). Middle panel, lysates from transfected 293 cells were immunoprecipitated with an anti-FLAG antibody and blotted using an anti-actin antibody. Actin was coimmunoprecipitated with FLAG-tagged Nd1-L. Lower panel, lysates from transfected cells were separated by SDS-PAGE and blotted using anti-kelch polyclonal antibodies. 293 cells transfected with FLAG-tagged Nd1-L (lane 1), FLAG-tagged TIAP (lane 2), and pcDNA (lane 3). C, in vitro association of the Nd1-kelch domain with F-actin. Co-sedimentation assays of F-actin incubated with GST protein, GST-Nd1-kelch protein, or actinin. Pellets were analyzed using Coomassie Blue-stained SDS-PAGE. GST protein without actin (lane 1), GST protein with actin (lane 2), GST-Nd1-kelch protein without actin (lane 3), GST-Nd1-kelch protein with actin (lane 4), and -actinin with actin as a positive control (lane 5). D, subcellular localization of Nd1-L in a transfected NIH3T3 cell. Cell scan micrographs of NIH3T3 cells stably transfected with FLAG-tagged Nd1-L expression vectors. Cells were stained with an anti-FLAG antibody and rhodamine-phalloidin. Left panel, FLAG-tagged Nd1-L in a transfected cell was detected with an anti-FLAG antibody followed by FITC-conjugated goat antibodies to mouse Ig (green). Middle panel, in the same cell as in A, actin fibers were visualized with rhodamine-conjugated phalloidin (red). Right panel, merged image of Nd1-L (green) and actin (red) yielding yellow. Scale bar, 10 µm.

To detect the subcellular localization of Nd1-L, we did double immunofluorescent staining using NIH3T3 cells stably transfected with FLAG-tagged Nd1-L. TRITC-labeled phalloidin was used to visualize F-actin in context with FITC-labeled anti-FLAG antibody in transfected cells. The positive staining of FLAG-tagged Nd1-L was found throughout cytoplasm. Furthermore, extensive colocalization of Nd1-L with actin stress fibers was observed (Fig. 4D). These results were also confirmed in a C1300 neuroblastoma cell line (data not shown).

Effect of Nd1 Overexpression on Cell Growth-- To gain further insight into functions of Nd1 proteins, we overexpressed FLAG-tagged Nd1-L or Nd1-S in NIH3T3 cells. Two of each Nd1-L (clones 8 and 12) and Nd1-S (clones 1 and 7) stable transfectants were obtained. These transfectants expressed exogenous Nd1 proteins or mRNAs in the presence of CdCl₂ (data not shown). When those transfectants were seeded on a culture plate and viable cell numbers were counted, their cell growth was slower than that of parental cells (Fig. 5A). This was not due to massive cell death because we found no difference in cell viability between transfectants and parental cells by MTT assay and trypan blue staining (data not shown).

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Effect of Nd1 overexpression on cell growth. A, growth curves of NIH3T3 cells (open circles) and the transfectants (closed triangles; Nd1-L clone 8, closed squares; Nd1-L clone 12, open triangles; Nd1-S clone 1, open squares; Nd1-S clone 7). Values are means (± S.D.) of three independent experiments. B, cell cycle analysis of NIH3T3 cells and Nd1-L transfectants (clone 12). C, phase contrast images of Nd1-L transfectants (clone 12). Arrows indicate cells with two nuclei. Scale bar, 100 µm. D, confocal micrograph of a clone 12 cell with two nuclei. Merged image of Nd1-L (green) and actin (red) yielding yellow. Scale bar, 10 µm.

To further examine a mechanism of growth delay caused by overexpression of Nd1-L, cell cycle analysis was done. The percentage of cells in G₂/M phase of the cell cycle increased in Nd1-L transfectants (Clone 8: G₁; 69.8%, S; 13.4%, G₂/M; 16.8%. Clone 12: G₁; 65.8%, S; 16.6%, G₂/M; 18.6%) compared with that of parental NIH3T3 cells (NIH3T3: G₁; 80.4%, S; 14.1%, G₂/M; 5.5%) (Fig. 5B), thereby suggesting an abnormality in the G₂/M transition. Furthermore, we occasionally observed Nd1-L transfectants with two nuclei in the cytokinesis phase, using a phase contrast microscope (Fig. 5C). The number of transfectants in cytokinesis (clone 12: 21 ± 3/100 cells) was larger than that of parental cells (NIH3T3: 2 ± 1/100 cells). Confocal microscopic analysis demonstrated that Nd1-L and actin colocalized surrounding the divided nuclei (Fig. 5D). Thus, overexpression of Nd1-L makes cell growth delay by affecting the transition of cytokinesis in NIH3T3 cells. In case of Nd1-S transfectants, disorganized actin cytoskeleton was observed by phalloidin staining (data not shown) and polyploidy was not observed.

The Role of Nd1-L in Stress Induced by Actin Destabilization-- Since Nd1-L binds to the actin cytoskeleton, we asked if overexpression of Nd1-L would affect functions related to actin organization other than cytokinesis. Cytochalasin D destabilizes actin stress fibers and disrupts the network organization of actin fibers (31, 32). When parental NIH3T3 cells were cultured in the presence of cytochalasin D, many of these cells became round and detached from the plates within 1 day after the treatment (Fig. 6A). Disassembly and disruption of actin fibers were observed in the NIH3T3 cells stained with phalloidin using a confocal microscope. In contrast, actin stress fibers in Nd1-L transfectants were maintained 4 days after cytochalasin D treatment. When viable cell numbers were counted, the number decreased to 20% of the original within 4 days of culture in case of parental NIH3T3 cells (Fig. 6B). However, the viable cell number did not decrease in Nd1-L transfectants. When those cells were treated with a microtubule stabilizing (taxol) or a destabilizing (nocodazole) reagent, we saw no significant difference in cell viability between NIH3T3 cells and transfectants. These results indicate that overexpression of Nd1-L protects cells from stress induced by actin depolimerization but not by microtubule dysfunction.

View larger version (68K): [in this window] [in a new window]   Fig. 6.   Effect of Nd1-L overexpression on stress induced by cytochalasin D. A, phase contrast images and confocal micrographs of parental NIH3T3 cells and Nd1-L transfectants (clone 12). Cells were seeded on plates and cultured with or without 1 µM cytochalasin D for 3 days and stained with rhodamine-conjugated phalloidin to visualize actin fibers. scale bars in phase contrast and confocal images indicate 100 and 10 µm, respectively. B, cells (1 × 105) were cultured with 1 µM cytochalasin D, 2 µM taxol, or 0.8 µg/ml nocodazole. Viable cell numbers were counted after various time points, as indicated. Open circles, NIH3T3 cells; closed triangles, clone 8; closed squares, clone 12. Values are means ± S.D. of three independent experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We isolated Nd1, a novel gene that encodes protein with two major structural elements, the N-terminal BTB/POZ domain and the C-terminal six repeats of the kelch motif. The BTB/POZ domain is found in several types of transcription factors and other kelch family proteins. This domain has been proposed to mediate protein-protein interactions. Some kelch family members such as human Mayven (29), ENC-1 (28, 33) and Drosophila kelch proteins (4) form a homodimer through their BTB/POZ domain. Indeed, the BTB/POZ domain of Nd1 can bind to each other as demonstrated using immunoprecipitation (Fig. 4A) and in yeast two-hybrid experiments (data not shown), thereby suggesting the dimerization of Nd1.

Kelch motif was found in several actin-associated proteins such as Drosophila kelch (2), - and -scruin in Limulus polyhemus (34), murine intracisternal A-particle-promoted placenta protein (35), murine and human ENC-1 (28, 33), spe-26 of C. elegance (36) and human Mayven (29), and in a series of other proteins. However, some kelch family proteins do not interact with actin. For example, Tea1, Schizosaccharomyces pombe kelch family member protein, localizes at the tip of growing cells and is thought to interact with the end of microtubules (16). Furthermore, some members of kelch proteins localize in the nucleus rather than cytoskeleton. Human NRP/B is found in the nuclear matrix and interacts with Rb protein (23). Its function is assigned to regulate the cell cycle during neuronal differentiation. Recombination activating protein 2 (RAG2) is a kelch family protein and localizes in the nucleus. The kelch motif of RAG2 is necessary to interact with RAG1 and plays a critical role in V(D)J recombination of immunoglobulin and T cell receptor genes (37). We demonstrate here that kelch repeats of Nd1-L directly associate with F-actin. Double immunofluorescent staining data also indicates that Nd1-L colocalizes with actin filaments in cytoplasm in vivo. Thus, Nd1-L is an actin-associated protein.

To determine the function of Nd1 protein, we overexpressed Nd1-L or Nd1-S in NIH3T3 cells. Two independent Nd1-L or Nd1-S transfectants grow slowly compared with parental cells. Nd1-S transfectants become round, and their actin stress fibers are disorganized. Since Nd1-S has a BTB/POZ domain but lacks a kelch motif, it is likely that Nd1-S binds to Nd1-L by disrupting actin filament cross-linking and works as a dominant negative fashion. In case of Nd1-L transfectants, accumulation of cells with two nuclei was frequently observed in morphological examinations. These cells represent late mitotic cells undergoing cytokinesis. Furthermore, Nd1-L and actin localized surrounding the divided nuclei. Thus, these data suggest that cytokinesis is disturbed by overexpressed Nd1-L. Depolymerization and reorganization of actin in the cell cortex are required to allow for assembly of the contractile ring in the process of cytokinesis. It is possible that Nd1-L regulates actin filaments assembly during cytokinesis by binding to them. Excess amounts of Nd1-L may interfere with dynamic movement of actin filaments during cell division thus resulting in a delay of cell growth. It is noteworthy that many cardiac muscle cells in which Nd1-L is abundant have two nuclei. A physiological role of Nd1-L in cardiac muscles needs to be elucidated.

Overexpression of Nd1-L protected against the actin disorganization induced by cytochalasin D. Cytochalasin D binds to the end of actin filaments like capping proteins and inhibits polymerization resulting in nucleation and shortening of actin filaments (31, 32). Parental NIH3T3 cells became round and detached from a plate within 1 day after cytochalasin D treatment. Nd1-L transfectants attached to the culture plate and retained a pseudopodia-like shape even 4 days after treatment. Furthermore, actin stress fibers were detected in Nd1-L transfectants after treatment of cytochalasin D, whereas few stress fibers were observed in parental NIH3T3 cells after treatment (Fig. 6A). However, both parental cells and transfectants died at the same time after treatment with nocodazole or taxol, which affects microtubule organization. Thus, Nd1-L bundles and reinforces actin fibers by binding to actin and modulates their properties. Since Nd1-L transfectants hardly detach from a culture dish by forming a pseudopodia-like structure, it is also possible that Nd1-L is involved not only in cytokinesis but also in a focal adhesion complex with actin and actinin (38-40) or talin (41).

A role of the actin cytoskeleton has been implicated in basic cellular functions, including cell migration in normal embryogenesis, differentiation, and response to environmental stress (42, 43). Overexpression of -actinin, another actin-binding protein, causes reduction in cell motility (44). -Actinin-deficient NIH3T3 cells formed tumors upon injection into nude mice, which suggests that modulations in actin-binding protein expression can affect cell motility and tumorigenic property of cells (44, 45). Mutations in actin-associated proteins such as desmin and titin are responsible for human inherited cardiomyopathy (46-48). Mutations in kelch family proteins have also been noted in human genetic diseases. Point mutations in RAG2 were identified in some cases of human severe combined immunodeficiency or Omenn syndrome (37, 49, 50). A new member of kelch family protein, gigaxonin, is mutated in a human autosomal recessive neurodegenerative disorders named giant axonal neuropathy (51). Since Nd1-L plays an important role in cell division and stabilization of actin filaments and is ubiquitously expressed, Nd1-L may play a critical role in fundamental cellular functions such as cell division by coordinating with the dynamics of actin cytoskeleton as a housekeeping gene. Mice overexpressing Nd1 or deficient in Nd1 will be a valuable tool to elucidate a physiological function of Nd1 proteins in vivo and to identify human genetic diseases related with Nd1.

	ACKNOWLEDGEMENTS

We extend our thanks to H. Satake for technical assistance and to K. Ujiie for secretarial services. Language assistance was provided by M. Ohara.

	FOOTNOTES

^* This work was supported in part by a grant-in aid for scientific research from the Ministry of Education, Science, Technology, Sports and Culture of Japan.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 nucleotide sequences for the Nd1-L and Nd1-S have been deposited in GenBank^TM under the accession numbers AB055737 and AB055738, respectively.

^¶ To whom correspondence should be addressed: Dept. of Developmental Genetics (H2), Graduate School of Medicine, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba City 260-8670, Japan. Tel.: 81-43-226-2182; Fax: 81-43-226-2183; E-mail: hatano@med.m.chiba-u.ac.jp.

Published, JBC Papers in Press, September 3, 2002, DOI 10.1074/jbc.M202596200

	ABBREVIATIONS

The abbreviations used are: RDA, representational difference analysis; RACE, rapid amplification of cDNA ends; HA, hemagglutinin; SMT, sheep metallothionein; GST, glutathione S-transferase; PBS, phosphate-buffered saline; FITC, fluorescein isothiocyanate.

	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:

				E. E. Dunham, E. A. Stevens, E. Glover, and C. A. Bradfield The Aryl Hydrocarbon Receptor Signaling Pathway Is Modified through Interactions with a Kelch Protein Mol. Pharmacol., July 1, 2006; 70(1): 8 - 15. [Abstract] [Full Text] [PDF]

				R. Fujii and T. Takumi TLS facilitates transport of mRNA encoding an actin-stabilizing protein to dendritic spines J. Cell Sci., December 15, 2005; 118(24): 5755 - 5765. [Abstract] [Full Text] [PDF]

				L. Heiska and O. Carpen Src Phosphorylates Ezrin at Tyrosine 477 and Induces a Phosphospecific Association between Ezrin and a Kelch-Repeat Protein Family Member J. Biol. Chem., March 18, 2005; 280(11): 10244 - 10252. [Abstract] [Full Text] [PDF]

				L. Fujimura, Y. Matsudo, M. Kang, Y. Takamori, T. Tokuhisa, and M. Hatano Protective role of Nd1 in doxorubicin-induced cardiotoxicity Cardiovasc Res, November 1, 2004; 64(2): 315 - 321. [Abstract] [Full Text] [PDF]

				S. Kaur, O. Norkina, D. Ziemer, L. C. Samuelson, and R. C. De Lisle Acidic duodenal pH alters gene expression in the cystic fibrosis mouse pancreas Am J Physiol Gastrointest Liver Physiol, August 1, 2004; 287(2): G480 - G490. [Abstract] [Full Text] [PDF]

				T. Hara, H. Ishida, R. Raziuddin, S. Dorkhom, K. Kamijo, and T. Miki Novel Kelch-like Protein, KLEIP, Is Involved in Actin Assembly at Cell-Cell Contact Sites of Madin-Darby Canine Kidney Cells Mol. Biol. Cell, March 1, 2004; 15(3): 1172 - 1184. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/46/44140    most recent M202596200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sasagawa, K. Articles by Hatano, M. Search for Related Content PubMed PubMed Citation Articles by Sasagawa, K. Articles by Hatano, M. Social Bookmarking                      What's this?