Identification of substrates for transglutaminase in Physarum polycephalum, an acellular slime mold, upon cellular mechanical damage

Authors


  • Database
    The nucleotide sequence of the Physarum polycephalum adenine nucleotide translocator is available in the DDBJ/EMBL/GenBank database under accession number AB259838

K. Hitomi, Department of Applied Molecular Biosciences, Graduate School of Bioagricultural Sciences, Nagoya University, Chikusa, Nagoya, 464-8601, Japan
Fax: +81 52 789 5542
Tel: +81 52 789 5541
E-mail: hitomi@agr.nagoya-u.ac.jp

Abstract

Transglutaminases are Ca2+-dependent enzymes that post-translationally modify proteins by crosslinking or polyamination at specific polypeptide-bound glutamine residues. Physarum polycephalum, an acellular slime mold, is the evolutionarily lowest organism expressing a transglutimase whose primary structure is similar to that of mammalian transglutimases. We observed transglutimase reaction products at injured sites in Physarum macroplasmodia upon mechanical damage. With use of a biotin-labeled primary amine, three major proteins constituting possible transglutimase substrates were affinity-purified from the damaged slime mold. The purified proteins were Physarum actin, a 40 kDa Ca2+-binding protein with four EF-hand motifs (CBP40), and a novel 33 kDa protein highly homologous to the eukaryotic adenine nucleotide translocator, which is expressed in mitochondria. Immunochemical analysis of extracts from the damaged macroplasmodia indicated that CBP40 is partly dimerized, whereas the other proteins migrated as monomers on SDS/PAGE. Of the three proteins, CBP40 accumulated most significantly around injured areas, as observed by immunofluoresence. These results suggested that transglutimase reactions function in the response to mechanical injury.

Abbreviations
ANT

adenine nucleotide translocator

Bio-Cd

biotinylated cadaverine

CBB

Coomassie Brilliant Blue R250

CBP40

40 kDa Ca2+-binding protein

DAPI

4′,6-diamidino-2-phenylindole

F-Cd

fluorescein cadaverine

HMC

Hepes-based magnesium and calcium buffer

KLH

keyhole limpet hemocyanin

PpANT

adenine nucleotide translocator from Physarum polycephalum

PpTGase

transglutaminase from Physarum polycephalum

PVDF

poly(vinylidene difluoride)

TGase

transglutaminase.

The transglutaminase (TGase; EC 2.3.2.13) enzyme family catalyzes the Ca2+-dependent crosslinking of the γ-carboxyamide group of glutamine residues and the ε-amino group of lysine residues or primary amines [1,2]. This reaction results in the formation of an isopeptide bond between two proteins and the covalent incorporation of polyamines into proteins. In mammals, the crosslinking activity of several TGase isozymes functions in blood coagulation, stabilization of extracellular matrix, apoptosis, and skin barrier formation [3–7].

Similar crosslinking reactions are observed in various organisms, from microorganisms to animals. TGases with papain-like characteristics, such as Ca2+-dependency and an active-center Cys residue, have been identified in vertebrates and arthropods [1,2,8,9]. In bacteria, yeasts, and lower invertebrates such as nematodes, genes encoding homologous proteins have not been found [2,9,10]. We, however, have reported that Physarum polycephalum, an acellular slime mold, is the evolutionarily lowest organism with a TGase that has a primary structure similar to that of TGases in mammals [11,12].

Physarum polycephalum, which belongs to the Mycetozoa, is a model eukaryote with a unique life cycle characterized by spores, amoebae, macroplasmodia, and microplasmodia. The plasmodium, used in this study, is a giant and multinucleated cell with a veined structure and no internal cell walls. So far, Physarum has been used mainly in studies on the cell cycle, inheritance of mitochondrial DNA, and cytoplasmic streaming [13–18]. Physarum is also an appropriate model organism for studies on responses to environmental stress. For example, in response to heat stress, Physarum enhances glycosylation of membrane sterol to induce its signal transduction system to synthesize heat shock proteins [19]. Also, Physarum TGase activity is induced upon exposure to ethanol or detergent, resulting in transamidation of proteins [20].

In mammals, there are several reports that TGase is activated in protective responses to environmental stimuli and contributes to wound healing in various cells [21–26]. In some of these events, remodeling and stabilization of extracellular matrix proteins by TGase resulted in repair of chemical and mechanical injury. However, TGase substrates and their potential roles in repair of damage in unicellular organism are unknown.

In this study, we further investigated the role of P. polycephalum TGase (PpTGase) in response to mechanical damage. Following mechanical damage, we observed TGase reaction products around the mechanically injured area. On the basis of these observations, we identified and characterized three preferred glutamine-donor TGase substrates: 40 kDa Ca2+-binding protein (CBP40)[27,28], Physarum actin [29], and a novel protein with high structural similarity to eukaryote adenine nucleotide translocator (ANT).

Results

Detection of TGase reaction products around injured areas

To investigate whether PpTGase is involved in the response to mechanical damage, we examined in situ enzymatic reactions in slime mold macroplasmodia following injury. As shown in Fig. 1, after cells were stabbed with a toothpick, fixed proteins into which fluorescein cadaverine (F-Cd) was incorporated by TGase catalysis were observed around the injured area. This reaction was completely blocked by several inhibitors of TGase, such as L-682.777, cystamine, and cadaverine. These results indicate that labeled primary amine was incorporated into several glutamine-donor substrates by activated TGase upon mechanical damage.

Figure 1.

 Incorporation of F-Cd into glutamine-donor substrates at injured sites in macroplasmodia. Macroplasmodia grown on a PVDF membrane were injured in the presence of F-Cd. After 3 min, the cells were fixed, and differential interference images (DIC) and fluorescent images (F-Cd) of the cells were obtained. The same experiment was performed in the copresence of 40 µm L-682.777, 20 mm cystamine or 20 mm cadaverine in F-Cd solution. The bar represents 200 µm.

Purification of potential PpTGase substrates upon mechanical damage

Next, we identified the glutamine-donor substrate proteins that incorporated primary amines in response to damage in macroplasmodia. Total cellular lysates were prepared from macroplasmodia damaged in the presence of biotinylated cadaverine (Bio-Cd). Depending on the time after injury, Bio-Cd was incorporated into several proteins (Fig. 2). In control cells with no damage (both at 10 s and 180 s), only nonspecific bands (marked at the right with asterisks) were observed; those bands probably represent endogenous biotin-conjugating and biotin-binding proteins. Furthermore, no specific incorporation was observed in the copresence of several inhibitors or in the absence of Bio-Cd. During the assay period, levels of expressed PpTGase remained equivalent, as indicated by immunoblotting (Fig. 2, lower panel). These results indicated that PpTGase catalyzed transamidation of several proteins acting as preferred glutamine-donor substrates when activated upon mechanical injury.

Figure 2.

 Detection of total cellular proteins that incorporated Bio-Cd upon mechanical damage. At time 0 s, growing macroplasmodia on an agar plate were injured in the presence of Bio-Cd. Total cellular extracts of macroplasmodia were prepared at the indicated periods. Samples were subjected to 10% SDS/PAGE and transferred to PVDF membranes. Top: Proteins incorporating Bio-Cd were detected using peroxidase-conjugated streptavidin. Samples from cells without damage (10 s and 180 s) and from damaged cells (180 s) in the presence of L-682.777 (40 µm), cystamine (20 mm) or cadaverine (20 mm), or in the absence of Bio-Cd, were prepared in parallel. The asterisks indicate no specific signals. Bottom: All samples were subjected to immunoblotting using a monoclonal antibody to PpTGase.

Next, we purified these candidate substrates. As they are likely to be attached to the plasma membrane, a soluble membrane fraction obtained by Triton X-100 treatment was subjected to purification. As shown in Fig. 3, three major proteins (p44, p40, and p33) were eluted as potential substrates, and these proteins were not obtained with the same procedure in the absence of Bio-Cd (lane 7). Using peroxidase-conjugated streptavidin, the eluted proteins were detected as biotin-incorporated proteins (Fig. 3B). In this fraction, there were other minor proteins as possible substrates, the amounts of which were not sufficient for the following analysis. The proteins in the gel were subjected to trypsinization and then to TOF MS analysis. On the basis of data in the database of molecular masses of fragmented proteins, p40 and p44 were identified as CBP40 [27,28] and Physarum actin [29,30], respectively, whereas p33 was a novel protein not found in the database.

Figure 3.

 Purification of proteins incorporating Bio-Cd from damaged slime mold. The total cellular extract, cytosolic fraction and Triton X-100 soluble membrane fraction were prepared from Physarum macroplasmodia injured in the presence of Bio-Cd. From the membrane fraction, proteins incorporating Bio-Cd were affinity-purified with streptavidin-sepharose. To compare them with nonspecifically bound proteins, the same procedure without addition of Bio-Cd was also performed. (A) CBB staining. (B) Detection of biotinylated proteins by peroxidase-conjugated streptavidin. In both panels, lanes are as follows: lane 1, total cellular extract; lane 2, cytosolic fraction; lane 3, Triton X-100 soluble fraction; lane 4, dialyzed Triton X-100 soluble fraction (applied sample); lane 5, unbound fraction; lanes 6 and 7, eluted fractions from extracts prepared in the presence and absence of Bio-Cd, respectively.

Purification and molecular cloning of a novel 33 kDa substrate protein

In order to identify p33, we purified the protein by affinity chromatography and SDS/PAGE. Because the N-terminus of the protein was blocked, purified p33 was treated with cyanogen bromide, and the resulting fragments were subjected to amino acid sequence analysis.

On the basis of the partial amino acid sequence of one fragment, a cDNA clone encoding p33 was obtained by 3′-RACE using degenerate primers: 5′-RACE resulted in 5′-nucleotide sequences that probably include the initiation codon ATG (Fig. 4). The complete sequence shows an ORF of 936 bp encoding 312 amino acids with a calculated molecular mass of 33 622 Da. The amino acid sequence deduced from the nucleotide sequence was highly homologous to that of the ANT seen in several eukaryotes, and we therefore designated the protein as PpANT (for P. polycephalum ANT). The amino acid sequence of PpANT was 50–77% identical to those of human (ANT1, NP_001142; ANT2, NP_001143; ANT3, NP_001627), mouse (ANT1, NP_031476; ANT2, NP_0031477), bovine (NP_777083), Caenorhabditis elegans (NP_001022799), Dictyostelium discoideum (XP_647166), Arabidopsis thaliana (NP_850541), Zea mays (CAA40781) and Saccharomyces cerevsiae (NP_009523) homologs (Fig. 5). From the PpANT primary structure, six possible membrane-spanning regions were deduced from the distribution of hydrophobic regions, as is observed in ANTs of other species. Although the initiation codon (ATG) was deduced from the alignment, recombinant protein produced from expression of the full-length cDNA in bacteria was of the predicted size (data not shown).

Figure 4.

 Nucleotide and deduced amino acid sequences of PpANT. The complete amino acid sequence of PpANT was deduced from the nucleotide sequence. The numbers of nucleotide and amino acid residues are shown on the left and right sides, respectively. The gray background indicates the fragment cleaved by cyanogen bromide treatment of the purified protein.

Figure 5.

 Multiple alignment of PpANT with several eukaryotic ANTs. Amino acid sequences were aligned using the default setting of clustal x, a multiple sequence alignment program. Amino acid residues common to all sequences are denoted by an asterisk above the sequences, whereas conservative residues are indicated by a colon (: high) or a period (. low).

It is known that eukaryote ANT is the most abundant protein in mitochondria [31]. We also investigated the cellular distribution of PpANT in Physarum macroplasmodia using a polyclonal antibody. The cell was counterstained with 4′,6-diamidino-2-phenylindole (DAPI) to visualize both the nucleus (Fig. 6, arrow) and mitochondrial nucleoid (Fig. 6, arrowhead). By phase-contrast (Fig. 6A) and DAPI fluorescence microscopy (Fig. 6B), mitochondria of macroplasmodia were observed as oval-shaped structures and each of them contained a rod-like mitochondrial nucleoid. Fluorescence immunostaining microscopy revealed that the staining patterns of PpANT coincided with the mitochondria, as was expected (Fig. 6C,D).

Figure 6.

 Immunolocalization of PpANT in Physarum macroplasmodia. A growing macroplasmodum was fixed and reacted with polyclonal antibody to PpANT and then developed by an Alexa Fluor 488-conjugated secondary antibody. Mitochondrial and nuclear DNAs were counterstained with DAPI. (A) Merged image of phase-contrast and DAPI staining. (B) DAPI staining image. (C) Immunostaining image obtained using antibody to PpANT. (D) Merged image of (B) and (C). The arrows and arrowheads indicate nucleus and mitochondria, respectively. Enlarged images are shown in the inset. The bar represents 5 µm.

Immunoblotting analysis of potential substrates upon mechanical damage

In order to find how these possible substrates reacted with PpTGase upon cellular injury, we performed immunoblotting of total cellular extracts (Fig. 7, left). The first identified substrate, CBP40, migrated as a 40 kDa protein, but the levels of a higher molecular mass band (80 kDa), probably corresponding to a dimer, increased over time. The slight band (80 kDa) with no damage was produced during the preparation of extracts. This crosslinked product was not observed in the presence of cystamine, suggesting that CBP40 was dimerized by TGase in response to mechanical damage. PpANT and actin were detected at the predicted monomeric size, without no possible dimer form.

Figure 7.

 Immunoblot analysis of potential PpTGase substrates upon cellular injury. Total cellular extracts were prepared at the indicated times (10–180 s) from damaged macroplasmodia growing on a plate. Upon injury, HMC buffer was added to the plate, and cells were stabbed with toothpicks several times. As a control, cystamine was added to block the TGase reaction. The right lanes of all blots contains purified protein, which incorporated Bio-Cd from injured macroplasmodia using streptavidin-sepharose chromatography (Fig. 3, lane 6). Samples were subjected to SDS/PAGE followed by CBB staining (right) and immunoblotting analysis using each polyclonal antibody (left): (A) anti-CBP40; (B) anti-Physarum actin; (C) anti-PpANT. The closed arrows indicate each protein. The open arrow indicates a possible CBP40 dimer.

Affinity-purified proteins incorporating Bio-Cd were recognized by respective antibodies (Fig. 7, eluted fraction), confirming that proteins were transamidated upon injury. Taken together, these results suggested that CBP40, actin, and PpANT are enzymatically modified by PpTGase in mechanically damaged macroplasmodia.

Cellular analysis of potential substrates in injured macroplasmodia

To investigate the localization of potential substrates around the injured area, each protein was analyzed by immunostaining in cells (Fig. 8). In the absence of cell damage (J, K, L), all proteins were stained uniformly. CBP40 protein was strongly stained around the injured area (D, G), suggesting that this protein accumulates or is aggregated upon damage. However, both Physarum actin and PpANT showed no apparent difference in staining pattern in injured versus noninjured areas (E, F, H, I).

Figure 8.

 Immunostaining of potential PpTGase substrates in macroplasmodia. A macroplasmodium grown on a PVDF membrane was injured by a toothpick (A–C; DIC, differential interference images). Then, fixed and permeabilized cells were immunostained with respective antibodies against CBP40 (D, G, J), Physarum actin (E, H, K), and PpANT (F, I, L) using an Alexa Fluor 488-conjugated secondary antibody. The indicated injured region is enlarged (G–I; box in panels D–F). Immunostaining analyses for uninjured areas are shown at the same scale in parallel (J–L). All immunostained signals are shown as stacked images in the vertical direction. The bars represent 10 µm and 100 µm.

Discussion

Although most eukaryotic cells and tissues exhibit protective elements, cells can suffer damage following environmental insult. To respond to mechanical damage, adaptive systems have been developed not only at the tissue level but also at the cellular level. Membrane resealing, for example, triggered by Ca2+ entry upon disruption, is a membrane-repair process allowing cells to survive [32,33]. Although it is likely that various molecules and mechanisms participate in responses to mechanical challenge, the process is not well understood.

TGases are Ca2+-dependent crosslinking enzymes, and are thus likely to function in such mechanisms [1,2]. Indeed, in mammals, it has been shown that TGases respond to environmental attack by participating in wound healing [21–26]. In fibroblasts, for example, TGase maintains tissue integrity by formation of an SDS-insoluble shell-like structure following rapid loss of Ca2+ homeostasis [23].

We have focused on the physiologic significance of TGase in Physarum, as this is the lowest known organism exhibiting a TGase similar to that expressed in mammals [11,12]. Upon mechanical damage, Physarum displayed TGase-dependent incorporation of a fluorescent-labeled primary amine into glutamine-donor substrate protein(s) (Fig. 1). The product was observed around the mechanically injured area, suggesting that Ca2+ influx activated a latent form of intracellular TGase since PpTGase is Ca2+-dependent as in the case for mammalian TGase [11]. The substrate proteins might localize around the membrane that activated TGase can access. Based on time-dependent transamidation, as shown in Fig. 2, several proteins underwent modification without change in the amount of PpTGase, indicating that endogenous TGase activity was stimulated by damage. Although unidentified minor proteins in the purified fraction may also be substrates, the further analyzed glutamine-donor substrates consisted of mainly three proteins: actin, CBP40, and PpANT. This observation is consistent with the fact that chemical damage of Physarum microplasmodia by treatment with ethanol or detergent results in transamidation of actin and CBP40 [20].

Mechanical damage also resulted in the crosslinking of CBP40 to form a covalently bound dimeric form, and enhanced its levels around the injured area. CBP40, which has four EF-hand motifs in the C-terminus and a putative α-helix domain in the N-terminus, aggregates reversibly in a Ca2+-dependent manner via the N-terminus in vitro[27]. TGase may contribute to self-assembly of CBP40, where the crosslinked dimer form acts as core to initiate further assembly. Although CBP40 orthologs in other organisms have not been reported, such a crosslinking reaction is reminiscent of clot formation in vertebrates.

Both actin and PpANT, identified as potential substrates, also incorporated Bio-Cd by transamidation upon mechanical damage, although significant aggregation or accumulation was not observed by immunostaining. In western blot analysis, actin and PpANT did not show apparent changes in molecular size following damage, suggesting that they are modified by transamidation or deamidation, as reported for several substrates [34–36]. Physarum actin, which is highly homologous to mammalian actin, is implicated as a force-generating system in actomyosin fibrils [29]. In mammals, actin, as both G-actin and F-actin is a favorable TGase 2 substrate in vitro[37,38]. In this study, the distribution of actin was not affected by injury in the presence or absence of a TGase inhibitor (Fig. 6, and data not shown). In Physarum, monomer actin might not be affected even after modification.

PpANT, another potential TGase substrate, was cloned for the first time in this study. On the basis of its considerable homology to ANTs in other eukaryotes and observation of its exclusive localization in mitochondria, it is likely that PpANT functions as an antiporter mediating ADP/ATP exchange in the slime mold. Although we could not show the localization of PpTGase in mitochondria, TGase activity was detected in the purified mitochondrial fraction in mammalian liver and brain [40]. Additionally, in TGase2-overexpressing cells, TGase2 has been reported to localize to mitochondria upon induction of apoptosis [41]. Determining whether transamidation by PpANT regulates ATP-translocating activity or induces apoptosis will require further study.

As shown in Fig. 3B, there were minor biotin-incorporating proteins present upon injury. As recovery from cellular damage might require more than three major substrates, further investigation of unidentified substrates and crosslinking reactions would be necessary. We have recently established a system to identify the TGase preferred substrate sequence with respect to mammalian TGases [42]. Applying this system to the identification of PpTGase preferred substrates should reveal other substrates and potentially define a network of substrates. Additionally, knockdown analyses of TGases and their substrates by an RNA interference method that has recently been established in this organism might be also useful [43].

Although little is known about the physiologic functions of TGases in nonmammalian species, there are several reports of TGases being essential for defense against environmental factors [8,44,45]. Cellular responses to mechanical damage are required for eukaryotes to maintain their homeostasis. In the horseshoe crab, for example, TGase is implicated in the formation of coagulin polymers upon aggregation of hemocytes, and it also crosslinks several chitin-binding proteins in the cuticle [8,46]. As evolutionarily lower organisms do not possess an acquired immune system, TGase activity may be particularly important in defending these organisms against environmental challenges. Further investigation of possible TGase substrates in the slime mold should provide insights into the responses of eukaryotic cells to mechanical damage.

Experimental procedures

Cell culture

Physarum macroplasmodia were basically grown on 1.5% agar plates containing MEA medium consisting of 0.165% mycological peptone (Oxoid, Basingstoke, UK), 1% malt extract (Oxoid), and 5 µg·mL−1 hemin (ICN Biomedicals Inc., Irvine, CA)[13,14]. In the case of observation of the fixed macroplasmodia, cells were grown on a poly(vinylidene difluoride) (PVDF) membranes (Millipore, Bedford, MA) located on the agar plate. Both cultures were grown in complete darkness at 25 °C.

Incorporation of F-Cd into PpTGase substrate in the damaged slime mold

Macroplasmodia cells grown on a PVDF membrane were transferred to a 35 mm dish containing Hepes-based magnesium and calcium buffer (HMC; 20 mm Hepes/NaOH, pH 7.4, 10 mm NaCl, 40 mm KCl, 2 mm CaCl2, 7 mm MgCl2). F-Cd (Invitrogen, Carlsbad, CA) was added to a final concentration of 0.1 mm, and then the cells were injured by stabbing them with a toothpick. After 3 min, cells on a PVDF membrane were washed with HMC buffer and then fixed at room temperature for 15 min in a solution of 10% trichloroacetic acid. The cells were then washed with NaCl/Pi buffer (10 mm sodium phosphate, pH 7.4, 150 mm NaCl) three times, and incubated in NaCl/Pi buffer containing 1.0% Triton X-100 for 30 min at room temperature. Cells were removed from the membrane, and located on a coverslip coated with 0.01% poly(l-lysine). After drying, these samples were mounted on a glass slide with antifading solution containing Mowiol 4-88 (Calbiochem, Darmstadt, Germany) and glycerol. Samples were analyzed under a confocal laser-scanning microscope (LSM5 PASCAL; Zeiss, Gottingen, Germany).

Cystamine (Sigma, St Louis, MO), cadaverine (Sigma), and L-682.777 (N-Zyme, product name: 1,3,4,5-tetramethyl-2-[(2-oxopropyl)thio]imidazolium chloride) were used to inhibit the enzymatic reaction by PpTGase.

Detection and purification of TGase substrates upon cellular damage to macroplasmodia

HMC buffer containing Bio-Cd at a final concentration of 0.2 mm was added to macroplasmodia growing on MEA agar plates. The cells were injured with a bundle of toothpicks several times, as described above. After various periods, the TGase reaction was halted by the addition of cystamine. From the cells homogenized with lysis buffer (20 mm Tris/Cl, pH 7.5, 100 mm NaCl, 2 mm 2-mercaptoethanol, 20 mm cystamine, 1 mm phenylmethylsulfonylfluoride, 25 ng·µL−1 leupeptin, and 1 µm pepstatin), total cell extract was prepared by solubilization with SDS-dye buffer and boiled. For detection of Bio-Cd incorporated into cellular proteins, the proteins were subjected to SDS/PAGE and blotted onto a PVDF membrane, which was then developed by peroxidase-conjugated streptavidin (Rockland, Gilbertsville, PA) and the chemiluminescent method using the Super Signal West Pico chemiluminescent substrate detection kit (Pierce, Rockland, IL).

For purification of potential TGase substrates, the damaged slime mold in the presence of Bio-Cd was harvested after 3 min. The cells were washed and suspended by lysis buffer. The harvested cells were homogenized and centrifuged at 10 000 g for 10 min using a SRX-4 centrifuge (TOMY) and TA-4 rotor. The unsolubilized fraction was treated with the TNE buffer (20 mm Tris/HCl, pH 7.5, 100 mm NaCl, 5 mm EDTA, 2 mm 2-mercaptoethanol) containing 2% Triton X-100, 20 mm cystamine and protease inhibitors for 1 h at 4 °C. The membrane fraction was obtained as a supernatant by centrifugation [10 000 g for 20 min using a SRX-4 centrifuge (TOMY) and TA-4 rotor, and 100 000 g for 30 min using TL100 centrifuge (Beckman) and TLA100.3 rotor]. The supernatant was dialyzed against TNE buffer overnight to remove unincorporated Bio-Cd, and then applied to a streptavidin-conjugated column previously equilibrated with the same buffer. After several washings with TNE buffer, the bound proteins were eluted with 1 mm Tris/Cl buffer (pH 8.0) containing 4% SDS buffer. The eluate was concentrated and subjected to SDS/PAGE following by Coomassie Brilliant Blue (CBB) staining. The protein bands of interest were excised and further analyzed by using standard MALDI-TOF MS methodology.

To identify p33 protein, the protein was excised from 12.5% SDS/PAGE gel and then subjected to carbamidomethylation using iodoacetoamide. The protein concentrated by acetone precipitation was dissolved in 70% formic acid, and treated with cyanogen bromide at room temperature for 24 h in the dark. The reaction product was separated on a 15% SDS/PAGE gel and transferred to a PVDF membrane. The cleaved protein bands were excised and sequenced by automated Edman degradation.

Molecular cloning of a novel 33 kDa protein

3′-RACE was performed with the RNA LA PCR Kit Ver.1.1 (TAKARA Biomedicals, Japan). Total RNA from macroplasmodia was obtained by the acid guanidium phenol chloroform method. The first-strand cDNA was synthesized using 1 µg of total RNA in a reaction mixture of 1.0 mm dNTPs, 16 U of RNasin, 14 U of AMV reverse transcriptase, and oligo dT-M4 adaptor primer in the supplied buffer. The resulting cDNAs were subjected to PCR with M13 primer M4 and the degenerate primer 5′-GCTGGAGCTGCT(A/T)(C/G)(A/T/G/C) (C/T)T(A/T/G/C)AC(A/T/G/C)TTTGT-3′, which was designed on the basis of the amino acid sequence AGAASLTFVY. Amplification conditions were as follows: 30 cycles at 95 °C for 30 s, 51 °C for 30 s, and 72 °C for 90 s.

The PCR products obtained from 3′-RACE was cloned into a TA-cloning vector, pCR 2.1-TOPO (Invitrogen, Carlsbad, CA), according to the manufacturer's instructions. The nucleotide sequences of the isolated clones were determined with an automated fluorescent sequencer, ABI PRISM 310 (PE Applied Biosystems, Foster City, CA), using a Bigdye terminator cycle sequencing ready reaction kit (PE Applied Biosystems).

In order to obtain 5′-terminal cDNA, 5′-RACE was performed using reverse transcriptase and RNA ligase, according to the manufacturer's protocols (5′-Full RACE Core Set; TAKARA Biomedicals). First-strand cDNA was synthesized from 1 µg of the poly(A)+ RNA, purified with an oligo(dT) cellulose column, using AMV reverse transcriptase XL with a specific primer, 5′- TAGAGACCAGTGATACCATC-3′ (antisense, nucleotide sequence number 577–596), and then phosphorylated by T4 polynucleotide kinase. After degradation of the template poly(A)+RNA with RNaseH at 30 °C for 1 h, the resulting single-strand cDNA was precipitated with ethanol and dissolved in 40 µL of a reaction mixture containing 20% poly(ethylene glycol) #4000, RNA ligation buffer, and 1 U of T4 RNA ligase. To change the cDNAs to circular and/or concatemer cDNAs, the reaction solution was incubated at 15 °C for 16 h. The cDNAs were directly used as a template for the first PCR amplification with primers 5′-GGTGAACGCCAGTTCAATGGC-3′ (S1, sense, 523–543) and 5′-CGGACTTGTTGTCGTTAGCCAAAC-3′ (A1, antisense, 485–508), which correspond to the cDNA sequence obtained by 3′-RACE. The reaction was carried out for 30 cycles with the following conditions: 94 °C for 30 s, 53 °C for 30 s, and 72 °C for 90 s. The resulting PCR product was diluted 1000-fold with sterile H2O, and a 1 µL aliquot was used as a template for the second nested PCR amplification with primers 5′-GCTTGCTGGATGTCTACAGAAAGACC-3′ (S2, sense, 542–567) and 5′-ACGAGTACGGGCGTAGTCGAG-3′ (A2, antisense, 466–486) under the same conditions. Additional 5′-RACE reactions using other primers [5′-AGCAGCGATGTTG-3′ (antisense, 411–423), 5′-GAATGTTCGCTGTCCCCAAG-3′ (sense, 362–381), 5′-GTCCTTGAAGGCGAAGTTGAG-3′ (antisense, 331–351), 5′-GCCTCCTACGGAAAGAAGTTC-3′ (sense, 385–405) and 5′-CTTGGGTGGGGAAGTAACGG-3′ (antisense, 309–328)] produced putative full-length cDNA. Cloning and nucleotide sequencing were carried out as described for 3′-RACE.

Finally, after completion of cloning of putative full-length cDNA, oligonucleotides encoding 5′- and 3′-ends were prepared (5′-CTGGATCCCGAGAAGAAGAACGACCTCAG-3′ and 5′-GATGCTCGAGTTATCCACCTCCGCCAGAG-3′), and used for PCR reaction to obtain directly full-length cDNA.

Polyclonal antibodies

Polyclonal antibody against Physarum actin was kindly provided by K. Furuhashi (Shiga University, Japan) [30]. Antibodies against PpTGase [12], and CBP40 [27] were prepared as described previously. Polyclonal antibody against PpANT was prepared by immunization of peptide conjugated with keyhole limpet hemocyanin (KLH; Sigma). On the basis of the deduced amino acid sequence, a peptide (YDSLKPALSPLENNPVALGC) corresponding to the amino acid sequence of region 199–217 with an additional Cys residue at the C-terminus was synthesized. Then, the Cys residue of the peptides was covalently crosslinked with KLH using m-maleimidobenzoil-N-hydroxysuccinimide ester, and used as immunogen to raise antibody in rabbit. By subcutaneous immunization of the peptide–KLH six times, antiserum was prepared. The antibody was affinity-purified from antisera using a column that immobilized the peptide.

Immunologic analysis of potential substrates from total cellular lysates

For western blotting, total cellular extracts were prepared from the injured macroplasmodia by stabbing with toothpicks as described above. The harvested cells were homogenized with lysis buffer, and then solubilized directly in SDS sample buffer. Next, the samples were subjected to SDS/PAGE and western blotting using PVDF membranes. Antibodies were reacted by standard methods, and immunosignals were detected by the chemiluminescent method as described above.

Immunostaining analysis

Macroplasmodia cells grown on a PVDF membrane were damaged and fixed as described above. After being washed with NaCl/Pi, cells were incubated in NaCl/Pi containing 1% BSA to prevent nonspecific binding for 1 h at 37 °C.

Then, the cells fixed by trichloroacetic acid solution were incubated in the presence of each polyclonal antibody. Subsequently, cells were incubated with Alexa Fluor 488-conjugated goat anti-rabbit serum (Molecular Probes). Samples were analyzed with a confocal laser-scanning microscope (Zeiss) as described above, using 488 nm and 505–530 nm filters. The software used was lsm image browser (Zeiss).

In the case of counterstaining of DNA (Fig. 6), cells were fixed by 3.7% formaldehyde for 15 min, and then subjected to the immunostaining reaction as described above. Before mounting of samples on a glass slide, DNA was counterstained with DAPI. Cells were observed under an epifluorescence microscope equipped with a phase-contrast objective (Olympus, Tokyo, Japan).

Acknowledgements

This work was supported by a Grant-in-Aid for Scientific Research (C) no. 14560063 (to K. Hitomi), Young Scientist Research grant no. 15000941 (to F. Wada), and a TOYOAKI Science Foundation grant (to K. Hitomi). We thank Dr K. Furuhashi (Shiga University, Japan) for providing us with antibody to Physarum actin. F. Wada and Y Sugimura are Japanese Society for the Promotion of Science (JSPS) Research Fellows.

Ancillary