Animals

A clonal strain of planarian, Dugesia japonica, established in our laboratory, was used. It was bred in boiled tap water at 22-24°C, and fed chicken liver twice a week. All worms were used for experiments after one week of starvation.

Cloning and DNA sequencing

Mouse antiserum against living planarians was used to screen a λZAP expression cDNA library of planarian. Immunological screening was performed according to the manufacture protocols (Stratagene). Forty-seven positive clones among 4 × 10⁴ cDNA clones were classified by cross-reactivity and restriction enzyme analyses. These positive clones were recloned into pBluescript by in vivo excision. Purified plasmid DNA was sequenced using a BcaBEST Dideoxy Sequencing Kit (Takara) or a Thermo Sequence Fluorescent Labeled Primer Cycle Sequencing Kit (Amersham). Analysis of the sequence was performed using the DDBJ on-line service (National Institute of Genetics, Japan) or the Wisconsin GCG Package program (Version 9.0, Genetics Computer Group, Madison, WI, USA).

In situ hybridization of sections and whole-mount specimens

Planarians were treated with 2% HCl for 5 min and fixed either in relaxant solution (1% HNO₃, 1.6% formaldehyde, 20 μM MgSO₄; Dawar, 1973) for 12-24 hr, or in 2% formaldehyde, 0.2% saturated picric acid solution for 30 min, followed by 2% formaldehyde, 0.2% saturated picric acid solution, 0.1% glutaraldehyde for 1 hr, and then 2% formaldehyde, 0.2% saturated picric acid solution for 2 hr. They were then dehydrated through a graded ethanol series, embedded in paraffin, sectioned horizontally, sagittally or transversely and then mounted on poly-L-lysine-coated slides. After de-paraffination in xylene and hydration through the ethanol series, the mounted specimens were rinsed in PBS and acetylated in 0.1 M triethanolamine, 0.09% HCl and 0.25% acetic anhydride for 15 min. They were rinsed in 2 × SSC and then incubated at 55°C for 30 min in 50% formamide, 5 × SSC and 10 mM DTT. Digoxygenin-labeled antisense probe derived from DjMHC-A or DjMHC-B was diluted to 50 ng-500 ng/ml in 50% formamide, 5 × SSC, 0.1 mg/ml yeast RNA, 0.1 mg/ml heparin, 0.1% Tween-20, 10 mM DTT and 10% dextran sulfate, denatured at 65°C for 15 min, and added to slides for hybridization at 55°C for 16-36 hr. After hybridization, the slides were washed in 50% formamide, 5 × SSC and 0.1% Tween-20 for 9-12 hr at 55°C, and twice in 2 × SSC at room temperature for 15 min. The slides were rinsed in Buffer-1 (0.1 M maleic acid, 0.15 M NaCl, 0.1% TritonX-100), incubated in Buffer-2 (1% Boehringer blocking reagent in Buffer-1) for 30 min and then in Buffer-2 containing alkaline-phosphatase (AP)-conjugated antidigoxygenin antibody (Boehringer) diluted 1:5000 at room temperature for 2 hr. They were rinsed in Buffer-1 and in TMN buffer (0.1 M NaCl, 0.1 M Tris-HCl, pH 9.5, 0.05 M MgCl₂) for 15 min. Signals were rendered detectable by incubation of the slides in TMN buffer containing 200 μg/ml NBT (Sigma) and 175 μg/ml BCIP (Sigma) for 2-3 hr. After the chromogenic reaction was completed, the slides were washed twice in TE buffer (10 mM Tris-HCl, pH 8, 1 mM EDTA), mounted in 50% glycerol in TE buffer and observed under a microscope. Double staining was performed using digoxigenin-labeled probe for MHC-B and biotin-labeled probe for MHC-A. Both signals were simultaneously detected using a TSA Kit (NEN Life Science Products) according to the procedures of Kato et al. (manuscript in preparation). Whole-mount in situ hybridization was carried out as described by Umesono et al. (1997) and Agata et al. (1998).

Dot blot hybridization

Non-labeled sense RNAs synthesized from DjMHC-A or DjMHC-B and yeast tRNA were diluted serially and blotted on nylon membranes (Hybond-N, Amersham) in duplicate. The RNA was fixed to the membrane by baking at 80°C for 2 hr and cross-linking by UV light. Both membranes were incubated in prehybridization solution at 55°C for 1 hr. They were then incubated with the probe at 50 ng/ml at 55°C for 16-36 hr. After hybridization, they were washed in the wash buffer at 55°C for 9-12 hr and washed twice in 2 × SSC at room temperature for 15 min. The hybridization signal was detected immunologically as described above.

Isolation and characterization of two planarian MHC genes

Forty-seven clones that reacted with anti-planarian antiserum were classified into two groups (A and B) by crossreactivity and restriction enzyme analyses. The longest clone of group A contained a 6111-bp insert with an ORF of 1958 amino acids. The longest clone of group B contained a 5388-bp insert with an ORF of 1743 amino acids. The nucleotide sequence data reported in this paper will appear in the DDBJ/EMBL/GenBank nucleotide sequence databases with the accession numbers AB015484 and AB015485.

Computer analysis of the DNA and amino acid sequences showed high similarity between the two genes (79.9% at the protein level) and similarity of both genes to myosin heavy chain genes of other animals (Fig. 1). The predicted amino acid sequence of the group-A gene shows 64.4% similarity to that of scallop and 53.1% similarity to that of human MHC. That of the group-B gene shows 64.2% and 52.8% similarity to those of scallop and human MHCs, respectively. Myosinspecific domains such as the ATP-binding site, actin-binding sites, two closely spaced reactive thiol groups; SH1 and SH2, and hinge were well conserved. Therefore, we named the group-A and -B genes “DjMHC-A” and “DjMHC-B”, respectively.

Fig. 1

Predicted amino acid sequences of DjMHC-A and DjMHC-B compared to those of clam (scallop) and human MHCs. Conserved amino acids are indicated by dashes. Dots indicate gaps for alignment. Myosin-specific domains of the MHCs are boxed. Underlines indicate amino acid sequences conserved among the striated-muscle MHCs.

To classify these two planarian MHCs, we compared them to striated-muscle-type and smooth-muscle-type MHCs of other organisms. Figure 2A shows a comparison of the amino acid sequences of the isoform-specific region among DjMHC-A, DjMHC-B and MHCs of various other organisms (Schuchert et al., 1993; Takahashi et al., 1992). This comparison indicated that both DjMHC-A and DjMHC-B genes are striatedmuscle-type MHC genes since they showed much higher similarities to striated-muscle-type than to smooth- or nonmuscletype MHCs. Phylogenetic tree analysis also suggested that both planarian MHC genes should be classified as striatedmuscle type. DjMHC-A and DjMHC-B are related more closely to each other than to any other MHC gene. Although these structural analyses suggest that both MHCs are striated-muscle-type MHCs, no striation has been observed in planarian muscles (MacRae, 1963; Morita, 1965; Hori, 1983).

Fig. 2

Comparison of amino acid sequences in the isoform-specific region, and phylogenic tree analysis. (A) Alignment of the sequences in the isoform-specific region (Schuchert et al., 1993). Complete sequences were aligned by the pileup program of the GCG Package and then the isoform-specific region was selected. Sequences were taken from the SWISS Plot and DDBJ/GenBank/EMBL sequence libraries. Dashes indicate identical residues. The consensus sequence of the striated-muscle-type MHCs is indicated. Capital letters indicate perfect conservation; lower case letters, high conservation; points, no conservation. (B) An unrooted phylogenetic tree of DjMHC-A, DjMHC-B and other MHCs. This tree was produced using the distances program of the GCG Package. The accession numbers and amino acid residues of the MHCs used in these analyses are as follows: amoebae, P05659; ascidian, D45163; blood fluke, L01634; cellular slime mold, P08799; chicken-skeletal, P13538; chicken-nonmuscleA, P14105; chicken-nonmuscleB, M93676, PID g212449; chicken-smooth, P10587; clam, P24733; fly-nonmuscle, Q99323; human-nonmuscle, P35579; human-skeletal, P11055; human-cardiac, P12883; jellyfish, X64057; nematode-MHC A, P12844; nematode-MHC B, P02566; nematode-MHC C, P12845; nematode-MHC D, P02567; plant, Z34293; rabbit-smooth, P35748; rat-cardiac, P02563; sea urchin, M55321; yeast, P08964. Accession numbers that begin with a “P” are from the SWISS Plot database; all others are from the DDBJ/GenBank/EMBL database. “10” indicates the phylogenetic distance expressed as substitutions per 100 amino acids.

Comparison between DjMHC-A and DjMHC-B

We next compared the structures of DjMHC-A and DjMHC-B. The overall identities between DjMHC-A and DjMHC-B are 74.9% and 79.9% at the DNA and protein levels, respectively. HarrProt analysis clearly indicated regions conserved between DjMHC-A and DjMHC-B (Fig. 3A). The head region is highly homologous between them (92.2%), but the light meromyosin (LMM) region of the tail is much less homologous (70.4%). Since the two genes show strong homology in the head region, we detected cross-reaction between them during routine hybridization procedures. Thus, we needed to establish conditions for avoiding cross-reaction during in situ hybridization. First, we used tail regions as probes, since these regions have low homology. However, the resultant hybridization signals were weaker than the signals obtained using full-length probes (data not shown). Therefore, we tried performing RNA hybridization at 55°C in a solution containing 50% formamide, and checked the cross-reactivity between DjMHC-A and DjMHC-B by serial dilution experiments (Fig. 3B). We obtained satisfactory results using these high stringency conditions, although slight cross-reactivity was detected.

Fig. 3

Comparison between structures of DjMHC-A and DjMHC-B. (A) HarrProt analysis was performed to compare the DjMHC-A and DjMHC-B genes. Unit size used for the comparison: 30, Dot plot matching number: 25. HMM, heavy meromyosin; LMM, light meromyosin; S1, subunit 1; S2, subunit 2. The table summarizes the identity between DjMHC-A and DjMHC-B in three regions. (B) Cross-reactivity between DjMHC-A and DjMHC-B in high stringency condition. The sense RNAs of DjMHC-A and DjMHC-B, and yeast tRNA were diluted serially and blotted on nylon membranes. Dot blot hybridization was performed using labeled antisense RNA probes: DjMHC-A (upper panel), DjMHC-B (lower panel).

Expression of DjMHC-A and DjMHC-B in normal adult planarians

To investigate the spatial distribution of planarian MHC gene transcripts in normal adult planarians, we performed in situ hybridization under the high stringency conditions described above. Strong expression of DjMHC-A was detected in pharynx muscles (Fig. 4A-a, b and d). DjMHC-A-positive signals were also observed in muscles surrounding the intestinal ducts (Fig. 4A-c and d; arrows), a subpopulation of bodywall muscles (Fig. 4A-c; arrowheads), several cells in the mesenchymal region around the base of the pharynx (Fig. 4A-c; asterisks), and pharynx-cavity muscles (Fig. 4A-d; arrowheads).

Fig. 4

Expression patterns of DjMHC-A and DjMHC-B in normal adult planarians. (A) Expression of DjMHC-A. (a) A whole-mount in situ hybridization. (b) A transverse section of the pharynx region. (c) A transverse section of the base of the pharynx. Arrowheads indicate expression in body-wall muscles. Arrows indicate expression in the muscles surrounding the intestinal ducts. Asterisks indicate expression in the cells in the mesenchymal space. (d) A high-magnification view of the pharynx region in a horizontal section. Arrows indicate expression in the pharynx-cavity muscle. Arrows indicate expression in the muscles surrounding the intestinal ducts. (B) Expression of DjMHC-B. (a-c) Arrows indicate expression in body-wall muscles. (a) A sagittal section. Arrowheads indicate transverse muscles. (b) A transverse section of the pharynx region. Arrowheads indicate transverse muscles. (c) A transverse section of the tail region. Short Arrows indicate horizontally oriented muscles. Open arrowheads indicate nucleus of the transverse muscles. Arrows indicate nucleus of the horizontally oriented muscles. (C) A diagram of the position of sections. I, intestinal duct. Scale bars, 0.5 mm (Aa) and 0.2 mm (Abd and Ba-c).

In contrast to the pattern seen with DjMHC-A, expression of DjMHC-B was detected in body-wall muscles (Fig. 4B-a, b and c; arrows), vertical muscles (Fig. 4B-a and b; arrow-heads) and horizontally oriented muscles in the ventral part of the body (Fig. 4B-c; short arrows), but not in pharynx muscles. Interestingly, we found that the nucleus of the vertical muscles is located in the dorsal region (Fig. 4B-c; open arrowheads), whereas the nucleus of the horizontally oriented muscles is located in the ventral region (Fig. 4B-c; arrowheads).

Since we detected expression of both genes in the body-wall-muscle layer and the mesenchymal region, we performed doubly probed in situ hybridizations to investigate whether or not DjMHC-A and DjMHC-B were expressed in the same cells. In Fig. 5, DjMHC-A- and DjMHC-B-positive cells are shown in green and red colors, respectively. Blue color indicates nuclear counter-staining using Hoechst 33342. No overlapping expression was observed. In the mesenchymal region, DjMHC-A- positive cells were distributed at the periphery of the intestinal duct, but DjMHC-B-positive cells were observed in the space between the intestinal ducts (Fig. 5B-D). Figure 5E-G shows higher magnification views of the body-wall layer. DjMHC-A and DjMHC-B expression was observed in cells in the same layer, but not in the same cells. These results clearly demonstrate that muscle cells of planarians can be classified into at least two types by their patterns of expression of MHC genes.

Fig. 5

Doubly stained in situ hybridization of sagittal sections using the DjMHC-A and DjMHC-B probes. Nuclei were counter-stained with Hoechst 33342 (blue). (A) A diagram of the position of sections. (B-D) Intestine and transverse muscle area. (B) Expression of DjMHC-A was observed in the muscles surrounding the intestinal duct (green). (C) Expression of DjMHC-B was observed in transverse muscles (red). (D) A double exposure of figures (B) and (C); DjMHC-A (green) and DjMHC-B (red). Dots indicate the intestinae. (E-G) Body-wall area. (E) Expression of DjMHC-A (green) (F) Expression of DjMHC-B (red). (G) A double exposure of figures (E) and (F); DjMHC-A (green) and DjMHC-B (red). I, intestinal duct. Scale bars, 50 μm.

Expression of DjMHC-A and DjMHC-B in regenerating planarians

We analyzed the expression patterns of the two MHC genes during regeneration. Intact planarians were cut behind the pharynx, and the resultant regenerating planarians were analyzed by in situ hybridization. During regeneration, the cells expressing each MHC gene in the body-wall-muscle layer increased specifically in the blastema region (Fig. 6A and B; the blastema region is surrounded by dots). Furthermore, DjMHC-B expressing cells were also increased in the region inside of the blastema region. Interesting views were obtained during the process of pharynx-cavity formation. After 5-6 days of regeneration, the pharynx cavity becomes open. Around that time, DjMHC-B-positive cells disappeared from the body-wall-muscle layer in the nascent cavity-forming region (Fig. 6D, asterisk), whereas DjMHC-A-positive cells were accumulated there (Fig. 6C, asterisk). Such complementary expression patterns suggest that DjMHC-A- and DjMHC-B-expressing cells may have distinct functions in the body-wall-muscle layer.

Fig. 6

Expression patterns of DjMHC-A and DjMHC-B during the process of regeneration from tail piece. (A, B) Sagittal sections 2 days after amputation. The blastema region is surrounded by dots. (A) Expression of DjMHC-A. (B) Expression of DjMHC-B. (C, D) Sagittal sections 5 days after amputation. Asterisks indicate the pharynx-cavity-forming region. (C) Expression of DjMHC-A. (D) Expression of DjMHC-B. Scale bars, 0.1 mm.