Red snow crab Chionoecetes japonicas (male, carapace width; 12 cm, weight; 725 g, collected at Ishikawa prefecture Japan on November in 2015) was purchased from Nakagawa Co., Ltd. (Japan). Sample was transported living and then stored at −80˚C.

The sequences of all primers used are presented in Table 1. Total RNA was extracted from the epidermis of the leg muscle of C. japonicus using ISOGEN II reagent (Nippon Gene, Tokyo, Japan) according to the manufacturer’s instructions. First-strand cDNA was synthesized using 500 ng of total RNA and oligo dT primers with PrimeScript II Reverse Transcriptase (RNase H-free) (Takara Bio, Shiga, Japan) according to the manufacturer’s instructions. Degenerate primers were designed for the reverse transcriptase-polymerase chain reaction (RT-PCR) from conserved sequences in insect and crustacean chitin deacetylases. In the first PCR, C. japonicus cDNA was used as a template and CDA F-1 and CDA R-1 were used as primers. The PCR conditions were as follows: 95˚C for 2 min, followed by 30 cycles of 95˚C for 30 s, 55˚C for 30 s, and 72˚C for 50 s. The primers used are listed in Table 1, and the primer combinations are shown in Figure 1.

For the 3' rapid amplification of cDNA ends (RACE), we designed primers specific to CjCDA (CjCDA 3'-1 and CjCDA 3'-2; Table 1) based on the detected sequences. We amplified cDNA fragments encoding the 3' region of CjCDA using

C. japonicus cDNA as the template and the primer pairs CjCDA 3'-1, 3' RACE and CjCDA 3'-2, 3' RACE. The PCR conditions were as follows: 95˚C for 2 min, followed by 30 cycles of 95˚C for 30 s, 58˚C for 30 s, and 72˚C for 80 s. For the 5' RACE, specific primers (CjCDA 5'-1 and CjCDA 5'-2; Table 1) were designed based on the nucleotide sequences obtained from the RT-PCR. Then, the cDNA fragments encoding the 5' regions of CjCDA were amplified by PCR. In the first PCR, the newly synthesized first-strand cDNA was used as a template, and AAP and CjCDA 5'-1 were used as the primers. In the nested PCR, the first PCR products were used as templates and AUAP and CjCDA 5'-2 were used as primers. The PCR conditions were as follows: 95˚C for 2 min, followed by 30 cycles of 95˚C for 30 s, 56˚C for 30 s, and 72˚C for 20 s. The nucleotide sequences of cDNA fragments containing a full-length open reading frame (ORF) were confirmed by PCR using the specific primers CjCDA Full F and CjCDA Full R (Table 1) and Platinum Pfx DNA Polymerase (Invitrogen, Carlsbad, CA).

The RT-PCR and 3' and 5' RACE amplification products as well as the full-length amplification products were subcloned into the pGEM-T Easy Vector (Promega, Madison, WI), according to the manufacturer’s instructions. The sequences were determined on an ABI PRISM 3130 genetic analyzer (Applied Biosystems, Foster City, CA) using the Big Dye Terminator v3.1 cycle sequencing kit (Applied Biosystems).

Total RNA was prepared from the muscle, epidermis, hepatopancreas, gills, shell, tendon, and intersegmental membrane as described in Section 2.2. Then, first-strand cDNA was amplified from the RNA isolated from each tissue as described above. For tissue-specific expression, we designed primers specific to CjCDA (CjCDA F and CjCDA R; Table 1) based on the detected sequences. Then, CjCDA was amplified using the first-strand cDNA as template and primers CjCDA F and CjCDA R (Table 1). The PCR conditions were as follows: 95˚C for 2 min, followed by 30 cycles of 95˚C for 30 s, 50˚C for 30 s, and 72˚C for 30 s. To determine the amount of total RNA in each tissue, we amplified β-actin mRNA using a specific primer pair as a control for normalization (Table 1).

To determine the relationship between the CDA isolated from the epidermis of C. japonicus and other CDAs from insects, crustaceans, and mollusks, we constructed a phylogenetic tree based on the sequences of enzyme precursors by the neighbor-joining method using ClustalW2 (http://www.ebi.ac.uk/Tools/msa/clustalw2/). A fungal CDA (GenBank: XM_007272808.1) was used as an outgroup.

From the epidermis of the leg muscle of red snow crab (Chionoecetes japonicas), we obtained a 2141-bp chitin deacetylase (CDA) gene containing a 1575-bp open reading frame (ORF) that encodes 525 amino acids. The sequence determined for the cDNA encoding the chitin deacetylase CjCDA was registered with the DNA Data Bank of Japan (DDBJ) database (accession no. LC342072). A poly(A) tail specific to eukaryotes was found at the 3'-terminus of CjCDA. Comparison of the deduced amino acid sequence of CjCDA (CjCDA) to CDAs in other organisms by BLAST showed the highest similarity (74%) with gastrolith protein 59 of C. quadricarinatus (CqCDA) [22] and the second highest similarity (72%) with chitin deacetylase1 of P. monodon (PmCDA) [23] . In addition, CjCDAshowed ~60% similarity was known CDAs in insects. The structural prediction for CjCDA showed an N-terminal signal peptide (amino acids 1 - 18), chitin-binding domain (CBD; amino acids 31 - 90), low-density lipoprotein receptor class A domain (LDL-A; amino acids 103 - 142), and glycoside hydrolase/deacetylase, beta/alpha-barrel (Glyco_hydro/deAcase_b/a-brl; amino acids 145 - 461) (Figure 2). Because amino acids 183 - 293 in the Glyco_hydro/deAcase_b/a-brl are presumed to be the NodB homology domain (NODB_dom) and several characteristic motifs, including TFDD, H[S/T]xxHP, RxP[Y/F], DxxDW, and GxxxFxx [13] , are present, CjCDA is considered to be a member of carbohydrate esterases family 4 (CE4). Comparison of CjCDA to the deduced amino acid sequences of CqCDA, PmCDA, and known CDAs in insects showed that CjCDA shares a very similar domain structure with these proteins (Figure 3). The CDAs in insects have been classified into five groups based on their structure (Figure 4) [24] . Groups I and II contain a signal peptide, CBD, LDL-A, and catalytic domain. Groups III and IV contain a signal peptide, CBD, and catalytic domain but no LDL-A. Group V contains only a signal peptide and catalytic domain, and these CDAs are specific to the midgut [24] . Because CjCDA contains a CBD, LDL-A, and catalytic domain, it can be classified into Group I. CqCDA and PmCDA share similar domain structure, and all crustacean

CDAs that have been discovered to date appear to belong to Group I.

Tissue expression analysis of CjCDA in C. japonicus by RT-PCR using the β-actin gene as a control revealed that CjCDA is expressed only in the epidermis of muscle (Figure 5), and no expression was observed in gills, muscle, or other examined tissues. Previous studies revealed that PmCDA in P. monodon is mainly expressed in the gills and functions in the immune response to the pathogen that causes white spot disease [23] and the CDA in the fruit fly Drosophila

melanogaster is involved in the control of trachea extension [21] . In addition, the CDAs of D. melanogaster and a species of locust, Locusta migratoria, are expressed in the cuticle [21] [25] . Because inhibition of CDA expression in a species of beetle (Tribolium castaneum) prevented ecdysis, CDA is thought to play an important role in ecdysis [24] . When arthropods molt, the epidermis separates from the cuticle and a new cuticle is secreted on the surface of the epidermis. Thus, we hypothesize that CjCDA has some function during C. japonicus molting.

As a result of phylogenetic analysis based on the deduced amino acid sequences of CDAs from various arthropods, including insects and crustaceans, as well as mollusks and the fungus Colletotrichum gloeosporioides, which was used as an outgroup, CDAs in arthropods were shown to belong in Groups I-V, as reported previously [24] . However, the CDAs from mollusks were not in Groups I-V. All CDAs from crustaceans, including CjCDA, belonged to Group I (Figure 6).

There are no reports of CDAs from crustaceans belonging to any Group other than Group I. However, because some species have multiple CDA isozymes, like T. castaneum, there is a possibility that C. japonicus may contain multiple isozymes with different structures, like T. castaneum.