BmCalpains are involved in autophagy and apoptosis during metamorphosis and after starvation in Bombyx mori.
The first identified calpains constituted a family of intracellular Ca-dependent cysteine proteases that have been associated with a wide variety of physiological processes and pathological conditions (Sorimachi et al. 1997; Suzuki et al. 1995). In vertebrates, eight different members of this family have been identified: the ubiquitous m (CAPN1), m (CAPN2), m/m, and the tissue-specific p94 (CAPN3), nCL-2 (CAPN4), CAPN5, CAPN6, and nCL-4 calpains. The discovery of new vertebrate calpains widens the potential physiological functions of members of the family. Two members, CAPN5 (also called htra3) (Dear et al. 1997; Matena et al. 1998; Mugita et al. 1997) and CAPN6 (Dear et al. 1997; Matena et al. 1998), differ from other members in that the C-terminal domains of both proteins, termed domain T, possess no homology to the calmodulin-like calcium-binding C-terminal domain of other calpains and, thus, the proteins may not be calcium-dependent. Furthermore, CAPN6 lacks residues believed to be critical for the active site and may lack protease activity. The 30K small subunit that is required by CAPN1 and CAPN2 for protease activity (Graham-Siegenthaler et al. 1994) may also not associate with these atypical calpains. In order to further investigate the complexity of the calpain family, we have attempted to identify other vertebrate members. Here we report a novel mouse gene encoding a calpain significantly different from all others identified. We searched the EST database at the National Center for Biotechnology Information (www.ncbi.nlm.nih.gov), using the BLAST algorithm (Altschul et al. 1990) with nucleic acid and protein sequences of known vertebrate calpains. The predicted protein product of mouse EST AA450755 (The WashU-HHMI Mouse EST Project) showed significant homology to m-calpain and had been reported to show greatest homology to the predicted nematode calpain T04A8.16. Using this EST sequence, specific oligonucleotides were designed and used in the RACE procedure (Frohman et al. 1988) to obtain 58 and 38 cDNA nucleotide sequence information, which was finally assembled into a single contig of 2661 nucleotides. One large open reading frame exists in this sequence that could encode a protein of 769 amino acids (Mr of 87.5 kDa). The amino acid sequence shows greatest homology to the predicted nematode calpain T04A8.16 (39.3%). Homology to the vertebrate calpains varied from 9.1% (chicken p94) to 11.7% (mouse Capn6, rat nCL-2). The homology to vertebrate calpains is likely to reflect functional conservation as the conserved amino acids are also highly conserved between all the vertebrate calpains (Fig. 1A). The predicted amino acid sequence contains the three amino acid residues (Cys246, His414, and Asn434) that are part of the calpain active site and are essential for m-calpain protease activity (Arthur et al. 1995). The C-terminal domain of the predicted protein, however, shows no significant homology to either the calmodulin-like domain IV of the ‘traditional’ calpains or to the domain T of CAPN5 and CAPN6. A search using the PROSITE database at the European Bioinformatics Institute (www.ebi.ac.uk) did not reveal any putative EF-hand (Cabinding) domains in the protein. We have termed the gene Capn7 in accordance with the approved mouse nomenclature, and the novel C-terminal domain, domain N. Phylogenetic analysis of representative members of each of the calpain paralogs showed that Capn7 is the most divergent of the vertebrate calpains identified to date (Fig. 2). Northern blot analysis with a limited number of mouse tissues revealed a single transcript of approximately 4 kb hybridizing with a P-labeled full-length Capn7 cDNA. Hybridization of the same probe under identical conditions to a Clontech Mouse RNA Master Blot containing 22 different mouse tissues RNAs revealed that the gene appears to be ubiquitously expressed with similar mRNA levels in all tissues examined (data not shown). A search of the GenBank DNA sequence database with the domain N amino acid sequence and the TBLASTN algorithm revealed, surprisingly, that this domain possesses significant homology to the domain III of other calpains. Maximum homology (24.1% in a 125 amino acid overlap) is to domain III of mouse Capn5 (Figure 1B). In fact, the homology between mouse Capn7 domain N and the domain III of other vertebrate calpains is higher than the homology between Capn7 domain III and the domain III of other calpains (maximum homology of 12.8% in a 127 amino acid overlap with mouse Capn6 domain III). Similarly, the nematode T04A8.16 domain N exhibits significant homology with domain III of vertebrate calpains (maximum homology of 18.8% in a 143-amino acid overlap with mouse Capn2 domain III). This may represent some functional conservation between domains III and N, because many of the conserved residues between the two are also conserved in domain III of all vertebrate calpains. It is possible that the domain N arose by a tandem duplication of the domain III-encoding DNA in an ancestral calpain gene, predating the divergence of the invertebrate and vertebrate lineages. There is, however, no similarity in location or number of intron/exon splice junctions between mouse Capn7 domain III and domain Nencoding DNA with domain III containing four splice sites and domain N only two (Figure 1B). Nevertheless, tandem duplication of domain III-encoding DNA may still have occurred in an ancestral gene but must then have been followed by divergence of the intron/exon splice sites. The absence of any obvious Ca-binding sites in Capn7 suggests that the protein may not bind Ca. Furthermore, the 30K calpain small subunit interacts with the large subunit of calpains through domain IV (Crawford et al. 1993), and, since Capn7 lacks this domain, it is unlikely to be able to bind the small subunit. The conservation of this domain N between vertebrates and invertebrates argues for an important function; the function of domain III remains unknown although recent evidence suggests that in Correspondence to: T.N. Dear