Purification, characterization, cDNA cloning and nucleotide sequencing of a cellulase from the yellow-spotted longicorn beetle, Psacothea hilaris Masahiro Sugimura 1 , Hirofumi Watanabe 1 , Nathan Lo 1 and Hitoshi Saito 2 1 National Institute of Agrobiological Sciences, Ibaraki, Japan; 2 Department of Applied Biology, Faculty of Textile Science, Kyoto Institute of Technology, Japan A cellulase (endo-b-1,4-glucanase, EC 3.2.1.4) was purified from the gut of larvae of the yellow-spotted longicorn beetle Psacothea hilaris by acetone precipitation and elution from gels after native PAGE and SDS/PAGE with activity staining. The purified protein formed a single band, and the molecular mass was estimated to be 47 kDa. The purified cellulase degraded carboxymethylcellulose (CMC), insoluble cello-oligosaccharide (average degree of polymerization 34) and soluble cello-oligosaccharides longer than cellotriose, but not crystalline cellulose or cellobiose. The specific activity of the cellulase against CMC was 150 lmolÆ min )1 Æ(mg protein) )1 . TLC analysis showed that the cellu- lase produces cellotriose and cellobiose from insoluble cello- oligosaccharides. However, a glucose assay linked with glucose oxidase detected a small amount of glucose, with a productivity of 0.072 lmolÆmin )1 Æ(mg protein) )1 .The optimal pH of P. hilaris cellulase was 5.5, close to the pH in the midgut of P. hilaris larvae. The N-terminal amino-acid sequence of the purified P. hilaris cellulase was determined and a degenerate primer designed, which enabled a 975-bp cDNA clone containing a typical polyadenylation signal to be obtained by PCR and sequencing. The deduced amino- acid sequence of P. hilaris cellulase showed high homology to members of glycosyl hydrolase family 5 subfamily 2, and, in addition, a signature sequence for family 5 was found. Thus, this is the first report of a family 5 cellulase from arthropods. Keywords: cDNA cloning; cellulase; endoglucanase; insect; purification. Cellulase (endo-b-1,4-glucanase) is a widespread enzyme in micro-organisms such as bacteria and fungi [1,2]. Until recently it was believed that cellulose digestion in animals was mediated by microbial cellulase activity in their intestine, and that no animals possessed endogenous cellulase. This traditional view of cellulase activity in animals was challenged by two reports of endogenous animal cellulase genes from plant-parasitic nematodes and a termite [3,4]. Since these discoveries, a number of other animal cellulase genes have been reported (summarized in [5]). Glycosyl hydrolases are categorized into 90 families according to amino-acid sequence similarity and hydropho- bic cluster analysis, and among them, cellulases are found in 14 families [6,7] (refer also to: http://afmb.cnrs-mrs.fr/ CAZY/index.html). Known animal cellulases belong to three glycosyl hydrolase families (GHFs): GHF 5 (plant- parasitic nematodes), GHF 9 (termites, cockroaches and crayfish) and GHF 45 (mussel and beetle). These three families are structurally unrelated and their evolutionary origins are likely to be independent. Larvae of the yellow-spotted longicorn beetle, Psacothea hilaris, feed on mulberry and fig trees, tunneling inside the stems and ingesting the living wood. The major constituent is cellulose (44.6%), followed by hemicellulose (28.5%); soluble sugars constitute only 4.7% of the dry weight of the wood [8]. The habitat of P. hilaris larvae suggests that they possess the ability to digest cellulosic materials. In fact, a variety of carbohydrase activities, including endo-b-1, 4-glucanase and b-glucosidase, have been detected in the gut of P. hilaris larvae and adults [8]. To clarify further the cellulase activity in P. hilaris,we purified, characterized and obtained the cDNA sequence of a protein from this species with cellulolytic activity. Materials and Methods Measurement of the pH in the gut juice of P. hilaris larvae P. hilaris larvae were anesthetized by immersion in ice water for 10 min and their guts, including their contents, were removed by dissection. The removed guts were washed in ice-cold distilled water to prevent contamination by body fluid and blotted on filter paper. Then, the guts were cut into three parts (anterior midgut, posterior midgut and hindgut) and transferred into 1.5 mL plastic centrifuge tubes. The samples in the tubes were centrifuged, and insoluble materials were discarded. The pH values of the recovered Correspondence to H. Watanabe, National Institute of Agrobiological Sciences, Tsukuba, Ibaraki 305-8634, Japan. Tel./Fax: + 81 29 8386108, E-mail: hinabe@affrc.go.jp Abbreviations: GHF, glycosyl hydrolase family; CMC, carboxy- methylcellulose; CBB, Coomassie Brilliant Blue. Enzymes: endo-b-1,4-glucanase (EC 3.2.1.4). (Received 30 January 2003, revised 22 May 2003, accepted 30 June 2003) Eur. J. Biochem. 270, 3455–3460 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03735.x supernatants were measured with a pen-type pH meter (model B-212; Horiba, Kyoto, Japan). Enzyme assay A carboxymethylcellulase (CMCase) assay was performed by measuring the amount of reducing sugars after incuba- tion of 100 lL 1% (w/v) CMC (standard molecular mass, 250 kDa; degree of carboxymethyl substitution, 0.7; Sigma- Aldrich) in 0.1 M sodium acetate (pH 5.5) with 20 lL sample at 37 °C for an appropriate time period. Reducing sugars were measured with tetrazolium blue (Sigma- Aldrich) as a chromogenic reagent, with glucose as a stand- ard, as described by Jue & Lipke [9]. Glucose production from insoluble cello-oligosaccharide by the P. hilaris cellu- lase was investigated with a commercial glucose assay kit (Glucose C-test; Wako Pure Chemicals), which utilizes glucose oxidase in an enzyme-linked colorization step, according to the instructions. Insoluble cello-oligosaccha- ride was prepared by the method of Sawano et al. [10], and the average degree of polymerization was  34. To test the degradation activity of P. hilaris cellulase against crystalline cellulose, Avicel (Merck) was used under the same condi- tions as the CMCase assay. Optimal pH for P. hilaris cellulase activity against CMC was determined with 0.1 M sodium acetate (4–6), sodium phosphate (5.5–7.5) or Tris/ HCl (7–9.5). Hydrolytic products from cello-oligosaccha- rides (cellobiose, cellotriose, cellotetraose and cellopentaose) were analyzed by TLC. Electrophoresis and activity staining SDS/PAGE was performed as described by Laemmli [11], and native PAGE was carried out in the same way except that SDS was excluded. Proteins were stained with CBB, and cellulase was visualized by activity staining on the gel. In the case of activity staining, the gel contained 0.1% (w/v) CMC and the conditions of sample treatment for SDS/ PAGE were changed from boiling for 5 min to incubating at 33 °C for 30 min. After being run, the gel was twice rinsed in distilled water with gentle shaking for 5 min, and soaked in 0.1 M sodium acetate buffer (pH 5.5) for 5–20 min. The gel was briefly rinsed with distilled water before staining with 0.2% (w/v) Congo red (Sigma-Aldrich) in water for 30 min. The excess dye was removed in 1 M NaCl with gentle shaking and replacement of the solution several times. Purification procedure P. hilaris larvae reared on a commercial artificial diet (Insecta LF; Nihon Nosan Kogyo, Yokohama, Japan) were dissec- ted, and whole guts were obtained. The guts from six larvae were homogenized in 4.5 mL sodium acetate buffer (0.1 M , pH 5.5) using a glass homogenizer, and centrifuged at 10 000 g for 10 min. The supernatant was mixed with 3 vol. cold ()20 °C) acetone by gentle stirring and then kept in a freezer ()35 °C) for 15 min. The supernatant was decanted after centrifugation at 20 000 g for 5 min, and the pellet was dissolved in a minimum volume of 62.5 m M Tris/HCl (pH 6.8). The suspension was centrifuged at 20 000 g for 5 min to remove insoluble materials, and then the super- natant was loaded on a gel for native PAGE and activity staining. After electrophoresis, the gel was rinsed twice in distilled water with gentle shaking for 5 min and divided into two parts: (a) a gel strip for activity staining; (b) the remaining gel sheet for elution of proteins. The gel sheet was stored at 4 °C until use, and the gel strip was subjected to activity staining. The position of cellulase activity was identified, and the corresponding position of the gel sheet was cut out, sliced into small sections ( 1 mm cubes) and transferred to 1.5 mL centrifugal micro tubes. The micro tubes were filled with distilled water and kept at 4 °C overnight to allow elution of the proteins from the gel sections. The tubes were centrifuged at 10 000 g for 5 min to sediment the gel sections, and the supernatants were transferred to a commercial disposable device for ultrafiltration (UltrafreeÒ-MC 10 000 NMWL centrifugal filter unit, Millipore), which had a polysulfone membrane with a 10-kDa exclusion size. The filter unit was centrifuged at 5000 g until the remaining volume decreased below 40 lL. The remaining solution, in which proteins were concentrated, was mixed with an equal volume of Laemmli Sample Buffer (Bio-Rad) without the addition of 2-mercaptoethanol before incubation at 33 °C for 30 min. Then, proteins were subjected to SDS/PAGE. After electrophoresis, proteins including the cellulase activity were eluted from the gel and concentrated as for native PAGE. The purity of the eluted proteins was checked by further SDS/PAGE with CBB staining. Analysis of N-terminal amino-acid sequence The purified cellulase from P. hilaris larvae was subjected to SDS/PAGE and transferred to a poly(vinylidene difluo- ride) membrane (Bio-Rad) in transfer buffer [48 m M Tris, 39 m M glycine, 0.0375% (w/v) SDS, pH 9.2] by using a TRANS-BLOTÒ SD Semi-dry Transfer Cell (Bio-Rad). Proteins were visualized with CBB, cut out, and subjected to gas phase protein sequencing (model LF-3400 DT; Beckman). cDNA cloning, genomic PCR and sequencing A QuickPrepÒ Micro mRNA Purification Kit (Amersham Bioscience) was used for isolation of mRNA from P. hilaris larval midguts. First-strand cDNA synthesis from the isolated mRNA and the following amplification of the target cDNA were performed with a SMART TM RACE cDNA Amplification Kit (Clontech) according to the manufacturer’s instruction except that SuperScript II reverse transcriptase (Invitrogen) was used. A degenerate oligonucleotide primer (5¢-GTICARGGIGTITGYATHG TIGAYG-3¢) was designed based on the N-terminal amino- acid sequence to amplify the cDNA for P. hilaris cellulase. RACE amplification of the 3¢-end was performed with this degenerate primer and an anchor primer corresponding to the anchor sequence combined to the 3¢-end of the oligo- dT primer for first-strand synthesis of cDNA [12], and 5¢-RACE amplification was performed with a gene-specific primer based on the sequence of the 3¢-fragment. The nucleotide sequence was determined by using a BigDyeTer- minator cycle sequencing kit and an ABI3700 automated DNA sequencer (Applied Biosystems). Sequence similarities were determined by a BLAST search (http://www.ncbi.nlm. nih.gov/BLAST/). Forward (F) and reverse (R) primers, 3456 M. Sugimura et al.(Eur. J. Biochem. 270) Ó FEBS 2003 1–24 (F), 209–230 (F), 289–311 (R), 446–469 (R), 651–680 (R), 667–688 (F), 713–740 (F), 781–807 (R) and 995–1013 (R), were designed from the cDNA sequence and used in various combinations of genomic PCRs of P. hilaris fat body tissue, extracted as described previously [13]. The conditions for PCR were 35 cycles of 94 °Cfor1min,52°C for 1 min, 72 °C for 3 min. Care was taken with the solutions, and pipettes were used to avoid potential contamination from previously prepared cDNAs. Protein assay Protein concentration was determined by using a protein assay kit (CoomassieÒ Plus Protein Assay Reagent; Pierce) with BSA as a standard. TLC analysis TLC was performed with silica gel 60 (Merck) in a solvent system of butan-1-ol/acetic acid/water (2 : 1 : 1, by vol.), and sugars were visualized by a heat treatment at 120 °Cfor 10 min after the spraying of 50% (v/v) H 2 SO 4 in methanol. Results Purification of P. hilaris cellulase from larval guts The gut homogenate of P. hilaris larvae was precipitated with acetone. The acetone treatment was effective in reducing sample volume while minimizing loss of cellulase activity. On native PAGE, although a number of proteins from the acetone-treated samples were detected by CBB staining, only one band was detected by activity staining (data not shown). After elution and concentration of the protein solution containing the cellulase activity after native PAGE, the solution was mixed with sample buffer for SDS/ PAGE without reducing agent and incubated at 33 °Cfor 30mintoloadanSDS/polyacrylamidegel.TheSDS/ PAGE analysis detected three proteins bands by CBB staining and one activity band by activity staining. Proteins including cellulase activity were eluted and concentrated from the unstained gel, and the protein solution was incubated with sample buffer for SDS/PAGE, adding reducing agent at 100 °C for 5 min. Then, the protein solution was again subjected to SDS/PAGE. A single 47-kDa protein band was detected by CBB staining, which indicated a successful purification of cellulase (Fig. 1). Optimal pH for P. hilaris cellulase activity and pH in the gut juice of P. hilaris larvae The effect of pH on P. hilaris cellulase activity was tested with 0.1 M sodium acetate, sodium phosphate and Tris/HCl buffers. For a particular pH, the use of different buffers did not markedly alter enzyme activity, nor did different concentrations of these buffers. The optimal pH for cellulase activity against CMC was 5.5. Although the cellulase activity was greatly reduced at pH values above 7, some activity remained at pH 9.5 (data not shown). No cellulase activity was observed at pH values less than 4.0. The digestive tract of P. hilaris larvae was found to be folded, and its total length was  1.5 times its body length. The boundary between the midgut and hindgut was indistinct, and no specialized hindgut structure such as that of symbiont-possessing insects such as termites and scarab- aeid beetles was observed. The gut contents were semisolid and scarcely flowed out upon dissection. The pH values estimated in the anterior midgut, posterior midgut and hindgut were 5.7, 5.9 and 7.7, respectively. Enzymatic degradation of cellulose and its derivatives by P. hilaris cellulase To investigate the ability of P. hilaris cellulase to degrade cellulose and its derivatives, CMC (soluble in water but carboxymethylated), insoluble cello-oligosaccharide (aver- age degree of polymerization 34), Avicel (crystalline cellu- lose), cellobiose, cellotriose, cellotetraose and cellopentaose were used as substrates. P. hilaris cellulase readily degraded CMC, and the specific activity was determined to be 150 lmolÆmin )1 Æ(mg protein) )1 . Insoluble cello-oligosaccha- ride was also readily degraded, and the degradation products were investigated by TLC. Cellobiose and cello- triose were detected but not glucose (Fig. 2A). However, with the use of a glucose assay kit, a small amount of glucose was detected as a degradation product from insoluble cello-oligosaccharide, and the glucose productivity was 0.072 lmolÆmin )1 Æ(mg protein) )1 . Both cellotetraose Fig. 1. SDS/PAGE analysis of the purified cellulase from larval gut juice of P. hilaris . Lane 1, molecular mass standards consisting of myosin (200 kDa), b-galactosidase (116 kDa), phosphorylase B (97.4 kDa), BSA (66 kDa), ovalbumin (45 kDa), carbonic anhydrase (31 kDa), trypsin inhibitor (21.5 kDa), lysozyme (14.4 kDa) and aprotinin (6.5 kDa). Lane 2, purified cellulase. Proteins were stained with CBB R-250. Ó FEBS 2003 Cellulase of the yellow-spotted longicorn beetle (Eur. J. Biochem. 270) 3457 and cellopentaose were degraded to cellobiose and cellotri- ose (Fig. 2B). Production of cellobiose and cellotriose from cellopentaose by degradation is natural but production of cellobiose and cellotriose from cellotetraose is abnormal. Cellotetraose should be degraded to produce two cellobiose molecules or a combination of a cellotriose and glucose. To obtain information about the missing glucose, glucose- digesting activity was tested. No glucose-digesting activity was detected in the purified P. hilaris cellulase, suggesting that contamination by enzymes that digest glucose was not responsible for the missing glucose. Transglycosylation activity, known as the reverse reaction of some endo- glucanases, was examined. Very low activity was detected in the purified P. hilaris cellulase (data not shown). Avicel and cellobiose were not degraded by P. hilaris cellulase, and cellotriose was partially degraded to cellobiose. Analysis of N-terminal amino-acid sequence, cDNA sequence and deduced amino-acid sequence The N-terminal amino-acid sequence of the purified P. hi- laris cellulase was analyzed. The sample blotted to a poly(vinylidene difluoride) membrane after SDS/PAGE was used, and 30 amino acids from the N-terminus were determined as follows: KDAAL ETVSK HGQLS VQGVD IVDES GEKVQ. A degenerate primer was designed based on the amino acids determined, and an  0.9-kbp 3¢-frag- ment was amplified. The flanking region for the 5¢-end of the cDNA was obtained by 5¢-RACE with a gene specific primer based on the sequence of the first fragment. A full- length of cDNA clone encoding a cellulase gene was obtained and sequenced. The nucleotide sequence was deposited in GenBank (accession number is AB080266). The cDNA encoding P. hilaris cellulase contained an ORF 975 bp long, starting with an ATG codon at position 38 and ending with a TAA codon at position 1014. A poly(A) tail and typical polyadenylation signal were found. Two poten- tial N-glycosylation sites, N270 and N300, were detected in the deduced amino-acid sequence. The ORF consisted of a protein of 325 amino acids. The molecular mass of P. hilaris cellulase was calculated to be 36.0 kDa, and the first 21 amino acids were predicted to be a signal sequence for secretion. The molecular mass of the mature enzyme was deduced to be 33.8 kDa. The P. hilaris cellulase consisted of a single catalytic module only and no carbohydrate binding module was found. BLAST searches with the deduced amino-acid sequence indicated that P. hilaris cellulase was closely related to nematode cellulases and some bacterial cellulases, those belonging to GHF5 subgroup 2. The overall identities and similarities of P. hilaris cellulase to GHF 5 subgroup 2 members were: 49% and 67% to Pseudomonas fluorescence CelE, 49% and 66% to Meloidogyne incognita MI-ENG1, respectively. A GHF 5 signature sequence, [LIV]-[LIVMFYWGA](2)-[DNEQG]- [LIVMGST]-x-N-E-[PV]-[RHDNSTLIVFY], was con- served in the deduced amino-acid sequence of P. hilaris cellulase (Fig. 3). The conserved potential proton donor and Fig. 2. Substrate specificity and degradation products of P. hilaris cellulase. TLC was performed with the solvent system butan- 1-ol/acetic acid/water (2 : 1 : 1, v/v/v.). Sugars were visualized by incubating the plate at 120 °C for 10 min after spraying with 50% (v/v) H 2 SO 4 in methanol. (A) Lane 1, standard sugars: glucose (G1), cellobiose (G2), cellotri- ose (G3) and cellotetraose (G4). Lane 2, insoluble cello-oligosaccharides. Lane 3, purified P. hilaris cellulase. Lane 4, insoluble cello-oligosaccharides treated with P. hilaris cellulase at 37 °C for 8 h. (B) Lane 1, standard sugars: G1, G2, G3, G4, cellopentaose (G5) and cellohexaose (G6). Lanes 2–5, G2–G5 treated with P. hilaris cellulase at 37 °Cfor 2h. Fig. 3. Comparison of amino-acid sequences for potential catalytic proton donor and nucleophile regions of P. hilaris cellulase (AB080266), Globodera rostochiensis ENG1 (GR-ENG1, AF004523), Meloidogyne incognita ENG1 (MI-ENG1, AF100549), Pseudomonas fluorescence CelE (CelE, X86798) and Erwinia chrysanthemi CelZ (CelZ, Y00540). Alignment was performed with the computer program CLUSTAL X (version 1.81) using the catalytic module of each cellulase. Residue numbers are given on the left of the sequences. Amino acids with similar groups of side chains and identical amino acids in sequences are indicated by Ô:Õ or Ô*Õ, respectively, above the sequence. ÔPÕ and ÔNÕ below the sequence indicate potential catalytic proton donor and nucleophile amino acids, respectively. 3458 M. Sugimura et al.(Eur. J. Biochem. 270) Ó FEBS 2003 nucleophile amino acids are also found in the sequence (Fig. 3). Genomic PCR experiments with a variety of different primers designed from the cDNA sequence resulted in the amplification of bands in each case. Sequencing of these bands showed that they matched the sequence of the cellulase cDNA completely; however, no introns were found in any case. Discussion Although optimal pH values for cellulase activity vary from acidic to alkaline [14–17], all animal cellulases reported until now have optimal activity under weak acidic conditions [4,18,19]. The optimal pH for the purified cellulase from the larval gut of P. hilaris against CMC was also 5.5. This is reasonable for the physiological function of cellulase activity in larval guts as the pH values in the anterior midgut, posterior midgut and hindgut were 5.7, 5.9 and 7.7, respectively. These results suggest that P. hilaris cellulase functions mainly in the midgut, which is generally thought of as a digestive and absorptive organ in the insect alimentary canal. The purified P. hilaris cellulase showed no degradation activity against crystalline cellulose, which suggests that P. hilaris larvae may utilize only the amorphous parts of cellulose materials ingested. The elongated (about 1.5 times its body length) digestive tract of P. hilaris presumably means that the ingested cellulose material is exposed to enzymatic digestion for long periods. Animals in general absorb sugars in monomeric forms such as glucose and fructose [20,21]. Although the degra- dation products of the P. hilaris cellulase were found to be cellotriose and cellobiose, glucose would be produced by b-glucosidase activity in the larval gut, which has been previously demonstrated in P. hilaris [8]. TLC analysis detected cellotriose and cellobiose after cellotetraose was treated with the purified P. hilaris cellulase (Fig. 2B). A molecule of cellotetraose should be degraded into two cellobiose molecules, or a cellotriose molecule and a glucose molecule, by a single catalytic event. Therefore, after cellotetraose is degraded, glucose equivalent to cello- triose should be detected. However, only cellotriose was detected by TLC. Similarly, after cellotriose was degraded by the P. hilaris cellulase, TLC detected cellobiose but not glucose (Fig. 2B). It is known that some endoglucanases possess transglycosylation activity as the reverse reaction [22]. Oikawa et al. [23] reported that the addition of acetone to the reaction buffer increased transglycosylation activity of the Rhodotorula glutinis cellulase. Observation of trans- glycosylation activity in the purified P. hilaris cellulase was attempted under various conditions, including the addition of acetone to the reaction mixture. Although transglycosy- lation activity was detected, it was not enough to explain the lack of glucose. However, transglycosylation is the most probable explanation for the missing glucose, because there are few other potential mechanisms for eliminating it. ThemolecularmassofP. hilaris cellulase deduced from its DNA sequence is 36.0 kDa. The apparent molecular mass of the purified P. hilaris cellulase was, however, estimated to be 47 kDa from its mobility on SDS/PAGE. N-Terminal amino-acid sequencing analysis indicated that thematureproteinofP. hilaris cellulase was a truncated form, which lacked a signal peptide composed of the first 21 amino acids. Therefore, the deduced molecular mass of mature cellulase protein is 33.8 kDa and the difference is 13.2 kDa. This inconsistency may be explained by a possible post-transcriptional modification at two potential N-glyco- sylation sites, N270 and N300, in the amino-acid sequence of P. hilaris cellulase. Alternatively, the mobility of the P. hilaris cellulase protein on SDS/PAGE may be different from those of the standard proteins used in the experiment. A cDNA encoding a cellulase, which belongs to GHF 45, has been cloned from a gut library from the phytophagous beetle, Phaedon cochleariae [24]. This GHF 45 cellulase has been confirmed to be expressed in the gut of P. cochleariae. P. cochleariae and P. hilaris are closely related species, belonging to the same superfamily, Chrysomeloidea. There- fore, a GHF 45 cellulase might have been expected from P. hilaris. However, the cellulase purified from P. hilaris in the current study was shown to belong to GHF 5. The activity staining indicated that there were no other CMCase activities except the GHF 5 cellulase in the larval gut of P. hilaris. The sensitivity of the activity staining used is high, and CMCase activity has been detectable with 5 lL of 1000 times diluted gut juice of P. hilaris larvae. Therefore, if a GHF 45 enzyme is present, its level of expression would be extremely low in P. hilaris larvae. Glycosyl hydrolases have been categorized into 90 families according to homologies of their amino-acid sequences, and cellulases are distributed into 14 families (http://afmb.cnrs-mrs.fr/CAZY/index.html). Most of these 14 families are composed of cellulases only, but some include other enzymes, such as xylanase and mannanase. P. hilaris cellulase was proposed to belong to GHF 5 and the signature sequence of GHF 5 was found in the amino- acid sequence (Fig. 3). GHF 5 is the largest GHF and includes endoglycosylceramidase, b-mannanase, exo-b-1,3- glucanase, endo-1,6-b-glucosidase, b-xylanase and some other enzymes, in addition to cellulase (endo-b-1,4-gluca- nase). On the basis of sequence homology, GHF 5 enzymes can be further divided into five subfamilies [1,25]. P. hilaris cellulase is closely related to subfamily 2 members, which is composed only of cellulases from bacteria, fungi and nematodes. It has been shown that the nematode cellulase genes contain several introns, and the positions are conserved in the deduced amino-acid sequences [26]. Genomic DNA of P. hilaris was amplified and sequenced to determine whether the cellulase gene contained introns; however, none were found, despite combinations with several primers. As intronless genes appear to be fairly common among insects [27,28], the lack of introns in the P. hilaris cellulase gene does not provide evidence for its recent horizontal transfer from a prokaryote. A discussion of the evolutionary origins of the cellulase enzymes in animals will be given elsewhere [29]. Acknowledgements We thank Ms. Sanae Wada for advice about rearing P. hilaris larvae. This work was supported by the Promotion of Basic Research Activities for Innovative Biosciences Fund from the Bio-oriented Technology Research Advancement Institution (BRAIN; Omiya, Saitama, 331-8537 Japan; www.brain.go.jp) and by the Pioneer Research Project Fund (No. PRPF-0022) from the Ministry of Ó FEBS 2003 Cellulase of the yellow-spotted longicorn beetle (Eur. J. Biochem. 270) 3459 Agriculture, Forestry and Fisheries of Japan. N.L. is supported by a Science and Technology Agency of Japan Postdoctoral Fellowship. References 1. Tomme, P., Warren, R.A. & Gilkes, N.R. (1995) Cellulose hydrolysis by bacteria and fungi. Adv. Microb. Physiol. 37, 1–81. 2. Be ´ guin, P. & Aubert, J.P. (1994) The biological degradation of cellulose. FEMS Microbiol. Rev. 13, 25–58. 3. Smant, G., Stokkermans, J.P., Yan, Y., de Boer, J.M., Baum, T.J., Wang, X., Hussey, R.S., Gommers, F.J., Henrissat, B., Davis, E.L., Helder, J., Schots, A. & Bakker, J. (1998) Endogenous cel- lulases in animals: isolation of beta-1, 4-endoglucanase genes from two species of plant-parasitic cyst nematodes. Proc. Natl. Acad. Sci. USA 95, 4906–4911. 4. Watanabe, H., Noda, H., Tokuda, G. & Lo, N. (1998) A cellulase gene of termite origin. Nature (London) 394, 330–331. 5. Watanabe, H. & Tokuda, G. (2001) Animal cellulases. Cell. Mol. Life Sci. 58, 1167–1178. 6. Henrissat, B.A. (1991) Classification of glycosyl hydrolases based on amino acid sequence similarities. Biochem. J. 280, 309–316. 7. Henrissat, B. & Bairoch, A. (1993) New families in the classifica- tion of glycosyl hydrolases based on amino acid sequence simila- rities. Biochem. J. 293, 781–788. 8. Scrivener, A.M., Watanabe, H. & Noda, H. (1997) Diet and carbohydrate digestion in the yellow-spotted longicorn beetle Psacothea hilaris. J. Insect Physiol. 43, 1039–1052. 9. Jue, C.K. & Lipke, P.N. (1985) Determination of reducing sugars in the nanomole range with tetrazolium blue. J. Biochem. Biophys. Methods 11, 109–115. 10. Sawano, M., Sakamoto, R. & Arai, M. (1988) Cellulases of Aspergillus aculeatus. Methods Enzymol. 160, 274–299. 11. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 224, 680–685. 12. Tokuda, G., Saito, H. & Watanabe, H. (2002) A digestive beta- glucosidase from the salivary glands of the termite, Neotermes koshunensis (Shiraki): distribution, characterization and isolation of its precursor cDNA by 5¢-and3¢-RACE amplifications with degenerate primers. Insect Biochem. Mol. Biol. 32, 1681–1689. 13. Tokuda, G., Lo, N., Watanabe, H., Slaytor, M., Matsumoto, T. & Noda, H. (1999) Metazoan cellulase genes from termites: intron/ exon structures and sites of expression. Biochim. Biophys. Acta 1447, 146–159. 14. Hurst,P.L.,Nielsen,J.,Sullivan,P.A.&Shepherd,M.G.(1977) Purification and properties of a cellulase from Aspergillus niger. Biochem. J. 165, 33–41. 15. Park, S.R., Cho, S.J., Kim, M.K., Ryu, S.K., Lim, W.J., An, C.L., Hong,S.Y.,Kim,J.H.,Kim,H.&Yun,H.D.(2002)Activity enhancement of Cel5Z. from Pectobacterium chrysanthemi PY35 by removing C-terminal region. Biochem. Biophys. Res. Commun. 291, 425–430. 16. Ito, S., Shikata, S., Ozaki, K., Kawai, S., Okamoto, K., Inoue, S., Takei, A., Ohta, Y. & Satoh, T. (1989) Alkaline cellulase for laundry detergent: production by Bacillus sp. KSM-635 and enzymatic properties. Agric. Biol. Chem. 53, 1275–1281. 17. Fukumori, F., Kudo, T. & Horikoshi, K. (1985) Purification and properties of a cellulase from alkalophilic Bacillus sp, 1139. J. Gen. Microbiol. 131, 3339–3345. 18. Be ´ ra-Maillet, C., Arthaud, L., Abad, P. & Rosso, M.N. (2000) Biochemical characterization of MI-ENG1, a family 5 endo- glucanase secreted by the root-knot nematode Meloidogyne incognita. Eur. J. Biochem. 267, 3255–3263. 19. Xu, B., Hellman, U., Ersson, B. & Janson, J.C. (2000) Purification, characterization and amino-acid sequence analysis of a thermo- stable, low molecular mass endo-beta-1,4-glucanase from blue mussel, Mytilus edulis. Eur. J. Biochem. 267, 4970–4977. 20. Levin, R.J. (1994) Digestion and absorption of carbohydrates – from molecules and membranes to humans. Am.J.Clin.Nutr.59, 690S–698S. 21. Turunen, S. & Crailsheim, K. (1996) Lipid and sugar absorption. In Biology of the Insect Midgut (Lehane, M.J. & Billingsley, P.F., eds), pp. 293–320. Chapman & Hall, London. 22. Rye, C.S. & Withers, S.G. (2000) Glycosidase mechanisms. Curr. Opin. Chem. Biol. 4, 573–580. 23. Oikawa, T., Tsukagawa, Y., Chino, M. & Soda, K. (2001) Increased transglycosylation activity of Rhodotorula glutinis endo-beta-glucanase in media containing organic solvent. Biosci. Biotechnol. Biochem. 65, 1889–1892. 24. Girard, C. & Jouanin, L. (1999) Molecular cloning of cDNAs encoding a range of digestive enzymes from a phytophagous beetle, Phaedon cochleariae. Insect Biochem. Mol. Biol. 29, 1129– 1142. 25. Be ´ guin, P. (1990) Molecular biology of cellulose degradation. Annu. Rev. Microbiol. 44, 219–248. 26. Yan, Y., Smant, G., Stokkermans, J., Qin, L., Helder, J., Baum, T., Schots, A. & Davis, E. (1998) Genomic organization of four beta-1,4-endoglucanase genes in plant-parasitic cyst nematodes and its evolutionary implications. Gene 220, 61–70. 27. Imamura, M. & Yamakawa, M. (2002) Molecular cloning and expression of a Toll receptor gene homologue from the silkworm, Bombyx mori. Biochim. Biophys. Acta 1576, 246–254. 28. Da Lage, J.L., Maczkowiak, F. & Cariou, M.L. (2000) Molecular characterization and evolution of the amylase multigene family of Drosophila ananassae. J. Mol. Evol. 51, 391–403. 29. Lo, N., Watanabe, H. & Sugimura, M. (2003) Evidence for the presence of a cellulase gene in the last common ancestor of bila- terian animals. Proc.R.Soc.Lond.B(in press). 3460 M. Sugimura et al.(Eur. J. Biochem. 270) Ó FEBS 2003 . Purification, characterization, cDNA cloning and nucleotide sequencing of a cellulase from the yellow-spotted longicorn beetle, Psacothea hilaris Masahiro. larvae and adults [8]. To clarify further the cellulase activity in P. hilaris, we purified, characterized and obtained the cDNA sequence of a protein from