Purification and cDNA cloning of a cellulase from abalone
Haliotis discus hannai
Ken-ichi Suzuki, Takao Ojima and Kiyoyoshi Nishita
Laboratory of Biochemistry and Biotechnology, Graduate School of Fisheries Sciences, Hokkaido University, Japan
A cellulase [endo-b-1,4-
D
-glucanase (EC 3.2.1.4)] was iso-
lated from the hepatopancreas of abalone Haliotis discus
hannai by successive chromatographies on TOYOPEARL
CM-650M, hydroxyapatite and Sephacryl S-200 HR. The
molecular mass of the cellulase was estimated to be
66 000 Da by SDS/PAGE, thus the enzyme was named
HdEG66. The hydrolytic activity of HdEG66 toward carb-
oxymethylcellulose showed optimal temperature and pH at
38 °C and 6.3, respectively. cDNAs encoding HdEG66 were
amplified by the polymerase chain reaction from an abalone
hepatopancreas cDNA library with primers synthesized on
the basis of partial amino-acid sequences of HdEG66. By
overlapping the nucleotide sequences of the cDNAs, a
sequence of 1898 bp in total was determined. The coding
region of 1785 bp located at nucleotide position 56–1840
gave an amino-acid sequence of 594 residues including the
initiation methionine. The N-terminal region of 14 residues
in the deduced sequence was regarded as the signal peptide as
it was absent in HdEG66 protein and showed high similarity
to the consensus sequence for signal peptides of eukaryote
secretory proteins. Thus, matured HdEG66 was thought to
consist of 579 residues. The C-terminal region of 453 residues
in HdEG66, i.e. approximately the C–terminal three quar-
ters of the protein, showed 42–44% identity to the catalytic
domains of glycoside hydrolase family 9 (GHF9)-cellulases

[18,19], crayfish [20], nematode [21], and mussel [22].
According to the criteria based on hydrophobic cluster
analysis [23], termite and crayfish cellulases are classified
into the GHF9 subfamily which includes the majority of
cellulases from plants, bacteria, and a slime mold [24].
Nematode cellulases are classified into GHF5 which
includes some bacterial and fungal cellulases [21]. On the
other hand, a thermostable and low molecular mass ( 20-
kDa) cellulase was recently isolated from blue mussel and
the primary structure was determined [13,22]. Origin of the
mussel cellulase was also investigated by genomic PCR
similar to the case of arthropod cellulases. According to the
primary structure analysis, the mussel cellulase is classified
into the GHF45 subfamily 2, being distinct from the
arthropod ones that are classified into GHF9. This leads us
to consider that molluscan cellulases possess somewhat
different properties and a different evolutionally origin from
arthropod ones. However, at present there is little informa-
tion about the biochemical properties and primary struc-
tures of molluscan cellulases to assess the fundamental
differences between molluscan and other invertebrate cellu-
lases.
Therefore, in the present study, we attempted to isolate a
cellulase from abalone Haliotis discus hannai which is one of
the most common and valuable herbivorous molluscs in
Japan, and determine its primary structure. In addition, we
investigated the existence of a cellulase gene in abalone
chromosomal DNA by genomic PCR.
Correspondence to T. Ojima, Laboratory of Biochemistry and
Biotechnology, Graduate School of Fisheries Sciences,

Cellulase activity was assayed in a 1-mL of reaction mixture
containing 0.5% CMC, 10 m
M
sodium phosphate (pH 7.0),
and an appropriate amount of enzyme at 30 °C. The
reducing sugar liberated by hydrolysis of CMC was
determined by the method of Nelson and Somogyi [25].
One unit of cellulase was defined as the amount of enzyme
that liberates reducing sugars equivalent to 1.0 lmol of
glucose per min under the conditions described above.
Temperature dependence of cellulase activity was assayed at
4–80 °C and pH 7.0. Thermal stability of cellulase was
assayed by measuring remaining activity of the enzyme that
had been incubated at 4–70 °C for 30 min. pH dependence
of cellulase activity was assayed at 30 °C in reaction mixtures
adjusted at pH 3.0–9.0 with 10 m
M
sodium phosphate.
Amino-acid sequencing
The N-terminal amino-acid sequence of intact enzyme was
determined with the sample electrically transferred to a
poly(vinylidene difluoride) membrane after SDS/PAGE
using an ABI 473 A protein sequencer (Applied Biosystems,
CA, USA). For the analysis of internal amino-acid
sequences, the enzyme was digested with 1/100 (w/w) of
lysylendopeptidase at 37 °C for 2 h. The fragments were
separated by HPLC (LP-1000, EYLA, Tokyo, Japan)
equipped with Mightysil RP-18 GP column (150 · 4.6 mm)
(KANTO CHEMICAL CO., INC, Tokyo, Japan) and
subjected to the protein sequencer.

Determination of protein concentration
Protein concentration was determined by the biuret method
[28] or the method of Lowry et al. [29] using bovine serum
albumin fraction V as a standard protein.
cDNA cloning
Construction of the cDNA library and cloning of cellulase
cDNA was achieved as follows: Total RNA was extracted
from 1 g of abalone hepatopancreas by the ganidinium
thiocyanate-phenol method [30] and mRNA was selected
with Oligotex-dT (30) from the total RNA according to the
manufacturer’s protocol. Double-stranded cDNA was syn-
thesized from the mRNA with a cDNA synthesis kit
(TaKaRa, Tokyo, Japan) and used as an abalone cDNA
library. cDNAs encoding abalone cellulase were amplified
by PCR from the cDNA library with degenerated primers
synthesized on the basis of partial amino-acid sequences of
the cellulase. PCR was carried out in a 50-lL of reaction
mixture containing 50 m
M
KCl, 10 m
M
Tris/HCl (pH 8.3),
2m
M
each of dATP, dTTP, dGTP and dCTP, 1.2 m
M
MgCl
2
,2pmolÆmL
)1

M
EDTA. After
772 K i. Suzuki et al.(Eur. J. Biochem. 270) Ó FEBS 2003
the extraction at 4 °C for 30 min, the extract was centri-
fuged at 10 000 g for 15 min. The supernatant was applied
to a TOYOPEARL CM-650M column (2.0 · 15 cm) pre-
equilibrated with 10 m
M
sodium phosphate (pH 7.0), and
proteins adsorbed were eluted with a linear gradient of
0–200 m
M
NaCl in 10 m
M
sodium phosphate (pH 7.0). As
shown in Fig. 1, three fractions showing hydrolytic activity
toward CMC, namely CM-I–III fractions, were eluted.
According to SDS/PAGE followed by zymography, the
CM-I and -II were found to contain 66 000, 75 000, and
100 000 Da proteins with cellulase activity in substantial
amounts. However, the CM-III fraction contained only the
66 000 Da protein in fairly high purity. Therefore, in the
present study, we focused on the 66 000-Da cellulase
(named HdEG66) in the CM-III fraction and attempted
to isolate it.
The CM-III fraction, i.e. fractions 19–24, was applied to a
hydroxyapatite column (1.5 · 20 cm) pre-equilibrated with
10 m
M
potassium phosphate (pH 7.0), and adsorbed pro-

quently, an amino-acid sequence of 14 residues was
identified as VDVTISNHWDGGFQ (Table 2). Then, in
order to analyze the internal amino-acid sequences, the
HdEG66 was digested with lysylendopeptidase and the
Fig. 1. TOYOPEARL CM-650M column chromatography of abalone
crude extract. Crude extract from abalone hepatopancreas was applied
to a TOYOPEARL CM-650M column (2.0 · 15 cm) and eluted with
a 0–0.2
M
NaCl linear gradient in 10 m
M
sodium phosphate (pH 7.0)
at a flow rate of 30 mLÆh
)1
. Each fraction contains 5.0 mL. The SDS
gel electrophoretic patterns of the sample before chromatography (Cr)
and fractions indicated by the arrows a–j are shown in the inset.
M, molecular mass markers; 15 K ¼ 15 000 Da etc.
Fig. 2. Purification of abalone cellulase by hydroxyapatite column
chromatography. The CM-III fraction in TOYOPEARL CM-650M
chromatography was applied to a hydroxyapatite column
(1.5 · 20 cm) and eluted with a 0.01–0.3
M
potassium phosphate
(pH7.0)ataflowrateof30mLÆh
)1
. Each fraction contains 5.0 mL.
The SDS gel electrophoretic patterns of fractions indicated by the
arrowsa–iareshownintheinset.Thefractionsc–ewerepooled.
Fig. 3. Purification of abalone cellulase by Sephacryl S-200 HR gel

the N-terminal 335 amino-acid sequence of the HdEG66.
Next, in order to obtain cDNAs encoding C-terminal region
of the HdEG66, a forward specific primer F2 was newly
synthesized and PCR was performed using the F2-R2
primer pair. Thus, Hd2-DNA of 477 bp encoding the
C-terminal 159 amino acids of the HdEG66 was amplified.
Finally, 5¢-and3¢-RACE PCRs were performed using
primers shown in Table 2, and Hd5RACE-DNA and
Hd3RACE-DNA for 5¢-and3¢-terminal regions were
amplified, respectively. By combining the nucleotide
sequences of the Hd5RACE-DNA, Hd1-DNA, Hd2-
DNA and the Hd3RACE-DNA in this order, the nucleo-
tide sequence of total 1898 bp was determined (Fig. 5). The
reliability of the nucleotide sequence was confirmed with
HdFull-DNA which was amplified with the specific primer
pair, FLF1–FLR1 (Table 2 and Figs 4 and 5). The
translational initiation codon ATG was found in nucleotide
positions from 56 to 58 and termination codon TAA from
1838 to 1840 (Fig. 5). In the 3¢-terminal region, a putative
polyadenylation signal sequence AATAAA and a
poly(A+) tail were found. These structural characteristics
indicate that the HdEG66 cDNA is not derived from
prokaryote like intestinal bacteria. The translational region
of 1785 bp gave an amino-acid sequence of 594 residues. All
the amino-acid sequences determined with lysylendopepti-
dase fragments, LP1–LP9, are found in the deduced
sequence, indicating that the thus cloned cDNAs are of
the HdEG66 protein. It is noteworthy that the N-terminal
15 residues in the deduced sequence are absent in the
HdEG66 protein. According to the sequence comparison

Specific activity
(unitsÆmg
)1
)
Total activity
(units)
Purification
(fold)
Yield
(%)
Crude extract 1470 0.15 220 1 100
CM-III fraction 13.4 2.95 40 20 18
Hydroxyapatite 4.14 4.00 16 27 7
Sephacryl S-200 1.06 13.9 15 93 7
774 K i. Suzuki et al.(Eur. J. Biochem. 270) Ó FEBS 2003
between the deduced sequence and the sequence of LP7
peptide was found at the amino-acid position 388. Namely,
the neighboring residue of LP7 toward the N-terminus
should be lysine because LP7 is a fragment produced by
lysylendopeptidase digestion. However, the corresponding
residue is not lysine but asparagine in both the deduced
sequence and the amino-acid sequence of LP8. We now
consider that this inconsistency has arisen from hetero-
geneity of the HdEG66, e.g. coexistence of proteins with
lysine and asparagine at the position 338 in the HdEG66
preparation.
Amplification of HdEG66 gene from abalone
chromosomal DNA
The existence of HdEG66 gene in the abalone chromosomal
DNA was examined by genomic PCR using primers, 3RAC

onine was absent in the purified HdEG66 protein and
showed the characteristic feature for signal peptides of
eukaryotic secretory proteins [32]. Therefore, this region was
regarded as the signal peptide that was cut away upon
secretion of the HdEG66. Accordingly, the matured
HdEG66 was considered to consist of 579 residues with
the calculated molecular mass of 63 196.88 Da.
By sequence comparison with other invertebrate and
bacterial cellulases, the C-terminal region of 453 residues in
the HdEG66 was regarded as the GHF9-type catalytic
domain i.e. it showed 44, 43, and 42% identity with the
corresponding regions of termite [18], crayfish [20], and
Thermomonospora fusca [34] cellulases, respectively (Fig. 6).
Further, the catalytically important residues in GHF9
cellulases [35–38], i.e. His506, Asp200, Asp203, Asp550 and
Glu559 in the HdEG66 sequence were all conserved
(Fig. 6). Based on these results, we conclude that the
HdEG66 is classified into GHF9. On the other hand,
HdEG66 was found to possess an extended N-terminal
region of 126 residues which is deficient in other invertebrate
cellulases (Fig. 6). This extended region showed sequence
identity of 27% with the CBM attached by a linker in
Cellulomonas fimi CenA [39]. The CBM of CenA belongs
to CBM family II, which is currently the largest among five
principal families, i.e. families I–IV and VI [1]. The family II
CBMs possess strictly conserved four tryptophans and
highly conserved two cysteines that form a disulfide bridge.
In case of N-terminal extended region of the HdEG66, three
out of the four tryptophans are conserved at residues 24, 43,
and 79, although the remaining one is substituted by

The Hdcel-1 DNA consisted of four exons and three introns
(Fig. 4B) and the coding sequence of the exons was
consistent with that of cDNA. This strongly suggests that
the Hdcel-1 DNA is a genomic fragment derived from
Fig. 5. The nucleotide and deduced amino-acid sequences of the HdEG66. Residue numbers for both nucleotide and amino acid are indicated in the
right of each row. The translational start codon ATG, termination codon TAA, and a putative polyadenylation signal AATAAA are boxed. A
putative signal peptide is indicated by a dotted underline. The amino-acid sequences determined with intact HdEG66 (N-terminus) and peptides
LP1–LP9 are indicated by lines under the amino-acid sequence. The positions of primers for PCR are indicated by lines above the nucleotide
sequence. The Asp338 that was suggested to be lysine from the LP7 sequence was double-boxed. Introns 1–3 indicates the positions of introns
revealed by the analysis of a genomic fragment Hdcel-1 DNA (see Fig. 4B).
776 K i. Suzuki et al.(Eur. J. Biochem. 270) Ó FEBS 2003
abalone chromosome not from intestinal symbiotes and that
the HdEG66 is an enzyme secreted by abalone itself. Further,
the position of intron-2 in the Hdcel-1 DNA was found to
coincide to the position of corresponding intron in termite
cellulase gene [19]. These results strongly suggest that the
HdEG66 and termite cellulase genes derive from a common
ancestral gene although the termite cellulase lacks CBM.
In the present study, we found the presence of cellulases
of approx. 75 000 and 100 000 Da as well as HdEG66 (see
Fig. 1). We are now attempting to purify these isoforms and
determine their primary structures in order to reveal the
structural characteristics of abalone cellulases and their
evolutionary relationships to other invertebrate and bacter-
ial cellulases.
References
1. Tomme, P., Warren, R.A.J. & Gilkes, N.R. (1995) Cellulose
hydrolysis by bacteria and fungi. Adv. Microbiol. Physiol. 37, 1–81.
2. Brummell, D.A., Lashbrook, C.C. & Bennett, A.B. (1994) Plant
endo-1,4-b-

Identical residues among the sequences are
indicated by asterisks. While the sequence of
N-terminal 126 residues except for the puta-
tive signal peptide region was aligned with the
CBM attached by a linker of Cellulomonas
fimi CenA [39]. Locations of four tryptophans
strictly conserved in family II CBMs [1] are
indicated by reverse triangles.
Ó FEBS 2003 cDNA cloning of abalone cellulase (Eur. J. Biochem. 270) 777
activity in the digestive system of the redclaw crayfish (Cherax
quadricarinatus). Aquaculture 180, 373–386.
8. Smant, G., Stokkermans, J.P.W.G., 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 cellulases in animals: isolation of b-1,4-
endoglucanase genes from two species of plant-parasitic cyst
nematodes. Proc.NatlAcad.Sci.USA95, 4906–4911.
9. Yokoe, Y. & Yasumasu, I. (1964) The distribution of cellulase in
invertebrates. Comp. Biochem. Physiol. 13, 323–338.
10. Marshall, J.J. (1973) Purification of a b-1,4-glucan hydrolase
(cellulase) from the snail, Helix pomatia. Comp. Biochem. Physiol.
44B, 981–988.
11. Marshall, J.J. & Grand, R.J.A. (1976) Characterization of a b-1,4-
glucan hydrolase from the snail, Helix pomatia. Comp. Biochem.
Physiol. 53B, 231–237.
12. Anzai, H., Nisizawa, K. & Matsuda, K. (1984) Purification
and characterization of a cellulase from Dolabella auricularia.
J. Biochem. 96, 1381–1390.
13.Xu,B.,Hellman,U.,Ersson,B.&Janson,J C.(2000)
Purification, characterization and amino-acid sequence analysis of

of a molluscan endo-b-1,4-glucanase gene from the blue mussel,
Mytilus edulis. Eur. J. Biochem. 268, 3718–3727.
23. Lemesle-Varloot, L., Henrissat, B., Gaboriaud, C., Bissery, V.,
Morgat, A. & Mornon, J.P. (1990) Hydrophobic cluster analysis:
procedures to derive structural and functional information from
2-D-representation of protein sequences. Biochimie 72, 554–574.
24. Henrissat, B. & Davies, G. (1997) Structural and sequence-based
classification of glycoside hydrolases. Curr. Opin. Struct. Biol. 7,
637–644.
25. Nelson, N. (1944) A photometric adaptation of the Somogyi
method for the determination of glucose. J. Biol. Chem. 153,
375–380.
26. Porzio, M.A. & Pearson, A.M. (1977) Improved resolution of
myofibrillar proteins with sodium dodecyl sulfate-polyacrylamide
gel electrophoresis. Biochim. Biophys. Acta 490, 27–34.
27. Be
´
guin, P. (1983) Detection of cellulase activity in polyacrylamide
gels using Congo red-stained agar replicas. Anal. Biochem. 131,
333–336.
28. Gornall, A.G., Bardawill, C.J. & David, M.M. (1949)
Determination of serum proteins by means of the biuret reaction.
J. Biol. Chem. 177, 751–766.
29. Lowry, O.H., Rosebrough, N.J., Farr, A.L. & Randall, R.J.
(1951) Protein measurement with the Folin phenol reagent. J. Biol.
Chem. 193, 265–275.
30. Chomczynski, P. & Sacchi, N. (1987) Single-step method of RNA
isolation by acid guanidinium thiocyanate-phenol-chloroform
extraction. Anal. Biochem. 162, 156–159.
31.Sambrook,J.,Fritsch,E.F.&Maniatis,T.(1989)Molecular