Biochemical and Biophysical Research Communications 307 (2003) 620–625

www.elsevier.com/locate/ybbrc

BBRC

A constitutively expressed 36 kDa exochitinase fromBacillus thuringiensis HD-1

Naresh Arora, Tarannum Ahmad, R. Rajagopal, and Raj K. Bhatnagar*

International Center for Genetic Engineering and Biotechnology, Aruna Asaf Ali Marg, P.O. Box 10504, New Delhi 1100 67, India

Received 5 June 2003

Abstract

A 36kDa chitinase was purified by ion exchange and gel filtration chromatography from the culture supernatant of Bacillus

thuringiensis HD-1. The chitinase production was independent of the presence of chitin in the growth medium and was produced

even in the presence of glucose. The purified chitinase was active at acidic pH, had an optimal activity at pH 6.5, and showed

maximum activity at 65 �C. Of the various substrates, the enzyme catalyzed the hydrolysis of the disaccharide 4-MU(GlnAc)2 most

efficiently and was therefore classified as an exochitinase. The sequence of the tryptic peptides showed extensive homology with

Bacillus cereus 36 kDa exochitinase. The 1083 bp open reading frame encoding 36 kDa chitinase was amplified with primers based on

the gene sequence of B. cereus 36 kDa exochitinase. The deduced amino-acid sequence showed that the protein contained an N-

terminal signal peptide and consisted of a single catalytic domain. The two conserved signature sequences characteristic of family 18

chitinases were mapped at positions 105–109 and 138–145 of Chi36. The recombinant chitinase was expressed in a catalytically

active form in Escherichia coli in the vector pQE-32. The expressed 36 kDa chitinase potentiated the insecticidal effect of the veg-

etative insecticidal protein (Vip) when used against neonate larvae of Spodoptera litura.

� 2003 Elsevier Inc. All rights reserved.

Keywords: Bacillus thuringiensis; Chitinase; Synergism

Bacillus thuringiensis is a gram positive, spore form-

ing, soil bacterium that forms parasporal crystals during

sporulation. A diverse range of crystal proteins is pro-

duced by different B. thuringiensis strains that are highly

specific for different insect larvae [1]. These features meet

the criteria of an ideal biocontrol agent and B. thurin-

giensis has been used as an insecticidal agent for decades[2]. However, resistance to crystal toxins by targetted

pests has been observed and various methods are being

employed to increase the efficacy of the insecticidal

crystal proteins and in controlling agricultural pests

[3,4].

The insecticidal crystal proteins ingested by the insect

and undergo site-specific proteolysis from the N and the

C terminus to generate active fragments. These activatedpolypeptides bind to the receptors in the midgut epi-

thelium and form ion channels, inducing osmotic lysis of

* Corresponding author. Fax: +91-11-2616-2316.

E-mail address: raj@icgeb.res.in (R.K. Bhatnagar).

0006-291X/03/$ - see front matter � 2003 Elsevier Inc. All rights reserved.

doi:10.1016/S0006-291X(03)01228-2

the epithelium that consequently kills the larvae [5]. The

midgut lumen is separated from the epithelium by a

protective structure consisting of a network of chitin

and proteins, called the peritrophic membrane [6]. It is

suggested that the accessibility of the toxins to the

midgut epithelium is restricted by the peritrophic ma-

trix. It is believed that chitinases disrupt the integrity ofthe peritrophic membranes, facilitating the contact be-

tween the activated toxins and receptors in the midgut

epithelium [7].

A synergistic action between insecticidal crystal pro-

teins and chitinase, which hydrolyzes the b-1,4 linkage inchitin, has been demonstrated to occur during co-ap-

plication of insecticidal protein containing spore sus-

pension along with chitinase [8]. Co-expression ofheterologous chitinase genes in B. thuringiensis has also

been demonstrated to increase the insecticidal activity of

the bacterium [9–11]. B. thuringiensis itself produces

chitinases and the role of these endogenous chitinases

has recently come under investigation. The involvement

N. Arora et al. / Biochemical and Biophysical Research Communications 307 (2003) 620–625 621

of endogenous chitinases during B. thuringiensis aizawai

infection of Spodoptera littoralis larvae was demon-

strated by the addition of a chitinase inhibitor, allos-

amidin that increased the LC50 values of the toxin [12].

Chitinases from two different strains of B. thuringi-

ensis, kenyae and pakistani, have been cloned and they

are very similar to each other in gene structure and

properties of the encoded protein [13,14]. We report here

the purification of a novel chitinase from B. thuringiensis

sp. HD-1, cloning of the corresponding gene, and its

heterologous expression in Escherichia coli. The syner-

gistic action of this chitinase along with the vegetative

insecticidal protein (Vip) against Spodoptera litura lar-

vae is also demonstrated.

Materials and methods

Bacterial strains and growth conditions. B. thuringiensis HD1 was

obtained from Bacillus Genetic Stock Centre, OH, USA. For chitinase

production, the bacterium was grown in G-medium [15] with 0.2%

glucose or 0.2% colloidal chitin at 30 �C with shaking. Colloidal chitin

was prepared according to Roberts and Selitrennikoff (1988) [16]. For

chromosomal DNA extraction, bacteria were grown in Luria–Bertani

(LB) medium. E. coli DH5a and M15 carrying the pGEMT-chi and

pQE-chi plasmids, respectively, were grown at 37 �C in LB medium for

expression and plasmid preparations.

Purification of Chi36 from the culture fluid of Bt-HD1. Protein from

the supernatant of a 500ml culture grown in G-medium supplemented

with glucose was concentrated by ammonium sulphate fractionation at

the following saturation concentration, 0–35%, 35–55%, and 55–95%.

Protein in each fraction was dissolved in 5ml of 50mM Tris–HCl, pH

7.5, and dialyzed overnight against the same buffer. A glycol-chitin

activity gel was done in each fraction to locate the chitinase activity

[16]. The active fraction from ammonium sulphate precipitation was

further resolved on a DE-52 anion exchange column previously

equilibrated with 50mM Tris–HCl buffer, pH 7.5. After washing un-

adsorbed proteins with the same buffer, chitinase was eluted with a

linear gradient of 0–0.75M KCl in same buffer at a flow rate of 1ml/

min. The active fractions were pooled, dialyzed and concentrated by

passing through amicom PM-30 membrane. The concentrated sample

was resolved on a Sephacryl S-100 4R gel filtration column. Elution

was performed at a flow rate of 1ml/min in 50mM Tris–HCl, pH 7.5,

containing 100mM NaCl. One milliliter fractions were collected and

individual fractions were analyzed for chitinase activity on a non-

denaturing PAGE.

Amino-acid sequence of Chi36. Sequencing of the tryptic digests of

the 36 kDa major protein of the active fraction after gel filtration was

performed at Biotechnology Research Laboratory, Medical University

of South Carolina, Charleston, USA.

Chitinase assays. Chitinase was assayed on a 10% non-denaturing

gel containing 0.01% glycol chitin by the method of Trudel and Asselin

[17], with minor modifications. Fluorometric substrates 4-methylum-

belliferyl b-DD-N ,N 0,N 00-triacetylchitotrioside and 4-methylumbelliferyl

b-DD-N ,N 0-diacetylchitobioside were obtained from Sigma. The reaction

mixture contained in a total volume of 100ll, 50lM of the substrate,

50mM phosphate buffer, pH 7.0, and an appropriate amount of the

enzyme. To find out the optimum pH of the reaction, the following

range of buffers were used: 50mM sodium acetate buffer, pH 4.0–5.5,

50mM sodium phosphate buffer, pH 6.0–7.5, 50mM Tris–HCl, pH

8.0–9.0, and sodium carbonate buffer, pH 10–11. The reaction mixture

was incubated at 30 �C for 20min and stopped by adding 2.4ml of

150mM glycine–NaOH buffer, pH 10.5. The release of free 4-meth-

ylumbelliferone was monitored by fluorescence spectroscopy with ex-

citation at 360 nm and measuring emission at 450 nm. The fluorimetric

intensity was converted to nanomolar of 4-methylumbelliferone re-

leased by preparing a standard curve of 4-methylumbelliferone. The

activity has been expressed as nmol of 4-methylumbelliferone released

min�1 mg�1 protein.

Insect bioassays. The insecticidal effect of Chi36 in combination

with the Vip was tested against S. litura neonate larvae by feeding the

larvae leaves coated with required concentrations of Vip and Chi36, or

Vip alone. Six neonate larvae were released on the leaves at each set of

concentration. The larvae on the leaves were incubated under con-

trolled conditions of temperature 25 �C, 70% relative humidity, and a

photoperiod of 12 h light:12 h dark. Observations were recorded after

48 h and the dose combination was replicated in five independent sets

of experiments.

Cloning and expression of chi36. Forward (50 GATGTTAA

ACAGGTTCAA 30) and reverse primers (50 TTATTTTTGCAAGG

AAAG 30) were designed based on homology to Bacillus cereus 36 kDa

chitinase, and the gene encoding for Chi36 was amplified by PCR using

Bt-HD1 genomic DNA as a template. The amplified product was

cloned into the PCR cloning vector, pGEMT easy, and was sequenced

completely. To express the polypeptide in E. coli, the gene was sub-

cloned into the expression vector, pQE-32 (Qiagen) and expressed as

an N-terminal fusion protein with 6� His tag in M-15 expression host.

Expression of recombinant chitinase was induced by adding IPTG at a

final concentration of 1mM to the actively growing culture of E. coli.

The culture was further grown for 2 h and the cells harvested by

centrifugation. The cells were disrupted by sonication and the

induced proteins were analyzed by resolving the proteins on 10%

SDS–PAGE.

Results

Production of Chi36 by B. thuringiensis HD1

The bacterium was grown in the following combi-

nation of carbon sources supplemented individually in

G-medium: glucose, glucose + chitin or only chitin.

Resolution of equal amounts of protein on a glycol-

chitin activity gel revealed that comparable level of

chitinase was produced in cultures independent of the

nature of the carbon source (Fig. 1). In all subsequentexperiments, the bacterium was grown in the presence of

glucose only. Time course for the accumulation of chi-

tinase was examined in cultures grown for 24, 36, 48, 60,

and 72 h in G-medium supplemented with glucose.

Highest chitinase activity was observed in the culture

grown for 72 h. Protein in the supernatant of 72 h grown

culture was fractionated by ammonium sulphate. Pro-

tein precipitating at 35–65% ammonium sulphate satu-ration contained nearly 80% of total chitinase activity.

Ammonium sulphate precipitated protein was subjected

to anion exchange chromatography and chitinase was

obtained in the wash fraction, whereas other proteins

eluted at higher salt concentrations. Further purification

of the protein was done by gel filtration chromatogra-

phy and a chitinase corresponding to a molecular mass

of 36 kDa was obtained (Fig. 2). To determine theamino-acid sequence of the protein, tryptic digests of the

protein were generated and the obtained sequence was

Fig. 1. Chitinase production in the presence of glucose and chitin. B.

thuringiensis HD-1 was grown in the presence of only 0.2% glucose,

only 0.2% chitin, 0.2% glucose+ 0.2% chitin, or without any carbon

source. Proteins in the culture supernatant were precipitated by am-

monium sulphate fractionation and equal amounts (5lg) were loadedon a glycol chitin activity gel, as described in Materials and methods.

The dark areas represent the clearing zones produced by the hydrolysis

of glycol chitin. (+) or ()) indicates the presence or absence of glucoseor chitin.

622 N. Arora et al. / Biochemical and Biophysical Research Communications 307 (2003) 620–625

analyzed for homologies at the NCBI server using the

BLAST application. The sequence of the tryptic pep-

tides matched the exochitinase from B. cereus.

Fig. 2. (A) Gel filtration chromatography of DE-52 purified protein.

The solid line indicates absorbance at 214 nm and the dashed line in-

dicates the chitinase activity on a glycol chitin activity gel. The clearing

zones were subjected to densitometric analysis and plotted where

100,000 pixels is equivalent to 1 absorbance unit. (B) (a) Coomassie

brilliant blue staining of the gel filtration purified active fractions. (b)

Activity staining of the active fractions. The dark zones represent the

hydrolysis of glycol chitin in the activity gel.

Properties and kinetics of Chi36

Chi36 is active on both polymeric as well as oligo-

saccharide substrates. Higher specific activity was ob-

tained with the disaccharide substrate as compared to

the trisaccharide substrate, which classifies chi36 as a

chitobiosidase or an exochitinase (Table 1). When tested

against the monomer, 4-MU N-acetylglucosaminide, no

activity was obtained showing that chi36 does not havean N-acetylglucosaminidase activity. The Km and Vmax

for the oligosaccharide substrates were calculated to be

30 lM and 45 nmol of product formed min�1 mg�1

protein, respectively, for the hydrolysis of 4 MU-(Glc-

NAc)2 and 16 lM and 7.2 nmol of product formed

min�1 mg�1 protein for 4 MU-(GlcNAc)3. To study the

effect of pH on chitinase activity, buffers of different pH

were used. The results are shown in Fig. 3. The enzymewas active at acidic pH with the highest activity at pH

6.5. At alkaline pH, considerable activity was lost. To

investigate the optimum temperature of chi36, the en-

zymatic reactions were performed at 25–65 �C in sodium

phosphate buffer, pH 6.5, using 4 MU-(GlcNAc)3. The

activity of chi36 increased with increase in temperature

(Fig. 4) and showed a high activity at 65 �C.

Fig. 3. Optimum pH of Chi36: activity of chitinase was measured with

4-MU (GlcNAc)3, final concentration 50 lM, in 50mM sodium acetate

buffer, pH 4.0–5.0, 50mM sodium phosphate buffer, pH 6.0–7.5,

50mM Tris–acetate, pH 8.0–9.0, and 50mM sodium carbonate buffer,

pH 10–11, at 30 �C for 20min.

Table 1

Substrate specificity and reaction rate of purified Chi36 from

B. thuringiensis HDI

Kinetic parameters Substrate

4-MU

(GlcNAc)

4-MU

(GlcNAc)2

4-MU

(GlcNAc)3

Km (lM)a 0 30 16

Vmax (nmol of 4-MU

formed/minute)a0 45 7.2

aReaction rate was measured with the oligosaccharide substrates

4-MU(GlcNAc)1-3, final concentration 50 lM, in sodium phosphate

buffer, pH 6.5 at 30 �C for 20 min.

Fig. 4. Temperature dependency of Chi36: activity of chitinase was

measured with 4-MU (GlcNAc)3, final concentration 50lM, in 50mM

sodium phosphate buffer, pH 6.5, at 25–65 �C for 1 h. The points

represent mean values of two independent measurements.

Fig. 5. The deduced amino-acid sequence of Bt Chi36: the signal

peptide is underlined. The conserved domains I and II are boxed and

the tryptic peptide is double underlined. Bold lettering indicates resi-

dues that are different from B. cereus 6E1 36 kDa exochitinase.

Fig. 6. Expression of recombinant Chi36 in E. coli: M 15 cells trans-

formed with the plasmid pQE-Chi36 were induced with 1mM IPTG

for 2 h. The induced and un-induced cultures were separated on 10%

SDS–PAGE and stained with Coomassie brilliant blue. M, marker; U,

un-induced; and I, induced.

N. Arora et al. / Biochemical and Biophysical Research Communications 307 (2003) 620–625 623

Cloning and heterologous expression of chi36

The homology search at NCBI BLAST revealed that

the amino-acid sequences of tryptic peptides matched

the exochitinase from B. cereus with 96% identity.

Forward and reverse primers were synthesized based on

the nucleotide sequence of B. cereus exochitinase and a

product of ca. 1000 bp was amplified from Bt-HD1 ge-

nomic DNA, and sequenced. BLAST search showedthat the cloned fragment had extensive homology with

B. cereus chitinase, the gene was therefore designated as

chi36. Residues depicted in bold are unique to Chi36

being reported here. Chi36 is 1083 bp in length and en-

codes a polypeptide of 39,407 Da, which is larger than

the size of the native protein observed on SDS–PAGE.

The Signal P site predicted a signal peptide from resi-

dues 1–27 and cleavage between residues Ala 27 and Ala28. Therefore, the cleavage of the signal peptide gener-

ates a 36 kDa protein, which has alanine at the mature

N-terminus. Conserved Domain Architecture Retrieval

Tool (CDART) at NCBI identified a carbohydrate me-

tabolism and transport domain containing the con-

served motifs characteristic of family 18 chitinases at

residues 105–109 and 138–145 (Fig. 5). The entire open

reading frame was cloned in the prokaryotic expressionvector, pQE 32 as a NcoI/SphI fragment. A novel in-

duced protein of 36 kDa was found in induced cultures

transformed with the recombinant pQE-chi36 plasmid,

which was absent in the un-induced cultures (Fig. 6).

The recombinant chitinase was localised in both the

soluble and insoluble fractions of the cell lysate. The

expressed recombinant chitinase was catalytically active

against polymeric and oligomeric substrates. In addi-tion, the recombinantly expressed Chi36 showed iden-

tical properties and kinetics to the native enzyme.

Synergism of insecticidal activity

Preliminary analysis had shown that the LC50 for Vip

towards S. litura larvae was 20 ng cm�2. To investigate if

the insecticidal activity of Vip was potentiated by the

addition of chitinase, partially purified Chi36 alone,

20 lg of Vip alone, and Vip in combination with dif-

ferent concentrations of chitinase was fed S. litura ne-onate larvae. A 30% decrease in the LC50 of vip

indicated that chi36 potentiated the toxic effect of vip

against S. litura larvae.

Discussion

It is well established that chitinase production in mostof the chitinase producing bacteria is inducible by chitin,

chito-oligosaccharides, or even N-acetylglucosamine

[18]. Low levels of chitinase are observed in cultures

624 N. Arora et al. / Biochemical and Biophysical Research Communications 307 (2003) 620–625

grown in the presence of glucose. However, in this study,high levels of Chi36 were present when the bacteria were

grown in the presence of glucose, suggesting that the

production of Chi36 is independent of the presence of

chitin. Normally, the induction of chitinase in Strepto-

myeces is dependent on the presence of chitin in the

growth medium. Interestingly, a single point mutation in

the promoter of chitinase shifts the synthesis from in-

ducible to constitutive mode [19].Of the various chitinases described from related Ba-

cillus sp., Bt Chi36 displayed extensive similarities to B.

cereus 6E1 exochitinase [20] as the two proteins shared

98% amino-acid identity (Fig. 5). But the two proteins

do not have identical properties. B. cereus exochitinase

activity for the trisaccharide was only 4.4% of the ac-

tivity with the disaccharide, whereas in Bt Chi36, 15%

relative activity was obtained with the trisaccharide.Also, B. cereus exochitinase lost activity at high tem-

peratures but Bt Chi36 was active upto 65 �C. Another

36 kDa chitinase, ChiA [20], from B. cereus CH shared

95% identity of amino acids but it differed from Bt

Chi36 in being an endochitinase. Amongst the Bacillus

circulans chitinases, A1, A2, B1, B2, C, and D [21], Bt

Chi36 shared 53% amino acid similarity with B. circu-

lans ChiD, which had a FnIII domain and a catalyticdomain. Bt Chi36 is distinct from the 74 kDa chitinases

characterized from B. thuringeinsis kenyae that showed

homology to B. cereus ChiB [13]. Therefore, it is possible

that the Bt Chi36 (present study) and 74 kDa chitinases

(B. thuringiensis kenyae) in B. thuringiensis correspond

to B. cereus ChiA and ChiB, respectively.

Btchi36 contains the conserved region characteristic

of family 18 chitinases that have the consensus sequence[DN]–G–[LIVMF]–[DN]–[LIVMF]–[DN]–x–E, where E

is the catalytic residue. A signal peptide is present at the

N-terminus that may be responsible for the secretion of

the protein. Btchi36 consists of a single catalytic domain

and lacks the chitin binding or the fibronectin domain

involved in binding chitin. In spite of the lack of the

chitin-binding domain, a high activity on a polymeric

substrate, glycol chitin, was obtained demonstratingthat the presence of the chitin binding domain is not an

absolute requirement for the hydrolysis of chitin. That

the absence of chitin binding domain does not affect the

catalytic efficiency of chitinase has been from other

bacterial chitinases also. The 36 kDa chitinase homolog

from B. cereus also lacks the chitin binding domain and

is also active on the polymeric substrate, CM-chitin-

RBV [21]. Also, ChiA1 from B. circulans, retained someactivity for chitin on deleting the chitin binding domain,

although considerable activity was lost [22].

The observed potentiation of insecticidal activity of

Vip in the presence of chitinase offers another tool to

enhance the application of Bt proteins. Chi36, when

used in combination with Vip enhanced the insecticidal

activity of Vip against neonate larvae of S. litura. Vip,

like the cry toxins, also targets the midgut of the insectand has to penetrate the peritrophic matrix to reach the

midgut epithelium. Scanning electron microscopy of

midguts of S. litura larvae fed on recombinant ChiAII

showed that the peritrophic membrane was perforated,

and when these larvae were fed ChiAII in combination

with CryIC, an increase in toxicity was obtained [7].

Similar synergistic effect of chitinase with cry toxins has

been demonstrated in other studies as well. The presentstudy suggests that the mechanism of the synergistic

effect is similar when Vip is used instead of cry toxins.

Chitinases in B. thuringiensis are likely to assist

pathogenesis of insects by hydrolyzing the chitin present

in the gut. The bacterium has probably adapted to a

constitutive, rather than inducible expression of chiti-

nase to meet this requirement.

Acknowledgments

We are grateful to the Bacillus Genetic Stock Center for providing

the strain Bacillus thuringiensis HD-1. This project was funded by the

Indo-Swiss grant.

References

[1] R. de Maagd, A. Bravo, N. Crickmore, How Bacillus thuringiensis

has evolved to colonize the insect world, Trends Genet. 17 (2001)

193–199.

[2] C.C. Beegle, T. Yamamoto, History of Bacillus thuringiensis

Berliner research and beyond, Can. Ent. 124 (1992) 587–616.

[3] F.J. Perlak, Modification of the coding sequence enhances plant

expression of insect control proteins, Proc. Natl. Acad. Sci. USA

88 (1991) 3324–3328.

[4] B.E. Tabashnik, F.R. Finson, N. Groeters, W.J. Moar, M.W.

Johnson, K. Luo, M.J. Adang, Reversal of resistance to Bacillus

thuringiensis in Plutella xylostella, Proc. Natl. Acad. Sci. USA

91 (1994) 4120–4124.

[5] A.A. Lorence, C. Darzon, A. Diaz, R. Lievano, R. Quintero, A.

Bravo, d-Endotoxins induce cation channels in Spodoptera

frugiperda brush border membranes in suspension and in planar

bilayers, FEBS Lett. 360 (1995) 217–222.

[6] R.W. Terra, The origins and function of the insect peritrophic

membrane and peritrophic gel, Arch. Insect Biochem. Physiol.

47 (2001) 47–61.

[7] A. Regev, M. Keller, N. Strizhov, B. Sneh, E. Prudovsky, I. Chet,

I. Ginzberg, Z. Koncz-Kalman, C. Koncz, J. Schell, A. Zilber-

stein, Synergistic activity of a Bacillus thuringiensis d-endotoxinsand a bacterial endochitinase against Spodoptera littoralis larvae,

Appl. Environ. Microbiol. 62 (1996) 3581–3586.

[8] W.A. Smirnoff, Comfirmation experimentales du potentiel du

complexe Bacillus thuringiensis chitinase pour la repression de al

tordeuse des bourgeons de espirette Choristoneura fumiferana,

Can. Entomol. 109 (1977) 351–358.

[9] S. Tantimavanich, S. Pantuwatana, A. Bhumiratana, W. Pan-

bangred, Cloning of a chitinase gene into Bacillus thuringiensis

subsp. aizwai for enhanced insecticidal activity, J. Genet. Appl.

Microbiol. 43 (1997) 341–347.

[10] N. Sirichrotpakorn, P. Rognoparut, K. Choosang, W. Ranbag-

red, Coexpression of chitinase and cry11Aa1 toxin genes in

Bacillus thuringiensis serovar israelensis, J. Invertebr. Pathol.

78 (2001) 106–109.

N. Arora et al. / Biochemical and Biophysical Research Communications 307 (2003) 620–625 625

[11] M. Lertcanawanichakul, C. Wiwat, Improved shuttle vectors for

expression of chitinase gene in Bacillus thuringiensis, Lett. Appl.

Microbiol. 2 (2001) 123–128.

[12] M.N. Sampson, G.W. Gooday, Invlovement of chitinases of

Bacillus thuringiensis during pathogenesis in insects, Microbiology

144 (2001) 2189–2194.

[13] J.E. Barboza-Corona, E. Nieto-Mazacco, R. Velazquez-Robledo,

R. Salcedo-Hernandez, M. Bautista, B. Jimenez, J.E. Ibarra,

Cloning, Sequencing and expression of the chitinase gene chiA74

from Bacillus thuringiensis, Appl. Environ. Microbiol. 69 (2003)

1023–1029.

[14] S. Thamthiankul, S. Suan-Ngay, S. Tantimavanich, W. Panbag-

red, Chitinase from Bacillus thuringiensis pakistani, Appl. Micro-

biol. Biotechnol. 56 (2001) 395–401.

[15] D. Wu, X.L. Cao, Y.Y. Bai, A.I. Aronson, Sequence of an operon

containing a novel endotoxin gene from Bacillus thuringiensis,

FEMS Microbiol. Lett. 81 (1991) 32–36.

[16] W.K. Roberts, C.P. Selitrennikoff, Plant and bacterial chitinases

differ in antifungal activity, J. Genet. Microbiol. 134 (1988)

169–17656; J. Genet. Microbiol. 134 (2001) 395–401.

[17] W. Trudel, A. Asselin, Detection of chitinase activity after

polyacrylamide gel electrophoresis, Anal. Biochem. 178 (1989)

362–366.

[18] T. Felse, T. Panda, Cloning and regulation of microbial

chitinase genes, Appl. Microbiol. Biotechnol. 51 (1999) 141–

151.

[19] I. Delic, P. Robbins, J. Westpheling, Direct repeat sequences are

implicated in the regulation of Streptomyeces chitinase promoters

that are subject to carbon catabolite control, Proc. Natl. Acad.

Sci. 89 (1992) 1885–1889.

[20] S. Wang, A. Moyne, G. Thottappilly, S. Wu, R.D. Locy, N.K.

Singh, Purification and characterization of Bacillus cereus exoch-

itinase, Enzyme Microb. Technol. 28 (2001) 492–498.

[21] N. Mabuchi, I. Hashizume, Y. Araki, Characterization of

chitinases excreted by Bacillus cereus CH, Can. J. Microbiol. 46

(2000) 370–375.

[22] T. Watanabe, W. Oyanagi, K. Suzuki, H. Tanaka, Chitinase

system of Bacillus circulans W-12 and importance of Chiti-

nase A1 in chitin degradation, J. Bacteriol. 172 (1990) 4017–

4022.