Expression and characterization of endochitinase C from Serratia marcescens BJL200 and its purification by a one-step general chitinase purification method

Article abstract of DOI:10.1271/bbb.70594

In this study we cloned, expressed, purified, and charaterized chitinase C1 from Serratia marcescens strain BJL200. As expected, the BJL200-ChiC1 amino acid sequence of this strain was highly similar to sequences of ChiC1 identified in two other strains of S. marcescens. BJL200-ChiC1 was overproduced in E. coli by the T7 expression system, and purified by a one-step hydrophobic interaction chromatography (HIC) with phenyl-sepharose. BJL200-ChiA and BJL200-ChiB had an approximately 30-fold higher k_cat and 15 fold-lower K_m than BJL200-ChiC1 for the oligomeric substrate 4-methylumbelliferyl-β-D-N- N′-N″-triacetylchitotrioside, while BJL200-ChiC1 was 10-15 times faster than BJL200-ChiB and BJL200-ChiA in degrading the polymeric substrate CM-chitin-RBV. BJL200-ChiC1 degradation of β-chitin resulted in a range of different chito-oligosaccharides (GlcNAc)₂ (main product), GlcNAc, (GlcNAc)₃, (GlcNAc)₄, and (GlcNAc)₅, indicating endo activity. The purification method used for BJL200-ChiC1 in this study is generally applicable to family 18 chitinases and their mutants, including inactive mutants, some of which tend to bind almost irreversibly to chitin columns. The high specificity of the interaction with the (non-chitinous) column material is mediated by aromatic residues that occur in the substrate-binding clefts and surfaces of the enzymes.

Full text of DOI:10.1271/bbb.70594

Biosci. Biotechnol. Biochem., 72 (3), 715–723, 2008
Expression and Characterization of Endochitinase C
from Serratia marcescens BJL200 and Its Purification
by a One-Step General Chitinase Purification Method
Bjørnar SYNSTAD, Gustav VAAJE-KOLSTAD, F. Henning CEDERKVIST, Silje F. SAUA,
y
Svein J. HORN, Vincent G. H. EIJSINK, and Morten SØRLIE
Department of Chemistry, Biotechnology, and Food Science, Norwegian University of Life Sciences,
P.O. Box 5003, N-1432 Aas, Norway
Received September 19, 2007; Accepted November 20, 2007; Online Publication, March 7, 2008
[doi:10.1271/bbb.70594]
In this study we cloned, expressed, purified, and
charaterized chitinase C1 from Serratia marcescens
strain BJL200. As expected, the BJL200-ChiC1 amino
acid sequence of this strain was highly similar to
sequences of ChiC1 identified in two other strains of
S. marcescens. BJL200-ChiC1 was overproduced in
E. coli by the T7 expression system, and purified by a
one-step hydrophobic interaction chromatography
(HIC) with phenyl-sepharose. BJL200-ChiA and
BJL200-ChiB had an approximately 30-fold higher kcat
and 15 fold-lower Km than BJL200-ChiC1 for the
oligomeric substrate 4-methylumbelliferyl-ꢀ-D-N-N0-
N00-triacetylchitotrioside, while BJL200-ChiC1 was 10–
15 times faster than BJL200-ChiB and BJL200-ChiA in
degrading the polymeric substrate CM-chitin-RBV.
BJL200-ChiC1 degradation of ꢀ-chitin resulted in a
range of different chito-oligosaccharides (GlcNAc)2
(main product), GlcNAc, (GlcNAc)3, (GlcNAc)4, and
(GlcNAc)5, indicating endo activity. The purification
method used for BJL200-ChiC1 in this study is generally
applicable to family 18 chitinases and their mutants,
including inactive mutants, some of which tend to bind
almost irreversibly to chitin columns. The high specific-
ity of the interaction with the (non-chitinous) column
material is mediated by aromatic residues that occur in
the substrate-binding clefts and surfaces of the enzymes.
sidic bonds in chitin is catalyzed by chitinases, which
are found in all types of organisms, from viruses to
humans. On the basis of sequence similarities, chitinases
are classified into two different families, families 18 and
19, that differ in structure and mechanism of catalysis.1)
The catalytic domains of family 18 chitinases have an
ðꢁ=ꢀÞ8 (TIM-barrel) fold.2–4) In addition to the catalytic
domain, many family 18 chitinases contain one or more
domains that are involved in binding to the substrate
(chitin-binding domains).5–7)
The soil bacterium Serratia marcescens produces up
to four different chitinases (A, B, C1, and C2), a chitin
binding protein (CBP21), and a chitobiase when grown
on chitin.8–12) The crystal structures of ChiA and ChiB
have been determined.3,4,13) Based on structural consid-
erations and the outcomes of a variety of experiments,
it has been suggested that ChiA and ChiB are exoen-
zymes that degrade chitin chains from opposite direc-
tions.4,10,14–17) Even so, both enzymes have shown endo
activity on the soluble substrate chitosan.18) Both
enzymes contain a deep substrate-binding groove, which
is extended by an aromatic surface of a putative chitin-
binding domain. The suggestion that these enzymes
degrade chitin chains from opposite ends was originally
based on the fact that the chitin-binding domains are
located on opposite sides of the catalytic center,4) and it
has recently been confirmed by elegant experiments with
end-labeled substrates by Hult et al.16)
Key words: Serratia marcescens; chitinase C; purifica-
tion; hydrophobic interaction chromatogra-
phy; endochitinase
The chiC gene has been described for two different
S. marcescens strains.10,19) Judged by the deduced
amino acid sequence, ChiC1 contains two chitin-binding
domains, which are located on the C-terminal side of the
molecule. ChiC2 is a proteolytic derivative of ChiC1
that can emerge under certain experimental conditions,
and that corresponds to the catalytic domain only.19,20)
The crystal structures of ChiC1 or ChiC2 are not known.
Chitin, a 1,4-ꢀ-linked polymer of N-acetyl-ꢀ-D-glu-
cosamine (GlcNAc), is the second most abundant
biopolymer in nature, and is a major structure compo-
nent of the exoskeleton of insects and crustaceans and
the cell walls of some fungi. Hydrolysis of the glyco-
y
To whom correspondence should be addressed. Tel: +47-64965902; Fax: +47-64945901; E-mail: morten.sorlie@umb.no
Abbreviations: 4MU-GlcNAc3, 4-methylumbelliferyl-ꢀ-D-N-N0-N00-triacetylchitotrioside; ChiA, chitinase A; ChiB, chitinase B; ChiC, chitinase C;
GlcNAc, N-acetyl-ꢀ-D-glucosamine; HIC, hydrophobic interaction chromatography

716
B. SYNSTAD et al.
Fig. 1. Sequence Alignment of the Catalytic Domains of ChiA (chitinase A from S. marcescens BJL200), ChiB (chitinase B from S. marcescens
BJL200), ChiC1 (chitinase C1 from S. marcescens BJL200), and Heva (Hevamine from Hevea brasiliensis).
Sequence numbering and secondary structure (SS) is based on the sequence and structure of ChiB.4) Conserved resides are highlighted in
black. Residues conserved in ChiA, ChiB, and hevamine, but not in ChiC1, are highlighted in gray. The stretches of residues constituting the
ꢁ=ꢀ-domain present in ChiA and ChiB, but lacking in ChiC1 and hevamine, are underlined. An x in the sequences indicates a small insertion in
hevamine.
Sequence alignments indicate that BJL200-ChiC differs
widely from BJL200-ChiA and BJL200-ChiB in that
it lacks the so-called ꢁ=ꢀ-domain, which makes up
one of the walls of the deep substrate-binding clefts in
the two exoenzymes (Fig. 1). Judged by its sequence,
the catalytic domain of BJL200-ChiC resembles that of
hevamine, a plant endochitinase with a shallow active
site cleft (Fig. 1).21)
on the hydrolysis of powdered chitin has also been
observed for several combinations of Serratia chitinases,
including BJL200-ChiC1.10,22) These observations indi-
cate that ChiA, B, and C1 have different roles in the
degradation of chitin, and suggests different modes of
action in the three enzymes. Indeed, recent studies
with chitosan as a substrate combined with work by
Hult et al.16) strongly indicate that ChiA and ChiB act
as processive exo-enzymes on chitin, whereas ChiC1 is
a non-processive endo-enzyme.
Synergistic effects on the rate of degradation of
colloidal chitin have been observed in combining
BJL200-ChiA and BJL200-ChiB.15,22) Clear synergism
We have cloned, purified, and characterized ChiA and

Purification and Enzymatic Properties of Endochitinase C
717
ChiB from S. marcescens BJL200.15,23,24) To obtain a
complete picture of the chitinolytic machinery of this
strain and to obtain more detailed insight into the
enzymatic properties of BJL200-ChiC1, we have cloned
and expressed its chiC. Previously, BJL200-ChiA and
BJL200-ChiB from this S. marcescens strain were
purified to homogeneity by an easy one-step procedure
involving hydrophobic interaction chromatography.15,23)
As the second part of this study, we wanted to determine
whether this potentially general purification method
would also be applicable to a chitinase quite different
from ChiA and ChiB, ChiC1 (see below). The avail-
ability of such a potentially general purification method
is important, since some chitinases, and especially
their mutants (G. Vaaje-Kolstad and V. G. H. Eijsink,
unpublished observations) are not easily purified by
common purification methods that are based on inter-
actions with the substrate (e.g., chitin beads). After
purification of BJL200-ChiC1 by this method, the
enzyme was characterized with respect to its activity
towards oligomeric (4MU-GlcNAc3) and polymeric
substrates (CM-chitin-RBV; ꢀ-chitin). In order to
provide a structural perspective on the ChiC catalytic
domain for discussion of the enzymatic mode of action,
the BJL200-ChiC1 sequence was aligned against a
structural alignment of BJL200-ChiA, BJL200-ChiB,
and hevamine.
polymerase (Promega, Madison, WI). The resulting PCR
product was cloned into TOPO TA cloning vector
(Invitrogen) and sequenced. For control of the sequence,
see DNA sequencing and sequence analysis below.
To verify the amplified chiC sequence and to ensure
that no insertions or deletions occurred during the
PCR reaction, the DNA sequence of the chiC gene with
upstream and downstream sequences was verified in
three different clones, yielding identical and correct
sequences. For overexpression of ChiC in E. coli using
T7 promoter vector pRSETB (Invitrogen), two primers,
CCGGGAATTCCATATGAGCACACAAATAACAC
(NdeIChiC1) and M13 forward primer, were used in a
PCR reaction using the TOPO clone as a template. This
PCR reaction produces the chiC gene with an in-frame
ATG and a NdeI site at the start of the gene. The
PCR product was digested with NdeI and XhoI and
ligated into pRSETB digested with the same enzymes.
The resulting plasmid (pRSETB-chiC1) was trans-
formed into E. coli BL21(DE3) star.
DNA sequencing and sequence analysis. The se-
quence of chiC was determined using the ABI PRISM
TM Dye Terminator Cycle Sequencing Ready Reaction
Kit and and an ABI PRISM 37878 DNA Sequencer
(Perkin Elmer Applied Biosystems, Foster City, CA).
Computer analysis of the nucleic acid sequence was
done using the program BioEdit (http://www.mbio.
ncsu.edu/BioEdit/bioedit.html). ORF analyses were
done with the translation tool of the ExPasy Proteomix
tools. BLAST searches were carried out with software
available from the National Center for Biotechnology
Information (NCBI) of USA. The sequence of the
BJL200-ChiC catalytic domain (BJL200-ChiC2, resi-
dues 1-327) was aligned with the profile of the structure-
based multiple sequence alignment of BJL200-ChiA,
BJL200-ChiB, and hevamine (see van Aalten et al.4)),
using Clustal-V.27) Due to limited sequence similarity
between the initial part of the sequence and the profile,
the alignment of the first 90 residues was guided by
multiple alignment of BJL200-ChiC2 and its 20 most
similar sequences (identified by a BLAST search).
Methods
Bacterial strains and culture conditions. Serratia
marcecens BJL200 was grown in Luria-Bertani (LB)
medium at 30 ꢀC for chromosomal DNA extraction.
Escherichia coli TOP 10 cells containing chiC in
plasmid TOPO TA were grown in LB medium supple-
mented with 100 mg/ml ampicillin for plasmid prepara-
tion. E. coli BL21(DE3) star cells (Invitrogen Carlsbad,
CA) transformed with the chiC gene cloned in pRSETB
were grown in NZCYM medium supplemented with
100 mg/ml of ampicillin for chitinase production. In the
latter two cases, the cultures were grown at 37 ꢀC.
DNA extraction and cloning of the chiC gene from
S. marcescens in E. coli. Chromosomal DNA from
S. marcescens was isolated according to the method
described by Saito and Miura,25) with some small
modifications.26) The chiC gene was amplified from
S. marcescens strain BJL200 chromosomal DNA by
PCR using two primers constructed based on nucleotide
sequences upstream of a putative promoter and down-
stream of putative inverted repeats of chiC from
S. marcescens strain KCTC2172.19) The sense primer
upstream of chiC was 50-CCACGCTTGAGCGACATT-
GAAACCATGC-30, and the antisense primer down-
stream of chiC was 50-GGGCAGCATTGCCGATT-
TATCAGCG-30. The amplification consisted of 25
cycles of denaturation at 94 ꢀC for 1 min, annealing
at 55 ꢀC for 1 min, and extension at 72 ꢀC using Tag
One-step purification of BJL200-ChiC1. In the
production of BJL200-ChiC1 in E. coli, BL21(DE3)
expression of chiC1 was induced by 0.4 mM isopropyl-b-
D-thiogalactoside (IPTG) when the OD600 was between
0.5 and 0.7. After growth for 150 min at 37 ꢀC, the
cultures were harvested. Harvested cells were fractio-
nated by the osmotic shock method,28) and BJL200-
ChiC1 was purified from the resulting periplasmic
extracts in a one-step purification step using a Phenyl-
Superose HR 5/5 column in an FPLC system (GE
Healthcare, Oslo, Norway) as previously reported by
Brurberg et al.15,23)
Enzymology. Determination of specific activities of
BJL200-ChiC1 fractions during purification was done

718
B. SYNSTAD et al.
using 4-methyllumbelliferyl-ꢀ-D-N,N0-diacetylchitobio-
side [4MU-(GlcNAc)2] (Sigma, St. Louis, MO) as
described previously.29,30) A 69-mM substrate solution
was used in the assays. The enzyme activity was
expressed as nmol 4MU released per s and mg protein.
Kinetic parameters for ChiA, B and, C1 were de-
termined using 4MU-(GlcNAc)3 by initial rate measure-
ments at various substrate concentrations. The reactions
were carried out in 125 ml of 50 mM citrate-phosphate
buffer, pH 6.3, with substrate concentrations ranging
from 2–8 mM (BJL200-ChiA), 1–10 mM (BJL200-ChiB),
and 40–120 mM (BJL200-ChiC1). The enzyme concen-
trations in assays were 0.15 nM, 0.2 nM, and 10 nM
for BJL200-ChiA, BJL200-ChiB, and BJL200-ChiC1
respectively. The reactions were stopped by adding
125 ml 1 M Glycin/NaOH, pH 10.5. The amount of
4-MU released was measured with a Cary Eclipse
Fluorescence Spectrophotometer (Varian Scientic
Instruments, Palo Alto, CA) with excitation at 365 nm
and emission at 460 nm, using a 4MU standard curve.
Product formation was monitored at four time points for
each reaction, and was found to be linear for at least
12 min for all substrate concentrations. While BJL200-
ChiA and BJL200-ChiC1 quantitatively liberate the
4MU group from 4MU-(GlcNAc)3, BJL200-ChiB
converts this substrate exclusively to 4MU-GlcNAc
and (GlcNAc)2. Hence, the activity of BJL200-ChiB
was monitored in a mixed assay based on measuring the
disappearance of substrate. To achieve this, substrate
remaining at the end of the reactions was quantitatively
converted to free 4MU by adding an overdose of 60 nM
BJL200-ChiA for 15–30 s, before stopping the reactions.
For pH dependency measurements, concentrations
of 30, 60, and 120 mM of 4MU-(GlcNAc)3 (substrate
inhibition was not observed until concentrations greater
than 220 mM) and 53 nM (pH 3.9 to 6.39) and 106 nM
(pH 7 to 9) of BJL200-ChiC1 were used. Aliquots were
taken out and analyzed at 13, 30, and 45 seconds (pH
levels, 3.5 and 3.9), 1, 2, and 3 min (pH levels, 4.4, 4.8,
and 5.3), and 2.5, 5.0, and 7.5 min (pH levels 6.3, 7.0,
8.0, and 9.0). Sample volumes were 50 ml, and the
reactions were stopped by addition to solution of 1.95 ml
of 0.2 M Na2CO3.
in an Eppendorf centrifuge. Absorbance in the super-
natants was measured at 550 nm and corrected for
adsorption in a control sample without chitinase. The
enzyme concentration was 2 nM for BJL200-ChiA and
BJL200-ChiB and 0.5 nM for BJL200-ChiC1. With these
enzyme concentrations and the actual substrate concen-
trations, we found a linear release increase in adsorption
with time for at least 12 min. The specific activities were
expressed as ꢀA550 minꢁ1 per pmole protein.
Hydrolysis of squid pen ꢀ-chitin (3 mm in size;
Seikagaku, Tokyo) was carried out at 37 ꢀC in 50 mM
sodium acetate buffer, pH 6.1.
Protein concentrations were determined by the meth-
od of Bradford31) using a protein assay from Bio-Rad.
Bovine Serum albumin (BSA) was used as the standard.
Chromatography of chito-oligosaccharides. Sizes and
anomers of chito-oligosaccharides were determined by
HPLC using a Tosoh TSK Amide 80 column (0:46 ꢂ
25 cm) with an Amide 80 guard-column. A 10 ml-sample
was injected onto the column, and chitin fragments were
eluted isocratically at 0.7 ml/min with 70% acetonitrile
at room temperature. UV absorbance of the chito-
oligosaccharides was measured at 210 nm, and the peaks
were identified using chito-oligosaccharide standards
purchased from Sigma (St. Louis, MO).
Results
Nucleotide sequence and deduced amino acid
sequence
The chiC gene with flanking sequences was isolated
from the genomic DNA of S. marcescens BJL200 by
PCR amplification using two primers homolog to a
upstream and downstream region of chiC from S. mar-
cescens KCTC2172.19) The resulting PCR product was
cloned into TOPO TA cloning vector. Three individual
clones from three individual PCR reactions were
sequenced to ensure that the obtained sequence, of
1,862 nucleotides, was correct. The same nucleotide
sequence was obtained in the three individual clones.
One open reading frame (ORF) of 1,440 nucleotides
encoding a protein of 480 amino acids was identified
in the sequenced DNA region. The degrees of identity
to the two previously described enzymes were 95%
and 96% at the DNA level and 93% and 98% at the
protein level for the enzymes from strain KCTC217219)
and strain 2170,20) respectively. This clearly indicates
that the gene cloned here is the counterpart of the two
previously described chiC genes. A possible Shine
Dalgarno sequence (GAGG) five bases upstream of the
ATG start codon and a putative terminator nine bases
downstream of the UAA stop codon were identified
(data not shown). These sequences are also found in the
upstream and downstream regions of the two other chiC
genes published for S. marcescens.19,20)
Km- and kcat-values were determined using hyperbolic
regression with the program Hyper (J.S. Easterby,
University of Liverpool, obtainable from homepage
http://www.liv.ac.uk./~jse/software.html).
Chitinase activity was also measured using the
polymeric substrate, carboxymethyl chitin Remazol
Brilliant Violet (CM-chitin-RBV, Loewe Biochemica,
Munich, Germany). A 5-ml reaction mixture containing
37.5 mM NaAcetate, pH 5.5, and 0.6 mg/ml CM-chitin-
RBV was preincubated at 37 ꢀC before the chitinase was
added to the solution. The reaction was monitored every
third min up to 15 min by transfer of a sample of 800 ml
to a new tube containing 200 ml 0.2 M HCl to stop the
reaction. After mixing, the samples were cooled on ice
for 10 min and centrifuged at maximum speed for 5 min
The sequences of BJL200-ChiC1 and the correspond-
ing gene chiC with upstream and downstream regions

Purification and Enzymatic Properties of Endochitinase C
719
Fig. 2. Purification of ChiC1.
Left panel, hydrophobic interaction chromatography. The picture shows the A280 trace and the gradient (- - - - - - -) from 100% buffer A (20 mM
Tris, 1 mM EDTA, and 0.5 M (NH4)2SO4, pH 8.0) to 100% buffer B (20 mM Tris and 4% isopropanol, pH 8.0). The first large peak in the
chromatogram (marked I) represents waste proteins, whereas the second peak contains ChiC1 (marked II). The ChiC1 peak is symmetrical and
shows good separation characteristics. The third peak (marked III) does not contain ChiC1 as controlled by SDS–page analysis. Right panel,
SDS–page gel analysis of ChiC1 purification. Lane 1, protein standard; lane 2, the periplasmatic extract; lane 3, purified protein from peak II.
Table 1. One-Step Purification of S. marcescens Chitinase BJL200-ChiC1 by Hydrophobic Interaction Chromatography
Volume
(ml)
Total protein
(mg)
Total activity
(nmol sꢁ1
Recovery
(%)
Specific activity
(nmol sꢁ1 mgꢁ1
)
Fraction
)
Periplasmatic extract
Peak fractions
2.0
1.7
2.92
0.31
4.4
3.8
100
86
1.5
12.3
from S. marcescens strain BJL200 have been deposited
in the EMBL Nucleotide Sequence Database under
accession no. AJ630582.
Overproduction and purification of BJL200-ChiC1
BJL200-ChiC1 was overproduced in E. coli by the
T7 expression system.32) After induction with IPTG,
the cells were fractionated by osmotic shocking. It has
been found that the osmotic shocking method used
here somehow leaves most of recombinantly produced
BJL200-ChiA (containg an N-terminal signal peptide for
sec-dependent secretion) or BJL200-ChiB (not contain-
ing a recognizable secretion signal24)) in the periplasmic
fraction, providing a chitinase-enriched starting material
for purification. Although BJL200-ChiC1 does not con-
tain any recognizable signal for secretion, approximately
88% of the chitinase activity was detected in the
periplasmic extract. BJL200-ChiC1 was purified from
this extract using hydrophobic interaction chromatog-
raphy (HIC) (Fig. 2). In a typical experiment, the
recovery of enzyme activity is between 80% and 90%
(Table 1). This is in good agreement with the yields of
BJL200-ChiA and BJL200-ChiB by the same purifica-
tion method.15,23) SDS–PAGE showed that the chitinase
peak contained a highly pure protein of approximately
50 kb (Fig. 2).
Comparison of the catalytic domain of BJL200-ChiC1
with other family 18 chitinases
A structural alignment of the catalytic domain of
BJL200-ChiC1 with the corresponding domains in
BJL200-ChiA and BJL200-ChiB from S. marcescens
and with hevamine is shown in Fig. 1. Generally, there
were very few fully conserved residues. The few regions
with full sequence conservation included the catalyti-
cally crucial SXGG sequence at the end of the third ꢀ-
strand and the DXXDXDXE motif with the catalytic
glutamic acid that spans the fourth ꢀ-strand. The main
difference among the four catalytic domains was the
absence of the so-called ꢁ=ꢀ-domain in hevamine and
BJL200-ChiC1.3) In BJL200-ChiA and BJL200-ChiB,
this domain is situated between the seventh ꢀ-strand and
ꢁ-helix, and it consists of residues 295–373 in BJL200-
ChiB and residues 447–509 in BJL200-ChiA. This
domain, comprising five ꢀ-sheets and two helixes,
makes up one of the sides of the deep substrate binding
cleft in BJL200-ChiA and BJL200-ChiB.
Since the results with BJL200-ChiC1 indicated
that the periplasmic extraction procedure combined

720
B. SYNSTAD et al.
22)
Table 2. Kinetic Parameters and Hydrolytic Activity of Serratia marcescens Chitinases Using 4MU-(GlcNAc)3
Substrates
and CM-Chitin-RBV as
4MU-(GlcNAc)3
CM-chitin-RBV
Km
kcat
(sꢁ1
kcat=Kꢁm1
Sp. Activity
(ꢀOD550 minꢁ1, pmoleꢁ1
)
Enzyme
(mM)
)
(sꢁ1 mM
)
BJL200-ChiA
BJL200-ChiB
BJL200-ChiC1
4:2 ꢃ 0:4
6:8 ꢃ 1:0
79:8 ꢃ 7:5
67:4 ꢃ 3
56:8 ꢃ 4:2
2:0 ꢃ 0:1
16
0:0035 ꢃ 0:0005
0:0047 ꢃ 0:0010
0:066 ꢃ 0:001
8.4
0.025
with hydrophobic interaction chromatography does
indeed provide a general method for purification of
family 18 chitinases, we applied the same one-step
procedure to a variety of chitinase variants, including
inactive mutants such as BJL200-ChiC1-E141Q,
BJL200-ChiB-E144Q, and BJL200-ChiA-E315Q, which
we could not easily purify using chitin-affinity chroma-
tography. All these mutants were successfully purified,
with yields and purities comparable to those of the wild-
type enzymes. BJL200-ChiB variants with mutations
in the chitin-binding domain or in aromatic residues in
the substrate-binding groove were also purified by
this method, but they eluted from the column earlier
than other BJL200-ChiB variants (see below for further
discussion).
longer oligomers are initially produced by BJL200-
ChiC1 is in accordance with the notion that this
enzyme is an endo-acting one. However, as the reaction
proceeds, the longer oligosaccharides are degraded and
the final product is a mixture of only GlcNAc and
(GlcNAc)2, as for BJL200-ChiA and BJL200-ChiB.33)
The pH dependency of 4MU-(GlcNAc)3 on BJL200-
ChiC1 was investigated (Table 3). The results show that
BJL200-ChiC1 has a pH optimum at 3.9 (16:1 ꢃ 3:0 sꢁ1
)
and that enzyme turnover decreases with increasing
pH. Substrate efficiency is also highest at pH 3.9
(0.114 sꢁ1 mMꢁ1), with subsequent decreases with in-
creasing pH. The enzyme was not stable for the time
window of the experiment at pH levels lower than 3.5,
and hence the kinetic parameters were not determined
at pH levels lower than this.
Enzymatic characterization of BJL200-ChiC1 and
comparison with BJL200-ChiA and BJL200-ChiB
The hydrolytic activities of BJL200-ChiA, BJL200-
ChiB, and BJL200-ChiC1 against various chitinous
substrates were determined. Extensive reports on com-
parative studies with a variety of substrates may be
found in the work of Horn et al.22,33) For the purpose of
this study, we compared activity of the three enzymes
towards an oligomeric substrate, 4MU-(GlcNAc)3, with
the activity towards a polymeric substrate, CM-chitin-
RBV. In addition, we studied product formation by
BJL200-ChiC1 from ꢀ-chitin.
Table 2 summarizes the activities of BJL200-ChiA,
BJL200-ChiB, and BJL200-ChiC1 against 4MU-
(GlcNAc)3 and CM-chitin-RBV. Both BJL200-ChiA
and BJL200-ChiB had higher activity against 4MU-
(GlcNAc)3 than did BJL200-ChiC1. When kinetic
parameters kcat and Km were determined using this
substrate, BJL200-ChiA and BJL200-ChiB had approx-
imately 30 fold higher kcat and 15 fold lower Km than
BJL200-ChiC1. On the other hand, the specific activity
of BJL200-ChiC1 towards the polymeric substrate
CM-chitin-RBV was approximately 10 and 15 times
higher respectively than that of BJL200-ChiB and
BJL200-ChiA.
Discussion
In this study, we cloned, expressed, and purified
ChiC1 from S. marcescens strain BJL200. The BJL200-
ChiC1 amino acid sequence resembled those of the
two previously described ChiCs from S. marcescens.
Generally, family 18 chitinases are highly conserved
between S. marcescens strains. For the BJL200 and
2170 strains, the sequence identities at the amino acid
level are 99.3%, 98.3%, and 98.3% for ChiA, ChiB, and
ChiC1 respectively. ChiC1 from Serratia marcescens
strain 2170 has been characterized by Suzuki et al.20)
This enzyme consists of a catalytic domain and two
putative C-terminal chitin-binding domains. The cata-
lytic domain lacks the ꢁ=ꢀ-domain, which makes up one
of the walls of the substrate binding clefts in ChiA and
ChiB. The absence of this domain and the consequent
shallowness of the predicted substrate-binding groove
are in accordance with the idea that ChiC1 is an
endochitinase, like hevamine.34)
The experimental results presented above also con-
firm the idea that BJL200-ChiC1 acts as an endo-
enzyme: BJL200-ChiC1 is much less active than the the
processive exo-chitinases towards oligomeric substrates,
whereas it is much more active towards complex
polymeric substrates, i.e., substrates that are likely
to be the primary target of the endo-acting member of
the chitinolytic enzyme machinery. Our results are in
agreement with those obtained by Suzuki et al.,10) who
reported that 2170-ChiC1 hydrolyzes (GlcNAc)4 more
slowly than do 2170-ChiA and 2170-ChiB.
As Fig. 3 shows, initial product mixtures formed
during degradation of ꢀ-chitin with BJL200-ChiC1
contained not only (GlcNAc)2 and GlcNAc, but also
the longer oligosaccahrides (GlcNAc)3, (GlcNAc)4, and
(GlcNAc)5. Degradation of ꢀ-chitin by BJL200-ChiA
and BJL200-ChiB initially yields GlcNAc2 and a small
amounts of GlcNAc and (GlcNAc)3.33) The fact that

Purification and Enzymatic Properties of Endochitinase C
721
(GlcNAc)
2
60
50
40
30
20
10
0
(GlcNAc)
3
(GlcNAc)
GlcNAc
4
(GlcNAc)
5
10
15
20
25
Retention time (min)
400
350
300
250
200
150
100
50
(GlcNAc)
2
GlcNAc
0
10
15
20
25
Retention time (min)
Fig. 3. HPLC Analysis of Degradation Products of ꢀ-Chitin by ChiC after 10 min (top) and 1 (bottom, lower panel) and 48 h (bottom, upper panel).
The ꢁ-anomer eluted before the ꢀ-anomer; thus, two peaks were observed for each chito-oligosaccharide. (There was no temperature control
of the column during experiments, and retention time and shape of the peaks can vary with column temperature).
Whereas ChiA and ChiB from Serratia marcescens
(both BJL200 and 2170) have broad neutral pH optima
for activity,10,15) this study indicates that BJL200-ChiC1
has a clear acidic pH optimum, at about pH 4.0.
Table 3. pH Dependency of 4MU-(GlcNAc)3 Hydrolysis by
BJL200-ChiC1
kcat
(sꢁ1
Km
kcat=Kꢁm1
(sꢁ1 mM
)
Moreover, BJL200-ChiC1 contains an Asn at position
209, whereas BJL200-ChiA and BJL200-ChiB contain
an Asp at the corresponding positions, 391 and 215
respectively (Fig. 1). Using site-directed mutagenesis,
Synstad et al. have shown for BJL200-ChiB that
mutation of Asp215 to asparagine leads to a drastic
acidic shift in the pH optimum of this enzyme.29)
Asp215 in BJL200-ChiB stabilizes the observed boat
conformation of the sugar in the ꢁ1 subsite during
hydrolysis by accepting a hydrogen bond from the O6
pH
)
(mM)
3.5
3.9
4.4
4.8
5.3
6.3
7.0
8.0
9.0
14:3 ꢃ 2:3
16:1 ꢃ 3:0
9:3 ꢃ 1:2
158 ꢃ 35
140 ꢃ 38
0.091
0.114
0.095
0.072
0.053
0.025
0.009
0.003
0.001
98:1 ꢃ 20:8
95:5 ꢃ 15:5
94:0 ꢃ 19:8
80:0 ꢃ 7:5
83:7 ꢃ 12:9
113:1 ꢃ 18:3
90:7 ꢃ 21:6
6:9 ꢃ 0:7
5:0 ꢃ 0:6
2:0 ꢃ 0:1
0:74 ꢃ 0:07
0:31 ꢃ 0:04
0:06 ꢃ 0:01

722
B. SYNSTAD et al.
hydroxy group of the sugar.29) In addition, Asp215
increases the pKa values of the Asp142-Glu144 diad,
which is a part of the conserved DXXDXDXE motif.29)
For BJL200-ChiB-D215N, the hydrogen bond to the
sugar moiety is retained while the increase in the pKa
values for the Asp142-Glu144 diad is abolished, result-
ing in a pH profile similar to that of BJL200-ChiC1.
Cloned ChiC1 protein from S. marcescens BJL200
was recovered from E. coli BL21(DE3) by extracting
periplasmic proteins using cold osmotic shock treatment
of the cells. The same procedure has worked very
well for BJL200-ChiA and BJL200-ChiB, as well as
for numerous BJL200-ChiB variants produced in our
laboratories.15,23) Remarkably, of these proteins, only
BJL200-ChiA contains a typical N-terminal signal
sequence for sec-dependent secretion. Although the
reasons for this are unclear, it is clear that the osmotic
shock method produces very good starting preparations
for purification of recombinant family 18 chitinases
produced in E. coli.
The use of perplasmic extracts as starting material
permitted the purification of BJL200-ChiC1 by a single
fast hydrophobic interaction chromatography step,
which has also been used successfully in the purification
of BJL200-ChiA and BJL200-ChiB.15,23) As for the
purification of BJL200-ChiA and BJL200-ChiB,
BJL200-ChiC1 was bound to the column material in
20 mM Tris-buffer with 0.5 M ammoniumsulphate, and
was eluted in a sharp symmertrical peak after the
concentration of buffer B, consisting of 20 mM Tris and
4% isopropanol, reached 100% in the chromatographic
system (Fig. 2). Thus, remarkably, this one-step chro-
matographic procedure appears to be a general method
of purifying family 18 chitinases from S. marcescens. It
is tempting to speculate that the interaction between
the chitinases and phenyl-sepharose somehow involves
stacking interactions between the aromatic phenyl ring
and the many aromatic residues that line the substrate-
binding grooves of the enzymes and that cover the
putative binding surfaces of the chitin-binding do-
mains.3,4,13,20) In accordance with this idea, it was
observed that BJL200-ChiB variants in which residues
Trp97 or Trp220 in the substrate-binding groove had
been replaced by alanine, as well as a variant truncated
at residue 446 (i.e., lacking the C-terminal chitin-
binding domain4)) eluted earlier from the column, at
approximately 80% buffer B (data not shown).
Aknowlegdement
This work was supported by a grant from the
Norwegian Research Council. We thank Xiaohong Jia
and Ellen Hasle Kokkim for technical assistance.
References
1) Henrissat, B., and Davies, G., Structural and sequence-
based classification of glycoside hydrolases. Curr. Opin.
Struct. Biol., 7, 637–644 (1997).
2) Matsumoto, T., Nonaka, T., Hashimoto, M., Watanabe,
T., and Mitsui, Y., Three-dimensional structure of the
catalytic domain of chitinase A1 from Bacillus circulans
WL-12 at a very high resolution. Proc. Jpn. Acad. Ser.
B-Phys. Biol. Sci., 75, 269–274 (1999).
3) Perrakis, A., Tews, I., Dauter, Z., Oppenheim, A. B.,
Chet, I., Wilson, K. S., and Vorgias, C. E., Crystal
structure of a bacterial chitinase at 2.3 A resolution.
Structure, 2, 1169–1180 (1994).
4) van Aalten, D. M. F., Synstad, B., Brurberg, M. B.,
Hough, E., Riise, B. W., Eijsink, V. G. H., and
Wierenga, R. K., Structure of a two-domain chitotriosi-
dase from Serratia marcescens at 1.9-angstrom resolu-
tion. Proc. Natl. Acad. Sci. USA, 97, 5842–5847 (2000).
5) Hashimoto, M., Ikegami, T., Seino, S., Ohuchi, N.,
Fukada, H., Sugiyama, J., Shirakawa, M., and Watanabe,
T., Expression and characterization of the chitin-binding
domain of chitinase A1 from Bacillus circulans WL-12.
J. Bacteriol., 182, 3045–3054 (2000).
6) Ikegami, T., Okada, T., Hashimoto, M., Seino, S.,
Watanabe, T., and Shirakawa, M., Solution structure of
the chitin-binding domain of Bacillus circulans WL-12
chitinase A1. J. Biol. Chem., 275, 13654–13661 (2000).
7) Watanabe, T., Ito, Y., Yamada, T., Hashimoto, M.,
Sekine, S., and Tanaka, H., The roles of the C-terminal
domain and type-III domains of chitinase A1 from
bacillus-circulans Wl-12 in chitin degradation. J. Bac-
teriol., 176, 4465–4472 (1994).
8) Fuchs, R. L., Mcpherson, S. A., and Drahos, D. J.,
Cloning of a Serratia-Marcescens gene encoding chiti-
nase. Appl. Environ. Microbiol., 51, 504–509 (1986).
9) Suzuki, K., Suzuki, M., Taiyoji, M., Nikaidou, N., and
Watanabe, T., Chitin binding protein (CBP21) in the
culture supernatant of Serratia marcescens 2170. Biosci.
Biotechnol. Biochem., 62, 128–135 (1998).
10) Suzuki, K., Sugawara, N., Suzuki, M., Uchiyama, T.,
Katouno, F., Nikaidou, N., and Watanabe, T., Chitinases
A, B, and C1 of Serratia marcescens 2170 produced by
recombinant Escherichia coli: enzymatic properties and
synergism on chitin degradation. Biosci. Biotechnol.
Biochem., 66, 1075–1083 (2002).
The one step hydrophobic interaction chormatography
procedure also allowed for the purification of inactive
mutants such as BJL200-ChiC1-E141Q, BJL200-ChiB-
E144Q, and BJL200-ChiA-E315Q. When chitin affinity
column chromatography, a standard chromatographic
method for the isolation of chitinases,10,12,19,20,35) is used,
these inactive mutants tend to become irreversibly
bound to the chitin beads of the column, and can be
eluted only as denatured proteins with a 0.3 M NaOH
solution (G: Vaaje-Kolstad and V. G. H: Eijsink,
unpublished observations).
11) Tews, I., Vincentelli, R., and Vorgias, C. E., N-
Acetylglucosaminidase (chitobiase) from Serratia
marcescens: gene sequence, and protein production and
purification in Escherichia coli. Gene, 170, 63–67
(1996).
12) Watanabe, T., Kimura, K., Sumiya, T., Nikaidou, N.,
Suzuki, K., Suzuki, M., Taiyoji, M., Ferrer, S., and
Regue, M., Genetic analysis of the chitinase system of
Serratia marcescens 2170. J. Bacteriol., 179, 7111–7117
(1997).

Purification and Enzymatic Properties of Endochitinase C
723
13) Papanikolau, Y., Prag, G., Tavlas, G., Vorgias, C. E.,
Oppenheim, A. B., and Petratos, K., High resolution
structural analyses of mutant chitinase A complexes
with substrates provide new insight into the mechanism
of catalysis. Biochemistry, 40, 11338–11343 (2001).
14) Aronson, N. N., Halloran, B. A., Alexyev, M. F.,
Amable, L., Madura, J. D., Pasupulati, L., Worth, C., and
Van Roey, P., Family 18 chitinase-oligosaccharide
substrate interaction: subsite preference and anomer
selectivity of Serratia marcescens chitinase A. Biochem.
J., 376, 87–95 (2003).
15) Brurberg, M. B., Nes, I. F., and Eijsink, V. G. H.,
Comparative studies of chitinases A and B from Serratia
marcescens. Microbiology, 142, 1581–1589 (1996).
16) Hult, E. L., Katouno, F., Uchiyama, T., Watanabe, T.,
and Sugiyama, J., Molecular directionality in crystalline
beta-chitin: hydrolysis by chitinases A and B from
Serratia marcescens 2170. Biochem. J., 388 (Pt. 3),
851–856 (2005).
gene product. FEMS Microbiol. Lett., 124, 399–404
(1994).
24) Brurberg, M. B., Eijsink, V. G. H., Haandrikman, A. J.,
Venema, G., and Nes, I. F., Chitinase B from Serratia
marcescens BJL200 is exported to the periplasm without
processing. Microbiology, 141, 123–131 (1995).
25) Saito, H., and Miura, K., Preparation of transforming
deoxyribonucleic acid by phenol treatment. Biochim.
Biophys. Acta, 72, 619 (1963).
˚
26) Synstad, B., Emmerhoff, O., and Sirevag, R., Malate
dehydrogenase from the green gliding bacterium
Chloroflexus aurantiacus is phylogenetically related to
lactic dehydrogenases. Arch. Microbiol., 165, 346–353
(1996).
27) Higgins, D. G., Bleasby, A. J., and Fuchs, R., Hot
sciences–clustal-V–
papers–computer-science
life
improved software for multiple sequence alignment.
Scientist, 8, 17 (1994).
28) Manoil, C., and Beckwith, J., A genetic approach to
analyzing membrane-protein topology. Science, 233,
1403–1408 (1986).
17) Uchiyama, T., Katouno, F., Nikaidou, N., Nonaka, T.,
Sugiyama, J., and Watanabe, T., Roles of the exposed
aromatic residues in crystalline chitin hydrolysis by
chitinase A from Serratia marcescens 2170. J. Biol.
Chem., 276, 41343–41349 (2001).
˚
29) Synstad, B., Gaseidnes, S., van Aalten, D. M. F., Vriend,
G., Nielsen, J. E., and Eijsink, V. G. H., Mutational and
computational analysis of the role of conserved residues
in the active site of a family 18 chitinase. Eur. J.
Biochem., 271, 253–262 (2004).
18) Sikorski, P., Sorbotten, A., Horn, S. J., Eijsink, V. G. H.,
˚
and Varum, K. M., Serratia marcescens chitinases with
tunnel-shaped substrate-binding grooves show endo
activity and different degrees of processivity during
enzymatic hydrolysis of chitosan. Biochemistry, 45,
9566–9574 (2006).
30) Vaaje-Kolstad, G., Houston, D. R., Rao, F. V., Peter,
M. G., Synstad, B., van Aalten, D. M. F., and Eijsink,
V. G. H., Structure of the D142N mutant of the family
18 chitinase ChiB from Serratia marcescens and its
complex with allosamidin. Biochim. Biophys. Acta,
1696, 103–111 (2004).
31) Bradford, M. M., A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing
principle of protein-dye binding. Anal. Biochem., 72,
248–254 (1976).
19) Gal, S. W., Choi, J. Y., Kim, C. Y., Cheong, Y. H., Choi,
Y. J., Lee, S. Y., Bahk, J. D., and Cho, M. J., Cloning of
the 52-kDa chitinase gene from Serratia marcescens
KCTC2172 and its proteolytic cleavage into an active
35-kDa enzyme. FEMS Microbiol. Lett., 160, 151–158
(1998).
20) Suzuki, K., Taiyoji, M., Sugawara, N., Nikaidou, N.,
Henrissat, B., and Watanabe, T., The third chitinase gene
(chiC) of Serratia marcescens 2170 and the relationship
of its product to other bacterial chitinases. Biochem. J.,
343, 587–596 (1999).
21) Terwisscha van Scheltinga, A. C., Armand, S., Kalk,
K. H., Isogai, A., Henrissat, B., and Dijkstra, B. W.,
Stereochemistry of chitin hydrolysis by a plant chiti-
nase/lysozyme and X-ray structure of a complex with
allosamidin: evidence for substrate assisted catalysis.
Biochemistry, 34, 15619–15623 (1995).
32) Studier, F. W., Rosenberg, A. H., Dunn, J. J., and
Dubendorff, J. W., Use of T7 RNA-polymerase to
direct expression of cloned genes. Methods Enzymol.,
185, 60–89 (1990).
33) Horn, S. J., Sørbotten, A., Synstad, B., Sikorski, P.,
˚
Sørlie, M., Varum, K. M., and Eijsink, V. G. H.,
Endo/exo mechanism and processivity of family 18
chitinases produced by Serratia marcescens. FEBS J.,
273, 491–503 (2006).
34) Terwisscha van Scheltinga, A. C., Hennig, M., and
˚
Dijkstra, B. W., The 1.8 A resolution structure of
22) Horn, S. J., Sørlie, M., Vaaje-Kolstad, G., Norberg,
˚
hevamine, a plant chitinase/lysozyme, and analysis of
the conserved sequence and structure motifs of
glycosyl hydrolase family 18. J. Mol. Biol., 262, 243–
257 (1996).
A. L., Synstad, B., Varum, K. M., and Eijsink, V. G. H.,
Comparative studies of chitinases A, B and C from
Serratia marcescens. Biocatal. Biotransfor., 24, 39–53
(2006).
35) Taira, T., Toma, N., and Ishihara, M., Purification,
characterization, and antifungal activity of chitinases
from pineapple (Ananas comosus) leaf. Biosci. Biotech-
nol. Biochem., 69, 189–196 (2005).
23) Brurberg, M. B., Eijsink, V. G. H., and Nes, I. F.,
Characterization of a chitinase gene (chiA) from Serratia
marcescens BJL200 and one-step purification of the