|
|
|
|||||||
|
|||||||||
(Received for publication, October 27, 1994) From the
In the commonly accepted mechanism for enzymatic hydrolysis of
cellulose, endo- The fungal and bacterial The catalytic domain of cellobiohydrolase II (CBH II), (
Figure 1:
Structural analysis of the catalytic domain
of Thermomonospora fuscaE2, a family 6
endo-
Figure 2:
Partial
amino acid sequence of CbhA
The
hypothesis described above implies that deletion of the loops covering
its active site would allow an exoglucanase to hydrolyze internal
Purified CbhA and CbhA Simultaneous determination of specific
fluidity ( Michaelis-Menten parameters for the
hydrolysis of 2`,4`-dinitrophenyl- Hydrolysis of insoluble cellulose
was assayed by incubation of 1.5 mg of acid-swollen CF1 cellulose (13) and 1.5 nmol of enzyme in 1.5 ml of 5 mM potassium phosphate, pH 7, at 37 °C. Aliquots were removed at
2, 6, and 24 h and centrifuged to sediment residual cellulose. The
resulting supernatants were incubated for 5 min at 100 °C to
prevent further hydrolysis of soluble sugar prior to analysis. Total
soluble sugar in supernatants was determined by the phenol sulfuric
acid assay(14) ; component sugars were analyzed by high
performance liquid chromatography(6) . Cellotetraose
hydrolysis was examined by incubation of 50 pmol of enzyme and 40 nmol
of cellotetraose in 75 µl of 5 mM potassium phosphate, pH
7, at 37 °C for 30 min. Reactions were stopped by incubation at 100
°C for 5 min and the products analyzed by high performance liquid
chromatography.
A disulfide bond occurs between
Cys
Figure 3:
Relationship between specific fluidity
(
The plot for CbhA
The rate of hydrolysis of insoluble
cellulose by CbhA
Figure 4:
Release of total soluble sugar from
acid-swollen cellulose by CbhA (
Figure 5:
High performance liquid chromatographic
analysis of products from cellulose (A) or cellotetraose (B). Soluble cello-oligosaccharides released from acid-swollen
cellulose were analyzed after hydrolysis by CbhA, CbhA
Volume 270,
Number 9,
Issue of March 3, 1995 pp. 4383-4386
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
-1,4-glucanase Activity of an Exocellobiohydrolase by
Deletion of a Surface Loop (*)
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
-1,4-glucanases randomly cleave glucosidic bonds
within glucan polymers, providing sites for attack by
exo-cellobiohydrolases (EC 3.2.1.91). It has been proposed that
hydrolysis by Trichoderma reesei cellobiohydrolase II is
restricted to the ends of cellulose polymers because two surface loops
cover its active site to form a tunnel. In a closely related
endoglucanase, E2 from Thermomonospora fusca, access
to the substrate appears to be relatively unhindered because the
carboxyl-proximal loop is shortened, and the amino-proximal loop is
displaced. The hypothesis was examined by deletion of a region in Cellulomonas fimi cellobiohydrolase A corresponding to part of
the carboxyl-proximal loop of T. reesei cellobiohydrolase II.
The mutation enhanced the endoglucanase activity of the enzyme on
soluble O-(carboxymethyl)cellulose and altered its activities
on 2`,4`-dinitrophenyl--D-cellobioside, insoluble
cellulose, and cellotetraose.
-1,4-glucanases involved in
cellulose degradation are usually comprised of two or more structural
and functional domains; a catalytic domain joined to a
cellulose-binding domain is a common arrangement(1) . The
enzymes range in size from about 20 to 120 kDa, but nearly all can be
classified into a few distinct families based on sequence identity of
their catalytic domains alone(1, 2) . Family 6,
originally called family B, contains eight known members; these include
both exoglucanases (cellobiohydrolases) and endoglucanases(3) . 
)a family 6 exo--1,4-glucanase from Trichoderma
reesei, contains a central 
/ barrel, similar to that of
triose-phosphate isomerase(4) . Two extensive surface loops
extend from the carboxyl end of the barrel and enclose the active site
to form a tunnel-shaped structure (Fig. 1). Amino acid sequence
alignment suggests that the two surface loops covering the CBH II
active site are shortened in the family 6 endoglucanases. These
observations led to the hypothesis that the basic difference between
exo- and endoglucanases is the accessibility of their active sites to
internal -1,4-glucosidic bonds in polymeric substrates. The
tunnel-shaped active site of an exoglucanase restricts hydrolysis to
-1,4-glucosidic bonds at the ends of cellulose
molecules(4) .
-Carbon
skeletons of the T. reesei CBH II and T. fuscaE2 catalytic domains. The views are chosen to illustrate
differences in the accessibilities of the two active sites. C and N, respectively, indicate the carboxyl- and
amino-proximal loops that cover the active site of CBH
II.
-1,4-glucanase, subsequently provided support for this
proposal(5) . Although the general topology of the E2
/ barrel is very similar to that of CBH II, the E2
active site is an open cleft structure (Fig. 1). The
carboxyl-proximal loop covering the CBH II active site
(Val
-Ala
, Fig. 2) is indeed
shortened in E2, whereas the amino-proximal loop
(Pro
-Asp
in CBH II), although present, is
bent back so that it no longer covers the active site.
C, a deletion mutant of CbhA, and its
alignment with corresponding regions of the catalytic domains of CbhA
and other known family 6 -1,4-glucanases. The region in CbhA that
was deleted is underlined. CbhA (GenBank L25809) and CenA
(GenBank M15823) are from C. fimi, CBH II (GenBank M16190)
from T. reesei, CEL3 (GenBank L24519) from Agaricus
bisporus, CelA (Swiss-Prot P26414) from Microbispora
bispora, Cel1 (Swiss-Prot P33682) from Streptomyces
halstedii; CasA (GenBank L03218) from Streptomyces sp.
KSM-9, and E2 (GenBank M73321) from Thermomonospora
fusca. CbhA, CBH II, and CEL3 are exocellobiohydrolases; CenA,
Cel1, CasA, and E2 are endoglucanases. Each sequence is numbered from the first residue of the mature enzyme; hyphens indicate gaps introduced to improve the alignment. Asterisks mark a pair of cysteine residues known to form one
of two disulfide bonds in the catalytic domains of CenA, CBH II, and E2; the arrow indicates the putative general base
catalyst(4) .
-1,4-glucosidic bonds in cellulose molecules. We tested this
prediction by constructing a mutant of cellobiohydrolase A (CbhA), a
family 6 exo--1,4-glucanase from Cellulomonas fimi, in
which residues corresponding to part of the CBH II carboxyl-proximal
active site loop are deleted.
Construction of a Plasmid Encoding
CbhA
The sequence of plasmid pALM1 (3) was changed
by site-directed mutagenesis to introduce NcoI and NheI sites (underlined) at the start of the cbhA gene, using the oligonucleotide
5`-CACCCGGAGGACCCCATGGCTAGCCTCGGCAAGCGAGCA-3`. This allowed insertion
of an NcoI-EcoRI fragment containing cbhA into the NcoI- and EcoRI-digested plasmid
pDAM2-1; this vector was chosen because it was known to give high
level expression of another C. fimi C-1,4-glucanase gene, cenD(6) . The resulting plasmid (pALM3) served as
template for further mutagenesis. Mutation of cbhA to cbhAC used the oligonucleotide
5`-CCCCCGGGTGAGTCCGACGGCGCATGCGACCCGACGTTCGTCTCGCCC-3`. The SphI site (underlined) was introduced as a silent
mutation to facilitate screening. The resulting plasmid, pALM3.2,
encoded CbhAC, a mutant in which CbhA residues corresponding to
part of the carboxyl-proximal active site loop of CBH II are deleted (Fig. 2).Preparation and Assay of Enzymes
CbhA(3) ,
CenA(7) , and Abg (8) were purified from Escherichia coli containing appropriate recombinant plasmids
by affinity chromatography on microcrystalline cellulose and
anion-exchange chromatography, as described previously. CbhAC was
purified from E. coli DH5 F` (pALM3.2), as described for
CbhA(3) . Purifications were monitored by SDS-polyacrylamide
gel electrophoresis(9) .C were
examined for the presence of free thiol groups by titration with
5,5`-dithiobis(2-nitrobenzoic acid)(10) . Fifty µl of
buffer (100 mM potassium phosphate, pH 7.3, 1 mM EDTA, 6 M guanidine hydrochloride) containing 3 mM 5,5`-dithiobis(2-nitrobenzoic acid) were added to 1 ml of buffer
containing 10 nmol of enzyme, and the mixture was incubated at 25
°C. Formation of nitrothiobenzoate was monitored from its
absorbance at 412 nm over a period of 1 h. The lower limit of detection
was 0.1 nmol of free thiol.
= 
) and
reducing sugar during Cm-cellulose hydrolysis has been
described(11) . Reaction mixtures contained 2 µM CbhA, 2 µM CbhAC, 3.6 nM CenA or 300
nM Abg, and 3.2% (w/v) Cm-cellulose (Sigma; low viscosity
grade; nominal degree of substitution, 0.7; nominal degree of
polymerization, 400) in 50 mM sodium citrate, pH 7; mixtures
were incubated at 37 °C.-D-cellobioside were
determined by spectrophotometric measurement of the rate of
dinitrophenolate release in 50 mM potassium phosphate, pH 7,
at 37 °C (12) ; substrate concentrations were from 0.2 to
10 
K
.Structural Representations
Coordinates for CBH II
were from the Protein Data Bank (acquisition number 3CBH); coordinates
for E2 were generously provided by Drs. Spezio, Wilson, and
Karplus.
General Characterization of CbhA
The
region SerC
-Met
of CbhA (Fig. 2) was deleted to give the mutant protein CbhAC. The
deleted residues correspond to 11 of the 
20-amino acid residue
carboxyl-proximal loop covering the active site in CBH II ( Fig. 1and Fig. 2). Purified CbhAC co-migrated with
CbhA on an SDS-polyacrylamide gel with an apparent molecular mass of
about 85 kDa, in agreement with its predicted molecular
mass(3) . No contaminating polypeptides were visible on a
heavily loaded gel (data not shown).
and Cys
in C. fimi CenA, a
family 6 endo--1,4-glucanase (Fig. 2); a second bond is
formed between Cys
and Cys
(7) .
Bonds occur between corresponding residues in CBH II (4) and E2(15) ; presumably, they are common to all family 6
catalytic domains because the 4 cysteine residues are strictly
conserved(3) . The 11-residue deletion of CbhAC is
immediately adjacent to Cys (Fig. 2); possible
interference with disulfide bond formation was examined using
5,5`-dithiobis(2-nitrobenzoic acid). No free thiol groups were detected
in either CbhAC or CbhA, indicating that all disulfide bonds are
correctly formed in the mutant and wild-type enzymes.Relative Exo- and Endohydrolytic Activities of CbhA and
CbhA
The preference of a C-1,4-glucanase for terminal
or internal glucosidic bonds can be evaluted from a plot of specific
fluidity () versus reducing sugar
production during hydrolysis of soluble Cm-cellulose. The hydrolysis of
any -1,4-glucosidic bond generates one new reducing sugar group,
but the random action of an endoglucanase results in a greater increase
in specific fluidity than attack by an exoglucanase(11) . The
slope of the plot for CbhA was 17.1
mlmmol (Fig. 3). The slope for CbhA is considered representative
of exoglucanase activity because a similar low slope (13.3
mlmmol) was obtained
with Abg (Fig. 3); Abg is an Agrobacterium sp.
-1,4-glucosidase that cleaves glucosyl residues from the
nonreducing end of cellobiose and longer -1,4-glucans by an
exclusive exohydrolytic mode of attack(16) . The data for CbhA
are consistent with a mechanism in which the enzyme preferentially
removes unsubstituted cellobiosyl residues from the ends of
Cm-cellulose polymers. We have previously shown that >95% of the
soluble sugar released from microcrystalline cellulose by CbhA is
cellobiose(3) . In contrast to CbhA and Abg, CenA gave a slope
of 98.0 mlmmol; a
similar high slope (88.1
mlmmol) was reported
for CenD, another C. fimi endo--1,4-glucanase(17) .
) and total reducing sugar during hydrolysis of
Cm-cellulose by CbhA, CbhAC, CenA, or Abg. The insets show the rate of reducing sugar production for each
enzyme.
C
had a slope of 44.5 mlmmol (Fig. 3). Therefore, deletion of amino acid residues
373-387 from CbhA enhances its endoglucanase activity. These data
support the hypothesis that the accessibility of the active site is a
major determinant of exo- or endohydrolytic activity in family 6
-1,4-glucanases, a conclusion hitherto based on structural
interpretations alone. Access to the CbhAC active site may be
still partially restricted, possibly by the amino-proximal loop,
because the slope of the CbhAC plot was lower than that for CenA.
However, bond cleavage preference may be influenced by additional
factors. For example, crystallographic data suggest differences in the
binding of small ligands to the four subsites in the active sites of
CBH II and E2(5) .Effects of Loop Deletion on Other Enzymatic
Properties
Both enzymes showed very slow turnover on
Cm-cellulose (Fig. 3, insets), but the rate of
hydrolysis by CbhAC (0.6 mol of
productminmol of
enzyme, calculated from reducing sugar production
after 30 min) was lower than that for CbhA (1.7
molminmol).
Similarly, the k
for CbhAC on
2`,4`-dinitrophenyl--D-cellobioside, a small soluble
substrate analog, was reduced by about 5-fold, although the K was unchanged (Table 1). The deletion in
CbhAC occurs only 2 amino acid residues from Asp
,
the putative general base catalyst, and results in the loss of several
charged residues (Fig. 2). Therefore, the reduced catalytic
efficiency of CbhAC is most easily explained by perturbation of
the active site environment.
C (56 mg of
producthµmol of
enzyme, calculated from total sugar solubilized
after 2 h) was also lower than that of CbhA (105
mghµmol) (Fig. 4), but analysis of the soluble products revealed a
further difference in the activities of the two enzymes. The CbhAC
reaction mixture contained a significant quantity of cellotetraose (Fig. 5A); in contrast, cellotetraose was not seen in
the CbhA reaction mixture, neither after 6 h of hydrolysis (Fig. 5A) nor at earlier times (data not shown). It
appears that cellotetraose accumulates in the reaction mixture because
CbhAC is unable to degrade this substrate efficiently (Fig. 5B). Under the same conditions, cellotetraose was
completely degraded to cellobiose by CbhA. Mechanistic interpretation
of these data is not yet possible, but they clearly show that deletion
of the loop has multiple effects on the activity of CbhA.
), CbhAC (), or CenA
(
).
C, or CenA
for 6 h (see Fig. 4); products from cellotetraose were analyzed
after hydrolysis by CbhA or CbhAC for 30 min. Glucose (1)
and cellobiose (2) were each resolved as single peaks;
cellotriose (3) and cellotetraose (4) were partially
resolved into their - and
-anomers.
)-1,4-glucosidase; CbhA, C. fimi cellobiohydrolase A;
CenA, C. fimi endoglucanase A; E2, T. fusca endoglucanase 2.
We thank Dr. Michael Spezio for advice and assistance.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Cartmell, E. Topakas, V. M-A. Ducros, M. D. L. Suits, G. J. Davies, and H. J. Gilbert The Cellvibrio japonicus Mannanase CjMan26C Displays a Unique exo-Mode of Action That Is Conferred by Subtle Changes to the Distal Region of the Active Site J. Biol. Chem., December 5, 2008; 283(49): 34403 - 34413. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-Y. Yao, K. L. Shrestha, Y.-J. Wu, H.-J. Tasi, C.-C. Chen, J.-M. Yang, A. Ando, C.-Y. Cheng, and Y.-K. Li Structural simulation and protein engineering to convert an endo-chitosanase to an exo-chitosanase Protein Eng. Des. Sel., September 1, 2008; 21(9): 561 - 566. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Proctor, E. J. Taylor, D. Nurizzo, J. P. Turkenburg, R. M. Lloyd, M. Vardakou, G. J. Davies, and H. J. Gilbert Tailored catalysts for plant cell-wall degradation: Redesigning the exo/endo preference of Cellvibrio japonicus arabinanase 43A PNAS, February 22, 2005; 102(8): 2697 - 2702. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-D. Bok, D. A. Yernool, and D. E. Eveleigh Purification, Characterization, and Molecular Analysis of Thermostable Cellulases CelA and CelB from Thermotoga neapolitana Appl. Envir. Microbiol., December 1, 1998; 64(12): 4774 - 4781. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |