The -carrageenase of Alteromonas fortis: a -helix … classified into four types, namely furcellaran, and kappa-, iota-, and lambda-carrageenans. κ-Carrageenan consists of repeated

Crystal structure of iota-carrageenase

1

The -carrageenase of Alteromonas fortis: a -helix fold-containing enzyme

for the degradation of a highly polyanionic polysaccharide.

GURVAN MICHELa,b, LAURENT CHANTALATa, ERIC FANCHONa, BERNARD

HENRISSATc, BERNARD KLOAREGb, AND OTTO DIDEBERGa*

aLaboratoire de Cristallographie Macromoléculaire, Institut de Biologie Structurale

Jean-Pierre Ebel, CNRS/CEA, 41, rue Jules Horowitz, 38027 Grenoble Cedex 1, FRANCE.

bStation Biologique de Roscoff, UMR 1931 (CNRS and Laboratoires Goëmar), Place

Georges Teissier, BP 74, 29682 Roscoff Cedex, FRANCE.

cArchitecture et Fonction des Macromolécules Biologiques, UMR 6098 (CNRS,

Universités d'Aix-Marseille I et II), 31 Chemin Joseph Aiguier, 13402 Marseille Cedex 20,

FRANCE.

Running title: Crystal structure of iota-carrageenase.

* Corresponding author: Phone: +33 (0)4 38 78 56 09

Fax: +33 (0)4 38 78 54 94

E-mail: [email protected]

Copyright 2001 by The American Society for Biochemistry and Molecular Biology, Inc.

JBC Papers in Press. Published on August 7, 2001 as Manuscript M100670200 by guest on M

arch 16, 2019http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


2

SUMMARY

Carrageenans are gel-forming hydrocolloids extracted from the cell walls of marine red

algae. They consist of D-galactose residues bound by alternate α-1,3 and β-1,4 linkages and

substituted by one (κ-carrageenan), two (ι-carrageenan), or three (λ-carrageenan) sulfate-ester

groups per disaccharide repeating unit. Both the κ- and ι-carrageenan chains adopt ordered

conformations, leading to the formation of highly-ordered aggregates of double-stranded

helices. Several κ-carrageenases and ι-carrageenases have been cloned from marine bacteria.

κ-Carrageenases belong to family 16 of the glycoside hydrolases, which essentially

encompasses polysaccharidases specialized in the hydrolysis of the neutral polysaccharides,

such as agarose, laminarin, lichenan and xyloglucan. In contrast, ι-carrageenases constitute a

novel glycoside hydrolase structural family. We report here the crystal structure of

Alteromonas fortis ι-carrageenase at 1.6 Å resolution. The enzyme folds into a right-handed

parallel β-helix of ten complete turns with two additional C-terminal domains. E245, D247, or

E310, in the cleft of the enzyme, are proposed as candidate catalytic residues. The protein

contains one sodium and one chloride binding sites, and three calcium binding sites, shown to

be involved in stabilizing the enzyme structure.

by guest on March 16, 2019

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


3

Introduction

Carrageenans are the main components of the cell walls of various marine red algae

(Rhodophyta), where they play a variety of structural (cell-cell cohesion and exchange

boundary) and signaling (cell-cell recognition) roles (1,2). They consist of linear chains of

galactopyranose residues in the D-configuration linked by alternating α(1→3) and β(1→4)

linkages. This regular structure is modified by 3,6-anhydro bridges and substitution with

sulfate ester groups. On the basis of the level and position of sulfate substitution, carrageenans

are classified into four types, namely furcellaran, and kappa-, iota-, and lambda-carrageenans.

κ-Carrageenan consists of repeated units of the disaccharide, 4-sulfate-O-1,3-β-D-

galactopyranosyl-1,4-α-3,6-anhydro-D-galactose, also known as neocarrabiose sulfate. At the

primary structure level, ι-carrageenan differs from κ-carrageenan in the presence at C2 on the

α-linked galactose residues of one additional sulfate substituent per repeating disaccharide

(Figure 1). ι-Carrageenans therefore contain two sulfate groups per repeat unit, i.e. one

anionic group per monosaccharide. Such a high linear charge density is reminiscent of that

seen in alginic acid, the main cell wall polysaccharide of brown algae, and in polygalacturonic

acid, the non-methylated component of higher plant pectins, which both contain one carboxyl

group per monomeric unit (1).

In aqueous solution, κ- and ι-carrageenans form thermoreversible gels (3) and are used

in a variety of industrial applications as gelling or thickening agents. The mechanism of

gelation involves the formation of double helices of carrageenan chains, followed by the

association of the double helices into a macromolecular three-dimensional network (4,5).

Gelation is promoted by the presence of cations, potassium in κ-carrageenan and calcium in ι-

carrageenan. By binding to the helices, either electrostatically or through specific binding

sites, these cations considerably reduce the charge density of the helices and enhance their

tendency to aggregate (6). The fine structure of ι-carrageenan in the solid state has been

studied in detail by X-ray diffraction of calcium carrageenate fibers (7). The polysaccharide

chains were shown to adopt a twisted-ribbon conformation with a 31 symmetry, stabilized by

the formation of double helices, themselves aggregated into larger clusters through the


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


4

coordinated binding of calcium ions (7). In the cell walls of red algae, carrageenans are laid

out as highly ordered molecules associated with the cellulose microfibrils or in the

microfibril-less intercellular matrix (8). ι-Carrageenans are mainly located in the outer,

cortical tissues whereas κ-carrageenans constitute the walls of the inner cortical and

medullary cells (9).

Enzymes that degrade carrageenans, namely κ-, ι- and λ-carrageenases, have been

isolated from various marine bacteria. They all are endo-hydrolases that cleave the internal

β(1→4) linkages of carrageenans, yielding products of the neocarrabiose series (10-13). Since

these galactan hydrolases display a strict substrate specificity, they obviously recognize the

sulfation pattern on the digalactose repeating unit and thus provide an opportunity for

investigating the structure-function relationships of hydrolases that degrade sulfated

polysaccharides. With this aim, we have undertaken the structural analysis of a representative

set of carrageenases. Recently, we reported the deduced amino acid sequences of ι-

carrageenase from Alteromonas fortis and Zobellia galactanovorans. They share no sequence

similarity with κ-carrageenases and, unlike β-agarases and κ-carrageenases, they are inverting

hydrolases. They represent the first members of a new family of glycoside hydrolases (14),

family 82 (13). In this context, a high-resolution three-dimensional structure for a ι-

carragenase would be pivotal in determining the fold prevailing in family 82 and in

delineating the molecular bases for the recognition of ι-carrageenan. To this end, we have

recently overexpressed and crystallized A. fortis ι-carrageenase (15) and we report here its

structure at 1.6 Å resolution.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


5

Experimental Procedures

Expression, purification and crystallization of native and selenomethionyl -

carrageenase

Full details of the expression and purification of native ι-carrageenase have been

described previously (15). Briefly, ι-carrageenase was expressed, using the pET20b vector

(Novagen), as an His-tagged fusion protein in the periplasm of E. coli BL21(DE3) strain and

purified by metal affinity chromatography on a column of Chelating Fast Flow Sepharose

(Pharmacia) loaded with NiSO4. The yield was 7-8 mg / l of culture medium and the purified

protein was concentrated to 6 mg / ml using a dialyzing concentrator (Amicon Co.). Single

crystals were obtained with polyethylene glycol and the presence of calcium ions appeared to

be crucial for crystallization. High quality crystals , typically 0.25 x 0.25 x 0.15 mm in

dimensions, were grown with 0.1 M sodium cacodylate at pH 6.5, 15-17 % polyethylene

glycol (MW 6,000) and 200 mM calcium acetate.

The seleno-L-methionine labeling procedure was identical to that described for the

expression of Se-Met-κ-carrageenase (16). As the expression yield was only 800 µg / l of

culture medium, the volume had to be scaled up to obtain a sufficient amount of protein.

During the concentration step, the Se-Met-ι-carrageenase appeared to be less soluble than the

native protein and could only be concentrated to 2 mg / ml. Crystallization of Se-Met-ι-

carrageenase was performed under similar conditions to those already reported (15), except

for the addition of 1 mM DTT and the replacement of Na cacodylate by imidazole, in order to

avoid the reaction between cacodylate and DTT (17). Under these conditions, small crystals

with a maximal size of 100 x 100 x 50 µm, were obtained in one month of equilibration,

whereas large native crystals appeared in a few days.

Data collection and processing

Crystals were successively soaked for 30 s in crystallization solutions in which the

glycerol concentration was increased by steps of 5 % to a final concentration of 20 %. A

single crystal was mounted on a loop, transferred to the goniometer head, and kept at 100 K in


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


6

a nitrogen stream. The MAD data were collected on a Se-met-ι-carrageenase crystal at three

wavelengths around the K absorption edge of selenium (beamline FIP / BM 30, ESRF-

Grenoble, France). A 345 mm MAR image plate detector was used. All intensity data were

integrated and reduced using DENZO / SCALEPACK (18). Base on unit cell dimensions two

molecules of protein are present in the asymmetric unit. Using the Patterson method and

SHELX 97 (19), the positions for 7 of the 16 selenium atoms expected were determined. This

partial structure was input into SHARP (20) for a first approximation of the phases. Five other

selenium atoms were located in the Fourier difference electron density map. The final partial

structure therefore contained 12 selenium atoms and the final phases from SHARP resulted in

a figure of merit of 0.43 to 2.3 Å resolution. Low (up to 2.2 Å) and high (up to 1.6 Å)

resolution data were also collected on a native ι-carrageenase crystal using a MAR CCD

detector (beamline ID14 EH2, ESRF Grenoble). All intensity data were integrated using

DENZO. At low temperature, the unit cell parameters were slightly different : a = 55.87, b =

90.07, c = 124.12 Å, α = γ = 90°, β = 93.53°. The low and high resolution data were merged

and reduced with SCALA (21).

Structure determination and refinement

Taking advantage of the non-crystallographic symmetry, the electron density was

improved by molecular averaging, histogram matching, and solvent-flattening (50 % solvent)

with DM (21). Using the native data, the phases were then extended to 1.6 Å resolution,

yielding a figure of merit of 0.75. Due to the high quality of the electron density map at 1.6 Å,

wARP (22) was used to build the model automatically, resulting in the assignment of 85 %

and 75 % of the main and side chain atoms, respectively. The rest of the model was build

manually using O (23). However, no clear electronic density was observed for residues 314-

333 and 341-350, which are not included in the final model. The refinement was performed

with CNS (24) using the native data set to 1.6 Å, with the maximum likelihood target

function. The program was set up to automatically compute a cross-validated σa estimate and

the weighting scheme between the X-ray refinement target and the geometric energy function.

Corrections for a flat bulk solvent and for anisotropy in the data were also applied. The σa


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


7

weighted maps obtained from the subsequent refinement models were used for further model

building. The first group of water molecules was added when peaks in the 2Fo-Fc density were

> 2 σ and had a stereochemistry compatible with at least one hydrogen bond with a protein

atom or another water molecule. In the final stages, the sigma cut-off was reduced to 1.0 σ

and water molecules with a B-factor > 60 Å2 were removed from the model. The final model

refined at 1.6 Å has an Rwork of 20.7 % and an Rfree of 22.3 % (25) and consists of 6,840

protein atoms, 1,110 water molecules, six glycerol molecules, six calcium, six sodium, and

two chloride ions. All these ions have an occupancy of 1 and their B factors refine to a value

close to neighboring protein atom B factors. The stereochemistry of the final structure was

evaluated using the PROCHECK program (26).

Calcium-dependent activity test and proteolytic digestion

ι-Carrageenase activity was assayed as follows. Three enzyme solutions were prepared

by diluting a stock solution of the purified enzyme (6 mg / ml) in 100 mM Tris-HCl, pH 7.2,

200 mM NaCl, and adding either 5 mM EGTA or 5 mM CaCl2. Aliquots (100 µl) were

incubated for 15 min. at 40 °C with 2 ml of substrate solution, consisting of 0.125% (w/v)

iota-carrageenan, 50 mM Tris-HCl, pH 7.2, 100 mM NaCl, with or without 5 mM CaCl2 and

the reaction mixture (200 µl) was assayed for reducing sugars (27), using boiled enzyme

blanks. One unit of enzyme activity (UA) is defined as the amount of enzyme which produces

an increase of 0.1 A237nm/min. in the reducing sugar assay.

Purified bovine pancreas trypsin (T1426, 10,000 BAEE units / mg protein) was

purchased from Sigma. Limited proteolysis of ι-carrageenase (6 mg / ml in 100 mM Tris-HCl,

pH 7.2, 200 mM NaCl) was performed using trypsin / ι-carrageenase ratios of 1:100 and 1:20

(wt/wt), in the presence or absence of 5 mM EGTA or 5 mM CaCl2. The samples were

incubated for 1h at 20°C, then the reaction was stopped by adding sodium dodecyl sulfate

(SDS) loading buffer and boiling the samples for 5 min at 100°C; the samples were then

loaded onto a 15% SDS polyacrylamide gel for electrophoresis and the gels stained with

Coomassie blue.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


8

Results

The overall structure of A. fortis -carrageenase

The three-dimensional crystal structure of ι-carrageenase lacking the signal peptide

(residues 1-27) was determined at 2.3 Å resolution by the MAD method, using a crystal of a

seleno-methionine (Se-Met)-substituted form of the enzyme (28). After phase extension with

a native data set at higher resolution, a high quality electron density map was obtained (Figure

2), allowing building and refinement of the model at 1.6 Å resolution. The crystallographic

statistics are shown in Table I. The asymmetric unit contains two mature ι-carrageenase

molecules, each containing amino acids 28-491. Residues 314-334 and 341-350 for molecule

A, and 313-334 and 341-351 for molecule B are not visible in the (2Fo-Fc) electron density

map and are presumed to exist in disordered or highly flexible conformations. Superposition

of molecules A and B reveals that the Cα atoms overlay with a rmsd of 0.21 Å.

Approximately 10 residues in each molecule presented clear alternate conformations. The

need to refine the occupancy for terminal atoms of several residues, such as aspartate,

glutamate, or methionine, suggests that a fraction of the protein population in the crystal has

been subjected to radiation damage (29).

The core of ι-carrageenase is folded into a right-handed parallel β-helix of ten complete

turns (Figure 3). This fold was first encountered in pectate lyase C from Erwinia

chrysanthemi (30). The lyase structure consists of three parallel β-sheets, PB1, PB2, and PB3.

PB2 and PB3 form planar surfaces almost perpendicular to each other, while PB1 is in the

form of a groove. In the lyase structure, the turns or loops (depending on the number of amino

acids inserted between consecutive β-helical strands) are referred to as T1 (PB1-PB2), T2

(PB2-PB3), and T3 (PB3-PB1). In β-helix proteins, the assignment of secondary structure

elements is based on the DSSP algorithm (31), with the additional criterion that any residues

with (Φ, ψ) angles in the left-handed α-helix region are not included in the β-strand. Based on

these rules, PB2 can be divided into two parts, and the ι-carrageenase β-helix consists of four

parallel β-sheets, PB1, PB2a, PB2b, and PB3, composed respectively of 10, 5, 11, and 10 β-


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


9

strands. These strands are relatively short, with an average number of 4.0, 2.4, 4.1, and 4.0

residues, respectively. Interestingly, like almost all β-helix proteins, ι-carrageenase contains

in its N-terminal region an amphipathic α-helix (residues 66-77) that shields the hydrophobic

core of the β-helix from the solvent.

The most striking difference between ι-carrageenase (Figure 3) and the eleven other β-

helix proteins of known structure is the presence, in the C-terminal region, of two large

additional domains (both 67 residues long). Domain A (residues 307-373) replaces the T1

turn between strands β32 and β35 (see legend of figure 3A for definition). Half of this domain

(residues 314-333 and 341-350) could not be built, as no clear electron density was observed.

The visible part of domain A features a sheet of two short anti-parallel β-strands, β33 and

β34, edged by one α-helix (residues 358-367), and has an average B value (37.8 Å2), twice

that of the β-helix core (17.8 Å2). At the border of the visible part of domain A is a

hydrophobic surface, suggesting that the non-visible residues complete a globular-shaped

domain. This domain is weakly bound to the β-helix by only four hydrogen bonds (E310 O -

K252 NZ, D358 OD1 - S442 OG, D358 OD2 -K443 N, and D362 OD1 - Y444 OH).

Moreover, the visible part of domain A makes no contact with neighboring molecules and is

located in a large solvent channel in the crystal. Domain B consists of residues 387-430,

located on a T3 turn that connect strands β36 and β39, and the C-terminal extension (residues

469-491). Mainly composed of long loops, it also features an anti-parallel β-sheet (β37-β38),

a 310 helix (residues 395-402), and a one-turn α-helix at the C-terminus. This domain, also

globular in shape, is folded around an independent hydrophobic core. In contrast to domain A,

this bulky structure is characterized by many side chain - side chain and side chain - main

chain hydrogen bonds, both within the domain and with the β-helix. These hydrogen bonds

have a clear stabilizing effect on domain B, since its average B-value (22.0 Å2) is close to that

of the β-helix core (17.8 Å2).

Each strand of PB2a is connected to PB2b by an asparagine residue in a left-handed α-

helix conformation. Residues in the αL conformation are also seen in the T2 turns in the

structure. The presence of these residues results in sharp bends of about 100° in the

polypeptide chain without disrupting the hydrogen bond pattern of the parallel β-strands. PB1


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


10

also presents a striking repetition of a structural irregularity : in each strand, one residue is in

a right-handed helix conformation, thus creating an alignment of nine β-bulges in the groove.

With the exceptions of S186 and the K464-T465 β-bulge in the C-terminus, which are

accessible to the solvent, all side chains in these β-bulges are aliphatic amino acid residues

and point towards the hydrophobic core of the β-helix.

The turns and loops between the β-strands are of different sizes and have different

conformations. Whereas almost all of the T2 turns consist of a single residue in the αL

conformation, the T1 and T3 turns are longer and more irregular, their sizes ranging from 1-

12 and 2-8 residues, respectively. At the N-terminal edge of the β-helix, the T3 loops form a

bulky protrusion above PB1. On the opposite side of this β−sheet, the T1 turn between strands

β15 and β18 folds into a β-sheet of two anti-parallel strands (β16 and β17), and the T1 turn

between β25 and β26 folds into a short α-helix. Strands β16 and β17 and this short α-helix

form a domain which is stabilized by a hydrophobic core (V191, L198, L250, M251, and

Y254) and hydrogen bonds between E193 OE2 - Y254 OH and L198 O - Q256 OE1. At the

surface, two hydrophilic networks (R136-D166-R194 and R202-W200-D227-R260-G257-

G258) firmly bind this domain to the β-helix core; these amino acids are also present in Z.

galactanovorans ι-carrageenase. On each edge of β-sheet PB1, the N-terminal T1 and T3

extensions and the protruding domains, A and B, in the C-terminal region create a long deep

cleft. This large channel, about 50 Å long and 10 Å wide, is probably the cleft that binds ι-

carrageenan chains.

A. fortis ι-carrageenase contains four disulfide bridges. The first, C269-C298, is located

within the β-helix core, connecting strand β28 to the T3 turn between strands β31 and β32.

The C336-C360 bridge, which links the 334-340 loop to helix α3, probably stabilizes domain

A. This disulfide bridge was opened by radiation damage, but its existence was confirmed by

the refinement of the structure (data not shown) from the data at 2.0 Å resolution obtained

with a less powerful beamline (15). However, even in this latter electron density map with an

intact disulfide bridge, the above-mentioned non-visible part of domain A could not be seen

either. The two other disulfide bridges, C408-C476 and C412-C484, which are strictly

conserved in the two ι-carrageenase sequences, bring the C-terminal extension of the protein


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


11

into close proximity with the main part of domain B. Finally, the ι-carrageenase structure

contains two chloride, six sodium, and six calcium binding sites (see below).

A major characteristic of β-helix proteins is their internal stability (mean B factor of

15.6 Å2), which results from numerous intra hydrogen bonds and side chain stacking of

identical or similar residues (32). Consistent with these observations, the ι-carrageenase

structure is dominated by aliphatic stacking in the β-helix interior. Four major stacks (I209-

L234-I267-V293-V383-V453, I229-I262-V288-V378, I187-I221-L242-V276-V302-V434,

and I113-V149-I117) form the hydrophobic core, but short stacks (V290-V380, V135-V167,

I187-I221, I172, and I206) also strengthen the β-helix cohesion. In the aligned β-bulges,

stacking of alanine residues (A241-A275-A301-A433), all in the αR conformation, is also

seen. In contrast to pectate lyases, however, clusters of aromatic residues are of minor

importance, involving only short clusters (F381-F451, F147-F175, F127-F162, F184-Y218).

On the surface of the protein, two rows of asparagine residues in the αL conformation (N102-

N137-N169-N203-N228-N261 and N233-N266-N293) flank β-sheet PB2, but few of these

amino acids are engaged in successive hydrogen bonds and only a short portion of asparagine

ladder (N102-N137-N163) can be identified. In contrast, as in F. heparinum chondroitin lyase

B (33), the ι-carrageenase β-helix features external basic stacks, which consist of a short

lysine pair (K373-K449) and a cluster of arginines (R136-R168-R202-R260).

-Carrageenase contains binding sites for sodium, chloride, and calcium.

The structure of ι-carrageenase reveals six sodium, two chloride, and six calcium ion

binding sites. Four of the sodium ions are located at the interface between molecules A and B

in the asymmetric unit and form two clusters related by the non-crystallographic symmetry. In

one cluster, sodium ions are bound to a carboxyl group, D54 or E108, and to four water

molecules, three of which are shared by the ions. Since ι-carrageenase elutes as a monomer on

gel filtration (data not shown), the presence of these four sodium binding sites is probably due

to crystal packing. The last two sodium binding sites are located in the same region of

molecules A and B in the interior of the β-helix core, 14-15Å from the catalytic residues. In

one molecule, the sodium ion is coordinated by five ligands with a trigonal bipyramidal


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


12

geometry and at an average distance of 2.35 Å ; these ligands are the oxygens of L159, I183,

and F184, the hydroxyl of T182, and one water molecule. T182, I183, and F184 are conserved

in Z. galactanovorans ι-carrageenase (13). Interestingly, F184 adopts a disallowed

conformation (φ = 78, ψ = 139). Each protein molecule binds one chloride ion, which is

located on the surface of the groove, slightly buried in a hydrophobic pocket, and 7 Å from

both the above-mentioned sodium ion and the catalytic residues. This hydrophobic cavity is

composed of the conserved residues F184, A185, L188, and Y284. The chloride ion is also in

contact with the amide group of the conserved Q222, the backbone amino group of A185, and

one water molecule. The activity of ι-carrageenase is known to depend on the presence of

sodium chloride and to decrease rapidly on either side of the optimum concentration of 100

mM NaCl; complete salt removal or NaCl concentrations higher than 500 mM result in

complete loss of activity (34). However, how the presence of sodium and chloride binding

sites in ι-carrageenase accounts for its activation by sodium chloride remains an open

question.

Interestingly, each ι-carrageenase molecule also has three calcium ion binding sites, all

remote from the active site. The first is located at the surface, between the second and third T2

turns. The calcium ion makes hydrogen bonds with the backbone oxygen and hydroxyl group

of S109, the side chain carbonyl of N145, the oxygen of G146, and four water molecules. As

these ligands are not present in Z. galactanovorans ι-carrageenase (13) this site appears to be

specific to A. fortis ι-carrageenase. The second and third sites, which display a hairpin

topology, are well conserved between the two ι-carrageenases. The second site precedes the

N-terminal α-helical cap (Figure 2). The calcium ion, buried inside the hairpin loop,

establishes short contacts (average distance 2.37 Å) with N58 O, D61 OD1, S63 O, D65 OD1

and OD2, and two water molecules. N58, D61, and D65 are strictly conserved in the two ι-

carrageenases. The hairpin loop is also structured and anchored to the β-helix by numerous

hydrogen bonds (N58 OD1- N60 N, G59 O - T91 N, D61 OD1 - S63 N, D64 OD1 - D65 N -

S66 N and OH, D65 OD1 - N58 N, and D65 OD2 - H93 N). The third calcium binding site is

in a more complex and irregular hairpin loop in the C-terminal region. It is in interaction with

T438 OH, D445 OD1, D445 OD2, D447 OD1, and one water molecule. This site is


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


13

strengthened by D436, which makes a hydrogen bond with T438 OH and the above-

mentioned water molecule. Only D436 and T438 are strictly conserved in the two enzymes,

while D445 and D447 are replaced by asparagine residues in Z. galactanovora ι-carrageenase.

The potential influence of calcium ions on the stability of ι-carrageenase was tested.

Compared to the activity in the purification buffer (10 mM Tris-HCl pH 7.2, 200 mM NaCl),

the addition of 5 mM CaCl2 had no effect on ι-carrageenase activity. In contrast, the trypsin

digestion pattern of ι-carrageenase was clearly affected by the presence of either EGTA or

CaCl2 (Figure 4). In the control buffer, trypsin digestion of ι-carrageenase (1:100 w/w)

resulted in the release of several peptides with sizes ranging from 5 to 53 kDa, but most of the

protein was undegraded; under the same conditions, this pattern was not affected by the

presence of 5 mM EGTA. A trypsin/ ι-carrageenase ratio of 1:20 resulted in greater protein

degradation, especially in the presence of EGTA when the protein was almost completely

degraded. In contrast, at either trypsin/ ι-carrageenase ratio, the addition of 5 mM CaCl2

significantly protected ι-carrageenase from proteolysis. Even though the enzyme was purified

in the absence of calcium, the marked differences seen in the digestion patterns in the

presence or absence of EGTA at the 1:20 trypsin/ ι-carrageenase ratio indicate that the

calcium binding sites were at least partly occupied in the protein, suggesting that they have a

high affinity for calcium. Taken together, these findings demonstrate that calcium is involved

in maintaining the structural integrity of the protein.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


14

Discussion

Comparison of -carrageenases with other right-handed parallel -helix proteins.

Only eleven protein structures are known to contain a right-handed parallel β-helix fold.

Most are pectin-degrading enzymes and employ a mechanism involving either β-elimination,

e.g. the pectate and pectin lyases of family 1 of the polysaccharide lyases (30,35,36), or

acid/base hydrolysis, e.g. Aspergillus niger rhamnogalacturonase (37) and polygalacturonase

(38) as well as Erwinia carotovora polygalacturonase (39), which constitute family 28 of the

glycoside hydrolases. Other right-handed parallel β-helix fold-containing proteins include

chondroitin lyase B from Flavobactrium heparinum (33), a virulence factor (P69 pertactin)

from the whooping cough agent, the bacterium Bordetella pertussis (40), and the tailspike

rhamnosidase from phage P22 (41), and a pectin methylesterase from Erwinia chrysanthemi

(42). It is worth noting thus that most β-helix proteins (pectate lyases, rhamnogalacturonases,

polygalacturonases, chondroitin lyases, and ι-carrageenases) are enzymes involved in the

depolymerization of acidic polysaccharides with high linear charge densities, from the

extracellular matrix of plants or animals. The exceptions include pectin lyases, which can

degrade the methylated form of polygalacturonic acid, the P22 tailspike protein, an

endorhamnosidase involved in the recognition and degradation of the neutral, rhamnose-rich

oligosaccharides known as O-antigen receptors (43), and P69 pertactin, the specificity of

which is unclear.

Although ι-carrageenase does not share sequence similarity with family 28 pectin

hydrolases, its overall structure shows similarities with the family 28 fold. As in these

rhamnogalacturonase and polygalacturonases, the ι-carrageenase β-helix fold consists of four

parallel β-sheets. All three of these glycoside hydrolases proceed with an inversion of the

anomeric configuration of their substrates. However, the pectin-degrading enzymes cleave α-

1,4 or α-1,2 glycosidic bonds, whereas ι-carrageenase cleaves β-1,4 glycosidic linkages. It

also appears that the family 28 pectinase catalytic residues, D180, D201, and D202 (38,44),

are not structurally conserved in ι-carrageenase. We thus conclude that ι-carrageenases


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


15

represent a genuinely new family of glycoside hydrolases, referred to as family 82 (13).

Hence, even though families 28 and 82 display the same protein fold, they act on different

types of glycosidic bonds (α vs. β), display different catalytic amino acids and, therefore, they

do not belong to the same clan of glycoside hydrolases (14).

Structure of the putative active site and recognition of -carrageenan chains.

ι-Carrageenase hydrolyses glycosidic bonds with a mechanism leading to overall

inversion of the anomeric configuration at the site of cleavage (13). In such an inverting

hydrolytic mechanism, the glycosidic bond is protonated by the acid/base catalyst, in order to

facilitate aglycon departure. This protonation step occurs concomitantly with the attack on the

anomeric carbon by a water molecule activated by a base residue, yielding a product with an

anomeric stereochemistry opposite to that of the substrate (45,46). In inverting glycoside

hydrolases, the acid/base and nucleophile catalysts consist of aspartate and/or glutamate

residues, their carboxyl groups being typically 10 Å apart (47). However, family 28

pectinases feature an active site topology with a short distance (5-6 Å) between the catalytic

residues (38,44).

In the structure of E. chrysanthemi pectate lyase C complexed with a plant cell wall

fragment, the oligosaccharide is found on PB1, defining the groove as the substrate-binding

site (48). As discussed below, the groove in ι-carrageenase also appears to be the binding site

for the ι-carrageenan chain. Of the acidic residues located in this cleft, only three, E245,

D247, and E310, are conserved between the two ι-carrageenase sequences. E245 and D247

are located at the bottom of the groove between domains A and B and their carboxyl groups

are separated by a distance of 4-5 Å. E310 belongs to domain A and its carboxyl group, which

points toward the cleft, is separated by a distance of about 10 Å from the carboxyl groups of

E245 and D247 (Figure 5). According to the DSSP program, E245, D247, and E310 are all

accessible to the solvent, with respective accessible surface values of 31, 46, and 61 Å2.

Interestingly, the entire region near these residues is well conserved in ι-carrageenases, and


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


16

we propose that the nucleophile and acid/base catalysts are a yet to be determined pair among

these three acidic amino acids.

The catalytic cleft of ι-carrageenase is not large enough to correctly accommodate the

double-stranded helix of ι-carrageenan for productive binding. Therefore, ι-carrageenase

presumably initiates its action on single-stranded chains of the amorphous zones of ι-

carrageenan gels. In κ-carrageenan, the glycosidic backbone contains alternating negatively

charged units of D-galactose-4-sulfate and hydrophobic 3,6-anhydro-D-galactose residues.

Consistent with these observations, the binding of κ-carrageenan by P. carrageenovora κ-

carrageenase has been shown to involve both ionic interactions with sulfate-ester substituents

and a specific hydrophobic environment for the 3,6-anhydro-D-galactose units (16). Since ι-

carrageenan bears a sulfate group at C2 of the 3,6-anhydro-D-galactose moiety, this α-1,4

linked residue is less hydrophobic than in the κ-carrageenans and substrate binding by ι-

carrageenases should mainly involve ionic interactions. This assumption is confirmed by the

paucity, in the active site cleft of ι-carrageenase, of aromatic residues, which are known to

interact with neutral saccharides through aromatic ring - pyranose ring stacking. The ι-

carrageenase groove contains no tryptophan residues and only three tyrosine residues, Y218,

Y224, and Y400. In contrast, four arginine residues (R125, R243, R303, and R353) and one

lysine residue (K122), which are also present in Z. galactanovorans ι-carrageenase, point

towards the cleft. Figure 6A shows the mapping of these conserved basic residues and of the

catalytic residues on the surface of ι-carrageenase.

Based on the 3D structure of the ι-carrageenan double helix (7), the distribution of

sulfate substituents along the ι-carrageenan chain is not homogeneous. Within the same

repeating unit, the sulfate groups flank the α-(1,3) glycosidic bond and are separated only by

4.65 Å. The distance between two such successive sulfate pairs is 11.5 Å, leaving the

immediate vicinity of the β-(1,4) linkages devoid of negative charges. Finally, two

neighboring pairs of sulfate-ester substituents on the same side of the glycosidic backbone are

separated by 16 Å (Figure 6B). This topology of the sulfate substituents on ι-carrageenan

shows a striking match with the distribution of the conserved basic amino acids in the ι-


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


17

carrageenase groove (Figure 6A). Residues R243 and R303, separated by 7 Å, could form salt

bridges with one ester-sulfate pair; this would position the uncleaved α-(1,3) linkage between

these two arginine residues and place the β-(1,4) linkage to be cleaved next to the potential

catalytic residues, E245, D247 and E310. This assumption is supported by the fact that R243

and R303 are tightly hydrogen bonded to E245 and D305, respectively (Figure 5). However,

we do not know whether R243 interacts with D-galactose-4-sulfate and R303 interacts with

3,6-anhydro-D-galactose-2-sulfate or vice versa. Thus, R243 and R303 participate either in

subsites -1 and -2 or subsites +1 and +2, and the directionality of the carbohydrate polymer

chain with respect to the protein structure remains to be identified. Two other basic amino

acids in the groove, K122 and R125, located 16 Å from R243 (Figure 6A) are in the correct

topology to bind the sulfate pair. Finally, the last basic residue, R353 is 10 Å from R303;

again in a correct position to establish a salt bridge with the sulfate-ester substituent located

11.5 Å from the sulfate-ester bound to R303.

Keywords: ι-carrageenase, glycoside hydrolase, MAD, sulfated galactan, X-ray

structure.

Acknowledgements

We wish to thank Dr Ed Mitchell for his invaluable help with the beamline ID14EH2

(ESRF), Dr Thierry Vernet for his support with the Se-Met-substituted protein expression, Dr

Anne-Marie DiGuilmi for her precious help with the proteolysis analysis, and Dr Tristan

Barbeyron for helpful discussion.

This work was supported by grants from the Action Concertée Coordonnée Sciences du

Vivant (N° V) and GDR 1002 of CNRS "Biology, Biochemistry and Genetics of Marine

Algae".


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


18

References

1. Kloareg, B. and Quatrano, R. S. (1988) Oceanogr. Mar. Biol. Annu. Rev. 26, 259-315

2. Potin, P., Bouarab, K., Kupper, F., and Kloareg, B. (1999) Curr. Opin. Microbiol. 2,

276-283

3. Rees, D. A. (1969) Adv. Carbohydr. Chem. Biochem. 24, 267-332

4. Morris, E. R., Rees, D. A., and Robinson, G. (1980) J. Mol. Biol. 138, 349-362

5. Viebke, C., Piculell, L., and Nilsson, S. (1994) Macromolecules 27, 4160-4166

6. Piculell, L., Svante, N., Viebke, C., and Zhang, W. (1994) Food hydrocolloids :

structures, properties and functions. Nishinari, K. & Doi, E. Eds., Plenum Press, New

York. 35-44

7. Arnott, S., Scott, W. E., Rees, D. A., and McNab, C. G. (1974) J. Mol. Biol. 90, 253-267

8. Gordon-Mills, E. M. and Mc Candless, E. L. (1977) Phycologia 17, 95-104

9. Zablackis, E., Vreeland, V., Doboszewski, B., and Laetsch, W. M. (1991) J. Phycol. 27,

241-248

10. Johnston, K. H. and Mc Candless, E. L. (1973) Can. J. Microbiol. 779-788

11. Barbeyron, T., Henrissat, B., and Kloareg, B. (1994) Gene 139, 105-109

12. Barbeyron, T., Gerard, A., Potin, P., Henrissat, B., and Kloareg, B. (1998) Mol. Biol.

Evol. 15, 528-537

13. Barbeyron, T., Michel, G., Potin, P., Henrissat, B., and Kloareg, B. (2000) J. Biol.

Chem. 275, 35499-35505


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


19

14. Henrissat, B. and Davies, G. (1997) Curr. Opin. Struct. Biol. 7, 637-644

15. Michel, G., Flament, D., Barbeyron, T., Vernet, T., Kloareg, B., and Dideberg, O.

(2000) Acta Crystallogr. D56, 766-768

16. Michel, G., Chantalat, L., Duée, E., Barbeyron, T., Henrissat, B., Kloareg, B., and

Dideberg, O. (2001) Structure 9, 513-525

17. Maignan, S., Guilloteau, J. P., Zhou-Liu, Q., Clement-Mella, C., and Mikol, V. (1998) J.

Mol. Biol. 282, 359-368

18. Otwinowski, Z. and Minor, W. (1997) Methods Enzymol. 276, 307-326

19. Sheldrick, G. M. (1997) Methods Enzymol. 276, 628-641

20. de La Fortelle, E. and Bricogne, G. (1997) Methods Enzymol. 276, 472-494

21. CCP4 (1994) Acta Crystallogr. D50, 760-763

22. Perrakis, A., Sixma, T. K., Wilson, K. S., and Lamzin, V. S. (1997) Acta Crystallogr.

D53, 448-455

23. Jones, T. A., Zou, J.-Y., Cowan, S. W., and Kjeldgaard, M. (1991) Acta Crystallogr.

A47, 110-119

24. Brünger, A. T., Adams, P. D., Clore, G. M., DeLano, W. L., Gros, P., Grosse-Kunstleve,

R. W., Jiang, J. S., Kuszewski, J., Nilges, M., Pannu, N. S., Read, R. J., Rice, L. M.,

Simonson, T., and Warren, G. L. (1998) Acta Crystallogr. D54, 905-921

25. Brünger, A. T. (1992) Nature 355, 472-475

26. Laskowski, R. A., MacArthur, M. W., Moss, D. S., and Thornton, J. M. (1993) J. Appl.

Crystallogr. 26, 283-291


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


20

27. Kidby, D. K. and Davidson, D. J. (1973) Anal. Biochem. 55, 321-325

28. Hendrickson, W. A., Horton, J. R., and LeMaster, D. M. (1990) EMBO J. 9, 1665-1672

29. Ravelli, R. B. and McSweeney, S. M. (2000) Structure Fold Des. 8, 315-328

30. Yoder, M. D., Keen, N. T., and Jurnak, F. (1993) Science 260, 1503-1507

31. Kabsch, W. and Sander, C. (1983) Biopolymers 22, 2577-2637

32. Yoder, M. D., Lietzke, S. E., and Jurnak, F. (1993) Structure 1, 241-251

33. Huang, W., Matte, A., Li, Y., Kim, Y. S., Linhardt, R. J., Su, H., and Cygler, M. (1999)

J. Mol. Biol. 294, 1257-1269

34. Greer, C. W. and Yaphe, W. (1984) Can. J. Microbiol. 30, 1500-1506

35. Pickersgill, R., Jenkins, J., Harris, G., Nasser, W., and Robert-Baudouy, J. (1994) Nat.

Struct. Biol. 1, 717-723

36. Mayans, O., Scott, M., Connerton, I., Gravesen, T., Benen, J., Visser, J., Pickersgill, R.,

and Jenkins, J. (1997) Structure 5, 677-689

37. Petersen, T. N., Kauppinen, S., and Larsen, S. (1997) Structure 5, 533-544

38. van Santen, Y., Benen, J. A., Schroter, K. H., Kalk, K. H., Armand, S., Visser, J., and

Dijkstra, B. W. (1999) J. Biol. Chem. 274, 30474-30480

39. Pickersgill, R., Smith, D., Worboys, K., and Jenkins, J. (1998) J. Biol. Chem. 273,

24660-24664

40. Emsley, P., Charles, I. G., Fairweather, N. F., and Isaacs, N. W. (1996) Nature 381, 90-

92


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


21

41. Steinbacher, S., Seckler, R., Miller, S., Steipe, B., Huber, R., and Reinemer, P. (1994)

Science 265, 383-386

42. Jenkins, J., Mayans, O., Smith, D., Worboys, K., and Pickersgill, R. W. (2001) J. Mol.

Biol. 305, 951-960

43. Steinbacher, S., Baxa, U., Miller, S., Weintraub, A., Seckler, R., and Huber, R. (1996)

Proc. Natl. Acad. Sci. U.S.A. 93, 10584-10588

44. Armand, S., Wagemaker, M. J., Sanchez-Torres, P., Kester, H. C., van Santen, Y.,

Dijkstra, B. W., Visser, J., and Benen, J. A. (2000) J. Biol. Chem. 275, 691-696

45. Koshland, D. E. (1953) Biol. Rev. Camb. Phylos. Soc. 28, 416-436

46. Sinnott, M. L. (1990) Chem. Rev. 90, 1171-1202

47. Davies, G. and Henrissat, B. (1995) Structure 3, 853-859

48. Scavetta, R. D., Herron, S. R., Hotchkiss, A. T., Kita, N., Keen, N. T., Benen, J. A.,

Kester, H. C., Visser, J., and Jurnak, F. (1999) Plant Cell 11, 1081-1092

49. Davies, G. J., Wilson, K. S., and Henrissat, B. (1997) Biochem. J. 321, 557-559

50. Kraulis, P. J. (1991) J. Appl. Crystallogr. 24, 946-950

51. Nicholls A. (1992) GRASP: graphical representation and analysis of surface properties.

New York: Columbia University

Footnote: Research Collaboratory for Structural Bioinformatics Protein Databank =

PDB # 1h80


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


22

Figures legends

Figure 1: Schematic diagram showing two disaccharide repeating units of -carrageenan

The two different glycosidic bonds in ι-carrageenan are labeled and further extensions of the

polymer at the non-reducing and reducing ends are indicated, respectively, by R' and R". As ι-

carrageenase cleaves the β-(1,4) glycosidic bond, its subsites for substrate binding are labeled

in accordance with the established nomenclature (49).

Figure 2: Solvent-flattened MAD electron density map at 1.6 Å resolution

Map contoured at 2.0 σ of the N-terminal calcium-binding hairpin loop. Calcium ion and

water molecules are indicated as yellow and red spheres, respectively. The oxygen, nitrogen

and carbon atoms in the protein are shown, in red, blue and yellow, respectively. This figure

was created using O (23).

Figure 3: Folding of A. fortis -carrageenase

A. Stereo view of the Cα trace of the protein. The N-terminus, C-terminus, and every

twentieth residue are labeled, while every tenth residue is marked with a black dot. The

secondary structures listed in the text are composed of the following residues :

β15 (187−189), β18 (200−204), β25 (240−245), β26 (258−260), β28 (267−270), β32 (302−30

4), β35 (378−381), β36 (383−386), β39 (434−436).

B. Ribbon representation of the structure. The β-helix core, domain A, and domain B are

shown, respectively, in blue, gold, and red. The small T1-extension, containing an anti-

parallel sheet (β16-β17) and an α-helix (α2), is shown in green. The red, yellow, and green

spheres represent sodium, calcium, and chloride ions, respectively. Figures 3 and 4 were

prepared using Molscript (50).

Figure 4: Protection of A. fortis -carrageenase from trypsin hydrolysis by calcium

binding


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


23

Proteins were separated by SDS-PAGE on a 15% polyacrylamide gel and detected by

Coomassie Blue staining. The number on the left indicates the size of the molecular weight

markers in kDa (Pharmacia), while the band corresponding to trypsin is indicated on the right.

Lane 1 : ι-carrageenase (4 µg) in the absence of trypsin. Lanes 2 to 4 : trypsin/ ι-carrageenase

ratio of 1:100 (wt/wt), lanes 5 to 7 : trypsin/ ι-carrageenase ratio of 1:20 (wt/wt). Lanes 2 and

5 : hydrolysis in control buffer (100 mM Tris-HCl, pH 7.2, 200 mM NaCl), lanes 3 and 6 :

hydrolysis in the presence of 5 mM EGTA, lanes 4 and 7 : hydrolysis in the presence of 5 mM

CaCl2.

Figure 5: Stereo view of the potential active site of A. fortis -carrageenase

The conserved residues of ι-carrageenase are shown in red, blue, and black for oxygen,

nitrogen, and carbon atoms, respectively. The other residues are shown in ribbon

representation. The green sphere represents a chloride ion. Residues E245, D247, and E310

are the potential catalytic amino acids, whereas H281 may maintain the correct pKa of E245.

The basic residues R243, R303, and R353 are probably involved in ι-carrageenan recognition.

Figure 6: -Carrageenan recognition by -carrageenase

A. View of the molecular surface of the ι-carrageenase groove. The potential catalytic

residues and the conserved basic amino acids are shown in red and blue, respectively. The

distances between clusters of basic residues are shown. Figure 6 was created using Grasp

(51).

B. Ball and stick representation of a ι-neocarrahexaose-sulfate (7). Oxygen, sulfur, and carbon

atoms are shown in red, yellow, and black, respectively. Distances between the sulfate

substituents are shown.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


24

Table I. Data reduction, phasing and refinement statistics

Peak Edge Remote NativeData collectionWavelength (Å) 0.9822 0.9837 0.9783 0.933Resolution (Å) 2.3 2.3 2.3 1.6Total data 99149 100329 100610 762046Unique data 56584 56722 56712 157467Redundancy 1.75 1.77 1.77 4.80Completeness (%) 89.3 (85.8) a 90.1 (85.6) 90.4 (85.5) 97.7 (96.1)

Rsym (%)b 5.0 (18.7) a 5.2 (22.0) 5.5 (21.8) 12.3 (20.3)

PhasingPhasing power Iso (Ano) (1.64) 1.83 (0.96) 1.77 (1.17)RCullis (centric) 0.58 0.60RCullis (ano) 0.69 0.81 0.79

FOMc before/after DM (to 1.6 Å) 0.43 (0.75)RefinementResolution range (Å) 20.0-1.6No. of unique reflections 156297Rwork (Rfree)

d 20.7 (22.3)

R.m.s.d.e bonds (Å) 0.005R.m.s.d. angles (°) 1.4R.m.s.d. NCS Cα atoms (Å) 0.21Quality of Ramachandran plotPercentage of residues in most favored regions 85.9Percentage of residues in additional allowed regions 13.4Percentage of residues in generously allowed regions 0.1Percentage of residues in disallowed regions 0.5 (F184 and H281, see text)Number of non-hydrogen atoms [average B-values (Å2)]Protein 6697 [19.5]Water 1100 [36.5]Ion 14 [20.5]Glycerol 36 [56.2]

aValues in parentheses correspond to the highest resolution shell: (2.35-2.30 Å) and (1.66-1.60 Å) for the seleno-methionine substituted and native proteins, respectively.bRsym = Σ|I –Iav| / ΣI, where the summation is over all symmetry equivalent reflections.cFOM, figure of merit.dR calculated on 5% of data excluded from refinement.

eR.m.s.d., root mean square deviation.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


25

Fig. 1 location page 3



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


26

Fig. 3 a location page 8

Fig. 3 b


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


27




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


28



http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

cleavage

β-(1,4)

β-(1,4)α-(1,3)

α-(1,3)

β-(1,4)

-2 -1 +1 +2

O

SO3-O OH

OOH

OOO

O O O

O

OSO3-

O

SO3-O OH

O

OSO3-OH

R' R''


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

460

N

C

300

400

380

480

240

180

220

440

60

140

40

420

160

260

280

100 200

360

80

340120

460

C

N

300

400

380

480

240

18060

220

140

40

440

420

160

260

280

100

80

200

120340

360


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

and Otto DidebergGurvan Michel, Laurent Chantalat, Eric Fanchon, Bernard Henrissat, Bernard Kloareg

for the degradation of a highly polyanionic polysaccharideThe iota-carrageenase of Alteromonas fortis: a beta-helix fold-containing enzyme

published online August 7, 2001J. Biol. Chem.

10.1074/jbc.M100670200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M100670200

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;M100670200v1&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2001/08/07/jbc.M100670200.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=early/2001/08/07/jbc&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2001/08/07/jbc.M100670200.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/

Documents

The -carrageenase of Alteromonas fortis: a -helix … classified into four types, namely furcellaran, and kappa-, iota-, and lambda-carrageenans. κ-Carrageenan consists of repeated