Structure of the monofunctional heme catalase DR1998from Deinococcus radioduransPatrıcia T. Borges, Carlos Fraz~ao, Cecılia S. Miranda, Maria A. Carrondo and Celia V. Rom~ao

Instituto de Tecnologia Quımica e Biologica, Universidade Nova de Lisboa, Oeiras, Portugal

Keywords

bacteria; clade 1; metalloprotein; radiation-

resistant; reactive oxygen species

Correspondence

C. V. Rom~ao, Instituto de Tecnologia

Quımica e Biologica, Universidade Nova de

Lisboa, Apartado 127, 2781-901 Oeiras,

Portugal

Fax: +351214433644

Tel: +351214469665

E-mail: cmromao@itqb.unl.pt

M. A. Carrondo, Instituto de Tecnologia

Quımica e Biologica, Universidade Nova de

Lisboa, Apartado 127, 2781-901 Oeiras,

Portugal

Fax: +351214433644

Tel: +351214469657

E-mail: carrondo@itqb.unl.pt

(Received 1 April 2014, revised 12 June

2014, accepted 24 June 2014)

doi:10.1111/febs.12895

Deinococcus radiodurans is an aerobic organism with the ability to survive

under conditions of high radiation doses or desiccation. As part of its pro-

tection system against oxidative stress, this bacterium encodes three mono-

functional catalases. The DR1998 catalase belongs to clade 1, and is

present at high levels under normal growth conditions. The crystals of

DR1998 diffracted very weakly, and the merged diffraction data showed

an Rsym of 0.308. Its crystal structure was determined and refined to 2.6 A.

The four molecules present in the asymmetric unit form, by crystallo-

graphic symmetry, two homotetramers with 222 point-group symmetry.

The overall structure of DR1998 is similar to that of other monofunctional

catalases, showing higher structural homology with the catalase structures

of clade 1. Each monomer shows the typical catalase fold, and contains

one heme b in the active site. The heme is coordinated by the proximal

ligand Tyr369, and on the heme distal side the essential His81 and Asn159

are hydrogen-bonded to a water molecule. A 25-A-long channel is the main

channel connecting the active site to the external surface. This channel

starts with a hydrophobic region from the catalytic heme site, which is fol-

lowed by a hydrophilic region that begins on Asp139 and expands up to

the protein surface. Apart from this channel, an alternative channel, also

near the heme active site, is presented and discussed.

Database

Coordinates and structure factors have been deposited in the Protein Data Bank in Europe

under accession code 4CAB

Introduction

Deinococcus radiodurans was the first organism to be

identified as being highly resistant to radiation, and its

genome was also the first genome of a radiation-resis-

tant organism to be sequenced [1,2]. It is now known

that several other organisms, such as Halobacteri-

um sp. and Pyrococcus furiosus, possess this property.

However, the organisms more closely related to D.

radiodurans are those that can survive to high doses of

radiation [3–6]. The mechanisms underlying this radia-

tion resistance have been studied by several groups,

and different hypotheses have been proposed over the

years, such as a condensed nucleoid structure, efficient

DNA repair pathways, and a higher cellular Mn/Fe

ratio [4,7–10]. In order to detoxify the reactive oxygen

species (ROS) formed under degrading conditions such

as exposure to radiation, D. radiodurans possesses an

enzymatic antioxidant system comprising superoxide

dismutases, peroxidases and catalases that target the

primary ROS, the superoxide radical and hydrogen

peroxide [11–16]. Initial studies on the antioxidant

Abbreviations

a.u., asymmetric unit; ESRF, European Synchrotron Radiation Facility; NCS, noncrystallographic symmetry; PDB, Protein Data Bank; ROS,

reactive oxygen species.

4138 FEBS Journal 281 (2014) 4138–4150 ª 2014 FEBS

activities of D. radiodurans extracts indicated that this

organism contains more than one catalase. In fact, the

KatA knockout mutant was shown to be only slightly

more sensitive to ionizing radiation than the wild-type,

indicating that the organism is still able to detoxify the

ROS formed [17–19]. Later, analysis of the D. radiodu-

rans genome revealed the presence of three catalases,

i.e. DR1998 (KatA), DRA0259, and DRA0146

[1,15,16], the first two of which are constitutively

expressed [17,19,20].

Catalases (hydrogen peroxide:hydrogen peroxide

oxidoreductase; EC 1.11.1.6) are widespread in aerobic

organisms of the three life domains, but are also pres-

ent in microaerophilic and anaerobic organisms such

as Campylobacter jejuni and Desulfovibrio gigas

[21,22]. These proteins can be divided into heme-con-

taining enzymes, including monofunctional catalases

or bifunctional catalase-peroxidases, and manganese

catalases. For reviews, see, for example, [23–27].Monofunctional catalases are further subdivided into

clade 1 and clade 2. Clade 1 consists of the smaller-

subunit catalases (55–69 kDa) with heme b as the

metal cofactor, which are present in plants, proteobac-

teria, and firmicutes. Clade 2 consists of catalases with

larger subunits (75–84 kDa), resulting from the pres-

ence of an additional C-terminal flavodoxin domain;

their prosthetic group is mostly heme d, although

some examples with heme b are also known. The

clade 2 catalases are mainly present in bacteria and

fungi. Clade 3 consists of small-subunit (43–75 kDa)

catalases containing heme b. These are widespread in

different archaeons, bacteria, and eukaryotes, and

some have NADPH as a cofactor [25,27,28].

The catalytic dismutation reaction of two hydrogen

peroxide molecules into one dioxygen and two water

molecules occurs in two stages. Briefly, in the first

stage, the heme iron (FeIII-porphyrin) is oxidized by

the first hydrogen peroxide molecule to form com-

pound I, which contains an oxo-FeIV porphyrin with a

p-cation radical, with the release of one water molecule

[29,30]. Compound I can be then reduced by a second

hydrogen peroxide molecule, with the release of one di-

oxygen molecule and of the second water molecule [31].

The three amino acids involved in the catalytic center

are conserved among catalases: the proximal ligand is

a tyrosine that coordinates the heme iron; a histidine

and an asparagine located on the opposite, distal side

of the heme plane also play a role in the catalytic mech-

anism. Although catalase was one of the first proteins

to be crystallized (in 1937), the first catalase crystal

structures only became available 40 years later: the

bovine liver [Protein Data Bank (PDB) 1CAT] and the

Penicillium vitale (PDB 4CAT) catalases [11–14].

The three catalases from D. radiodurans are mono-

functional, each belonging to a different clade: DR1998

(60 kDa) is in clade 1; DRA0259 (84 kDa) is in clade 2;

and DRA0146 is a smaller 40-kDa catalase, proposed to

be more related to eukaryotic catalases from plants than

to bacterial catalases [1,15]. The sequence identity

between DR1998 and DRA0259 is 25%, whereas

DRA0146 has only 6% sequence identity with either of

the other two catalases (Fig. 1). Nevertheless, some resi-

dues are conserved among the three proteins, namely

the tyrosine proximal ligand and the distal histidine

(Fig. 1). DR1998 is the only catalase to have been bio-

chemically characterized, and shows a specific catalase

activity of 68 800 U mg1 but no peroxidase activity

[32]. This protein is present at higher levels in D. radio-

durans, suggesting an important role in the detoxifica-

tion of ROS [20,33].

This article reports the first crystal structure of a

catalase from the radiation-resistant bacterium D. ra-

diodurans; the DR1998 structure was solved and

refined at 2.6 A resolution. The structural analysis

revealed an overall fold and quaternary structure that

are common among catalases. At present, several

crystal structures of monofunctional catalases from the

three clades are available, of which only two are

clade 1 monofunctional catalases: those from the

proteobacterium Pseudomonas syringae (PDB 1M7S)

and from the firmicute Exiguobacterium oxidotolerans

(PDB 2J2M) [34,35]. Therefore, the DR1998 crystal

structure contributes to elucidate the details of struc-

tural variability among the different clades of mono-

functional catalases.

Results and Discussion

Structure determination and quality

The DR1998 crystal structure was determined by molec-

ular replacement with four molecules in the asymmetric

unit (a.u.) [36]. Structural refinement proceeded

smoothly, and led to Rwork/Rfree of 0.2224/0.2436 and a

final Rfactor of 0.2363, with all available diffraction data.

However, rmsd values of bond distances and bond

angles (Table 1) are typical of structures with lower res-

olution. This is not too surprising, and could be attrib-

utable to the unusual low signal-to-noise ratio of the

original diffraction data (see Experimental procedures).

There are 564 solvent molecules in the structure, includ-

ing 10 chloride ions (Table 1). The assignment of the

chloride ions was based on a comparative analysis of

their atomic displacement parameters with the neigh-

boring atoms, and also on the presence of close neigh-

bors with positive charges.

4139FEBS Journal 281 (2014) 4138–4150 ª 2014 FEBS

P. T. Borges et al. DR1998 catalase X-ray structure

The four molecules in the a.u. (chains A, B, C,

and D) are arranged as two dimers, which, by crys-

tallographic symmetry (chains A0, B0, C0, and D0),form two typical catalase tetramers: tetramer 1,

formed by chains A, B, A0, and B0; and tetramer 2,

formed by chains C, D, C0, and D0 [36]. Superimposi-

tion of the independent molecules showed a low rmsd

of ~ 0.20 A between the superimposed Ca atoms, as

expected for a crystal structure refined with medium-

resolution data and noncrystallographic symmetry

(NCS) restraints.

There was clear electron density for all chains, except

for the first 30 residues and a loop around Gly520. As

the N-terminus up to residue 29 contains six glycines

and 12 hydrophilic residues, the absence of electron den-

sity at all N-termini is most probably due to conforma-

tional disorder. Whereas chains A and D showed weak

density (below 1r contour level) on residue 522,

chains B and C also showed weak density over its near-

est neighbors, residues 520–523 and 521–522, respec-

tively. Diffraction data showed a remarkably low

Wilson B-factor of 25 A2 for the working resolution

limit, but it compared nicely with the range of the mean

B-values of each chain, 23–29-A2 (Table 1). The C-ter-

minal parts of all chains (A, B, C, and D), residues 456–536, showed atomic displacement parameters of

~ 40 A2, indicating their more flexible nature.

The Ramachandran plot showed that 94.9% of the

residues have conformational main chain angles within

the favored regions, and no outlier residues were

observed.

The rmsd between the superimposed Ca atoms of

tetramer 1 and tetramer 2 is 0.24 A. Therefore, for

simplicity, only tetramer 1 will be analyzed in the sub-

sequent sections.

Overall structure of DR1998

The monomer of DR1998 adopts the highly con-

served catalase fold, which is divided into four struc-

tural regions (Figs 1 and 2): the N-terminal arm

(residues 30–82) contains only one a-helix and His81,

one of the essential catalytic residues known as distal

histidine; the second region (residues 83–377) containsone antiparallel eight-stranded b-barrel, with topology

b1b4b5b6b7A,Bb8b9b10 (strand b7 includes a b-bulge,and is therefore referred to as b7A,B), and 10 a-helices(a2–a11); the wrapping loop (residues 378–455) con-

nects the b-barrel domain with the C-terminal

domain, and has almost no secondary structure

elements, only a small helix, a12; and finally, the

C-terminal region (residues 456–536) is a helical

domain composed of four consecutive a-helices(a13a14a15a16) (Figs 1 and 2).

Fig. 1. Multiple amino acid sequence alignment of D. radiodurans catalases. The DR1998 secondary structure and the solvent accessibility

of the amino acids are shown above the alignment. A darker shade of blue denotes higher solvent accessibility. The different structural

domains are indicated, namely: N-terminal arm, b-barrel domain, wrapping loop, and C-terminal helical domain. The amino acids involved in

the main channel are represented by #, and those directly involved in the catalytic center by ↑. The black boxes represent the strictly

conserved residues among the D. radiodurans catalases. The sequence of DRA0259 is only represented up to residue 557.

4140 FEBS Journal 281 (2014) 4138–4150 ª 2014 FEBS

DR1998 catalase X-ray structure P. T. Borges et al.

DR1998 oligomerizes as a homotetramer with 222

point-group symmetry, which is highly conserved

among catalases [25,26]. The dimensions of the

DR1998 tetramer along the three consensually defined

orthogonal two-fold axes, P, Q, and R [13], are

approximately 95, 70, and 95 A, respectively (Fig. 3).

Superposition of either the DR1998 monomer or tet-

ramer with monofunctional catalases with known crys-

tal structures (Table 2) shows that both the tertiary and

quaternary structures of DR1998 are very similar to

those of the other family members. Ca superposition of

monomers gave, on average, rmsd values of 1.27 A with

clade 1, 1.52 A with clade 2 and 1.45 A with clade 3

structures, respectively, whereas the Ca superposition of

tetramers resulted in average rmsd values of 1.08, 1.48,

and 1.37 A, respectively. As expected, the lowest rmsd

values were obtained for the phylogenetically most clo-

sely related sequences, those of the clade 1 catalases

from P. syringae (PDB 1M7S) and E. oxidotolerans

(PDB 2J2M) [34,35].

DR1998 tetramer 1 includes a total of 273 hydrogen

bonds between its four subunits, of which almost

80% are formed between the R-related and Q-related

subunits, with the remaining 20% resulting from

interactions between P-related subunits (Fig. 3). This

Table 1. Data collection, processing and refinement statistics.

Numbers in parentheses refer to the highest-resolution shell.

Data collection

Beamline ESRF ID-29

Detector PILATUS 6M

Wavelength (A) 0.82656

Data processing XDS

Space group C2221

Unit cell parameters (A) a = 97.33, b = 311.88,

c = 145.63

Resolution (A) 49.4–2.6 (2.76–2.6)

Number of observations 352 748 (48 698)

Unique reflections 68 047 (10 656)

Completeness (%) 99.3 (97.5)

Multiplicity 5.2 (4.6)

Mosaicity (°) 0.049

CC1/2 (%)a 97.1 (75.4)

Rsym (%)b 30.8 (83.8)

Rmeas (%)c 34.3 (96.2)

Rpim (%)d 12.0 (27.4)

<I/r(I)> 5.3 (1.7)

Wilson B-factor (A2) 25

Number of molecules in

a.u.

4

VM (A3 Da1) 2.28

Estimated solvent content (%) 46

Refinement statistics

Rfactor (%)e 23.6

Rwork (%)e 22.2

Rfree (%)f 24.4

rmsd for bond lengths (A) 0.006

rmsd for bond angles (°) 0.663

Average chain B-factor (A2) 25, 28, 24, 29

Number of residues 2028

Number of solvent waters 554

Number of solvent chloride ions 10

Ramachandran plot

Residues in favored regions (%) 94.9

Residues in allowed regions (%) 100

Residues in disallowed regions (%) 0

PDB code 4CAB

a CC1/2 = correlation between intensities from random half-datasets

[69]. b Rsym = Σhkl Σi|Ii(hkl) <I(hkl)>|/Σhkl

Σi Ii(hkl), where Ii(hkl) is the observed intensity and <I(hkl)> is the

average intensity of multiple observations from symmetry-related

reflections [52]. c Rmeas = Σhkl[N/(N(hkl) 1)]1/2 Σi|Ii(hkl) <I(hkl)>|/Σhkl Σi Ii(hkl), where N(hkl) is the data multiplicity, Ii(hkl) is the

observed intensity, and <I(hkl)> is the average intensity of multiple

observations from symmetry-related reflections. It is an indicator of

the agreement between symmetry-related observations [53].d Rpim = Σhkl[1/(N(hkl) 1)]1/2 Σi|Ii(hkl) <I(hkl)>|/Σhkl Σi Ii(hkl), where

N(hkl) is the data multiplicity, Ii(hkl) is the observed intensity, and

<I(hkl)> is the average intensity of multiple observations from

symmetry-related reflections. It is an indicator of the precision of

the final merged and averaged data set [54]. e Rfactor = Σ|Fobs Fcalc|/Σ Fobs, where Fobs and Fcalc are the amplitudes of

the observed and the model calculated structure factors, respec-

tively. It is a measure of the agreement between the experimen-

tal X-ray diffraction data and the crystallographic model. f Rwork

refers to the actual working dataset used in refinement, and Rfree

refers to a cross-validation set that is not directly used in refine-

ment and is therefore free from refinement bias.

Fig. 2. Cartoon representation of the DR1998 monomer (chain B).

The four structural regions are shown in different colors: N-terminal

arm (red), b-barrel domain (green), wrapping loop (blue), and

C-terminal helical domain (orange). The heme is represented by

sticks and the heme iron as a black sphere.

4141FEBS Journal 281 (2014) 4138–4150 ª 2014 FEBS

P. T. Borges et al. DR1998 catalase X-ray structure

indicates that, in the tetramer, the major interactions

are between the R-related and Q-related subunits, as

confirmed by the relative interface areas between

P-related, Q-related and R-related subunits of 2211,

4412, and 5908 A2, respectively [37,38]. The N-termi-

nal arm and the b-barrel domain are the regions that

contribute most to the total number of hydrogen

bonds in the tetramer, most of which are between

these two domains (29%) and those involving the

wrapping loop (48%).

Heme active site and heme pocket

The crystal structure of DR1998, shows that each

monomer contains one heme b group. Each heme is

deeply buried inside the tetramer, and is mainly

surrounded by residues from the b-barrel domain

(Figs 2 and 4). The distances between the irons of

each monomer are 29 A for the P-related subunits,

35 A for the R-related subunits, and 44 A for the Q-

related subunits. The bond distances mentioned in the

following description are the mean values of the calcu-

lated equivalent bond distances from chains A, B, C,

and D.

The iron atoms in each heme are pentacoordinated

(Fig. 4) by the four nitrogen atoms from the porphyrin

ring, at a mean distance of 2.1 A, and the phenolate

oxygen atom from the proximal Tyr369 ligand, at a

distance of 2.3 A. The iron atom from the heme group

is displaced from the plane of the pyrrole nitrogen

atoms by 0.3 A towards the oxygen atom of the

Tyr369 phenolate. Besides coordinating the heme iron,

the Tyr369 phenolate oxygen atom forms two hydro-

gen bonds, with Arg365 NE (3.1 A) and NH2 (2.9 A).

Arg365 NH1 is hydrogen-bonded to Asn229 OD1

(2.4 A), Asn229 ND2 is hydrogen-bonded to Asp359

OD2 (2.9 A), and Asp359 OD1 is hydrogen-bonded

(2.9 A) to a solvent water molecule, W2 (Fig. 4),

located at the bottom of a cavity that is open to the

solvent. This hydrogen-bonded network is conserved

among the monofunctional catalases, and it has been

proposed to be involved in the stabilization of the elec-

trostatic field at the catalytic center in the course of

redox catalysis [39].

Both Arg365 and Asp359 are strictly conserved resi-

dues; however, Asn229 is usually a histidine in other

catalases (Table 2). Another known exception occurs

in clade 2, in the catalase from Neurospora crassa

(PDB 1SY7) (Table 2) [40].

On the heme distal side, the imidazole ring of the

essential catalytic His81 is nearly parallel to the

heme pyrrole IV ring, in what is normally termed

the His-IV orientation. This feature is common to

the catalases in clade 1, whereas, in those in clade 3,

the heme plane is rotated by 180°, resulting in a

His-III orientation [25]. His81 NE2 is 5.1 A distant

from the iron atom, and 2.7 A away from water

W1, whereas His81 ND1 is hydrogen-bonded to

Ser125 OE (2.8 A) and also to Thr126 O (2.8 A).

W1 is the only water present in the main channel

described below, and is observed for all chains in

the crystal structure of DR1998 (Fig. 4). The His-IV

orientation, together with the hydrogen bonds with

His81 ND1, favors the imidazole tautomer with a

proton on ND1, whereas NE2 is oriented towards

the iron atom that participates in the catalytic trans-

fer of the hydrogen peroxide, as proposed by Fita

and Rossmann [41]. Also on the heme distal side,

A

B

C

Fig. 3. The DR1998 oligomeric structure: views along the

crystallographic two-fold axis Q (A) and noncrystallographic two-

fold axes R (B) and P (C). Monomer A is in red, monomer B is in

blue, monomer A0 is in green, and monomer B0 is in orange. AB

and A0B0 are R-related dimer subunits; AB0 and BA0 are P-related

dimer subunits; and AA0 and BB0 are Q-related dimer subunits. Left

panels show peptide chains as cartoons, hemes as black sticks,

and iron as black spheres. Right panels show the corresponding

solvent-accessible surfaces.

4142 FEBS Journal 281 (2014) 4138–4150 ª 2014 FEBS

DR1998 catalase X-ray structure P. T. Borges et al.

Asn159, which is considered to be an essential

residue for the catalytic mechanism, is located at a

distance of 3.3 A from water W1.

The heme propionates form hydrogen bonds with

three conserved arginines in the monofunctional cata-

lases (Arg78, Arg123 and Arg376 in DR1998)

(Table 2). Carboxylate oxygens O1D and O2D from

propionate III are hydrogen-bonded to Arg376 NE

(2.7 A) and Arg376 NH2 (2.6 A), whereas O1A and

O2A from propionate IV are hydrogen-bonded to

Arg78 NE (3.1 A) and Arg123 NH1, respectively

(2.7 A) (Fig. 4).

Table 2. Structural alignment between DR1998 monomer and related monofunctional catalase X-ray crystallography structures. The

corresponding residues involved on the proximal heme side, heme propionate groups, heme distal side, main heme channel and NADPH-

binding region are presented. Only one PDB code was chosen from each organism. Clade 1: D. radiodurans (PDB 4CAB), P. syringae (PDB

1M7S) [34], and E. oxidotolerans (PDB 2J2M) [35]. Clade 2: Es. coli (PDB 1GGE) [46], Pe. vitale (PDB 2IUF) [57], and N. crassa (PDB 1SY7)

[40]. Clade 3: beef liver (PDB 7CAT) [70], S. cerevisiae (PDB 1A4E) [43], human erythrocyte (PDB 1DGF) [39], Micrococcus lysodeikticus

(PDB 1HBZ) [56], Proteus mirabilis (PDB 1M85) [71], Enterococcus faecalis (PDB 1SI8) [72], Helicobacter pylori (PDB 2IQF) [57],

Vibrio salmonicida (PDB 2ISA) [73], Hansenula polymorpha (PDB 2XQ1) [74], Corynebacterium glutamicum (PDB 4B7F), and

Pseudomonas aeruginosa (PDB 4E37). The conserved residues among all the catalases are in bold type. In the structures with PDB codes

1M85, 2ISA, and 4E37, it was observed a methionine sulfone (Omt), at the position of Val80 in DR1998 (4CAB).

Clade 1 Clade 2 Clade 3

4CAB 1M7S 2J2M 1GGE 2IUF 1SY7 7CAT 1A4E 1DGF 1HBZ 1M85 1SI8 2IQF 2ISA 2XQ1 4B7F 4E37

Proximal heme side

Tyr369 Tyr358 Tyr339 Tyr415 Tyr351 Tyr379 Tyr357 Tyr355 Tyr358 Tyr343 Tyr337 Tyr337 Tyr339 Tyr337 Tyr348 Tyr353 Tyr338

Arg365 Arg354 Arg335 Arg411 Arg347 Arg375 Arg353 Arg351 Arg354 Arg339 Arg333 Arg333 Arg335 Arg333 Arg344 Arg349 Arg334

Asn229 His220 His199 His275 His211 Asn239 His217 His215 His218 His203 His197 His197 His199 His197 His208 His201 His198

Asp359 Asp348 Asp329 Asp405 Asp341 Asp369 Asp347 Asp345 Asp348 Asp333 Asp327 Asp327 Asp329 Asp327 Asp338 Asp343 Asp328

Heme distal side

His81 His78 His56 His128 His64 His92 His74 His70 His75 His61 His54 His54 His56 His54 His65 His71 His65

Asn159 Asn150 Asn129 Asn201 Asn137 Asn165 Asn147 Asn143 Asn148 Asn133 Asn127 Asn127 Asn129 Asn127 Asn138 Asn143 Asn128

Ser125 Ser116 Ser95 Ser167 Ser103 Ser131 Ser113 Ser109 Ser114 Ser99 Ser93 Ser93 Ser95 Ser93 Ser104 Ser109 Ser94

Thr126 Ser117 Thr96 Thr168 Thr104 Thr132 Thr114 Thr110 Thr115 Thr100 Thr94 Thr94 Thr96 Thr94 Thr105 Thr110 Thr95

Heme propionate groups

Arg78 Arg75 Arg53 Arg125 Arg61 Arg89 Arg71 Arg67 Arg72 Arg58 Arg51 Arg51 Arg53 Arg51 Arg62 Arg68 Arg52

Arg123 Arg114 Arg93 Arg165 Arg101 Arg129 Arg111 Arg107 Arg112 Arg97 Arg91 Arg91 Arg93 Arg91 Arg102 Arg107 Arg92

Arg376 Arg365 Arg346 Arg422 Arg358 Arg386 Arg364 Arg362 Arg365 Arg350 Arg344 Arg344 Arg346 Arg344 Arg355 Arg360 Arg345

Heme vinyl group

Val228 Val219 Val198 Ile274 Val210 Val238 Ser216 Gly214 Ser217 Ser202 Ser196 Ser196 Ser198 Ser196 Gly207 Ser212 Ser197

Main heme channel

Val80 Val77 Val55 Val127 Val63 Val91 Val73 Pro69 Val74 Pro60 Omt53 Val53 Val55 Omt53 Val64 Pro70 Omt54

Val127 Val118 Val97 Val169 Val105 Val133 Val115 Val111 Val116 Val101 Val95 Val95 Val97 Val95 Val106 Val111 Val96

Pro140 Pro131 Pro110 Ile182 Val118 Val146 Pro128 Pro124 Pro129 Val114 Ile108 Pro108 Pro110 Ile108 Pro119 Val124 Ile109

Phe164 Phe155 Phe134 Phe206 Phe142 Phe170 Phe152 Phe148 Phe153 Phe138 Phe132 Phe132 Phe134 Phe132 Phe143 Phe148 Phe133

Phe165 Phe156 Phe135 Phe207 Phe143 Phe171 Phe153 Phe149 Phe154 Phe139 Tyr133 Phe133 Phe135 Phe133 Phe144 Phe149 Tyr134

Phe172 Phe163 Phe142 Phe214 Phe150 Phe178 Phe160 Phe156 Phe161 Phe146 Phe140 Phe140 Phe142 Phe140 Phe151 Phe156 Phe141

Ile176 Val167 Val146 Val218 Ile154 Ile182 Ile164 Ile160 Ile165 Ile150 Asn144 Ile144 Ile146 Asn144 Ile155 Ile160 Asn145

Asp139 Asp130 Asp109 Asp181 Asp117 Asp145 Asp127 Asp123 Asp128 Asp113 Asp107 Asp107 Asp109 Asp107 Asp118 Asp123 Asp108

Gln179 Phe170 Leu149 Val221 Val157 Gly185 Gln167 Gln163 Gln168 Gln153 Val147 Gln147 Gln149 Val147 Gln158 Gln163 Val148

His129 His120 His99 Gly171 Gly107 Gly135 Gly117 Gly113 Gly118 Gly103 Gly97 Gly97 Gly99 Gly97 Gly108 Gly113 Gly98

Arg138 Arg129 Arg108 Arg180 Arg116 Arg144 Arg126 Arg122 Arg127 Arg112 Arg106 Arg106 Arg108 Arg108 Arg117 Arg122 Arg107

NADPH-binding pocket

His205 Arg196 Asn175 His251 His187 His215 His193 His191 His194 His179 His173 His173 Tyr175 His173 His184 His189 His174

Ser212 Ser203 Thr182 Ser258 Ala194 Ser222 Ser200 Ser198 Ser201 Gly186 Ser180 Ser180 Ser182 Ser180 Ser191 Gly196 Ser181

Trp214 Glu205 Glu184 Arg260 His196 Arg224 Arg202 Arg200 Arg203 Arg188 Arg182 Arg182 Arg184 Arg182 Arg193 Arg198 Arg183

Gln224 Asp215 Arg194 Glu270 Asn206 Gln234 Asp212 His210 Asn213 Asn198 His192 His192 Asp194 His192 Asn203 Asp208 His193

Glu248 Lys239 Val218 Lys294 Lys230 Thr258 Lys236 Lys234 Lys237 Ile222 Arg216 Lys216 His218 Val216 Ile227 Lys232 Lys217

Trp314 Trp305 Trp284 Ile360 Val296 Trp324 Trp302 Trp300 Trp303 Ile288 Trp282 Val282 Trp284 Trp282 Trp293 Trp298 Trp283

Arg316 Asp307 Glu286 Glu362 Glu298 Glu326 His304 Gln302 His305 Gln290 His284 Gln284 Thr286 His284 His295 Gln300 His285

Gln457 Gln440 Gln423 His507 Ser442 Gln466 Gln441 Val444 Gln442 Gln429 Ser420 Thr418 Gln428 Ser420 Gln438 Gln456 Ser422

4143FEBS Journal 281 (2014) 4138–4150 ª 2014 FEBS

P. T. Borges et al. DR1998 catalase X-ray structure

Heme main channel

The dismutation reaction occurs at the heme distal

site, which is accessed from the surface through a

25 A long channel. In the tetramer, these channels run

parallel to the R-axis, and their entries are 20 A apart

(Fig. 5).

This main channel can be divided into two parts.

The first, a hydrophobic region near the catalytic site

is, 10 A in length with a diameter of 4 A, and is lined

mainly by the hydrophobic residues Val80, Val127,

Pro140, Phe164, Phe165, Phe172, Leu175, and Ile176

(Fig. 5). The second channel region is more hydro-

philic, starts on Asp139, and is less conserved in the

family, with some residues, e.g. His129, Arg138,

Asn188, Asn190, and Asn265, running through a fur-

ther length of 15 A with a diameter ranging between 4

and 15 A (Fig. 5).

The electrostatic charge distribution on the first part

of this channel is influenced by the negative charge of

Asp139 at the top of the hydrophobic region, and the

positive charge of the metal site at the bottom. This

feature allows fast substrate access to the iron, and

explains the high values for the fast-turnover kcat(range 7 9 104 to 1.6 9 106 s1) found in catalases

[26,42].

This region has been regarded as a molecular ruler,

optimized to select hydrogen peroxide. Mutations in

the monofunctional catalase of Sacharomyces cerevisiae

(PDB 1A4E) designed to enlarge the hydrophobic

channel region resulted in lower enzymatic activity and

increased peroxidase activity. This was attributed to

the facilitation of active site access by larger substrates

with a concomitant decrease in hydrogen peroxide

selectivity.

Fig. 4. The heme catalytic center of DR1998 (chain B). The heme

iron is coordinated by the proximal ligand Tyr369, which forms a

hydrogen bond network with three other residues: Arg365,

Asn229, and Asp359. Arg78, Arg123 and Arg376, are hydrogen-

bonded with the heme propionates. On the heme distal side, two

residues are represented, His81 and Asn159, which are conserved

among catalases, and are hydrogen-bonded to a water molecule

(W1). The amino acids and the heme are represented as sticks,

the heme iron as a black sphere, and waters W1 and W2 as red

spheres. The monomer is colored as in Fig. 2.

A B

Fig. 5. (A) Overall representation of the

main channels in the DR1998 tetramer.

These approximately 25 A-long channels

are viewed along the Q-axis (top) and the

R-axis (bottom). The channels are

represented as black mesh, and connect

the protein surface with each heme active

site. Monomers are represented as a Ca

trace, and colored as in Fig. 3. Heme

groups are represented as black sticks,

and iron atoms as black spheres. (B) The

main channel structure in DR1998

(chain B). Chains A and B are represented

as Ca traces (colored as in Fig. 3). The

side chains of the amino acids that line

the channel are represented as sticks,

with carbon atoms colored red for chain A

and white for chain B. The heme iron is

represented as a black sphere, and water

W1 as a red sphere.

4144 FEBS Journal 281 (2014) 4138–4150 ª 2014 FEBS

DR1998 catalase X-ray structure P. T. Borges et al.

Lateral channel

Besides the above-described main channel, a lateral

channel connecting the catalytic center to the external

surface has been described for other monofunctional

catalases, e.g. that from S. cerevisiae [43,44]. In the

vicinity of the active site, the solvent-accessible surface

of DR1998 shows a second channel (lateral channel)

(Fig. 6). This is a 16-A-deep cavity that reaches the

hydrophobic Val228, located close to the vinyl group

linked to pyrrole I. Three water molecules are located

within this channel in each of the four independent

protein chains in the crystal structure. This channel

has been reported to be involved in the exit/entry of

molecules [45]. However, in DR1998, Val228 blocks

access from the cavity to the active site. The homolo-

gous residues in other catalases are Val/Ile for clades 1

and 2 and Ser/Gly for clade 3 (Table 2). It has been

proposed that these residues (Table 2), with an unfa-

vorable φ,w conformation [34,40,46], could have a sig-

nificant role in catalysis. Routine analysis of

conformational φ and w angles in macromolecules was

pioneered by PROCHECK [47]; meanwhile, not only have

the quantity and quality of data in the PDB increased

enormously, but also the accuracy and specificity of

Ramachandran plots have been steadily improved

[48,49]. A re-examination of the above-mentioned

structures with MOLPROBITY [50], which uses a dataset

of 1.6 million residues with backbone atomic displace-

ment parameters of < 30 A2, showed that, in fact, all

of those residues have a conserved conformation, and

that they are all located in the allowed regions of the

Ramachandran (φ,w) plot. Therefore, the previous

suggestion regarding their catalytic role was based on

outdated estimates of the favorable φ,w conforma-

tions, and is no longer applicable.

In clade 3 small-subunit catalases, this lateral chan-

nel has been described as part of a bifurcation with

another channel that runs from the essential aspara-

gine in the heme distal side up to the NADPH-binding

pocket. A search for the presence of this type of chan-

nel in DR1998 was unsuccessful, as the corresponding

region is blocked by the C-terminus (residues 530–536)of the R-related subunit (Figs 6 and 7). Although

NADPH binding was only observed in some clade 3

catalases, the corresponding binding pocket is com-

pletely accessible from the exterior in clade 1 and

clade 3 catalases, and the amino acids that participate

in NADPH binding are present in catalases from the

different clades (Table 2). In the case of clade 2 cata-

lases, such as the Escherichia coli protein, this pocket

is blocked by residues 585–590 (Fig. 7). However, an

R260A mutation located in the binding pocket region

showed, in this case, an increase in catalytic activity,

suggesting faster diffusion of substrates or reaction

products through that alternative protein channel [45].

In DR1998, Trp214 is located in a structurally equiva-

lent position to Arg260 in Es. coli, and is the single

example of a tryptophan at this position in all of the

catalases included in Table 2. Therefore, in DR1998,

this part of the lateral channel is blocked by the C-

terminal residues 530–536 from the R-related subunit

and also by Trp214. Nevertheless, the function of

alternative channels in DR1998 warrants further

investigation.

Concluding remarks

D. radiodurans is a radiation-resistant organism with a

genome that encodes three monofunctional catalases,

one from each of the three known clades. This article

describes the crystal structure of the clade 1 catalase

DR1998. A sequence alignment between DR1998

(clade 1) and DRA0259 (clade 2) shows that most of

the residues important for catalysis are conserved.

However, some of these residues are not conserved in

DRA0146 (Fig. 1). The present structure was deter-

mined from a protein crystal with a diffraction data

Fig. 6. Main channel (MC) and lateral channel (LC) representation.

The channels are drawn as black meshes. The DR1998 monomers

are depicted as cartoons, colored red for chain A and blue for

chain B. The amino acids Asn159, Trp214, Val228 and Tyr369 and

the heme group are represented as sticks. The heme iron is

represented as a black sphere.

4145FEBS Journal 281 (2014) 4138–4150 ª 2014 FEBS

P. T. Borges et al. DR1998 catalase X-ray structure

signal-to-noise ratio of only 2.3, which is quite unu-

sual. Despite these weak diffraction data, the precision

indicators of the merged data, processed mosaic

spread, Wilson B-factor and refined structure atomic

displacement parameters showed lower values than

those usually found from datasets recorded with crys-

tals diffracting to a similar resolution. Such an accom-

plishment would have been unthinkable a few years

ago, and was only made possible by the combined use

of high-brilliance synchrotron radiation sources, pixel-

type detectors and data collection strategies leading to

a high multiplicity of equivalent measurements of the

same independent reflection set.

The analysis of the DR1998 structure highlights the

structural conservation of the oligomerization state, a

tetramer with 222 point-group symmetry, as well as of

the monomer architecture described as the catalase

fold. The hemes are buried within the tetramer, and

the residues involved in the catalytic active center are

also conserved, namely the proximal ligand Tyr369,

and the distal ligands His81 and Asn159. The main

channel that is considered to allow access of the

hydrogen peroxide to the active site begins at the heme

pocket, and continues through a narrow hydrophobic

region that is followed by a wider and more hydro-

philic region that starts at Asp139, and that allows the

access of solvent molecules from the external surface.

An alternative channel was also identified, starting at

the external surface and approaching (but not reach-

ing) the heme through a channel that ends in Val228.

The DR1998 structure is the first crystal structure of

a catalase from the radiation-resistant bacterium

D. radiodurans. It will be important in the future to

determine the crystal structures of the other two cata-

lases from this bacterium, and to complement these

with kinetic studies. This will not only contribute

towards understanding the detoxification mechanisms

for hydrogen peroxide in D. radiodurans, but also

allow a structural comparison between the three differ-

ent catalases, which are representatives of each of the

three known clades of the monofunctional catalases.

Experimental procedures

Protein purification and crystallization

DR1998 was obtained from D. radiodurans as previously

described [36]. A 300 L culture of D. radiodurans was

grown at 30 °C to an attenuance at 600 nm of 2.0, in M53

medium containing 0.5% (w/v) yeast extract, 0.5% (w/v)

glucose, 0.5% (w/v) NaCl, and 1% (w/v) casein. The cells

were harvested and resuspended in 20 mM Tris/HCl

(pH 7.2). The cells were then broken in a French press at

9000 p.s.i. The soluble fraction was separated from the

membrane pellet by ultracentrifugation at 182 000 g for

6.5 h, and dialyzed at 4 °C in 20 mM Tris/HCl (pH 7.2).

The purification protocol was performed at 4 °C and

pH 7.2. The soluble fraction was applied to a DEAE-Fast

Flow column (XK 50/30, 10 mLmin1) equilibrated with

20 mM Tris/HCl, and a 0–1 M linear NaCl gradient in the

same buffer was applied. The fraction containing the heme

UV-visible fingerprint (peak at 400 nm) eluted at 550 mM

NaCl, and was dialyzed against 20 mM Tris/HCl. This frac-

tion was then loaded onto a Q-Sepharose HP column

(XK 26/10, 2 mLmin1) equilibrated with 20 mM Tris/HCl,

and eluted with a 0–1 M linear NaCl gradient. The heme-

containing fraction eluted at 600 mM NaCl. A final purifica-

tion step was performed in order to remove the remaining

contaminants. The heme-containing fraction was applied to

a Q-Sepharose HiTrap HP column (5 mL, 1.5 mLmin1)

equilibrated with 20 mM Tris/HCl, and eluted with a 0–1 M

linear sodium acetate gradient. The final fraction containing

DR1998 eluted at 600 mM sodium acetate.

SDS/PAGE was used to confirm the purity of the pro-

tein, which was then used in crystallization trials as

described in [36]. Crystals were obtained at room tempera-

ture with the hanging-drop vapor diffusion method, by

Fig. 7. View of the ‘NADPH-binding pocket’ that is present in

some clade 3 catalases. The protein chains are drawn in cartoon

representation. DR1998 is colored red (chain A) and blue (chain B);

beef liver catalase (PDB 7CAT) is colored yellow, with its NADPH

highlighted in gray; Es. coli catalase (PDB 1GGE) is colored green.

Amino acids are represented as sticks, with carbon atoms in gray

for DR1998, yellow for beef liver catalase, and green for Es. coli

catalase. The residue numbering is from DR1998; the

corresponding residues for the other catalases are presented in

Table 2.

4146 FEBS Journal 281 (2014) 4138–4150 ª 2014 FEBS

DR1998 catalase X-ray structure P. T. Borges et al.

mixing 1.0 lL of 12 mgmL1 purified native protein

solution with 1.0 lL of reservoir solution, containing 0.2 M

magnesium acetate, 0.1 M sodium cacodylate (pH 6.5)

and 20% poly(ethylene glycol) 8000. Crystals (approxi-

mately 250 9 70 9 15 lm) were cryoprotected with the

reservoir solution supplemented with 25% glycerol prior to

flash-cooling in liquid nitrogen [36].

X-ray diffraction, crystal structure determination,

and refinement

Diffraction data were measured at 100 K at beamline ID29

of the European Synchrotron Radiation Facility (ESRF,

Grenoble, France) by the use of X-rays with a wavelength

of 0.826 A and a PILATUS 6M detector. Data collection,

processing, crystallographic characterization and structure

determination have been described previously [36], and are

summarized in Table 1.

Crystals of DR1998 diffracted so faintly that, even when

an X-ray beam with a diameter of 30 lm at ESRF beam-

line ID29 was used [51], the individual (unmerged) intensi-

ties showed a signal-to-noise ratio of only 2.3. Owing to a

data multiplicity of 5.2, the final merged data showed a

final signal-to-noise ratio of 5.3 for a resolution up to

2.6 A, where the mean diffraction intensity approached the

noise level (Table 1). However, the signature of the very

weak original data was apparent in the unusually

high merging statistics, Rsym [52], Rmeas [53], and Rpim [54]

(Table 1), as compared with PDB data [55] obtained at a

similar resolution, showing <Rsym/merge> of 0.10(6) (http://

rcsb-auto-check.rutgers.edu/dev/pdbitem/index.html).

Additionally, the mosaic spread and Wilson B-values of

0.05° and 25 A2, respectively, are comparable to those usu-

ally found in protein crystals diffracting to higher resolu-

tion. The low mosaic spread is certainly influenced by the

small beam divergence of ESRF station ID29, where the

crystal sample is positioned ~ 15 m away from the beam

primary slits, suggesting that the initial, unusually low sig-

nal-to-noise ratio of the diffraction was the main contribu-

tor to the high resolution limit discrepancy. As the

diffraction intensity at low resolution was already near its

detection limit, and despite a low Wilson B-value, even a

small decay caused by a modest resolution increment (up

to 2.6 A) was enough to bring the diffraction intensity

below its detection threshold.

Initial phases were determined by molecular replacement

with an ensemble of six catalase monomers from different

organisms showing the highest sequence identities (34–50%)

with DR1998 [36]: PDB entries 2J2M from E. oxidotolerans

[35], 1M7S from P. syringae [34], 1GWE from Micrococ-

cus luteus [56], 1DGF from Homo sapiens [39], 2IUF from

Penicillium vitale [57], and 1SY7 from N. crassa [40].

As the crystal structure contained four DR1998 mono-

mers in the a.u., NCS between them was expected, and 3%

of the data in thin-resolution shells were set aside for Rfree

monitoring [36]. The TLSMD server (http://skuld.bmsc.wash-

ington.edu/~tlsmd) [58] was used to define polypeptide

chain regions for translation, libration and screw refine-

ment of atomic displacement parameters. PHENIX.REFINE [59]

was used for structure refinement of atomic positional,

atomic isotropic and translation, libration and screw

atomic displacement parameters, and included standard ste-

reochemical and NCS restraints. During refinement, Rfree

was used to steer the relative weights of stereochemical

restraints versus experimental data in the minimized func-

tion. Structure refinement was performed in iteratively

repeated cycles of protein and solvent updating and refine-

ment, alternated with inspection of rA-weighted 2|Fo| |Fc| and |Fo| |Fc| electron density maps for manual

model improvement with COOT [60]. The stereochemistry of

the refined structure was analyzed with MOLPROBITY [50].

Diffraction data and refinement statistics are presented in

Table 1. Structure factors and associated structure coordi-

nates were deposited in the Protein Data Bank in Europe

[61]. The analysis of molecular channels was performed

with CAVER [62,63] and MOLE [64].

Sequence alignment

A multiple amino acid sequence alignment of the D. radio-

durans catalases was perfomed with CLUSTALX [65] and

edited with GENEDOC [66].

Figures

Figures were prepared with PYMOL [66,67,68].

Acknowledgements

This work was supported by grants and fellowships

from Fundac~ao para a Ciencia e Tecnologia under the

following projects: PTDC/BIA-PRO/100365/2008

(C. V. Rom~ao) and PEst-OE/EQB/LA0004/2013. P. T.

Borges is the recipient of PhD grant SFRH/BD/85106/

2012; C. S. Miranda was supported by a BI fellowship

within FCT grant PTDC/BIA-PRO/100365/2008; and

C. V. Rom~ao is the recipient of SFRH/BPD/94050/

2013. We thank the ESRF (Grenoble, France) for sup-

port during the X-ray diffraction data collection, and

P. Matias for critically reviewing the manuscript.

Author contributions

C.V.R., C.F. and P.T.B. designed research; C.V.R.,

C.F., P.T.B., C.S.M. performed research; C.V.R., C.F.

and P.T.B. analyzed data; and C.V.R., C.F., P.T.B.

and M.A.C. wrote the paper.

4147FEBS Journal 281 (2014) 4138–4150 ª 2014 FEBS

P. T. Borges et al. DR1998 catalase X-ray structure

References

1 White O, Eisen JA, Heidelberg JF, Hickey EK,

Peterson JD, Dodson RJ, Haft DH, Gwinn ML,

Nelson WC, Richardson DL et al. (1999) Genome

sequence of the radioresistant bacterium Deinococcus

radiodurans R1. Science 286, 1571–1577.

2 Anderson AW, Nordan HC, Cain RF, Parrish G &

Duggan D (1956) Studies on a radio-resistant

micrococcus. 1. Isolation, morphology, cultural

characteristics, and resistance to gamma radiation. Food

Technol 10, 575–578.

3 Confalonieri F & Sommer S (2011) Bacterial and

archaeal resistance to ionizing radiation. In Cost

Chemistry Cm0603 – Melusyn Joint Meeting: Damages

Induced in Biomolecules by Low and High Energy

Radiations, J Phys: Conf Ser 261, 012005.

4 Cox MM & Battista JR (2005) Deinococcus

radiodurans – the consummate survivor. Nat Rev

Microbiol 3, 882–892.

5 Gao N, Ma B-G, Zhang Y-S, Song Q, Chen L-L &

Zhang H-Y (2009) Gene expression analysis of four

radiation-resistant bacteria. Genomics Insights 2, 11–22.

6 Omelchenko MV, Wolf YI, Gaidamakova EK,

Matrosova VY, Vasilenko A, Zhai M, Daly MJ,

Koonin EV & Makarova KS (2005) Comparative

genomics of Thermus thermophilus and Deinococcus

radiodurans: divergent routes of adaptation to

thermophily and radiation resistance. BMC Evol Biol

5, 57.

7 Daly MJ, Gaidamakova EK, Matrosova VY, Vasilenko

A, Zhai M, Leapman RD, Lai B, Ravel B, Li SM,

Kemner KM et al. (2007) Protein oxidation implicated

as the primary determinant of bacterial radioresistance.

PLoS Biol 5, e92.

8 Daly MJ, Gaidamakova EK, Matrosova VY, Vasilenko

A, Zhai M, Venkateswaran A, Hess M, Omelchenko

MV, Kostandarithes HM, Makarova KS et al. (2004)

Accumulation of Mn(II) in Deinococcus radiodurans

facilitates gamma-radiation resistance. Science 306,

1025–1028.

9 Levin-Zaidman S, Englander J, Shimoni E, Sharma

AK, Minton KW & Minsky A (2003) Ringlike structure

of the Deinococcus radiodurans genome: a key to

radioresistance? Science 299, 254–256.

10 Zimmerman JM & Battista JR (2005) A ring-like

nucleoid is not necessary for radioresistance in the

Deinococcaceae. BMC Microbiol 5, 17.

11 McPherson A (1991) A brief history of protein crystal

growth. J Crystal Growth 110, 1–10.

12 Murthy MR, Reid TJ III, Sicignano A, Tanaka N &

Rossmann MG (1981) Structure of beef liver catalase.

J Mol Biol 152, 465–499.

13 Reid TJ III, Murthy MR, Sicignano A, Tanaka N,

Musick WD & Rossmann MG (1981) Structure and

heme environment of beef liver catalase at 2.5 A

resolution. Proc Natl Acad Sci USA 78, 4767–4771.

14 Sumner JB & Dounce AL (1937) Crystalline catalase.

Science 85, 366–367.

15 Makarova KS, Aravind L, Wolf YI, Tatusov RL,

Minton KW, Koonin EV & Daly MJ (2001) Genome

of the extremely radiation-resistant bacterium

Deinococcus radiodurans viewed from the perspective

of comparative genomics. Microbiol Mol Biol Rev 65,

44–79.

16 Slade D & Radman M (2011) Oxidative stress

resistance in Deinococcus radiodurans. Microbiol Mol

Biol Rev 75, 133–191.

17 Chou FI & Tan ST (1990) Manganese(II) induces cell

division and increases in superoxide dismutase and

catalase activities in an aging deinococcal culture.

J Bacteriol 172, 2029–2035.

18 Markillie LM, Varnum SM, Hradecky P & Wong KK

(1999) Targeted mutagenesis by duplication insertion

in the radioresistant bacterium Deinococcus

radiodurans: radiation sensitivities of catalase (katA)

and superoxide dismutase (sodA) mutants. J Bacteriol

181, 666–669.

19 Wang P & Schellhorn HE (1995) Induction of

resistance to hydrogen peroxide and radiation

in Deinococcus radiodurans. Can J Microbiol 41,

170–176.

20 Lipton MS, Pasa-Tolic L, Anderson GA, Anderson DJ,

Auberry DL, Battista JR, Daly MJ, Fredrickson J,

Hixson KK, Kostandarithes H et al. (2002) Global

analysis of the Deinococcus radiodurans proteome by

using accurate mass tags. Proc Natl Acad Sci USA 99,

11049–11054.

21 Day WA Jr, Sajecki JL, Pitts TM & Joens LA (2000)

Role of catalase in Campylobacter jejuni intracellular

survival. Infect Immun 68, 6337–6345.

22 Dos Santos WG, Pacheco I, Liu MY, Teixeira M,

Xavier AV & LeGall J (2000) Purification and

characterization of an iron superoxide dismutase and a

catalase from the sulfate-reducing bacterium

Desulfovibrio gigas. J Bacteriol 182, 796–804.

23 Barynin VV & Whittaker JW (2009) Manganese

Catalase. In Handbook of Metalloproteins, 3.

24 Chelikani P, Fita I & Loewen PC (2004) Diversity of

structures and properties among catalases. Cell Mol

Life Sci 61, 192–208.

25 Diaz A, Loewen PC, Fita I & Carpena X (2012) Thirty

years of heme catalases structural biology. Arch

Biochem Biophys 525, 102–110.

26 Mate MJ, Murshudov G, Bravo J, Melik-Adamyan W,

Loewen PC & Fita I (2006) Heme-Catalases in

Handbook of Metalloproteins. (Messerschmidt A, Huber

R, Poulos T & Wieghardt K, eds) pp. 486–502, John

Wiley & Sons Ltd, Oxford, England.

4148 FEBS Journal 281 (2014) 4138–4150 ª 2014 FEBS

DR1998 catalase X-ray structure P. T. Borges et al.

27 Zamocky M, Furtmuller PG & Obinger C (2008)

Evolution of catalases from bacteria to humans.

Antioxid Redox Signal 10, 1527–1548.

28 Zamocky M, Gasselhuber B, Furtmuller PG & Obinger

C (2012) Molecular evolution of hydrogen peroxide

degrading enzymes. Arch Biochem Biophys 525, 131–144.

29 Dolphin D, Forman A, Borg DC, Fajer J & Felton RH

(1971) Compounds I of catalase and horse radish

peroxidase: pi-cation radicals. Proc Natl Acad Sci USA

68, 614–618.

30 Groves JT, Haushalter RC, Nakamura M, Nemo TE &

Evans BJ (1981) High-valent iron–porphyrin complexes

related to peroxidase and cytochrome P-450. J Am

Chem Soc 103, 2884–2886.

31 Alfonso-Prieto M, Biarnes X, Vidossich P & Rovira C

(2009) The molecular mechanism of the catalase

reaction. J Am Chem Soc 131, 11751–11761.

32 Kobayashi I, Tamura T, Sghaier H, Narumi I,

Yamaguchi S, Umeda K & Inagaki K (2006)

Characterization of monofunctional catalase KatA

from radioresistant bacterium Deinococcus radiodurans.

J Biosci Bioeng 101, 315–321.

33 Karlin S & Mrazek J (2001) Predicted highly expressed

and putative alien genes of Deinococcus radiodurans and

implications for resistance to ionizing radiation damage.

Proc Natl Acad Sci USA 98, 5240–5245.

34 Carpena X, Soriano M, Klotz MG, Duckworth HW,

Donald LJ, Melik-Adamyan W, Fita I & Loewen PC

(2003) Structure of the clade 1 catalase, CatF of

Pseudomonas syringae, at 1.8 A resolution. Proteins 50,

423–436.

35 Hara I, Ichise N, Kojima K, Kondo H, Ohgiya S,

Matsuyama H & Yumoto I (2007) Relationship

between the size of the bottleneck 15 A from iron in

the main channel and the reactivity of catalase

corresponding to the molecular size of substrates.

Biochemistry 46, 11–22.

36 Borges PT, Miranda CS, Santos SP, Carita JN, Frazao

C & Romao CV (2014) Purification, crystallization and

phase determination of the DR1998 haem b catalase

from Deinococcus radiodurans. Acta Crystallogr F Struct

Biol Commun 70, 659–662.

37 Krissinel E & Henrick K (2007) Inference of

macromolecular assemblies from crystalline state. J Mol

Biol 372, 774–797.

38 McDonald IK & Thornton JM (1994) Satisfying

hydrogen bonding potential in proteins. J Mol Biol 238,

777–793.

39 Putnam CD, Arvai AS, Bourne Y & Tainer JA (2000)

Active and inhibited human catalase structures: ligand

and NADPH binding and catalytic mechanism. J Mol

Biol 296, 295–309.

40 Diaz A, Horjales E, Rudino-Pinera E, Arreola R &

Hansberg W (2004) Unusual Cys-Tyr covalent bond in

a large catalase. J Mol Biol 342, 971–985.

41 Fita I & Rossmann MG (1985) The active center of

catalase. J Mol Biol 185, 21–37.

42 Switala J & Loewen PC (2002) Diversity of properties

among catalases. Arch Biochem Biophys 401, 145–154.

43 Mate MJ, Zamocky M, Nykyri LM, Herzog C, Alzari

PM, Betzel C, Koller F & Fita I (1999) Structure of

catalase-A from Saccharomyces cerevisiae. J Mol Biol

286, 135–149.

44 Zamocky M, Herzog C, Nykyri LM & Koller F (1995)

Site-directed mutagenesis of the lower parts of the

major substrate channel of yeast catalase A leads to

highly increased peroxidatic activity. FEBS Lett 367,

241–245.

45 Sevinc MS, Mate MJ, Switala J, Fita I & Loewen PC

(1999) Role of the lateral channel in catalase HPII of

Escherichia coli. Protein Sci 8, 490–498.

46 Melik-Adamyan W, Bravo J, Carpena X, Switala J,

Mate MJ, Fita I & Loewen PC (2001) Substrate flow in

catalases deduced from the crystal structures of active

site variants of HPII from Escherichia coli. Proteins 44,

270–281.

47 Morris AL, MacArthur MW, Hutchinson EG &

Thornton JM (1992) Stereochemical quality of protein

structure coordinates. Proteins 12, 345–364.

48 Kleywegt GJ & Jones TA (1996) Phi/psi-chology:

Ramachandran revisited. Structure 4, 1395–1400.

49 Lovell SC, Davis IW, Arendall WB III, de Bakker PI,

Word JM, Prisant MG, Richardson JS & Richardson

DC (2003) Structure validation by Calpha geometry:

phi, psi and Cbeta deviation. Proteins 50, 437–450.

50 Chen VB, Arendall WB III, Headd JJ, Keedy DA,

Immormino RM, Kapral GJ, Murray LW, Richardson

JS & Richardson DC (2010) MolProbity: all-atom

structure validation for macromolecular

crystallography. Acta Crystallogr D Biol Crystallogr 66,

12–21.

51 de Sanctis D, Beteva A, Caserotto H, Dobias F,

Gabadinho J, Giraud T, Gobbo A, Guijarro M, Lentini

M, Lavault B et al. (2012) ID29: a high-intensity highly

automated ESRF beamline for macromolecular

crystallography experiments exploiting anomalous

scattering. J Synchrotron Radiat 19, 455–461.

52 Arndt UW, Crowther RA & Mallett JF (1968) A

computer-linked cathode-ray tube microdensitometer

for x-ray crystallography. J Sci Instrum 1, 510–516.

53 Diederichs K & Karplus PA (1997) Improved R-factors

for diffraction data analysis in macromolecular

crystallography. Nat Struct Biol 4, 269–275.

54 Weiss M (2001) Global indicators of X-ray data

quality. J Appl Crystallogr 34, 130–135.

55 Bernstein FC, Koetzle TF, Williams GJ, Meyer EF Jr,

Brice MD, Rodgers JR, Kennard O, Shimanouchi T &

Tasumi M (1977) The Protein Data Bank: a computer-

based archival file for macromolecular structures. J Mol

Biol 112, 535–542.

4149FEBS Journal 281 (2014) 4138–4150 ª 2014 FEBS

P. T. Borges et al. DR1998 catalase X-ray structure

56 Murshudov GN, Grebenko AI, Brannigan JA, Antson

AA, Barynin VV, Dodson GG, Dauter Z, Wilson KS

& Melik-Adamyan WR (2002) The structures of

Micrococcus lysodeikticus catalase, its ferryl

intermediate (compound II) and NADPH complex.

Acta Crystallogr D Biol Crystallogr 58, 1972–1982.

57 Alfonso-Prieto M, Borovik A, Carpena X, Murshudov

G, Melik-Adamyan W, Fita I, Rovira C & Loewen PC

(2007) The structures and electronic configuration of

compound I intermediates of Helicobacter pylori and

Penicillium vitale catalases determined by X-ray

crystallography and QM/MM density functional theory

calculations. J Am Chem Soc 129, 4193–4205.

58 Painter J & Merritt EA (2006) Optimal description of a

protein structure in terms of multiple groups

undergoing TLS motion. Acta Crystallogr D Biol

Crystallogr 62, 439–450.

59 Adams PD, Afonine PV, Bunkoczi G, Chen VB, Davis

IW, Echols N, Headd JJ, Hung LW, Kapral GJ,

Grosse-Kunstleve RW et al. (2010) PHENIX: a

comprehensive Python-based system for

macromolecular structure solution. Acta Crystallogr D

Biol Crystallogr 66, 213–221.

60 Emsley P & Cowtan K (2004) Coot: model-building

tools for molecular graphics. Acta Crystallogr D Biol

Crystallogr 60, 2126–2132.

61 Velankar S, Alhroub Y, Alili A, Best C, Boutselakis

HC, Caboche S, Conroy MJ, Dana JM, van Ginkel G,

Golovin A et al. (2011) PDBe: Protein Data Bank in

Europe. Nucleic Acids Res 39, D402–D410.

62 Chovancova E, Pavelka A, Benes P, Strnad O,

Brezovsky J, Kozlikova B, Gora A, Sustr V, Klvana

M, Medek P et al. (2012) CAVER 3.0: a tool for the

analysis of transport pathways in dynamic protein

structures. PLoS Comput Biol 8, e1002708.

63 Petrek M, Otyepka M, Banas P, Kosinova P, Koca J &

Damborsky J (2006) CAVER: a new tool to explore

routes from protein clefts, pockets and cavities. BMC

Bioinformatics 7, 316.

64 Petrek M, Kosinova P, Koca J & Otyepka M (2007)

MOLE: a Voronoi diagram-based explorer of molecular

channels, pores, and tunnels. Structure 15, 1357–1363.

65 Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F

& Higgins DG (1997) The CLUSTAL_X windows

interface: flexible strategies for multiple sequence

alignment aided by quality analysis tools. Nucleic Acids

Res 25, 4876–4882.

66 Nicholas KB, Nicholas HB Jr & Deerfield DWII (1997)

GeneDoc: Analysis and Visualization of Genetic

Variation in EMBNEW.NEWS 4 14.

67 DeLano WL (2002) The PyMOL Molecular Graphics

System. http://www.pymol.org.

68 The PyMOL Molecular Graphics System, Version

1.5.0.4 Schr€odinger, LLC.

69 Karplus PA & Diederichs K (2012) Linking

crystallographic model and data quality. Science 336,

1030–1033.

70 Fita I & Rossmann MG (1985) The NADPH binding

site on beef liver catalase. Proc Natl Acad Sci USA 82,

1604–1608.

71 Gouet P, Jouve HM & Dideberg O (1995) Crystal

structure of Proteus mirabilis PR catalase with and

without bound NADPH. J Mol Biol 249, 933–954.

72 Hakansson KO, Brugna M & Tasse L (2004) The

three-dimensional structure of catalase from

Enterococcus faecalis. Acta Crystallogr D Biol

Crystallogr 60, 1374–1380.

73 Riise EK, Lorentzen MS, Helland R, Smalas AO,

Leiros HK & Willassen NP (2007) The first structure of

a cold-active catalase from Vibrio salmonicida at 1.96

A reveals structural aspects of cold adaptation. Acta

Crystallogr D Biol Crystallogr 63, 135–148.

74 Pena-Soler E, Vega MC, Wilmanns M & Williams C

(2011) Structural features of peroxisomal catalase from

the yeast Hansenula polymorpha. Acta Crystallogr D

Biol Crystallogr 67, 690–698.

4150 FEBS Journal 281 (2014) 4138–4150 ª 2014 FEBS

DR1998 catalase X-ray structure P. T. Borges et al.