Functional and mechanistic studies on
catalase-peroxidase
Doctoral Thesis
Jutta Vlasits
(November 2009)
Department of Chemistry, Division of Biochemistry
University of Natural Resources and Applied Life Sciences
Muthgasse 18, 1190 Vienna
meinen Eltern
allen ehemaligen und zukünftigen
DissertantenInnen
Table of Contents
Abstract
3
Introduction
7
Bifunctional catalase-peroxidases
Chapter one
24
Hydrogen peroxide oxidation by catalase-peroxidase follows
a non-scrambling mechanism
Chapter two
30
Trapping the redox intermediates in the catalase cycle of
catalase-peroxidases from Synechocystis PCC 6803,
Burkholderia pseudomallei, and Mycobacterium tuberculosis
Chapter three
42
Probing hydrogen peroxide oxidation kinetics of wild-type
Synechocystis catalase-peroxidase (KatG) and selected variants
Chapter four
63
Disrupting the H-bond network in the main access channel of
catalase-peroxidase: effect on the redox thermodynamics
of the Fe(III)-Fe(II) couple
Chapter five
96
Putative substrate binding site(s) in bifunctional catalase-peroxidases
1
Appendix one
114
Intracellular catalase/peroxidase from the phytopathogenic fungus
Magnaporthe grisea: expression analysis and biochemical
characterization of the recombinant protein
Appendix two
130
Protein- and solvent-based contributions to the
redox thermodynamics of lactoperoxidase and eosinophil peroxidase
2
Abstract
Catalase-peroxidases (KatGs) are unique bifunctional enzymes with predominant catalatic
and substantial peroxidatic activity found in Bacteria, Archaea and Eukarya. Together with
ascorbate peroxidases (APX) and cytochrome c peroxidases (CcP) they belong to class I of the
superfamily of plant, bacterial, and fungal heme-containing peroxidases (EC 1.11.1.7).
Despite their striking sequence and structural similarities to APX and CcP KatGs are the only
heme peroxidases capable of both reduction and efficient oxidation of hydrogen peroxide with
rates similar to typical (monofunctional) catalases. The present work clearly demonstrated that
H2O2 dismutation to water and molecular oxygen in KatG follows a non-scrambling
mechanism with the two oxygen atoms in the evolving O2 coming from the same H2O2
molecule. Gas chromatography–mass spectrometry analysis using 16O- and 18O-labeled H2O2
showed the formation of
18
O2 (m/e = 36) and
16
O2 (m/e = 32) but not
16
O18O (m/e = 34)
indicating that in the two-electron oxidation step of H-O-O-H the O–O bond is not cleaved.
Apart from catalyzing the same overall dismutation reaction (2 H2O2  2 O2 + 2 H2O) as
monofunctional catalases, catalase-peroxidases follow an alternative catalatic mechanism.
Distinct differences between bovine liver catalase and KatG regarding spectral features of
redox intermediates could be monitored using multi-mixing stopped-flow UV-Vis
spectroscopy. Depending on pH two different low-spin intermediates that dominate during
H2O2 degradation could be trapped. Recent rapid-freeze-quench electronic spin resonance
spectroscopy demonstrated the formation of a protein-based radical on the KatG-typical distal
adduct that persists during peroxide turnover. This peculiar post-translational modification in
the active site of KatG - consisting of covalently linked Trp122, Tyr249 and Met275
(Synechocystis numbering) - rapidly reduces the porphyryl cation radical usually found in
redox intermediates (compound I) of heme peroxidases. In the presence of millimolar
hydrogen peroxide KatG is oxidized to compound I that rapidly converts to oxoferrous heme
(compound III) that co-exists with the radical formed at the covalent adduct. The latter
guarantees both the release of O2 as well as rapid turnover of compound III to ferric KatG. In
(monofunctional) peroxidases, however, the decay of compound III is extremely slow and
superoxide is released instead of O2. The importance of the Trp122-Tyr249-Met275 adduct in
3
the bifunctional activity was reflected by the fact that mutation of either Trp122 or Tyr249
completely converted a bifunctional KatG to a monofunctional peroxidase.
In order to follow the H2O2 oxidation reaction by both KatG and monofunctional catalases a
new method was developed that avoids enzyme cycling after substrate addition, which is a
general phenomenon in dismutating oxidoreductases. In the sequential stopped-flow study the
heme proteins were oxidized to compound I by peroxoacetic acid and finally mixed with H2O2
in the presence of cyanide. The latter inhibited enzyme cycling by trapping the proteins in
their low-spin complexes with KD values in the micromolar region. This, for the first time
enabled calculation of apparent bimolecular H2O2 oxidation rates of both KatG and a
monofunctional catalase. In case of KatG the rates most probably reflect the transition of
oxyferrous KatG to the ferric form. This study underlined the proposed catalatic mechanism
and clearly showed that the transition of compound III to native KatG is more than 3 orders of
magnitude faster than in monofunctional peroxidases.
Another structural feature, apart from the covalent adduct, that enables KatG to be
catalatically active, is the long and narrow substrate channel with an extensive and rigid Hbonding network. Interestingly, similar to manipulation of the distal adduct, mutations of
residues involved in this H-bonding network decrease the catalase activity whereas the
peroxidase activity remains unaffected or is even enhanced. Important residues for
maintenance of this ordered matrix of water molecules were shown to be the catalytic residues
His123 and Arg119, as well as Asp152 (controls channel entrance to heme cavity) and Glu253
(entrance of main substrate channel). This was demonstrated by kinetic analysis in
combination with temperature dependent spectroelectrochemical investigations. The latter
study gave a valuable inside in the redox thermodynamics of KatG and underlined the
importance of the channel architecture for KatG catalysis.
Finally, based on structural analysis and multiple sequence alignment, a putative peroxidase
substrate binding and oxidation site was analysed by site-directed mutagenesis and kinetic
investigations. Various mutants were designed and probed for oxidation of the artificial
substrates guaiacol or ABTS that are known to be oxidized by KatG but are too large for
entering the main access channel. Unfortunately, manipulation of this putative binding site did
not significantly alter the bifunctional activity of KatG. Thus, both the nature of the
endogenous peroxidase substrate as well as its binding site remain unknown.
4
Abstract
Katalase-Peroxidasen (KatGs) sind bifunktionale Oxidoreduktasen (EC 1.11.1.7) mit
vorwiegend katalatischer, aber auch beachtlicher Peroxidaseaktivität. Sie gehören
gemeinsam mit Ascorbat-Peroxidasen (APX) und Cytochrom c Peroxidasen (CcP) zur
Klasse I der Superfamilie Häm-hältiger Peroxidasen aus Pflanzen, Pilzen und Bakterien.
Trotz ihrer Homologie und strukturellen Ähnlichkeit mit APX und CcP, sind KatGs die
einzigen Peroxidasen mit nennenswerter Katalaseaktivität und daher ideale Modellenzyme
für das Studium der enzymkatalysierten Wasserstoffperoxid-Dismutation.
In
dieser
Arbeit
konnte
mittels
Gaschromatographie
in
Kombination
Massenspektrometrie gezeigt werden, dass im Zuge der Dismutation von
markiertem H2O2
18
O2 (m/e = 36) und
16
O2 (m/e = 32), aber kein
16
16
O- und
mit
18
O-
O18O (m/e = 34)
gebildet wird. Dies zeigt, dass bei der Oxidation des zweiten Peroxidmoleküls die O-O
Bindung intakt bleibt. Weiters konnte durch Stopped-flow Spektroskopie in Kombination
mit Elektronenspinresonanz-Spektroskopie die elektronische Natur der im Katalasezyklus
involvierten
Redoxintermediate
geklärt
und
somit
erstmals
ein
plausibler
Reaktionsmechanismus für die Gesamtreaktion (2 H2O2  2 O2 + 2 H2O) postuliert
werden, der dem KatG-spezifischen distalen kovalenten Addukt Trp122-Tyr249-Met275
die zentrale Rolle zuweist. Ähnlich wie in monofunktionalen Peroxidasen durchläuft auch
KatG in Gegenwart millimolarer Konzentration die Redoxintermediate Compound I und
Compound III. Im Zuge dieser Redoxkaskade wird jedoch in KatG rasch ein
Oxidationsäquivalent vom Porphyrinring zum Addukt verschoben und dies garantiert eine
effiziente und rasche Freisetzung von O2.
Die Kinetik der H2O2 Oxidationsreaktion konnte erstmals durch eine neue Methode
analysiert werden. Katalase-Peroxidase wurde mittels Peressigsäure zu Compound I
oxidiert und dieses Redoxintermediat im sequenziellen Stopped-flow Modus mit H2O2 in
Gegenwart von Cyanid inkubiert. Dadurch kommt es zur Ausbildung eines stabilen lowspin Komplexes der gebildeten Fe(III) Form von KatG und die Folgereaktion im
Katalasezyklus wird unterdrückt. Dadurch konnten erstmals sowohl für KatG als auch für
eine monofunktionale Katalase die H2O2-Oxidationsraten bestimmt werden. Zudem wurde
die Bedeutung des Adduktradikals dadurch unterstrichen, dass in monofunktionalen
Peroxidasen zwar dieselbe Reaktionsfolge beschritten wird, jedoch durch das Fehlen der
KatG-spezifischen post-translationalen Modifikation der Übergang Compound III zum
5
nativen Enzym um mehr als drei Zehnerpotenzen langsamer abläuft, abgesehen davon dass
Superoxid und nicht molekularer Sauerstoff freigesetzt wird.
Weiters konnte gezeigt werden, dass – neben dem distalen kovalenten Addukt – die
Architektur des Substratkanals für die effiziente H2O2 Oxidation entscheidend ist. Im
Gegensatz zu monofunktionalen Peroxidasen ist der Substratkanal in KatG eng und lang
und zeichnet sich durch ein ausgedehntes und rigides H-Brückennetzwerk aus. Mittels
ortsspezifischer
Mutagenese,
kinetischer
Analyse
und
Temperatur-abhängiger
spektroelektrochemischer Untersuchungen, konnte klar gezeigt werden, dass nur die
Oxidation aber nicht die Reduktion von H2O2 bei Manipulation dieses Substratkanals in
Mitleidenschaft gezogen wird. Wichtig für die Stabilisierung dieser geordneten
Wassermatrix sind die katalytischen Aminosäuren His123 und Arg119, als auch Asp152
(reguliert
den
Eintritt
des
Kanals
in
das
aktive
Zentrum)
und
Glu253
(Substratkanaleingang). Neben Trp122-Tyr249-Met275 (Addukt) sind diese Aminosäuren
in prokaryotischen und eukaryotisch KatGs hoch konserviert.
Basierend auf Strukturanalyse in Kombination mit multipler Sequenzanalyse wurde
weiters eine mögliche Peroxidasesubstratbindungs- und -oxidationsstelle mittels
Mutagenese
in
Kombination
mit
kinetischen
Methoden
untersucht.
Als
Peroxidasesubstrate wurden die artifiziellen Einelektronendonoren Guaiacol oder ABTS
herangezogen, da sie für den Substratkanal zu groß sind, aber dennoch von KatG oxidiert
werden. Zahlreiche Mutanten wurden rekombinant hergestellt, spektroskopisch analysiert
und hinsichtlich ihrer bifunktionellen Aktivität getestet. Die beobachteten Effekte waren
jedoch nicht signifikant unterschiedlich zum Wildtyp Protein. Somit bleiben sowohl die
chemische Natur als auch die Peroxidasesubstratbindungsstelle in KatG weiterhin
unbekannt.
6
introduction
Bifunctional catalase-peroxidases
Bifunctional catalase-peroxidases
Introduction
Hydrogen peroxide is one of the most frequently occurring reactive oxygen species
in the biosphere. It emerges either in the environment or as a by-product of aerobic
metabolism, i.e. by oxygen activation (e.g. superoxide formation and dismutation) in
respiratory and photosynthetic electron transport chains and as product of enzymatic
activity, mainly by oxidases. Both, excessive hydrogen peroxide as well as its
decomposition product hydroxyl radical that is formed in a Fenton-type reaction, are
harmful for almost all cell components. Thus, its rapid and efficient removal is of essential
importance for all aerobically living prokaryotic1 and eukaryotic cells2. On the other hand,
hydrogen peroxide can act as a second messenger in signal transduction pathways, mainly
for immune cell activation, inflammation, cellular proliferation and apoptosis3-5. There is
evidence that this signaling function of hydrogen peroxide was acquired rather late in
evolution. In unicellular organisms H2O2 mainly stimulates production of antioxidants and
ROS-removing and repairing enzymes, whereas in multicellular organisms (both animals
and plants) it is involved in activation of signaling pathways 5,6.
Hydroperoxidases (catalases and peroxidases) are ubiquitous “housekeeping”
oxidoreductases capable of the heterolytic cleavage of the peroxidic bond predominantly
in hydrogen peroxide (H-O-O-H) but also in some small organic peroxides (R-O-O-H. e.g.
ethyl hydroperoxide or acetyl hydroperoxide). Especially catalases have the additional
striking ability to evolve molecular oxygen (O2) by oxidation of hydrogen peroxide. Thus,
enzymes with catalase activity degrade hydrogen peroxide by dismutation.
Several gene families evolved in the ancestral genomes capable of H2O2
dismutation. The most abundant are heme-containing enzymes that are spread among
Bacteria, Archaea and Eukarya. They are divided in two main groups, namely typical or
“monofunctional” catalases (E.C. 1.11.1.6, hydrogen peroxide – hydrogen peroxide
oxidoreductase) and catalase-peroxidases. Both types of heme enzymes exhibit high
catalase activities, but have significant differences including absence of any sequence
similarity and very different active site, tertiary and quaternary structures. Enzymatic
classification of bifunctional catalase-peroxidases is not clear since besides their catalatic
activity (E.C. 1.11.1.6) they exhibit a peroxidase activity similar to conventional
peroxidases (EC 1.11.1.7, hydrogen peroxide – donor oxidoreductase). Both protein
8
families are present in prokaryotic and eukaryotic genomes. Non-heme manganesecontaining catalases (Mn-catalases) constitute a third (minor) group of catalatically active
enzymes. Mn-catalases (E.C. 1.11.1.6), initially referred to as pseudo-catalases, are
present only in bacteria. The so far known sequences of all enzymes from these three
protein families are collected and annotated in PeroxiBase7 (http://peroxidase.isb-sib.ch).
Catalase versus peroxidase activity
Continuous and effective degradation of H2O2 is indispensable for aerobic life8.
This is essential for both removal of excessive H2O2 and for strict regulation of its
concentration in signaling pathways5. As mentioned above nature has developed three
protein families that can perform its dismutation at reasonable rates. The overall catalatic
reaction includes the degradation of two molecules of hydrogen peroxide to water and
molecular oxygen (Reaction 1).
2 H2O2 → 2 H2O + O2
Reaction 1
Reaction 1 is the main reaction catalyzed by typical catalases, catalase-peroxidases and
Mn-containing catalases. Additionally, several heme-containing proteins, including most
peroxidases, metmyoglobin, and methemoglobin have been observed to exhibit a low level
of catalatic activity9. From these enzymes multifunctional chloroperoxidase from the
ascomycete Caldariomyces fumago has the greatest reactivity as a catalase10. Due to its
very limited distribution in nature this protein will not be discussed here in more detail.
In catalases and catalase-peroxidases two distinct stages can be distinguished in the
catalatic reaction pathway. The first stage involves oxidation of the heme iron using
hydrogen peroxide as substrate to form compound I (Reaction 2)11. The oxygen-oxygen
bond in peroxides (R-O-O-H) is cleaved heterolytically with one oxygen leaving as water
and the other remaining at heme iron (Reaction 2). Generally, compound I is a redox
intermediate two oxidizing equivalents above the resting [i.e. ferric, Fe(III)] state of the
enzyme. In monofunctional catalases, compound I is an oxoiron(IV) porphyrin -cation
radical species [+Por Fe(IV)=O], which is reduced back to the ferric enzyme by a second
molecule of hydrogen peroxide with the release of molecular oxygen and water (Reaction
3). Thus, in a catalase cycle H2O2 acts as oxidant (Reaction 2) and reductant (Reaction 3).
Compound I can also undergo an intramolecular one-electron reduction with or without a
proton resulting in the formation of an alternative compound I (Reaction 4) that is
9
catalatically inactive. Here, the protein moiety (AA) donates the electron that quenches
the porphyryl radical. Reaction 4 is responsible for the decrease of catalase activity with
time since the formed intermediate returns to the resting enzyme very slowly. A role of
NADPH, which binds to some clades of monofunctional catalases, as two-electron donor
for the transition of +AA Por Fe(IV)-OH back to ferric catalase has been discussed.
Por Fe(III) + H2O2  +Por Fe(IV)=O + H2O
+
Reaction 2
Por Fe(IV)=O + H2O2  Por Fe(III) + H2O + O2
Reaction 3
AA +Por Fe(IV)=O + H+  +AA Por Fe(IV)-OH
Reaction 4
A lively debate exists about the question whether catalase-peroxidases also follow
Reactions 2 & 3. As in monofunctional catalases, organic peroxides (e.g. acetyl
hydroperoxide) can oxidize catalase-peroxidases to a compound I-like species according
to Reaction 2, which is rapidly transformed to a protein radical species12,13 with spectral
UV-Vis signatures dissimilar to those described in monofunctional catalases14. Thus, in
the light of these data a number of alternative reaction schemes for catalase-peroxidases
have been proposed, that consider a rapid quenching of the porphyryl radical by an
electron from the protein moiety. These alternative compound I species have been
suggested to be +MYW Fe(IV)-OH or more generally +AA Fe(IV)-OH, with MYW being
the catalase-peroxidase typical distal side covalent adduct (see below) and AA being an
alternative amino acid near the heme. In any case, also in catalase-peroxidases the
catalatic cycle has two distinct phases with H2O2 acting as oxidant and reductant. But in
contrast to monofunctional catalases this unique compound I species must be able to
oxidize H2O2, i.e. it participates in the catalase cycle.
The mechanism of H2O2 degradation by non-heme Mn-catalases follows a
completely different scheme. The overall reaction (Reaction 1) also holds for these
binuclear metalloproteins and again H2O2 acts as oxidant and reductant. But the
mechanism of the two individual reaction steps is completely different.
Catalases and catalase-peroxidases can also catalyze a peroxidatic reaction
(Reaction 5) in which electron donors (AH2) are oxidized via one-electron transfers
releasing radicals (AH). Thus, a peroxidase cycle includes three reactions, i.e. compound
I formation (Reaction 2), compound I reduction to compound II, an oxoiron(IV) species
[PorFe(IV)=O] (Reaction 6) and compound II reduction back to the resting state (Reaction
7). Generally, the peroxidatic reaction of monofunctional catalases is weak but in
10
bifunctional catalase-peroxidases Reaction 5 is important. Similar to the discussion of the
catalase cycle there is also a dispute about the electronic structure of redox intermediates
of catalase-peroxidases in the peroxidase cycle15.
H2O2 + 2 AH2  2 H2O + 2 AH
Reaction 5
Por+Fe(IV)=O + AH2  Por Fe(IV)=O + AH
Reaction 6
Por Fe(IV)=O + AH2  Por Fe(III) + H2O + AH
Reaction 7
Peroxidatic reactions can also reduce compound I directly in a two-electron
reaction back to the ferric enzyme (Reaction 8). Especially mammalian and yeast catalases
have been described to oxidize ethanol or other short-chain aliphatic alcohols to
acetaldehyd or corresponding aldehydes8,16.
Por+Fe(IV)=O + CH3CH2OH  Por Fe(III) + CH3CHO + H2O
Reaction 8
Distribution and phylogeny of catalase-peroxidases
Bifunctional catalase-peroxidase (KatG) has raised considerable interest, since it
represents the only peroxidase with a reasonable high catalatic activity around neutral pH
(Reaction 1) besides a usual peroxidase activity (Reaction 6). Its overall fold and active
site architecture is typical peroxidase-like, which is underlined by their striking sequence
homologies to other members of class I of the plant, fungal and (archae)bacterial heme
peroxidase superfamily17. Together with these class I peroxidases (i.e. ascorbate
peroxidase and cytochrome c peroxidase) KatGs have the Pfam accession number
PF00141. The InterPro accession number IPR000763 is more specific only for catalaseperoxidase.
PeroxiBase currently (August 2008) contains 329 sequences of katG genes
and their evolution was analyzed recently18. The most important output of this
investigation is that katG genes are distributed in about 40% of bacterial genomes and
sometimes even closely related species differ in possessing katG genes of different origin
or even do not possess any katG gene. Two different evolutionary lines of katG genes also
exist in eukaryotes: one in algae and one in fungi and both are expected to originate from
lateral gene transfer of corresponding bacterial genomes living in close relationship with
the eukaryotes18,19. Phylogenetic analysis reveals that the closest neighbour of all fungal
katG genes is the katG gene of Flavobacterium johnsonii. In fungal KatGs two clades are
11
found, one encoding cytosolic and the other secreted enzymes. The occurrence of
extracellular fungal KatGs is underlined by the prediction of signal sequences.
Structure(s) and reaction mechanisms of catalase-peroxidases
The four available crystal structures of the KatGs Haloarcula marismortui
20
(1ITK) , Burkholderia pseudomallei (1MWV)21, Mycobacterium tuberculosis (1SJ2)22
and Synechococcus PCC 7942 (1UB2)23 revealed that the homodimeric heme b enzymes
have proximal and distal conserved amino acids at almost identical positions as in other
class I peroxidases (Figure 1). In particular both the triads His/Trp/Asp (His279, Trp330
and Asp389; Burkholderia numbering) and His/Arg/Trp (His112, Arg108, Trp111) are
conserved (Figure 2). Moreover, the proximal His is hydrogen-bonded to the carboxylate
side chain of the nearby Asp residue which, in turn, is hydrogen-bonded to the nitrogen
atom of the indole group of the nearby Trp residue (Figure 1). However, the X-ray
structures also revealed features unique to KatG. In the vicinity of the active site, novel
covalent bonds are formed among the side chains of three distal residues including the
conserved Trp111 (Figures 1 and 2). In particular, both X-ray crystallization data (Table
1) and mass spectrometric analysis24,25 have confirmed the existence of a covalent adduct
between Trp111, Tyr238 and Met264 (Figure 1). Exchange of either Trp111 or Tyr238
prevents crosslinking, whereas exchange of Met264 still allowed the autocatalytic
covalent bond formation between Trp111 and Tyr23825,26.
In addition to an extra C-terminal copy resulting from gene duplication17 that lacks
the prosthetic group but supports the architecture of the active site27, other KatG-typical
features are three large loops, two of them show highly conserved sequence patterns28 and
constrict the access channel of H2O2 to the prosthetic heme b group at the distal side. The
channel is characterized by a pronounced funnel shape and a continuum of water
molecules. At the narrowest part of the channel, which is similar but longer and more
restricted than that in other monofunctional peroxidases, two highly conserved residues,
namely Asp141 (Figures 1 & 2) and Ser324 control the access to distal heme side.
Together with a conserved glutamate residue at the entrance both acidic residues seem to
be critical for stabilizing the solute matrix and orienting the water dipoles in the channel.
Exchange of both residues affects the catalase but not the peroxidase activity. It is
interesting to note that in typical catalases similar acidic residues participate in
stabilization and orientation of the solute matrix in the main access channel for H2O2.
12
His112
(H123)
Asn142
(N153)
Met264
(M275)
Trp111
(W122)
Asp141
(D152)
Arg108
(R119)
Tyr238
(Y249)
His279
(H290)
Trp330
(W341)
Asp389
(D402)
Figure 1. Detailed view of active site residues of catalase-peroxidase from Burkholderia pseudomallei
(Synechocystis numbers in parentheses) with the distal active residues Met264, Tyr238 and Try111, which
are covalently linked, and the catalytically active His112, Arg108 and Asp141 that contribute to the
stabilization of the H-bonding network of the access channel. On the proximal side the conserved amino
acid triad His279, Trp330, and Asp389 is depicted. The Figure was constructed using the coordinates
deposited in the Protein Data Bank (accession code 1MWV).
13
(A) 2714SspCP1
2479SYspCP
3590BBFL71
3610BBFL72
2678FtCP1
3651FthCP1
3571MtuCP1
2303BpCP1
3313BpCP1b
2290BfCP1
3617FjCP1
5229FoCP2
5228FoCP1
2338GzCP2
5374NhaeCP
2337MagCP2
5459HjCP2
5353FoCP3
3415GmoCP1
2477GzCP1
5354GzCP3
5797TatCP1
5799TviCP1
2539HjCP1
1881AfumCP
3412NfCP1
3409AclCP1
3394AflCP1
3448AorCP1
5224AnCP1
3413AteCP1
2182PmCP1
5796TstCP1
1905AniCP
3408CgCP1
5373PanCP1
2288MagCP1
2181NcCP1
3449PnoCP1
5798PtritC
2538MgCP1
2327UmCP1
1960BgCP
(B) 2714SspCP1
2479SYspCP
3590BBFL71
3610BBFL72
2678FtCP1
3651FthCP1
3571MtuCP1
2303BpCP1
3313BpCP1b
2290BfCP1
3617FjCP1
5229FoCP2
5228FoCP1
2338GzCP2
5374NhaeCP
2337MagCP2
5459HjCP2
5353FoCP3
3415GmoCP1
2477GzCP1
5354GzCP3
5797TatCP1
5799TviCP1
2539HjCP1
1881AfumCP
3412NfCP1
3409AclCP1
3394AflCP1
3448AorCP1
5224AnCP1
3413AteCP1
2182PmCP1
5796TstCP1
1905AniCP
3408CgCP1
5373PanCP1
2288MagCP1
2181NcCP1
3449PnoCP1
5798PtritC
2538MgCP1
2327UmCP1
1960BgCP
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
-TGAVTASAHTGNRQWWPIQIDLGLLHQHHPASNPLGDTFDYPAAFAGLDLEALKADLAALMTDSQDWWPADWGHYGGLFIRMAWHSAGTYRGADGRGGA
-MHGANTTGQNGNLNWWPNALNLDILHQHDRKTNPMDDGFNYAEAFQQLDLAAVKQDLHHLMTDSQSWWPADWGHYGGLMIRMAWHAAGTYRIADGRGGA
---SMTKGEGKTNQQWYPNSLDLSVLRQNSDMSNPLGEEFDYKKEFESLDYDALKKDIELALTDSKDWWPADFGNYGGLFIRMAWHSAGTYRTGDGRGGT
---PNTAGRGTRNKDWWPNQLRLNVLRQNAPKSDPMDPDFNYEEAFKSLDMNALRKDLTALMTDSQDWWPADYGHYGGFFVRMAWHSAGTYRVGDGRGGS
---AIAEDTTTKNDNLSPQSVDLSPLRNLNKLDSPMDKDYNYHQAFKKLDTEQLKKDMQDLLTQSQDWWPADFGNYGPFFIRLSWHDAGTYRIYDGRGGA
---AIAEDTKTKNDNLSPQSVDLSPLRNLNKLDSPMDKDYNYHQAFKKLDTEQLKKDMQDLLTQSQDWWPADFGNYGPFFIRLSWHDAGTYRIYDGRGGA
HMKYPVEGGGNQ--DWWPNRLNLKVLHQNPAVADPMGAAFDYAAEVATIDVDALTRDIEEVMTTSQPWWPADYGHYGPLFIRMAWHAAGTYRIHDGRGGA
----QAAGNGTSNRDWWPNQLDLSILHRHSSLSDPMGKDFNYAQAFEKLDLAAVKRDLHALMTTSQDWWPADFGHYGGLFIRMAWHSAGTYRTADGRGGA
----QAAGNGTSNRDWWPNQLDLSILHRHSSLSDPMGKDFNYAQAFEKLDLAAVKRDLHALMTTSQDWWPADFGHYGGLFIRMAWHSAGTYRTADGRGGA
----HAAGGGTTNRDWWPKQLRLDLLSQHSGKSNPLDGGFNYAEAFRSLDLAAVKKDLAALMTDSQDWWPADFGHYGPLFIRMAWHSAGTYRIGDGRGGA
---DNQAASGTKNNDWWPKQLKVNILRQNSSLSNPLSKDFDYAEAFKTLDLEAVKKDLHVLMTDSQDWWPADFGHYGGLFIRMAWHSAGTYRVHDGRGGA
RVASNQAGGGTRSRDFWPCALRLDVLRQFSPQYNPLGADFDYTEAFKSLDFAALKKDLNALLTDSQDWWPADHGNYGGLFIRMSWHSAGTYRAMDGRGGS
RVASNQAGGGTRSRGFWPCALRLDVLRQFSPQYNPLGADFDYAEAFKSLDYESLKRDLKALLTDSQDWWPADHGSYGGLFIRMSWHSAGTYRAMDGRGGA
RIASNQAGGGTRSTDFWPCQLRLDVLRQFSPQFNPLGEDFDYAEAFKSLDFEALKKDIHALLTESQDWWPADFGHYGGLFIRMAWHSAGTYRAMDGRGGG
RVAGNAAGAGTRSRDFWPCQLRLDVLRQFSPQINPLGEEFDYVEAFKTLDYEALKKDLHSLITDSQDWWPADFGHYGGLFIRMAWHSAGTYRAIDGRGGG
AVKSNQAGGGTRSHDWWPCQLRLDVLRQFQPSQNPLGGDFDYAEAFQSLDYEAVKKDIAALMTESQDWWPADFGNYGGLFVRMAWHSAGTYRAMDGRGGG
PRKSTVAGGGTRSRQWWPCELNLGVLRQFSEKSNPLGSDFDYASAFGKIDVAQLKKDITAVQTTSQDWWPADFGDYGPFFIRLAWHNAGTYRAVDGRGGA
---PNVAGHGVTHQKWWPEALKTNILRQHSAVTNPYGENFNYAEAFNSIDYNELKEDLRKVCRDSQDWWPADFGHYGGLFVRMSWHAVGTYRVFDGRGGG
---PNVAGHGVTHQKWWPEALKTNILRQHSAVTNPYGENFNYAEAFNSIDYNELKEDLRKVCRDSQDWWPADFGHYGGLFVRMSWHAVGTYRVFDGRGGG
---PNVAGHGVTHQKWWPESLKTNILRQHSNVTNPYGENFSYAEAFKSLDYNALKEDLRKLMTDSQDWWPADFGHYGGLM----------------------PNVAGHGVTHQKWWPESLKTNILRQHSNVTNPYGENFSYAEAFKSLDYNALKEDLRKLMTDSQDWWPADFGHYGGLMVRLSWHSVGTYRVFDGRGGG
---INAAGGGTRNKDWWPESLKLNILRQHTNVTNPLGDDFDYAAAFKTLDYDAVKKDLTALMTDSQDFWPADFGHYGGLFVRLAWHSTGTYRVFDGRGGG
---VNVAGGGTRNRDWWPESLKLNILRQHTSVTNPLGADFDYAAAFKTLDYAAVKKDLTALMTDSQDWWPADFGHYGGLFIRLAWHSAGTYRVFDGRGGG
---INAAGHGTRNKDWWPESLKLNILRQHTDVTNPLGPDFDYPAAFKTLDYDAVKKDLTALMTDSQDWWPADFGHYGGLFIRLAWHSAGTYRVTDGRGGG
--QPNFAGGGTSNKDWWPDRLKLNILRQHTAVSNPLDADFDYAAAFNSLDYEGLKKDLRALMTDSQDWWPADFGHYGGLFIRMAWHSAGTYRVFDGRGGA
--QPNFAGGGTSNKDWWPDRLKLNILRQHTAVSNPLDADFDYAAAFKSLDYEALKKDLRALMTDSQDWWPADFGHYGGLFIRMAWHSAGTYRVFDGRGGA
--RSNVAGGGTRNTDWWPEQLKLNILRQHTTVTNPLGADFDYAAAFNTLDYDALKKDLTALMTDSQEWWPADFGHYGGLFIRMAWHSAGTYRVFDGRGGA
--KPNVAGSGTQNRDWWPDQLKLNILRQHTTVSNPLDPDFDYAAAFNSLDYYALKKDLQDLMTDSQDWWPADFGHYGGLFIRMAWHSAGTYRTFDGRGGG
--KPNVAGSGTQNRDWWPDQLKLNILRQHTTVSNPLDPDFDYAAAFNSLDYYALKKDLQDLMTDSQDWWPADFGHYGGLFIRMAWHSAGTYRTFDGRGGG
--NANVAGGGTRNTDWWPDQLNLGILRQHAPASNPFERDFDYTAAFNSLDYYALKKDLHALMTDSQDWWPADFGHYGGLFIRMAWHSAGTYRVFDGRGGG
---VNFAGGGTRNKDWWPDQLRLNILRQHTSASNPLDADFDYAAAFNSLDYNALKKDLEALMTDSQDWWPADFGHYGGLFIRMAWHSAGTYRVFDGRGGA
--KANVGGAGTSNQDWWPDRLKLNILRQNNPVSNPLGEEFDYAAAFNSLDYFALKKDIQDLMTDSQDWWPADFGHYGGLFIRMAWHSAGTYRVADGRGGG
--RSNAAGGGTSNRDWWPDRLKLNVLRQHNPVSNPLGEEFDYAAAFNSLDYFALKKDLHDLMTDSQDWWPADFGHYGGLFIRMAWHSAGTYRVFDGRGGG
--NANIGGGGQNNRDWWPDDLKLNILRQHNSVSNPLDKGFDYTAAFNSLDYFGLKRDLEALMTDSQDWWPADFGHYGGLFIRMAWHSAGTYRVFDGRGGG
---ANVAGGGTRNNDWWPNQLRLNILRQHQPASSPYSKEFDYAAAFKSLDYEALKKDITAVMTDSQDWWPADFGHYGGLFIRMAWHSAGTYRVFDGRGGG
--SANVAGGGTRNIDWWPNQLRLNILRQHTAASDPFHKEFNYAAAFKSLDYNALKKDLTDLMTNSQDWWPADFGHYGGLFIRMAWHSAGTYRVFDGRGGG
---ANVAGGGTRNRDWWPNTLKLNILRQHTEATNPYDPNFDYAEAFKSLDYEGLKKDLRALMTDSQEYWPADFGHYGGLFVRMAWHSAGTYRVMDGRGGG
---SNVGGGGTRNHDWWPAQLRLNILRQHTPVSNPLDKDFDYAAAFKSLDYEGLKKDLTKLMTDSQDWWPADFGHYGGLFIRMAWHSAGTYRVTDGRGGG
--TANVAGGGTKIKDWWPNELPVSVLRQHDPRQNPLTSDFNYAEEFKKLDYNALKKDLTALMTDSQDWWPADFGHYGGLLVRMAWHSAGTYRVTDGRGGG
--VPNVAGSGTRNTDWWPNELRTNILRQHDPRQNPLGPDFNYVEEFKKLDYDALKKDLNALMTDSQDWWPADFGHYGGLFIRMAWHSAGTYRVTDGRGGG
--LANVGGGGTRNRDWWPNELNTKILRQHTAATDPFGKQFDYPAAFKSLDYNGLKKDLNDLMTDSKDFWPADFGHYGGFFVRMAWHSAGTYRVADGRGGG
--RVPAAGFGTKNSDWWPNAVKLNVLRQHQAKSDPFNAEFDYAAAFNSLDYDALKKDLTHLMTDSQDWWPADYGHYGGFFIRMSWHAAGTYRVQDGRGGG
---ANVAGGGTHNKDWWPETLKLDALRQHTAESNPLGKDFDYASAFKTLDYEGLKKDLKDLMTDSQDWWPADFGHYGGFFVRMAWHSAGTYRSIDGRGGG
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
110
137
157
125
117
104
122
126
106
106
115
145
145
145
150
155
157
150
149
89
150
108
107
107
111
111
111
110
110
111
104
111
112
109
105
108
105
105
107
108
107
113
165
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
GHGNQRFAPLNSWPDNTNLDKARRLLWPLKRKYGNAISWADLIILSGNVALESMGFRTFGFAGGRTDIWQPEEDVFWGKETEWLS--DERHTTEG----ATGNQRFAPLNSWPDNVNLDKARRLLWPIKKKYGNKLSWGDLIILAGTMAYESMGLKVYGFAGGREDIWHPEKDIYWGAEKEWLASSDHRYGSEDRE--REGKQRFAPQNSWADNANLDKARRLLWPVKQKYGQKISWADLMILAGNVSFENMGFETLGYAGGREDTWEPQNNVYWGSESEML--GDE---RFNED--CSGSQRFAPLNSWPDNGNLDKARLLLWPIKQKYGKSISWADLMILTGNIAMESMGLKKLGFAGGREDIWEPEQDIYWGSETEWL--GND--ARYENG--NRGQQRFSPLNSWPDNVNLDKARQLLWPIKQKYGDAVSWSDLIVLAGTVSLESMGMKPIGFAFGREDDWQGDD-TNWGLSPEEIMSSN---VRDG----NRGQQRFSPLNSWPDNVNLDKARQLLWPIKQKYGDAVSWSDLIVLAGTVSLESMGMKPIGFAFGREDDWQGDD-TNWGLSPEEIMSSN---VRDG----GGGMQRFAPLNSWPDNASLDKARRLLWPVKKKYGKKLSWADLIVFAGNCALESMGFKTFGFGFGRVDQWEPDE-VYWGKEATWLGD-----ERYSGKRDGEGQQRFAPLNSWPDNANLDKARRLLWPIKQKYGRAISWADLLILTGNVALESMGFKTFGFAGGRADTWEPE-DVYWGSEKIWL---------------GEGQQRFAPLNSWPDNANLDKARRLLWPIKQKYGRAISWADLLILTGNVALESMGFKTFGFAGGRADTWEPE-DVYWGSEKIWL---------------GRGQQRFAPLNSWPDNVSLDKARRLLWPIKQKYGQKISWADLLILTGNVALETMGFKTFGFAGGREDTWEPDLDVYWGNEKTWLG--GD--VRYGKGAAG
GAGQQRFAPLNSWPDNVSLDKARRLLWPIKQKYGQKISWADLMILTGNVALESMGFKTFGFAGGRADVWEPDESVYWGSETTWLG--GD--ERYNNGSDG
GMGQQRFAPLDSWPDNQNLDKARRLLWPIKQKYGSKISWADLMVLAGNVALEHSGFETLGFAGGRADTWEADESIYWGAESTFVPKGND--VRYNGSTDI
GMLARQPEP------------------GQGSPSALKISWADMIVLAGNVALEHSGFETLGFAGGRADTWEADESIYWGAESTFVPKGND--VRYNGSTDI
GMGQQRFAPLNSWPDNQNLDKARRLIWPIKQKYGNKISWADLMLLCGNIALESSNFKTLGFAGGRPDTWQADESIYWGAETTFVPKGND--VRYNGSTDI
GMGQQRFAPLNSWPDNQNLDKARRLLWPIKQKYGNKISWADLILLAGNVALESMDFEPIGFGAGRADTWQSDESVYWGAETTFVPKGND--VRYNGSKDV
GMGQQRFAPLNSWPDNQNLDKARRLIWPIKQKYGNKISWADLMLLTGNVALENMGFKTLGFGGGRADTWQSDEAVYWGAETTFVPQGND--VRYNNSVDI
GQGQQRFAPLNSWPDNANLDKARRLLWPIKQKYGSALSWADLFVFAGNTAMENMGFPTYGFGFGREDTWQSDEGIYWGGEQEMFPEALNNVVRYNGSTDF
RQGQQRFAPLNSWPDNVSLDKARRLLWPVKQKYGNKVSWADLIVMAGNVALEDMGFKTIGFAAGRPDTWEADEATYYGGEDTWL--GND--VRYSDGHPG
RQGQQRFAPLNSWPDNVSLDKARRLLWPVKQKYGNKVSWADLIVMAGNVALEDMGFKTIGFAAGRPDTWEADEATYYGGEDTWL--GND--VRYSDGHPG
--GQQRFAPLNSWPDNVSLDKARRLLWPVKQKYGNKISWADLIVLAGNVALEDMGFKTLGFAGGREDTWEADEATYYGGEDTWL--GND--VRYSNGNKG
RQGQQRFAPLNSWPDNVSLDKARRLLWPVKQKYGNKISWADLIVLAGNVALEDMGFKTLGFAGGREDTWEADEATYYGGEDTWL--GND--VRYSNGNKG
AQGQQRFAPLNSWPDNVSLDKARRLLWPIKAKYGNKLSWADLIVLSGNVALESMGFKTIGFAGGRSDTWEADEAAYWGGETTWL--GND--VRYAHGFAG
GQGQQRFAPLNSWPDNVSLDKGRRLLWPIKQKYGNKLSWADLFILSGNVALESMGFKTLGFAGGRSDTWEADESAYWGGENTWL--GND--VRYAHGFAG
GQGQQRFAPLNSWPDNVSLDKARRLLWPIKQKYGNKLSWADLLILSGNVALESMGFKTLGFAGGRPDTWEADEAAYWGGETTWL--GND--VRYAHGFEG
GQGQQRFAPLNSWPDNVSLDKARRLLWPIKQKYGNKISWADLLILTGNVALESMGFKTFGFAGGRPDTWEADEATYWGRETTWL--GND--ARYAKGFSG
GQGQQRFAPLNSWPDNVSLDKARRLLWPIKQKYGNKISWADLMILTGNVALESMGFKTFGFAGGRPDTWEADEATYWGRETTWL--GND--ARYAKGFSG
GQGQQRFAPLNSWPDNVSLDKARRLLWPIKQKYGDKISWADLMILTGNVALESMGFKTFGFAGGRADTWEADESAYWGRETTWL--GND--ARYEKGFSG
GQGQQRFAPLNSWPDNVSLDKARRLLWPIKQKYGNKISWADLMILTGNVALESMGFKTFGFAGGRKDTWEADESVYWGGETTWL--GND--VRYSHGFAG
GQGQQRFAPLNSWPDNVSLDKARRLLWPIKQKYGNKISWADLMILTGNVALESMGFKTFGFAGGRKDTWEADESVYWGGETTWL--GND--VRYSHGFAG
GQGQQRFAPLNSWPDNASLDKARRLLWPIKQKYGAKISWADLMLLAGNVALESMGFKTYGFSGGRADTWEADESVYWGGESTWM--GND--VRYSDGFPG
GQGQQRFAPLNSWPDNASLDKARRLLWPIKQKYGNKISWADLMILAGNVALESMGFKPFGFSGGRADTWEADESVYWGGEKTWFPKGND--VRY-----GGGQQRFAPLNSWPDNVGLDKARRLLWPIKQKYGNKISWADLLLLTGNVALESMGFKTFGFSGGRADTWEVDESANWGGETTWL--GND--VRYSGG--GQGQQRFAPLNSWPDNVSLDKARRLLWPIKQKYGNKISWADLLLLAGNVALESMGFKTFGFAGGRADTWEVDESANWGGETTWL--GND--VRYSGGNAG
GQGQQRFAPLNSWPDNVSLDKARRLLWPIKQKYGSKISWADLLILAGNVALESMGFKTFGFAGGRSDTWEADQSVFWGGEKEWL--GND--VRYLNG--SQGQQRFAPLNSWPDNVSLDKARRLLWPVKQKYGDKISWADLLLLTGNVALESMGFKTFGFAGGRPDTWEADESAYWGGEKTWL--GND--VRYGQGNEG
GQGQQRFAPLNSWPDNVSLDKARRLLWPIKQKYGNKISWADLMLLTGNVALESMGFKTFGFAGGRPDTWEADESAYWGGETTWL--GNE--ARYAHGQEG
GQGQQRFAPLNSWPDNVSLDKARRLLWPIKQKYGNKISWADLMLLTGNVALEDMGFKTFGFAGGRPDTWEADESTYWGGETTWL--GNE--VRYSSGNEG
GEGQQRFAPLNSWPDNVSLDKARRLLWPIKQKYGNKISWSDLLLLTGNVALESMGFKTFGFAGGRPDTWEADESVYWGAETTWL--GNE--DRYSEGQEG
GDGQQRFAPLNAWPDNVSLDKARRLLWPIKQKYGNKISWADLMLLTGNVALESMGCETFGFAGGRPDTFQSDESIYWGGEDTWL--GND--VRYSNGNKG
GDGQQRFAPLNAWPDNVSLDKARRLLWPIKQKYGNKISWADLMLLTGNVALESMGCPTFGFAGGRADTFQSDESVYWGGETEWL--GSD--VRYADGAKG
GEGQQRFAPLNSWPDNANLDKARRLLWPIKQKYGNKISWADLFLLTGNVAIESMGLPTFGFAGGRADTWEADDSVYWG-ETTWL--GNE--VRYSDGKEG
GEGQQRFAPLNSWPDNGNLDKARRLLWPIKQKYGNKISWADLLLLAGNVALESMGFKTFGFAGGRADTWEADQSTYWGGETTWL--AND--VRYEEGTKGQGQHRFAPLNSWPDNGNLDKARRLLWPIKQKYGNKISWADLYLLTGNVAIESMGGKTFGFACGRPDTWEADDATFWGNETKWL--GND--ARYKNGSK-
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
203
234
249
218
208
195
215
209
189
202
211
243
225
243
248
253
257
246
245
183
246
204
203
203
207
207
207
206
206
207
196
204
208
202
201
204
201
201
203
204
202
208
260
Figure 2. Selected parts of the multiple sequence alignment of 43 catalase-peroxidases (KatGs). 32
complete fungal sequences are presented together with closely related bacterial counterparts. Similarity
scheme is identical with Figure 1. The substitution matrix and the alignment algorithm were the same as
used in recent experimental work on this topic18. (A) Distal side of the heme group with the catalytic triad.
(B) Distal side of heme with important hydrogen bond location. (C) Proximal side of the prosthetic heme
group with the essential histidine. (D) Proximal side of heme with important hydrogen bond location.
Catalytically important residues are marked with an arrow. Abbreviations for Linnean names and ID
numbers correspond to PeroxiBase nomenclature7 (see http://peroxibase.isb-sib.ch/ for all details).
14
(C) 2714SspCP1
2479SYspCP
3590BBFL71
3610BBFL72
2678FtCP1
3651FthCP1
3571MtuCP1
2303BpCP1
3313BpCP1b
2290BfCP1
3617FjCP1
5229FoCP2
5228FoCP1
2338GzCP2
5374NhaeCP
2337MagCP2
5459HjCP2
5353FoCP3
3415GmoCP1
2477GzCP1
5354GzCP3
5797TatCP1
5799TviCP1
2539HjCP1
1881AfumCP
3412NfCP1
3409AclCP1
3394AflCP1
3448AorCP1
5224AnCP1
3413AteCP1
2182PmCP1
5796TstCP1
1905AniCP
3408CgCP1
5373PanCP1
2288MagCP1
2181NcCP1
3449PnoCP1
5798PtritC
2538MgCP1
2327UmCP1
1960BgCP
(D) 2714SspCP1
2479SYspCP
3590BBFL71
3610BBFL72
2678FtCP1
3651FthCP1
3571MtuCP1
2303BpCP1
3313BpCP1b
2290BfCP1
3617FjCP1
5229FoCP2
5228FoCP1
2338GzCP2
5374NhaeCP
2337MagCP2
5459HjCP2
5353FoCP3
3415GmoCP1
2477GzCP1
5354GzCP3
5797TatCP1
5799TviCP1
2539HjCP1
1881AfumCP
3412NfCP1
3409AclCP1
3394AflCP1
3448AorCP1
5224AnCP1
3413AteCP1
2182PmCP1
5796TstCP1
1905AniCP
3408CgCP1
5373PanCP1
2288MagCP1
2181NcCP1
3449PnoCP1
5798PtritC
2538MgCP1
2327UmCP1
1960BgCP
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
-------------------------------ALDQPLAAVEMGLVYVNPEG-PHGHPDPVASGPDVRDTFARMGMTMEETVALVAGGHTFGKCHGAA---------------------------------SLENPLAAVQMGLIYVNPEG-VDGHPDPLCTAQDVRTTFARMAMNDEETVALTAGGHTVGKCHGNS--------------------------------RELETPLAASQMGLIYVNPEG-PNGNPDPVLAAHDIRQTFGRMGMNDEETVALIAGGHTLGKTHGAS---------------------------------ELEGMLGAAHMGLIYVNPEG-PNGKPDPVGSAHDIRETFGRMAMNDYETVALIAGGHTFGKAHGASD--------------------------------KLAPAYAATQMGLIYVNPEG-PDGKPDIKGAASEIRQAFRAMGMTDKETVALIAGGHTFGKTHGAVPED
-------------------------------KLAPAYAATQMGLIYVNPEG-PDGKPDIKGAASEIRQAFRAMGMTDKETVALIAGGHTFGKTHGAVPED
--------------------------------LENPLAAVQMGLIYVNPER-PNGNPDPMAAAVDIRETFRRMAMNDVETAALIVGGHTFGKTHGAG--------------ELSGGPNSRYSGD-----RQLENPLAAVQMGLIYVNPEG-PDGNPDPVAAARDIRDTFARMAMNDEETVALIAGGHTFGKTHGAG--------------ELSGGPNSRYSGD-----RQLENPLAAVQMGLIYVNPEG-PDGNPDPVAAARDIRDTFARMAMNDEETVALIAGGHTFGKTHGAG--DNDDGGVIVADEEKHGEEVSRDNSG-----RNLENPLGAVQMGLIYVNPEG-PDGNPDPLAAAHDIRETFARMAMNDEETVALIAGGHTFGKTHGAG------------VPKDHGVVSADDDADGKVHSRNLEKPLAAVQMGLIYVNPEG-PDGNPDPILAAKDIRDTFGRMAMNDEETVALIAGGHTFGKTHGAA--YERAD--------------------------KLEKPLGATHFGLIYVNPEG-PDGSSDPKASALDIRTAFGRMGMDDEETAALIIGGHTLGKTHGAV--YERAD--------------------------KLEKPLGATHFGLIYVNPEG-PDGSSDPKASALDIRTAFGRMGMDDEETAALIIGGHTLGKTHGAV--YERAD--------------------------KLEKPLGATHFGLIYVNPEG-PNGTPDPKASALDIREAFGRMGMDDEETAALIIGGHTLGKTHGAV--VDRAD--------------------------KLEKPLGAVSLGLIYVNPEG-PDANSNPADSALDIRVNFDRMGMNDEETVALIVGGHTFGKTHGAV--NARAD--------------------------KLEKPLAATHMGLIYVNPEG-PNGTPDPAASAKDIREAFGRMGMNDTETVALIAGGHAFGKTHGAV--TERAD--------------------------KLESPLGAVSMGLIYVDPRG-PNGQPDTRASALDIRETFGRMGMNDEETVALIAGGHAFGKTHGAV--TTKP------GATDSDQPPHKNIHT-----RELEKPLAAAHHGLIYVNPEG-PDGNPDPVAAARDIRETFGRMAMNDEETVALIAGGHTVGKTHGAG--TTKP------GATDSDQAPHKNIHT-----RELEKPLAAAHHGLIYVNPEG-PDGNPDPVAAARDIRETFGRMAMNDEETVALIAGGHTVGKTHGAG--TTKP------GATDSDQTEHKDIHN-----RDLEKPLAAAHHGLIYVNPEG-PDGNPDPVAAAHDIRETFGRMAMNDEETVALIAGGHTVGKTHGAG--TTKP------GATDSDQTEHKDIHN-----RDLEKPLAAAHHGLIYVNPEG-PDGNPDPVAAAHDIRETFGRMAMNDEETVALIAGGHTVGKTHGAG--ASKD-----HGVVDSDERKHTDVHS-----RELEAPLGAAHMGLIYVNPEG-PDGNADALAAARDIRITFARMAMNDEETVALIAGGHTFGKTHGAG--PTSDG----HAVLETDEKSHTDVHS-----RDLESPLAAAHMGLIYVNPEG-PDANADPISAARDIRITFARMAMNDEETVALIAGGHTFGKTHGAG--TKPGS----HGVVDSDEKRHTDIHS-----RELETPLAAAHMGLIYVNPEG-PDASGDPVAAARDIRVTFGRMAMNDEETVALIAGGHTFGKTHGAG--SDKRG----SLIADEES--HKTTHS-----RELETPLAAAHMGLIYVNPEG-PDGNPDPVAAAHDIRDTFGRMAMNDEETVALIAGGHTFGKTHGAA--SDKRG----TLIADEDS--HKTTHS-----RELETPLAAAHMGLIYVNPEG-PDGNPDPIAAAHDIRDTFGRMAMNDEETVALIAGGHTFGKTHGAA--SDKQG----TVIADEAS--HKTTHS-----RELENPLAAAHMGLIYVNPEG-PDGNPDPVAAAHDIRVTFGRMAMNDEETVALIAGGHTFGKTHGAA--SSKHG----AVIADEAS--HRDIHS-----RELEKPLAAAHMGLIYVNPEG-PDGNPDPVAAARDIRTTFARMAMNDEETVALIAGGHTFGKTHGAA--SSKHG----AVIADEAS--HRDIHS-----RELEKPLAAAHMGLIYVNPEG-PDGNPDPVAAARDIRTTFARMAMNDEETVALIAGGHTFGKTHGAA--VTKHG----ALSGDEPP--HRNIHT-----RDLEKPLAASHMGLIYVNPEG-PDGNPDPVAAARDIRTTFGRMGMNDEETVALIAGGHSFGKTHGAA------------------P--NDDIYT-----RDLENPLAASHMGLIYVNPEG-PNGNPDPKAAARDIRVTFGRMAMNDEETVALIAGGHSFGKTHGAS----------------KAD--HKDIHN-----RDLDKPLAAAHMGLIYVNPEG-PDGNPDPIAAAKDIRTTFGRMAMNDEETVALIAGGHTFGKTHGAG--IDAHG----VLSGQEAA--HKDIHN-----RDLDKPLAAAHMGLIYVNPEG-PDGTPDPVAAAKDIRVTFGRMAMNDEETVALIAGGHTFGKTHGAG---------------------------------ELDNPLAASHMGLIYVNPEG-PNKNPDPVLAAKDIRITFGRMAMNDEETVALIAGGHTFGKTHGAG--VAGQG----VVDGDESKKGHRDIHS-----RDLESPLAAAHMGLIYVNPEG-PDGIPDPVAAGRDIRTTFGRMAMNDEETVALIAGGHTFGKTHGAA--IAGKG----IVSGDESKKNHTDIHN-----RDLESPLAAAHMGLIYVNPEG-PDGNPDPVAAARDIRVTFGRMAMDDEETVALIAGGHTFGKTHGAA--HKESG----VIDGSESKKGHKDIHT-----RDLEKPVSAAHMGLIYVNPEG-PDGIPDPVAAARDIRTTFSRMAMNDEETVALIAGGHTVGKTHGAA--HEGHG----VVQGDESKKQHTDIHN-----RDLQSPLASSHMGLIYVNPEG-PDGIPDPVASAKDIRVTFGRMAMNDEETVALIAGGHSFGKTHGAG--VSGEG----VVDGDQHKMDHKDIHS-----RDLEQPVAAAHMGLIYVNPEGKPDGVPDPIAAARDIRTTFHRMAMNDEETAALIIGGHSFGKTHGAA--VKGEG----IMDGDQHKTDKSEPHTS----RNLEQPLAAAHMGLIYVNPEG-PEGIPDPVAAAHDIRTTFGRMAMNDAETAALIIGGHSFGKTHGAA--LTGDG----ILDGDQSKKQHTDIHG-----REMEEPIAAAHMGLIYVNPEG-PDGKPDPVASARDIRTTFGRMAMNDAETVALIAGGHAFGKTHGAG-------------NGGDIN----DLKN-----RNLDHALAASHMGLIYVNPEG-PNGEPDPVAAAHDIRTTFGRMAMNDEETVALIAGGHTFGKTHGAG--------------------DPKDIYT-----RQLEKPLSAVHMGLIYVNPEG-PDGIPDPVASARDIRTTFRRMAMNDEETVALIAGGHTFGKTHGAA---
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
268
299
315
284
276
263
279
288
268
293
297
313
295
313
318
323
327
331
330
268
331
290
290
290
292
292
292
291
291
292
269
279
293
267
288
291
288
288
291
292
289
284
333
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
TTADLSLRHDPLMEPVARRFHQDQDAFADAFARAWFKLTHRDLGPRALYLGADVPEEI-QIWQDPVPALDHPLIGAVEIKALKQKLL-ATGCSVGALVAT
TDADMAMIKDPIYRQISERFYREPDYFAEVFAKAWFKLTHRDLGPKSRYLGPDVPQED-LIWQDPIPPVDYTLS-EGEIKELEQQIL-ASGLTVSELVCT
LVTDLSLIKDPAYLKISKRFYENPEEFDAAFAKAWFKLTHRDMGPSSTYLGPEVPQEQ-FIWQDPIPARDYKLVSTSDINTLKGKIK-ASGLTTNELVTT
STADIALKTDPEYLKISQHFRENPDEFEDAFAKAWFKLTHRDMGPTTRYLGNDIPEKE-ELWQDPIPSVTHELINDEDVASLKSQIL-DSGLTIQELVSV
FTTDLALKEDDGFNKYTQEFYNNPEEFKEEFAKAWFKLTHRDMGPKSRYIGPWIPEQN-FIWQDPVPAADYKQVSTQDIAQLEQDII-NSGLTNQQLIKT
FTTDLALKEDDGFNKYTQEFYNNPEEFKEEFAKAWFKLTHRDMGPKSRYIGPWIPEQN-FIWQDPVPAADYKQVSTQDIAQLEQDII-NSGLTNQQLIKT
LATDLSLRVDPIYERITRRWLEHPEELADEFAKAWYKLIHRDMGPVARYLGPLVPKQT-LLWQDPVPAVSHDLVGEAEIASLKSQIR-ASGLTVSQLVST
LTTDLSLRFDPAYEKISRRFHENPEQFADAFARAWFKLTHRDMGPRARYLGPEVPAEV-LLWQDPIPAVDHPLIDAADAAELKAKVL-ASGLTVSQLVST
LTTDLSLRFDPAYEKISRRFHENPEQFADAFARAWFKLTHRDMGPRARYLGPEVPAEV-LLWQDPIPAVDHPLIDAADAAELKAKVL-ASGLTVSQLVST
LTTDLSLRFDPVYEKISRRFLENPDQLADAFARAWFKLTHRDMGPRARYLGPEVPAEE-LNWQDPIPALDHPLVNEQDIASLKQKIL-ASGLSVSQLVST
LTTDLSLRLDPEYEKISRRFLENPDQFADAFSRAWFKLTHRDMGPRARYLGPDVPQEV-LLWQDPIPEVNHKLIDENDIKQLKEKIL-NSGLSISQLVAA
LTSDLALINDPSYLKICKRWHDNPKELNAAFARAWYKLLHRDLGPVSRYLGPEVAKEK-FIWQDPLPERKGDIIGEADISSLKSAILSADGLDVSKLVST
LTSDLALINDPSYLKICKRWHDHPKEMNQAFARAWYKLLHRDLGPVSRYLGPEVPKEK-FIWQDPLPERKGDIISEGDISSLKSAILSTDGLTVSKLVST
LTSDLALINDDSYHKICKRWLKNPEEMNEAFAKAWYKLLHRDLGTASRYLGPDVPKEK-FIWQDPLPERKGDVITDEDISSLKTAILGADGLDVSKLVST
LTSDLALIHDPEYLKICKRWQKNPEELRKAFGRAWFKLLHRDLGPVSRYQGPEVPKEK-FTWQDPLPERKGDVIGDDDIATLKAELLAAKGLGVSKLVST
LTSDLALINDPEYLKISQRWLEHPEELADAFAKAWFKLLHRDLGPTTRYLGPEVPKES-FIWQDPLPAREGDLIDDADVDKLKAAILSTDGLDVSKLAST
LTSDLGLRDDPIYNNISRTFLNDFDYFTEKFGLAWYKLLHRDMGPIARYLGPEVPKKAPLIWQDPLPTPAYTTLAASDISSLKQQILAAPGLNVSNLVTT
LTTDLSLRFDPEYEKISRRFLENPDQFNEAFAKAWFKLTHRDMGPRDRYLGPEVPKEV-FLWQDPVPERDYKLADDGDISAIKNEIL-KSGIDVSKLVST
LTTDLSLRFDPEYEKISRRFLENPDEFNEVFAKAWFKLTHRDMGPRDRYLGPEVPKEV-FLWQDPVPERDYKLVDDGDISAIKNEIL-KSGVDVSKLVST
LTTDLSLRFDPAYEKISRRFLENPQEFNDAFAKAWFKLTHRDMGPRDRYLGPEVPKEI-FSWQDPVPERDYKVVDDGDISAIKNEIL-KSGIDVSKLVST
LTTDLSLRFDPAYEKISRRFLENPQEFNDAFAKAWFKLTHRDMGPRDRYLGPEVPKEI-FSWQDPVPERDYKVVDDGDISAIKNEIL-KSGIDVSKLVST
LTTDLSLRVDPAYEKISRRFLENPDQFAEVFAKAWFKLTHRDMGPSTRYVGPEIPKEQ-FSWQDPIPAVNYTLINDSDVAALKKEVL-ASGVTPSKLIST
LTTDLSLRFDPVYEKISRRFLEHPDQFAEAFAKAWFKLTHRDMGPLSRYVGPEIPKEV-FSWQDPIPAVDYTLVNDSDIATLKSQIL-ATGVSPSKLIST
LTTDLALRFDPAYEKISRRFLEHPDQFADAFAKAWFKLTHRDMGPLSRYVGPEVPKEE-FAWQDPIPPVNFALINESDIAALKKEIL-NSGVAPSKLIST
LTTDLSLRFDPAYEKIARRFLEHPDQFADAFARAWFKLTHRDMGPRARYLGPEVPSEV-LIWQDPIPAVNHPLVDASDIAALKDEIL-ASGVPPRSFIST
LTTDLSLRFDPAYEKIARRFLEHPDQFADAFARAWFKLTHRDMGPRARYLGPEVPSEV-LIWQDPIPAVNHPLVDASDIATLKDEIL-ASGVPFRSFIST
LTTDLSLRFDPAYEKISRRFLENPDQFADAFARAWFKLTHRDMGPRARYLGPEVPGEV-LLWQDPIPAVNHALIDTVDTAALKRDVL-ATGVNPSKFIST
LTTDLSLRFDPAYEKISRRFLENPDQFAEAFARAWFKLTHRDMGPRARYIGPEVPAEE-LSWQDPIPAVNHPVISETDIAALKRDIL-ATGVDPSKFIST
LTTDLSLRFDPAYEKISRRFLENPDQFAEAFARAWFKLTHRDMGPRARYIGPEVPAEE-LSWQDPIPAVNHPVISETDIAALKRDIL-ATGVDPSKFIST
LTTDLSLRYDPAYEKIARRFLDHPDEFADAFSRAWFKLLHRDMGPRTRYIGPEAPTED-LIWQDPIPAVNHTLVDANDIAALKRTIL-DTGLNKSNFVST
LTTDLALRYDPAYEKIARRFLENPDQFADAFARAWFKLTHRDMGPRARYVGPEVPSEV-LIWQDPIPAVNHPLVDASDIASLKQAIL-NSGVDRSKFVST
LTTDLSLRYDPIYEKISRRFLEHPDQFADAFARAWFKLLHRDLGPRALYIGPEVPAEV-LPWQDPVPAVDHPLISNEDASALKQRIL-ASGVKPSSLIST
LTTDLALRFDPVYEKISRRFLENPDQFADAFARAWFKLLHRDVGPRSLYLGPEVPSEV-LPWQDPVPAVNHPLIGNEDIAALKQHIL-GTGVNPSNFIST
LTTDLSLRYDPEYEKISRRFLENPDQFADAFARAWFKLTHRDVGPRVLYQGPEVPSEV-LIWQDPVPPLDHPVIDNDDIATLKKAIL-NSGISHTDLFST
LTTDLALRFDPEYEKISRRFLENPDQFADAFARAWFKLLHRDLGPRSRWLGPEIPSEV-LIWEDPVPAVNHPLVDEQDVTTLKRAIL-ATGVAPAKLIST
LTTDLSLRFDPGFEKISRRFLEHTDQFADAFARAWFKLLHRDMGPRSRWLGPEIPSEV-LLWEDPLPPLDHPVIDNNDIAAIKREIL-ATGLAPQKLIST
LTTDLSLRFDPEYEKISRRFLENPEQFKDAFARAWFKLLHRDMGPRSRWLGPEVPKET-LLWEDPIPTPDHPIIDGSDVDSLKKAIL-ATGVAPSKLIQT
LTTDIALRMDPAYDKICRDYLANPDKFADAFARAWFKLLHRDMGPRTRWIGPEVPSEI-LPWEDYIPPVDYQIIDDNDIAALKKEIL-ATGVAPKKLIFV
LTTDMSMRMDPGFEKITRRWLDHPQELHDAFIRAWFKLLHRDMGPRSRWVGPEIPKEV-LLWEDPVPEVDHALVEESDISALKKQIL-EAGIEPSKLIRT
LTTDLAMRMDPAYEKITRRWLEHPQELHDTFVRAWFKLLHRDMGPRSRWLGPEVPKEV-LIWEDPVPQPEGELIGESDIATLKKQVL-DAGVEPAKLIRT
LTSDLALRFDPEYEKISRDYLENPDKLADAFTRAWFKLLHRDMGPRSRWVGPEIPKEV-LIWEDPVPSVNHPLVDEKDVAALKKAII-ATGVEPTALISA
LTTDLSLRYDPAYEKISRRFLENHDEFADAFARAWFNCST------VTWV----PKEI-LIWEDPVPTADYALVDDRDLAGLKQAIF-ATGVEPSKFLAT
LTTDLALRHDKEYEKISLRFLENPDQFADAFARAWFKLLHRDMGPRSRWLGPEIPKEE-LIWEDPIPEIDHPIISQEDINNLKKEIL-SSGVGHNKLIQT
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
465
495
510
481
474
461
475
483
463
488
492
509
491
509
514
519
524
526
525
463
526
485
485
485
487
487
487
486
486
487
464
474
488
462
483
486
483
483
486
487
484
469
528
Figure 2. Continued. 15
This extensive distal site hydrogen-bonding network causes KatGs to differ from
typical peroxidases. Typically, the catalatic but not the peroxidatic activity is very
sensitive to mutations that disrupt this network29,30. Moreover, the integrity of this network
is crucial for the formation of distinct protein radicals that are formed upon incubation of
KatG with peroxides13,31. Mutational analysis clearly underlined the importance of the
KatG-typical covalent adduct Trp-Tyr-Met. Its disruption significantly decreased the
catalase but not the peroxidase activity. For instance, upon exchange of Tyr238 by Phe the
resulting KatG variant completely lost its capacity to oxidize H2O2 and was transformed to
a typical peroxidase31. Similarly all other mutations so far performed in the heme cavity or
substrate channel of a KatG affected the oxidation and not the reduction reaction of
hydrogen peroxide15, which is the initial reaction in all heme hydroperoxidases.
Both KM and kcat values of KatGs are significant lower compared to typical
catalases. Apparent values range from 3,500-6,000 s-1 and 3.7-8 mM15. In contrast to
typical catalases and Mn-catalases, the catalase activity of KatG has a sharp maximum
activity around pH 6.5.
KatG is a bifunctional enzyme. It oxidizes typical artificial peroxidase substrates
like o-dianisidine, guaiacol or ABTS. The pH profile of the peroxidase activity of KatG
has its maximum around pH 5.5 independent of the nature of most donors. Compound I
(formed with peroxoacetic acid) reduction by monosubstituted phenols and anilines has
been shown to depend on the substitution effect on the benzene ring and follows the
Hammet equation32 similar to typical peroxidases. Additionally, KatGs have been reported
to have also halogenation33 and NADH oxidase34 activity.
The naturally occurring peroxidase substrate is unknown. In physiological
conditions it is well known that KatG from Mycobacterium tuberculosis can activate the
anti-tuberculosis drug isoniazid35 but otherwise the physiological function of these
enzymes remains unknown. Due to the restricted access, only small peroxidase substrates
can enter the main entrance channel. A second access route, found in monofunctional
peroxidases, approximately in the plane of the heme, is blocked by the KatG-typical loops.
However, another potential route that provides access to the core of the protein, between
the two domains of the subunit, has been described21. It has been speculated that this could
be the binding site for substrates with extended, possibly even polymeric character. In any
case neither from prokaryotic nor from eukaryotic KatGs the naturally occurring oneelectron donor(s) is known nor is the function of a high peroxidase activity in a catalatic
enzyme. A reasonable role at low peroxide concentration could be the reduction of
16
alternative and catalatically inactive compounds I & II species to ferric KatG (similar to
NADPH in clade 3 catalases). Potential role of catalase-peroxidases in H2O2 signaling
remains elusive and needs to be investigated on both transcriptional and translational
levels. This is an actual topic mainly for eukaryotic catalase-peroxidases as there can exist
possible signaling pathways mainly during the interaction between host and pathogen.
Hydrogen peroxide reduction (i.e. compound I formation) follows Reaction 2,
which – in contrast to typical catalases – is immediately followed by Reaction 4 forming
the catalatically active intermediate
+
AA Por Fe(IV)-OH. One role of the Trp-Tyr-Met
adduct could be the (at least transient) radical site (+MYW Fe(IV)-OH) that quenches the
porphyryl radical. Similar to typical peroxidases, compound I formation (i.e. the
heterolytic cleavage of H2O2) is clearly assisted by the conserved distal residues His112
and Arg10815 but the exact role of distal amino acids in H2O2 oxidation is unknown. The
classical compound I, observed in virtually all heme peroxidases and catalases, has a
radical located on the porphyrin. However, in some enzymes a migration of an electron
from the protein to the heme takes place, thereby quenching the prophyrin radical and
producing a protein radical. Jakopitsch et al.14 showed - upon using stopped-flow
spectroscopy - that the catalase mechanism in KatG from three different sources
(Synechocysits PCC 6803, Burkholderia pseudomallei, and Mycobacterium tuberculosis)
diverges from the typical monofunctional enzyme pathway exemplified by bovine liver
catalase. Unlike monofunctional catalases, upon addition of millimolar concentrations of
H2O2 to ferric KatG was immediately transformed to an intermediate with characteristics
of oxyferrous heme14,36,37 at pH < 6.5, with absorbance maxima at 415 nm, 545 nm and
580 nm, whereas at pH > 7.0 another low-spin species was formed with maxima at 418
nm and 520 nm. These spectral features represent the dominating redox intermediate
during H2O2 degradation.
Analysis using stopped-flow spectroscopy in combination with rapid-freezequench electronic spin resonance spectroscopy clearly demonstrated the formation of
a protein-based radical on the distal side Trp-Tyr-Met adduct that co-exists with
hydroxoferryl heme. It persits during peroxide turnover and thus was proposed to be
the catalytically competent intermediate in the catalase cycle of wild-type KatG38,39. Based
on a reaction scheme proposed earlier14 and the recent findings by Suarez et al.38 it is now
possible to propose the catalatic mechanism of KatG (Figure 3).
17
OH
OH
H2O2
H
O
O
Fe(III)
H
Fe(III)
H2O
•
OH
O
H
O
H
OH
O
•+
Fe(IV)
Fe(IV)
H2O2
•
•
O
O
H2O
H
OO H
H
O
H
O
H
O2
Fe(IV)
Fe(II)
•
OH
O
O2
•
O2
Fe(III)
⎯
Fe(III)
Figure 3. Reaction scheme of the hydrogen peroxide reduction and oxidation reaction catalyzed by
catalase-peroxidase. It summarizes experimental data published by Jakopitsch et al14, Deemagarn et al.40,
Suarez et al.38, and Zhao et al.39.
18
In detail, the first substrate hydrogen peroxide enters the main channel into the
heme site by binding between Arg108 and Asp141 (Burkholderia pseudomallei
numbering), i.e. one of the two substrate paths proposed by Deemagarn et al.40.
Compound I formation is initiated by moving of the H2O2 molecule to interact with
Arg108 and His112. During compound I formation two electrons coming from the iron
and the porphyrin are transferred to the oxygen, thereby catalyzing the heterolytic
cleavage of H-O-O-H and forming an oxoferryl heme with a porphyrin cation radical. This
(classical) compound I is rapidly converted to an alternative isoelectronic species with
hydroxoferryl heme and the second oxidation equivalent at the Trp-Tyr-Met adduct
(Figure 3). The second substrate H2O2 enters the heme cavity on a proposed second path40
between Asp141 and the main chain atoms of Ile237, Tyr238 and Val239 (Burkholderia
pseudomallei numbering). It finally interacts with the residues Trp111 and His112 and
reduces compound I to oxyferrous heme [Fe(II)-O2 ↔ Fe(III)-O2●-], which decomposes to
ferric enzyme and superoxide, the latter being quantitatively oxidized at the adduct radical
closing the adduct shell and releasing dioxygen14,38. This reaction mechanism follows the
well known redox interconversion of monofunctional peroxidases in the presence of high
concentrations of hydrogen peroxide: ferric peroxidase  compound I  hydroxyferryl
 compound III [Fe(II)-O2 ↔ Fe(III)-O2●-] however, the decay of compound III is
extremely slow and superoxide is released instead of O2. In KatG the turnover of the latter
reaction is significantly enhanced due to the presence of the adduct radical that also
guarantees the release of only dioxygen instead of superoxide.
This reaction scheme is also underlined by the fact that variants that lack the KatGtypical covalent link have lost the catalatic activity. The variants form compound III
rapidly, but its turnover [i.e. conversion to Fe(III)] is extremely slow. So there is a clear
relationship between the stability of the oxyenzyme form, adduct radical formation and
catalase activity39. Of the three distal side mutants, the decay rates of the oxyferrous
enzyme was shown to be Met255Ala > Tyr229Phe > Trp107Phe39 (Mycobacterium
numbering), which is also the order of their catalase activities. Thus, the adduct radical
itself or some unique structural feature of this species, or both, contribute to the rapid
decay of wild-type oxyenzyme.
In any case, though catalyzing the same overall dismutation reaction (2 H2O2  2
H2O + O2) catalases and catalase-peroxidases follow a different reaction mechanism. As a
consequence the catalase activity of KatG should be attributed as pseudo-catalatic.
19
References
1
Cabiscol, E.; Tamarit, J.; Ros, J. Oxidative stress in bacteria and protein
damage by reactive oxygen species. Int. Microbiol. 2000, 3, 3-8.
2
Kirkman, H. N.; Gaetani, G. F. Mammalian catalase: a venerable enzyme
with new mysteries. Trends Biochem. Sci. 2007, 32, 44-50.
3
Peus, D.; Meves, A.; Vasa, R. A.; Beyerle, A.; O'Brien, T.; Pittelkow, M. R.
H2O2 is required for UVB-induced EGF receptor and downstream signaling pathway
activation. Free Radic. Biol. Med. 1999, 27, 1197-1202.
4
Rahman, I.; Adcock, I. M. Oxidative stress and redox regulation of lung
inflammation in COPD. Eur. Respir. J. 2006, 28, 219-242.
5
Veal, E. A.; Day, A. M.; Morgan, B. A. Hydrogen peroxide sensing and
signaling. Mol. Cell. 2007, 26, 1-14.
6
Sen, C. K.; Roy, S. Redox signals in wound healing. Biochim. Biophys. Acta
2008, 1780, 1348-1361.
7
Passardi, F.; Theiler, G.; Zamocky, M.; Cosio, C.; Rouhier, N.; Teixera, F.;
Margis-Pinheiro, M.; Ioannidis, V.; Penel, C.; Falquet, L.; Dunand, C. PeroxiBase: the
peroxidase database. Phytochemistry 2007, 68, 1605-1611.
8
Zamocky, M.; Koller, F. Understanding the structure and function of
catalases: clues from molecular evolution and in vitro mutagenesis. Prog. Biophys. Mol.
Biol. 1999, 72, 19-66.
9
Nicholls, P.; Fita, I.; Loewen, P. C. Enzymology and structure of catalases.
Adv. Inorg. Chem. 2001, 51, 51-106.
10
Thomas, J. A.; Morris, D. R.; Hager, L. P. Chloroperoxidase. VII. Classical
peroxidatic, catalatic, and halogenating forms of the enzyme. J. Biol. Chem. 1970, 245,
3129-3134.
11
Fita, I.; Rossmann, M. G. The active center of catalase. J. Mol. Biol. 1985,
185, 21-37.
12
Chouchane, S.; Girotto, S.; Yu, S.; Magliozzo, R. S. Identification and
characterization of tyrosyl radical formation in Mycobacterium tuberculosis catalaseperoxidase (KatG). J. Biol. Chem. 2002, 277, 42633-42638.
13
Ivancich, A.; Jakopitsch, C.; Auer, M.; Un, S.; Obinger, C. Protein-based
radicals in the catalase-peroxidase of Synechocystis PCC6803: a multifrequency EPR
20
investigation of wild-type and variants on the environment of the heme active site. J. Am.
Chem. Soc. 2003, 125, 14093-14102.
14
Jakopitsch, C.; Vlasits, J.; Wiseman, B.; Loewen, P. C.; Obinger, C. Redox
intermediates in the catalase cycle of catalase-peroxidases from Synechocystis PCC 6803,
Burkholderia pseudomallei, and Mycobacterium tuberculosis. Biochemistry 2007, 46,
1183-1193.
15
Smulevich, G.; Jakopitsch, C.; Droghetti, E.; Obinger, C. Probing the
structure and bifunctionality of catalase-peroxidase (KatG). J. Inorg. Biochem. 2006, 100,
568-585.
16
Vetrano, A. M.; Heck, D. E.; Mariano, T. M.; Mishin, V.; Laskin, D. L.;
Laskin, J. D. Characterization of the oxidase activity in mammalian catalase. J. Biol.
Chem. 2005, 280, 35372-35381.
17
Welinder, K. G. Superfamily of plant, fungal, and bacterial peroxidases.
Curr. Opin. Struct. Biol. 1992, 2, 388-393.
18
Passardi, F.; Zamocky, M.; Favet, J.; Jakopitsch, C.; Penel, C.; Obinger, C.;
Dunand, C. Phylogenetic distribution of catalase-peroxidases: are there patches of order in
chaos? Gene 2007, 397, 101-113.
19
Zamocky, M.; Jakopitsch, C.; Vlasits, J.; Obinger, C. Fungal catalase-
peroxidases - a novel group of bifunctional oxidoreductases. J. Biol. Inorg. Chem. 2007,
S97.
20
Yamada, Y.; Fujiwara, T.; Sato, T.; Igarashi, N.; Tanaka, N. The 2.0 A
crystal structure of catalase-peroxidase from Haloarcula marismortui. Nat. Struct. Biol.
2002, 9, 691-695.
21
Carpena, X.; Loprasert, S.; Mongkolsuk, S.; Switala, J.; Loewen, P. C.; Fita,
I. Catalase-peroxidase KatG of Burkholderia pseudomallei at 1.7A resolution. J. Mol. Biol.
2003, 327, 475-489.
22
Bertrand, T.; Eady, N. A.; Jones, J. N.; Jesmin; Nagy, J. M.; Jamart-
Gregoire, B.; Raven, E. L.; Brown, K. A. Crystal structure of Mycobacterium tuberculosis
catalase-peroxidase. J. Biol. Chem. 2004, 279, 38991-38999.
23
Wada, K.; Tada, T.; Nakamura, Y.; Kinoshita, T.; Tamoi, M.; Shigeoka, S.;
Nishimura, K. Crystallization and preliminary X-ray diffraction studies of catalaseperoxidase from Synechococcus PCC 7942. Acta Crystallogr. D Biol. Crystallogr. 2002,
58, 157-159.
21
24
Donald, L. J.; Krokhin, O. V.; Duckworth, H. W.; Wiseman, B.;
Deemagarn, T.; Singh, R.; Switala, J.; Carpena, X.; Fita, I.; Loewen, P. C. Characterization
of the catalase-peroxidase KatG from Burkholderia pseudomallei by mass spectrometry. J.
Biol. Chem. 2003, 278, 35687-35692.
25
Jakopitsch, C.; Kolarich, D.; Petutschnig, G.; Furtmuller, P. G.; Obinger, C.
Distal side tryptophan, tyrosine and methionine in catalase-peroxidases are covalently
linked in solution. FEBS Lett. 2003, 552, 135-140.
26
Ghiladi, R. A.; Knudsen, G. M.; Medzihradszky, K. F.; Ortiz de
Montellano, P. R. The Met-Tyr-Trp cross-link in Mycobacterium tuberculosis catalaseperoxidase (KatG): autocatalytic formation and effect on enzyme catalysis and
spectroscopic properties. J. Biol. Chem. 2005, 280, 22651-22663.
27
Baker, R. D.; Cook, C. O.; Goodwin, D. C. Properties of catalase-
peroxidase lacking its C-terminal domain. Biochem. Biophys. Res. Commun. 2004, 320,
833-839.
28
Zamocky, M.; Regelsberger, G.; Jakopitsch, C.; Obinger, C. The molecular
peculiarities of catalase-peroxidases. FEBS Lett. 2001, 492, 177-182.
29
Heering, H. A.; Indiani, C.; Regelsberger, G.; Jakopitsch, C.; Obinger, C.;
Smulevich, G. New insights into the heme cavity structure of catalase-peroxidase: a
spectroscopic approach to the recombinant Synechocystis enzyme and selected distal
cavity mutants. Biochemistry 2002, 41, 9237-9247.
30
Jakopitsch, C.; Ivancich, A.; Schmuckenschlager, F.; Wanasinghe, A.; Poltl,
G.; Furtmuller, P. G.; Ruker, F.; Obinger, C. Influence of the unusual covalent adduct on
the kinetics and formation of radical intermediates in Synechocystis catalase peroxidase: a
stopped-flow and EPR characterization of the MET275, TYR249, and ARG439 variants. J.
Biol. Chem. 2004, 279, 46082-46095.
31
Obinger,
Jakopitsch, C.; Auer, M.; Ivancich, A.; Ruker, F.; Furtmuller, P. G.;
C.
Total
conversion
of
bifunctional
catalase-peroxidase
(KatG)
to
monofunctional peroxidase by exchange of a conserved distal side tyrosine. J. Biol. Chem.
2003, 278, 20185-20191.
32
Regelsberger, G.; Jakopitsch, C.; Engleder, M.; Ruker, F.; Peschek, G. A.;
Obinger, C. Spectral and kinetic studies of the oxidation of monosubstituted phenols and
anilines by recombinant Synechocystis catalase-peroxidase compound I. Biochemistry
1999, 38, 10480-10488.
22
33
Jakopitsch, C.; Regelsberger, G.; Furtmuller, P. G.; Ruker, F.; Peschek, G.
A.; Obinger, C. Catalase-peroxidase from Synechocystis is capable of chlorination and
bromination reactions. Biochem. Biophys. Res. Commun. 2001, 287, 682-687.
34
Singh, R.; Wiseman, B.; Deemagarn, T.; Donald, L. J.; Duckworth, H. W.;
Carpena, X.; Fita, I.; Loewen, P. C. Catalase-peroxidases (KatG) exhibit NADH oxidase
activity. J. Biol. Chem. 2004, 279, 43098-43106.
35
Hydrogen
Zhao, X.; Yu, H.; Yu, S.; Wang, F.; Sacchettini, J. C.; Magliozzo, R. S.
peroxide-mediated
isoniazid
activation
catalyzed
by
Mycobacterium
tuberculosis catalase-peroxidase (KatG) and its S315T mutant. Biochemistry 2006, 45,
4131-4140.
36
Chouchane, S.; Lippai, I.; Magliozzo, R. S. Catalase-peroxidase
(Mycobacterium tuberculosis KatG) catalysis and isoniazid activation. Biochemistry 2000,
39, 9975-9983.
37
Jakopitsch, C.; Wanasinghe, A.; Jantschko, W.; Furtmuller, P. G.; Obinger,
C. Kinetics of interconversion of ferrous enzymes, compound II and compound III, of
wild-type Synechocystis catalase-peroxidase and Y249F: proposal for the catalatic
mechanism. J Biol Chem 2005, 280, 9037-9042.
38
Suarez, J.; Ranguelova, K.; Jarzecki, A. A.; Manzerova, J.; Krymov, V.;
Zhao, X.; Yu, S.; Metlitsky, L.; Gerfen, G. J.; Magliozzo, R. S. An oxyferrous
heme/protein-based radical intermediate is catalytically competent in the catalase reaction
of Mycobacterium tuberculosis catalase-peroxidase (KatG). J. Biol. Chem. 2009, 284,
7017-7029.
39
Zhao, X.; Yu, S.; Ranguelova, K.; Suarez, J.; Metlitsky, L.; Schelvis, J. P.;
Magliozzo, R. S. Role of the oxyferrous heme intermediate and distal side adduct radical
in the catalase activity of Mycobacterium tuberculosis KatG revealed by the W107F
mutant. J Biol Chem 2009, 284, 7030-7037.
40
Deemagarn, T.; Wiseman, B.; Carpena, X.; Ivancich, A.; Fita, I.; Loewen,
P. C. Two alternative substrate paths for compound I formation and reduction in catalaseperoxidase KatG from Burkholderia pseudomallei. Proteins 2007, 66, 219-228.
23
chapter one
Hydrogen peroxide oxidation by catalase-peroxidase follows
a non-scrambling mechanism
Jutta Vlasits, Christa Jakopitsch, Manfred Schwanninger, Peter Holubar
and Christian Obinger
FEBS Letters 581 (2007), 320 – 324
FEBS Letters 581 (2007) 320–324
Hydrogen peroxide oxidation by catalase-peroxidase follows
a non-scrambling mechanism
Jutta Vlasitsa, Christa Jakopitscha, Manfred Schwanningera, Peter Holubarb, Christian Obingera,*
b
a
Department of Chemistry at BOKU, University of Natural Resources and Applied Life Sciences, Muthgasse 18, A-1190 Vienna, Austria
Department of Biotechnology at BOKU, University of Natural Resources and Applied Life Sciences, Muthgasse 18, A-1190 Vienna, Austria
Received 9 November 2006; revised 6 December 2006; accepted 19 December 2006
Available online 5 January 2007
Edited by Judit Ovádi
Abstract Despite catalyzing the same reaction (2H2O2 fi
2H2O + O2) heme-containing monofunctional catalases and
bifunctional catalase-peroxidases (KatGs) do not share sequence
or structural similarities raising the question of whether or not
the reaction pathways are similar or different. The production
of dioxygen from hydrogen peroxide by monofunctional catalases has been shown to be a two-step process involving the redox
intermediate compound I which oxidizes H2O2 directly to O2. In
order to investigate the origin of O2 released in KatG mediated
H2O2 degradation we performed a gas chromatography–mass
spectrometry investigation of the evolved O2 from a 50:50
mixture of H218O2/H216O2 solution containing KatGs from
Mycobacterium tuberculosis and Synechocystis PCC 6803.
The GC–MS analysis clearly demonstrated the formation of
18
O2 (m/e = 36) and 16O2 (m/e = 32) but not 16O18O (m/
e = 34) in the pH range 5.6–8.5 implying that O2 is formed by
two-electron oxidation without breaking the O–O bond. Also
active site variants of Synechocystis KatG with very low catalase
but normal or even enhanced peroxidase activity (D152S,
H123E, W122F, Y249F and R439A) are shown to oxidize
H2O2 by a non-scrambling mechanism. The results are discussed
with respect to the catalatic mechanism of KatG.
2006 Federation of European Biochemical Societies. Published
by Elsevier B.V. All rights reserved.
Keywords: Catalase-peroxidase; Hydrogen peroxide
dismutation; Dioxygen release; Gas chromatography–mass
spectrometry analysis
1. Introduction
The four available crystal structures of catalase-peroxidases
(KatGs) of Haloarcula marismortui (1ITK), Burkholderia
pseudomallei (1MWV), Mycobacterium tuberculosis (1SJ2)
and Synechococcus PCC 7942 (1UB2) [1–4] revealed that the
organization of their active site is similar to those of cytochrome c peroxidase [5] and ascorbate peroxidase [6].
*
Corresponding author. Fax: +43 1 36006 6059.
E-mail address: [email protected] (C. Obinger).
Abbreviations: KatG, catalase-peroxidase; SynKatG, catalase-peroxidase from Synechocystis; MtbKatG, catalase-peroxidase from Mycobacterium tuberculosis; BLC, bovine liver catalase; GC–MS, gas
chromatography–mass spectroscopy; SIM, selected ion monitoring;
SOD, superoxide dismutase; EDTA, ethylene-diamine-tetraacetic acid;
NBT, nitro blue tetrazolium
Although belonging to the heme peroxidase superfamily of
plant, fungal and bacterial enzymes [7] KatGs [EC 1.11.1.7]
catalyze the same reaction as monofunctional catalases [EC
1.11.1.6] namely the dismutation of hydrogen peroxide to
dioxygen and water (2H2O2 fi 2H2O + O2). There are no
sequence or structural similarities between the two enzyme
classes and in contrast to monofunctional catalases the catalatic mechanism of KatGs is under discussion [8,9].
In monofunctional catalases the overall reaction follows two
distinct states in the reaction pathway. The first stage involves
oxidation of the heme iron using hydrogen peroxide as substrate to form compound I, an oxoiron(IV) porphyryl radical
(Por+FeIV‚O) species (reaction (1)). The second stage, or
reduction of compound I, employs a second molecule of peroxide as electron donor providing two oxidation equivalents
(reaction (2)). Gas chromatography–mass spectrometry (GC–
MS) analyses have revealed the exclusive formation of 18O2
and 16O2 from a 50:50 mixture of H216O2 and H218O2 indicating that in monofunctional catalases O2 is formed by two-electron oxidation of H2O2 without breaking the O–O bond
[10,11]. Reaction (2) was proposed to include proton abstraction by the distal histidine, hydride-ion transfer to compound
I and, finally, liberation of dioxygen and water [8,11]
PorFeIII þ H2 O2 ! Porþ FeIV @O þ H2 O
ð1Þ
Porþ FeIV @O þ H2 O2 ! Por FeIII þ O2 þ H2 O
ð2Þ
In chloroperoxidase the catalase reaction has been shown to
proceed also via a non-scrambling mechanism, whereas in a
reaction with labeled and unlabeled m-chloroperoxybenzoic
acid a scrambling mechanism for the release of O2 was established [12]. With m-chloroperoxybenzoic acid the oxygen atoms
of O2 derived from different molecules suggesting that at least
one oxygen atom came from compound I (Por+FeIV‚O) [12].
A clear picture of the reaction pathway of KatG has remained elusive. The determination of crystal structures of
KatGs [1–4] has led to the identification of several KatG-specific residues, subsequently confirmed by site-directed mutagenesis studies [9]. Such unique features have suggested a
number of unusual mechanisms controlling the catalase reaction. Moreover, the spectroscopic signatures of the redox intermediates are different from monofunctional catalases and the
exact electronic nature of the redox intermediate in the catalatic cycle is unknown [9].
Thus, in order to investigate whether KatG mediates the direct two-electron oxidation of H2O2 to O2 a GC–MS analysis
was performed using 16O- and 18O-labeled H2O2. Catalaseperoxidases from two different organisms (Mycobacterium
0014-5793/$32.00 2006 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.febslet.2006.12.037
25
J. Vlasits et al. / FEBS Letters 581 (2007) 320–324
tuberculosis and Synechocystis PCC 6803) and the most interesting Synechocystis KatG variants with diminished catalatic
activity have been investigated. For comparison, the enzymatic
action of bovine liver catalase (BLC, representative of monofunctional catalases) was elucidated under identical assay conditions. The results in all cases unambiguously show that the
O2 evolved originates from intact O–O bonds in hydrogen peroxide. The catalatic mechanism of KatG is discussed.
2. Materials and methods
2.1. Materials
Concentrations of H2O2, obtained as a 30% solution from Sigma,
and H218O2 (90% 18O), obtained as a 2% solution from ICON ISOTOPES, were determined by using a molar extinction coefficient of
e240 = 39.4 M1 cm1. Additionally, the concentration of H2O2 was
determined titrimetrically by oxidation with KMnO4 in acidified soluþ
2þ
þ 5O2 ðgÞþ
tion according to 2MnO
4 þ 5H2 O2 þ 6H ! 2Mn
8H2 O with 1 ml KMnO4 (0.002 mol l1) corresponding to 34.02 lg
H2O2 [13]. Titrisol potassium permanganate solution, [KMnO4] =
0.02 mol l1 was obtained from Merck. The 2 ml screw top vials with
septa, Viton were obtained from Supelco. Xanthine oxidase from buttermilk, xanthine, Cu–Zn superoxide dismutase from bovine erythrocytes, horse heart cytochrome c and nitro blue tetrazolium chloride
(NBT) were obtained from Sigma.
Bovine liver catalase (BLC) was obtained from Sigma and the concentration was calculated using the extinction coefficient of e406 = 1.2 ·
105 M1 cm1. Recombinant catalase-peroxidase from Synechocystis
and Mycobacterium tuberculosis were produced in E. coli and purified
as previously reported [14,2]. The concentrations of Synechocystis and
M. tuberculosis catalase-peroxidase (KatG) were calculated from a molar extinction coefficient of e406 = 1.0 · 105 M1 cm1. Mutagenesis,
expression and purification of the Synechocystis KatG variants
Y249F, W122F, H123E, D152S and R439A were described previously
[15–19].
2.2. Sample preparation
All solutions (total volume: 100 ll) and vials used in the assays were
made oxygen-free by flushing with nitrogen gas (oxygen <3 ppm) and
storing in a glove box (Meca-Plex, Neugebauer) with a positive pressure of nitrogen (25 mbar).
Hydrogen peroxide stock solutions were mixed with 100 mM phosphate/citrate buffer (pH 5.6), phosphate buffer (pH 7.0) and Tris–HCl
(pH 8.5) to a final hydrogen peroxide concentration of 50 mM H216O2
and 50 mM H218O2 (25 C). After initiating the reaction by enzyme
addition the vials were sealed gas-tight and discharged from the glove
box. The actual concentration of enzyme depended on the particular
enzyme under investigation, its kcat-value of catalatic turnover, and
its potential inactivation by high [H2O2]/[enzyme] ratios. In detail,
the concentration of BLC, wild-type Synechocystis and M. tuberculosis
KatG was 10 lM while that of Synechocystis variants was 20 lM
(D152S), 40 lM (R439A and Y249F), and 80 lM (W122F and
H123E), respectively.
2.3. Gas chromatography–mass spectrometry
Fifteen minutes after commencing the reaction, a gas-tight syringe
was used to inject 20 ll of the headspace gas onto a Hewlett–Packard
5890 II gas chromatograph (GC, Agilent Technologies) equipped with
a particle trap (2.5 m · 0.32 mm) and a HP-Plot MoleSieve capillary
column (30 m · 0.32 mm internal diameter · 12 lm film thickness) that
allowed a clear separation of oxygen and nitrogen. Helium was used as
carrier gas at an average velocity of 53.9 cm s1 corresponding to a
flow rate of 2.25 ml min1 (inlet pressure of 34.5 kPa). The injection
port was maintained at 71 C and the oven at 35 C.
Mass analysis was performed in the selected ion monitoring (SIM)
mode using a Hewlett–Packard 5971 A Mass-Selective Detector
(MSD) (Agilent Technologies). The selected masses were m/e = 28
(14N2), m/e = 30 (15N2), m/e = 32 (16O2), m/e = 34 (17O2), m/e = 36
(18O2 and 36Ar), m/e = 38 (38Ar), and m/e = 40 (40Ar). Due to the content of Argon in air (0.934 v/v%) and its isotope distribution [36Ar
(0.337%), 38Ar (0.063%), 40Ar (99.60%)] its contribution to m/e = 36
321
was negligible. Peak integration, performed with MSD ChemStation,
demonstrated a linear correlation between m/e = 32 peak area and
H216O2 or m/e = 36 peak area and H218O2 (in the concentration range
50 mM – 200 mM) dismutated by BLC. Furthermore, quantification of
dioxygen released by enzymatic activity was compared to that produced when H2O2 was oxidized by a small excess of KMnO4. The
amount of KMnO4 that was required for complete oxidation of hydrogen peroxide was judged by getting the acidic solution to turn slightly
violet.
2.4. Polarographic measurements
Dioxygen release from H2O2 (1–10 mM) was determined polarographically using a Clark-type electrode (YSI 5331 oxygen probe) inserted into a stirred thermostated water bath (YSI 5301B). To cover
the pH range 5.6–8.5, 50 mM phosphate/citrate, 50 mM phosphate
or 50 mM Tris–HCl buffers containing 100 lM ethylene-diamine-tetraacetic acid (EDTA) were used. All reactions were performed at 30 C,
initiated by addition of KatG and performed in the presence and
absence of superoxide dismutase (10 lg/ml).
2.5. Detection of superoxide
In order to assess putative superoxide release in the reaction of
KatG with hydrogen peroxide UV–Vis spectroscopy was applied to
follow superoxide mediated cytochrome c and nitro blue tetrazolium
reduction at 550 and 560 nm, respectively. Assay conditions were
50 mM buffer (pH 5.6–8.5) containing 100 lM EDTA, H2O2 (1–
5 mM) and either 20 lM cytochrome c or 100 lM NBT. Reactions
were initiated by adding KatG (10–50 nM). Xanthine oxidase was used
as a positive control for superoxide production. The assay conditions
were 50 mM buffer (pH 5.6–8.0) containing 100 lM EDTA, 50 lM
xanthine and either 20 lM cytochrome c or 100 lM NBT. Reactions
were initiated by adding xanthine oxidase (0.01–0.1 U).
3. Results and discussion
One method of approach to the study of enzymes with catalatic activity is to focus on their reaction product molecular
oxygen. Usually the enzymatic activity is followed polarographically and the turnover numbers given in Table 1 are
based on such measurements using a Clark-type electrode
[8,14–19]. Table 1 shows that the kcat value of the monofunctional catalase BLC is about two orders of magnitude higher
than that of KatGs. Nevertheless, KatGs are the only heme
peroxidases from the superfamily of plant, fungal and bacterial
enzymes, which exhibit peroxidase and substantial catalase
activities similar to monofunctional catalases (Table 1). But
the catalase activity is poorly characterized and it has been unknown so far whether the two O-atoms in each O2 liberated
came from the same H2O2 molecule or from two separate
H2O2 molecules. The mechanistic question was answered by
using two differently labeled molecules of hydrogen peroxide
(H216O2 and H218O2) simultaneously as substrates for KatG,
and using GC–MS to measure the liberated O2 in the headspace.
Precondition was a clear separation of the main components
of air, N2 (78.084 v/v%) and O2 (20.946 v/v%). In Fig. 1A it is
shown that the MoleSieve capillary column effectively separated O2 (2.03 min) from N2 (2.7 min). Analysis of the headspace of a reaction mixture containing 100 mM H2O2 in
100 mM phosphate buffer, pH 7.0, prepared in the glove box
and in the absence of an enzyme, gave a dominating N2-peak
and about 0.5–0.8% O2 most probably deriving from syringe
discharge from the glove box, transport to and injection onto
the gas chromatograph (thin line). The change in relative peak
areas upon total dismutation of 100 mM H2O2 by 10 lM catalase-peroxidase from Synechocystis (SynKatG) (sampling of
26
322
J. Vlasits et al. / FEBS Letters 581 (2007) 320–324
18
Enzyme
kcat (s )
m/z 32 ( O2) (% O2)
m/z 36 ( O2) (% O2)
BLC
MtbKatG
SynKatG
D152S
H123E
W122F
Y249F
R439A
212 000 [8]
6000 [4]
3500 [14]
200 [18]
2 [17]
– [16]
6 [15]
170 [19]
48
48
47
46
54
55
48
46
52
52
53
54
46
45
52
54
The enzymes were incubated with 50 mM H218O2 and 50 mM H216O2
in 100 mM phosphate buffer, pH 7.0, for 15 min in the glove box.
Enzyme concentrations used for GC–MS analyses: 10 lM (BLC,
MtbKatG, SynKatG), 20 lM (D152S), 40 lM (R439A and Y249F),
and 80 lM (W122F and H123E), respectively. 20 ll of head-space were
analyzed by GC–MS as described in Section 2. Areas of selected ion
chromatograms m/e = 32 and m/e = 36 are given in percentage of total
oxygen concentration [100% = (m/e = 32) + (m/e = 34) + (m/e = 36)].
All experiments were repeated at least three times and the mean values
are given. Average errors on m/e = 32 and m/e = 36 values are ±5%.
head-space 15 min after addition of SynKatG) is evident
(Fig. 1A, bold line). The release of 50 mM O2 from 100 mM
H2O2 led to a significant increase of O2 concentration in the
head-space as demonstrated by the significant increase in peak
area at 2.03 min. Independent of the mechanism of O2 release
from H2O2 (enzymatic dismutation or chemical oxidation by
KMnO4) there was a linear correlation between the H2O2 concentration in the assay and the calculated peak area at
2.03 min (inset to Fig. 1A). With the exception of the variants
W122F and H123E, under the substrate to protein ratios used
in these assays enzyme inactivation could be disregarded [14]
and H2O2 was completely degraded before sampling.
Fig. 1B shows the mass analysis in the SIM mode of the
2.03 min peak deriving from sampling the head-space 15 min
after addition of 10 lM SynKatG to a mixture of 50 mM
H216O2 and 50 mM H218O2 in 100 mM phosphate buffer, pH
7.0. Very similar results within experimental error were obtained with MtbKatG (10 lM) and BLC (10 lM). In any case,
two peaks of equal areas for 18O2 (m/e = 36) and 16O2(m/e = 32) with no indication of 16O18O (m/e = 34) formation
were found. This clearly demonstrates that O2 is formed by
two-electron oxidation of H2O2 without breaking the O–O
bond. In this respect there is no difference between KatG
and monofunctional catalases.
The pH profiles of catalatic activity for KatGs are quite different from those for monofunctional catalases. The latter’s
catalatic activities are essentially pH-independent from pH 5
to 10 [20], whereas KatGs show a sharp optimum between
pH 6 and 7 [9]. Performing the assays described above at pH
5.6 and 8.5 and analyzing the head-space by GC–MS gave very
similar results as shown in Fig. 1B indicating that the H2O2
oxidation follows the retention mechanism in the whole activity range of KatG. Thus, the drop in activity both at acidic and
alkaline pH must be due to other mechanisms. These could include proton abstraction before oxidation of H2O2 with the
help of an acid–base catalyst at the distal heme site and/or protonation of the active compound I species and/or pH-induced
changes in its electronic structure (see below).
Analysis of the crystal structures [1–4] and mass spectrometric data [21,22] has demonstrated the presence of two covalent
m/e = 32 (area×106)
16
7
N2
(2.70 min)
25
6
abundance (a.u.×105)
1
A
5
y = 0.1079x + 0.5284
20
15
10
5
y = 0.0534x + 0.6636
0
4
0
50
100
150
H2O2 (mM)
200
250
3
O2 (2.03 min)
2
1
0
1.5
B
2
2.5
time (min)
3
3.5
10
9
m/e = 32
8
abundance (a.u.×104)
Table 1
Enzymatic parameters and results of GC–MS analyses of bovine liver
catalase (BLC), catalase peroxidase from Mycobacterium tuberculosis
(MtbKatG), catalase-peroxidase from Synechocystis (SynKatG) and
variants with diminished catalatic activity
m/e = 36
7
6
5
4
3
2
m/e = 34
1
0
1.7
1.9
2.1
time (min)
2.3
2.5
Fig. 1. (A) Representative total ion chromatograms showing the clear
separation of O2 (retention time 2.03 min) and N2 (retention time
2.70 min) obtained with a MoleSieve capillary column. The thin line
represents the chromatogram of 20 ll sample taken from the headspace of a mixture prepared in the glove box and containing 100 mM
H216O2 in 100 mM phosphate buffer, pH 7.0. The bold line represents
the chromatogram of 20 ll taken from the head-space 15 min after
hydrogen peroxide degradation was started by addition of 10 lM
SynKatG. The inset depicts the linear correlation between the peak
area of m/e = 32 and the H216O2 concentration. Black line: H2O2
dismutation by BLC (conditions as in (A)). Grey line: H2O2 oxidation
by slight excess of KMnO4 in acidic solution. For details in mass
analysis and peak integration see Section 2. (B) Selected ion
chromatograms of m/e = 32 (bold black line), m/e = 34 (grey line),
and m/e = 36 (thin black line) of the GC peak at 2.03 min. 20 ll from
the head space were taken after reaction of 10 lM Synechocysits KatG
with 50 mM H218O2 and 50 mM H216O2 in 100 mM phosphate buffer,
pH 7.0.
bonds between three amino acid side chains, W122, Y249, and
M275 (Synechocystis numbering). This M-Y-W crosslink
appears to be a characteristic common to all KatGs and has
been demonstrated to be essential for the catalase activity
[9,15,16,19]. Interestingly, this adduct can be associated with
a KatG-typical arginine (R439) [2]. Its association with the
tyrosinate ion on the adduct is also required for optimum catalatic rates [19,23]. Another important residue at the distal
heme cavity is D152, which is part of the substrate channel
and has its side chain carboxyl group pointing toward the
heme pocket. It participates in maintaining a rigid and
27
J. Vlasits et al. / FEBS Letters 581 (2007) 320–324
extended hydrogen-bond network, which is important for the
H2O2 oxidation reaction [18,24]. Variants that had these residues exchanged (D152S, W122F, Y249F, and R439A) showed
a significantly reduced catalase (see Table 1) but a normal or
even enhanced peroxidase activity [9]. Thus, we included these
variants in this mechanistic investigation. Additionally, H123E
was probed since the distal H123 belongs to the very few amino acids, which are important in both the peroxidase and catalase activity because of its function in compound I formation
[17]. All these KatG variants are interesting since they exhibit
an altered kinetics of interconversion as well as spectral signatures of redox intermediates. This indicates that the electronic
nature of the active compound I species might be different as
could be the mechanism of H2O2 oxidation [9].
In order to see whether these variants also follow the retention mechanism, the dismutation of a 50:50 mixture of labeled
and unlabeled H2O2 was performed and the isotopic distribution of the reaction product O2 was analyzed by mass spectrometry. Higher enzyme concentrations were used (Table 1)
because of the diminished turnover. With the exception of
H123E and W122F in all assays the substrate was degraded
completely within 15 min of reaction. In case of H123E and
W122F about 55% and 65% of H2O2 were metabolized as
judged from the total 2.03 min O2 peak area. Nevertheless,
as Table 1 shows it is evident that with all these variants the
oxygen atoms of O2 derive from the same substrate molecule.
Although in some cases the catalatic activity is extremely low
and the variant is known to behave like a monofunctional
peroxidase (e.g. Y249F) the mechanism of H2O2 dismutation
follows the non-scrambling mechanism as in the wild-type
enzyme.
It has to be mentioned that a non-scrambling mechanism
could also result from one-electron oxidation steps of hydrogen peroxide by compound I and/or compound II. In this
alternative mechanism superoxide would be a reaction product
that dismutates to dioxygen and hydrogen peroxide in a
pH-dependent manner. If superoxide is a product in the degradation of hydrogen peroxide by catalase-peroxidase, its dismutation would not necessarily be rate limiting. Nevertheless, this
putative pathway can be rejected for several reasons. Firstly,
during H2O2 degradation by KatG neither cytochrome c nor
NBT reduction was observed (not shown) and secondly, superoxide dismutase (SOD) did not affect the kinetics of oxygen release mediated by KatG (not shown).
From the above experiments we deduce that catalase-peroxidases oxidize H2O2 in a two-electron oxidation step with both
oxygen atoms deriving from the same hydrogen peroxide molecule. This non-scrambling mechanism is independent of pH
and is not affected by manipulation of highly-conserved and
important catalatic residues. Principally, there are two possible
mechanisms of the formation of O2 following this retention
mechanism. (A) An ionic mechanism via initial proton abstraction with the help of an acid–base catalyst followed by a hydride-ion removal from H2O2 and release of O2. (B) A
hydrogen atom transfer from H2O2 to the ferryl species to yield
a radical intermediate as has been proposed for the alkaline
hydroxylation by cytochrome P450 [25]. Based on experiments
with monofunctional catalases and myoglobin variants and
deuterium isotope effects on the catalatic activities it has been
shown that the ionic mechanism is followed when a general
acid–base catalyst is available at the distal cavity at a proper
position [11]. In case of KatG two distal side residues have
323
been proposed to function as proton acceptor of the second
H2O2 molecule, namely W122 [16] and D152 [18,26]. Not clear
is the electronic nature of KatG compound I that is active in
H2O2 oxidation. The differences in the spectral features of
reaction intermediates formed during H2O2 dismutation by
monofunctional catalases and KatGs are significant [8,9]. In
monofunctional catalases the classical compound I is active
(i.e. oxoiron(IV)porphyryl radical) whereas in KatG – due to
electron transfer from the protein to the iron – alternative
compound I species (e.g. oxoiron(IV) protein radical or even
doubly oxidized protein species) have been suggested to be
catalatically active [8,9]. In any case, the present investigation
unequivocally suggests, that the non-scrambling mechanism is
followed generally by catalatic active heme enzymes and seems
to be independent of the actual electronic structure of compound I.
Acknowledgements: This work was supported by the Austrian Science
Funds (FWF Project P18751). We thank Prof. Peter Loewen from the
Department of Microbiology at the University of Manitoba for providing recombinant KatG from Mycobacterium tuberculosis.
References
[1] Yamada, Y., Fujiwara, T., Sato, T., Igarashi, N. and Tanaka, N.
(2002) The 2.0 Å crystal structure of catalase-peroxidase
from Haloarcula marismortui. Nat. Struct. Biol. 9, 691–695.
[2] Carpena, X., Loprasert, S., Mongkolsuk, S., Switala, J., Loewen,
P.C. and Fita, I. (2003) Catalase-peroxidase KatG of Burkholderia pseudomallei at 1.7 Å resolution. J. Mol. Biol. 327, 475–489.
[3] Wada, K., Tada, T., Nakamura, Y., Kinoshita, T., Tamoi, M.,
Shigeoka, S. and Nishimura, K. (2002) Crystallization and
preliminary X-ray diffraction studies of catalase-peroxidase from
Synechococcus PCC 7942. Acta Crystallogr. D Biol. Crystallogr.
58, 157–159.
[4] Bertrand, T., Eady, N.A.J., Jones, J.N., Jesmin, D., Nagy, J.M.,
Jamart-Gregoire, B., Raven, E.L. and Brown, K.A. (2004) Crystal
structure of Mycobacterium tuberculosis catalase-peroxidase. J.
Biol. Chem. 279, 38991–38999.
[5] Finzel, B.C., Poulos, T.L. and Kraut, J. (1984) Crystal structure
of yeast cytochrome c peroxidase refined at 1.7 Å resolution. J.
Biol. Chem. 259, 13027–13036.
[6] Patterson, W.R. and Poulos, T.L. (1995) Crystal structure of
recombinant pea cytosolic ascorbate peroxidase. Biochemistry 34,
4331–4341.
[7] Welinder, K.G. (1992) Superfamily of plant, fungal and bacterial
peroxidases. Curr. Opin. Struct. Biol. 2, 388–393.
[8] Nicholls, P., Fita, I. and Loewen, P.C. (2001) Enzymology and
structure of catalases. Adv. Inorg. Chem. 51, 51–106.
[9] Smulevich, G., Jakopitsch, C., Droghetti, E. and Obinger, C.
(2006) Probing the structure and bifunctionality of catalaseperoxidase (KatG). J. Inorg. Biochem. 100, 568–585.
[10] Jarnagin, R.C. and Wang, J.H. (1958) Investigation of the
catalytic mechanism of catalase and other ferric compounds with
doubly 18O-labeled hydrogen peroxide. J. Am. Chem. Soc. 80,
786–787.
[11] Kato, S., Ueno, T., Fukuzumi, S. and Watanabe, Y. (2004)
Catalase reaction by myoglobin mutants and native catalase:
mechanistic investigation by kinetic isotope effect. J. Biol. Chem.
279, 52376–52381.
[12] Hager, L.P., Doubek, D.L., Silverstein, R.M., Hargis, J.H. and
Martin, J.C. (1972) Chloroperoxidase. IX. The structure of
compound I. J. Am. Chem. Soc. 94, 4364–4366.
[13] Schulze G. and Simon J. (1989) Bestimmung von Wasserstoffperoxid. In: Jander Jahr Maßanalyse, 15, Auflage, de Gruyter, p. 168.
[14] Jakopitsch, C., Ruker, F., Regelsberger, G., Dockal, M., Peschek,
G.A. and Obinger, C. (1999) Catalase-peroxidase from the
cyanobacterium Synechocystis PCC 6803: cloning, overexpression
in Escherichia coli, and kinetic characterization. Biol. Chem. 380,
1087–1096.
28
324
[15] Jakopitsch, C., Auer, M., Ivancich, A., Rüker, F., Furtmüller, P.G.
and Obinger, C. (2003) Total conversion of bifunctional catalaseperoxidase (KatG) to monofunctional peroxidase by exchange of a
conserved distal side tyrosine. J. Biol. Chem. 278, 20185–20191.
[16] Regelsberger, G., Jakopitsch, C., Rüker, F., Switala, J., Loewen,
P. and Obinger, C. (2001) The role of distal tryptophan in the
bifunctional activity of catalase-peroxidases. Biochem. Soc.
Trans. 29, 99–105.
[17] Regelsberger, G., Jakopitsch, C., Engleder, M., Rüker, F.,
Peschek, G.A. and Obinger, C. (2000) Effect of distal cavity
mutations on the formation of compound I in catalase-peroxidases. J. Biol. Chem. 275, 22854–22861.
[18] Jakopitsch, C., Auer, M., Regelsberger, G., Jantschko, W.,
Furtmüller, P.G., Rüker, F. and Obinger, C. (2003) Distal site
aspartate is essential in the catalase activity of catalase-peroxidases. Biochemistry 42, 5292–5300.
[19] Jakopitsch, C., Ivancich, A., Schmuckenschlager, F., Wanasinghe,
A., Furtmüller, P.G., Rüker, F. and Obinger, C. (2005) Influence
of the unusual covalent adduct on the kinetics and formation of
radical intermediates in Synechocystis catalase peroxidase: a
stopped-flow and EPR characterization of the MET275,
TYR249, and ARG439 variants. J. Biol. Chem. 279, 46082–46095.
[20] Goldberg, I. and Hochman, A. (1989) Three different types of
catalases in Klebsiella pneumoniae. Arch. Biochem. Biophys. 268,
124–128.
J. Vlasits et al. / FEBS Letters 581 (2007) 320–324
[21] Donald, L.J., Krokhin, O.V., Duckworth, H.W., Wiseman, B.,
Deemagarn, T., Singh, R., Switala, J., Carpena, J.X., Fita, I. and
Loewen, P.C. (2003) Characterization of the catalase-peroxidase
KatG from Burkholderia pseudomallei by mass spectrometry. J.
Biol. Chem. 278, 35687–35692.
[22] Jakopitsch, C., Kolarich, D., Petutschnig, G., Furtmüller, P.G.
and Obinger, C. (2003) Distal side tryptophan, tyrosine and
methionine in catalase-peroxidases are covalently linked in
solution. FEBS Lett. 552, 135–140.
[23] Carpena, X., Wiseman, B., Deemargarn, T., Singh, R., Switala, J.,
Ivancich, A., Fita, I. and Loewen, P.C. (2005) A molecular switch
and electronic circuit modulate catalase activity in catalaseperoxidases. EMBO Rep. 6, 1156–1162.
[24] Jakopitsch, C., Droghetti, E., Schmuckenschlager, F., Furtmüller,
P.G., Smulevich, G. and Obinger, C. (2005) Role of the main
access channel of catalase-peroxidase in catalysis. J. Biol. Chem.
280, 42411–42422.
[25] Nesheim, J.C. and Lipscomb, J.D. (1996) Large kinetic isotope
effects in methane oxidation catalyzed by methane monooxygenase: evidence for C–H bond cleavage in a reaction cycle
intermediate. Biochemistry 35, 10240–10247.
[26] Deemagarn, T., Wiseman, B., Carpena, X., Ivancich, A., Fita, I.
and Loewen, P.C. (2007) Two alternative substrate paths for
compound I formation and reduction in catalase-peroxidase
KatG from Burkholderia pseudomallei. Proteins 66, 219–228.
29
chapter two
Trapping the redox intermediates in the catalase cycle of catalaseperoxidases from Synechocystis PCC 6803, Burkholderia
pseudomallei, and Mycobacterium tuberculosis
Christa Jakopitsch, Jutta Vlasits, Ben Wiseman, Peter C. Loewen
and Christian Obinger
Biochemistry 46 (2007), 1183 – 1193
Biochemistry 2007, 46, 1183-1193
1183
Redox Intermediates in the Catalase Cycle of Catalase-Peroxidases
from Synechocystis PCC 6803, Burkholderia pseudomallei, and
Mycobacterium tuberculosis†
Christa Jakopitsch,‡ Jutta Vlasits,‡ Ben Wiseman,§ Peter C. Loewen,§ and Christian Obinger*,‡
Department of Chemistry, DiVision of Biochemistry, BOKU-UniVersity of Natural Resources and
Applied Life Sciences, Muthgasse 18, A-1190 Vienna, Austria, and Department of Microbiology, UniVersity of Manitoba,
Winnipeg, MB R3T 2N2 Canada
ReceiVed NoVember 2, 2006; ReVised Manuscript ReceiVed December 8, 2006
ABSTRACT:
Monofunctional catalases (EC 1.11.1.6) and catalase-peroxidases (KatGs, EC 1.11.1.7) have
neither sequence nor structural homology, but both catalyze the dismutation of hydrogen peroxide (2H2O2
f 2H2O + O2). In monofunctional catalases, the catalatic mechanism is well-characterized with
conventional compound I [oxoiron(IV) porphyrin π-cation radical intermediate] being responsible for
hydrogen peroxide oxidation. The reaction pathway in KatGs is not as clearly defined, and a comprehensive
rapid kinetic and spectral analysis of the reactions of KatGs from three different sources (Synechocystis
PCC 6803, Burkholderia pseudomallei, and Mycobacterium tuberculosis) with peroxoacetic acid and
hydrogen peroxide has focused on the pathway. Independent of KatG, but dependent on pH, two lowspin forms dominated in the catalase cycle with absorbance maxima at 415, 545, and 580 nm at low pH
and 418 and 520 nm at high pH. By contrast, oxidation of KatGs with peroxoacetic acid resulted in
intermediates with different spectral features that also differed among the three KatGs. Following the rate
of H2O2 degradation by stopped-flow allowed the linking of reaction intermediate species with substrate
availability to confirm which species were actually present during the catalase cycle. Possible reaction
intermediates involved in H2O2 dismutation by KatG are discussed.
All aerobically growing organisms have to deal with
reactive oxygen species (ROSs), e.g., superoxide radicals,
hydrogen peroxide, and hydroxyl radicals. Nature has
evolved specialized enzymes for the degradation of the
different ROSs to protect the cells. The obvious role of
catalases is to dismutate hydrogen peroxide into water and
oxygen. There is a high degree of diversity among catalatically active enzymes, and according to their primary and
quaternary structure, subunit size, and prosthetic group, they
have been divided into four subgroups: monofunctional
heme b- or d-containing catalases, bifunctional heme bcontaining catalase-peroxidases, nonheme catalases, and,
finally, proteins with minor catalatic activity like monofunctional peroxidases (1). Here, we focus on the heme
b-containing monofunctional catalases and catalase-peroxidases (KatGs).1 Monofunctional catalases are found in all
kingdoms of life, whereas KatGs are found only in archaea,
bacteria, and fungi (2).
Although both types of heme enzymes exhibit high
catalatic activities, there are significant differences, including
†
This work was supported by the Austrian Science Funds (FWF
Project P18751). This work was also supported by grants from the
Natural Sciences and Engineering Research Council of Canada (to
P.C.L.) and by the Canadian Research Chair Program (to P.C.L.).
* To whom correspondence should be addressed. E-mail:
[email protected]. Phone: +43-1-36006-6073. Fax: +431-36006-6059.
‡
BOKU-University of Natural Resources and Applied Life Sciences.
§
University of Manitoba.
the absence of any sequence homology and very different
tertiary and quaternary structures, including the active site
residues (Figure 1). The most highly conserved part in
catalases is an eight-stranded antiparallel β-barrel domain
with six R-helical insertions in the turns between the strands.
The internal parts of this domain harbor essential distal side
residues His74, Ser113, and Asn147 (BLC numbering) and
the proximal heme iron ligand Tyr357 (3). In contrast, KatGs
contain 20 R-helices per monomer, 10 in the N-terminal
domain and 10 in the C-terminal domain organized in a
manner very similar to that of other class I peroxidases (4,
5). Heme is bound only to the N-terminal domain, and the
function of the duplicated C-terminal domain is still under
discussion (6). The active side residues include Arg108,
His112, and Trp111 (BpKatG numbering) on the distal side
of the heme and the proximal ligand His279 (Figure 1D). A
covalent adduct involving Trp111, Tyr238 [situated on a loop
not found in the other class I peroxidases but highly
conserved in KatGs (7)], and Met264 has been shown to be
essential for the catalatic activity of KatGs (8-13).
Catalase or peroxidase cycles are initiated by the H2O2mediated oxidation of the native ferric enzyme to com1
Abbreviations: KatG, catalase-peroxidase; SynKatG, catalaseperoxidase from Synechocystis PCC 6803; BpKatG, catalase-peroxidase
from B. pseudomallei; MtbKatG, catalase-peroxidase from M. tuberculosis; WT, wild-type; CCP, cytochrome c peroxidase; BLC, bovine
liver catalase; PAA, peroxoacetic acid; mCPB, m-chloroperbenzoic acid;
CT, charge transfer; EPR, electronic paramagnetic resonance; QM/MM,
quantum mechanical/molecular mechanical.
10.1021/bi062266+ CCC: $37.00 © 2007 American Chemical Society
Published on Web 01/13/2007
31
1184 Biochemistry, Vol. 46, No. 5, 2007
Jakopitsch et al.
FIGURE 1: (A) Monomer of BLC (4BLC). (B) Monomer of KatG of B. pseudomallei (1MWV). (C) Active side residues in BLC. (D)
Active side residues in KatG. The figures were constructed using the coordinates in the Protein Data Bank.
pound I. Generally, compound I is a redox intermediate two
oxidizing equivalents above the resting state. This reaction
causes the release of one water molecule and coordination
of the second oxygen atom to the iron center (14, 15). In
monofunctional catalases, compound I is an oxoiron(IV)
porphyrin π-cation radical species (Por•+FeIVdO) (16) which
is reduced back to the ferric enzyme by a second molecule
of hydrogen peroxide with the release of oxygen and water
(14). In the reaction of BLC with peroxoacetic acid, a
tyrosine radical can be formed by migration of an electron
to the porphyrin π-cation radical. However, the high rate of
turnover suggests that the tyrosyl radical plays no role in
the catalatic reaction (16).
A lively debate about the catalatic mechanism in KatGs
continues. Like that of BLC, reaction of KatGs with organic
peroxides generates an oxoiron(IV) porphyrin π-cation
radical species, which rapidly transforms to a protein radical
species (17-19), but with spectral UV-vis signatures
dissimilar to those described in monofunctional catalases and
peroxidases (1, 8-11). Because of the high intrinsic catalase
activity generating molecular oxygen, it has been difficult
to trap the spectroscopic signatures of the H2O2-generated
catalatic intermediates. In this paper, we have used stoppedflow techniques to characterize the UV-vis signatures of
the dominant catalatic intermediates of KatGs from three
different sources (Synechocystis PCC 6803, Mycobacterium
tuberculosis, and Burkholderia pseudomallei) at pH 5.6, 7.0,
and 8.5 and to monitor H2O2 degradation under identical
conditions. So far, almost all mechanistic studies have been
conducted with these three enzymes, which share all KatGtypical structural features but also showed some differences
in radical transformation when treated with organic peroxides
(17-19). Significant differences in comparison to BLC are
documented and discussed in terms of possible reaction
schemes.
MATERIALS AND METHODS
Reagents. Standard chemicals and biochemicals of the
highest available grade were obtained from Sigma. Hydrogen
peroxide was from Sigma, and its concentration was deter-
32
Catalatic Intermediates of Catalase-Peroxidase
Biochemistry, Vol. 46, No. 5, 2007 1185
mined using an extinction coefficient at 240 nm of 39.4 M-1
cm-1. To eliminate H2O2 from commercial peroxoacetic acid
(PAA), BLC (5 nM) was added to the buffered PAA stock
solutions. BLC was obtained from Sigma and used without
further purification. An extinction coefficient at 404 nm of
105 000 M-1 cm-1 was used for the determination of protein
concentration. Recombinant catalase-peroxidase from Synechocystis was produced in Escherichia coli and purified as
previously reported (20). B. pseudomallei and M. tuberculosis
(Mtb) KatGs were produced and purified as previously
described (5).
Steady-State Kinetics. Catalase activity was determined
polarographically using a Clark-type electrode (YSI 5331
oxygen probe) inserted into a stirred thermostated water bath
(YSI 5301B). To cover the pH range of 4.0-9.0, 50 mM
citrate-phosphate, 50 mM phosphate, or 50 mM Tris-HCl
buffer was used. All reactions were performed at 30 °C (37
°C in the case of BpKatG) and started by addition of KatG.
One unit of catalase is defined as the amount that decomposes
1 µmol of H2O2/min at pH 7 and 25 °C.
Transient-State Kinetics. Transient-state measurements
were taken using a model SX.18MV stopped-flow spectrophotometer and a PiStar-180 circular dichroism spectrometer
(Applied Photophysics Ltd.) equipped with a 1 cm observation cell. Calculation of pseudo-first-order rate constants (kobs)
from experimental time traces was performed with a SpectraKinetic workstation (version 4.38) interfaced with the
instrument. The substrate concentrations were at least 5 times
that of the enzyme to allow determination of pseudo-firstorder rate constants. Second-order rate constants were
calculated from the slope of the linear plot of the pseudofirst-order rate constants versus substrate concentration. To
follow spectral transitions, a model PD.1 photodiode array
accessory (Applied Photophysics Ltd.) connected to the
stopped-flow machine together with XScan diode array
scanning software (version 1.07) was utilized. The kinetics
of oxidation of ferric enzymes by hydrogen peroxide and
organic peroxides were followed in the single mixing mode.
The first data point was recorded 1.3 ms after mixing, and
2000 data points were accumulated. Sequential mixing
stopped-flow analysis was used to assess compound I
reduction (preformed with peroxoacetic acid) by hydrogen
peroxide. In detail, 3 µM enzyme was mixed with 100-200
µM peroxoacetic acid (final concentration), and after a
defined delay time, compound I was mixed with hydrogen
peroxide. All stopped-flow measurements were taken at 25
°C, and at least three determinations were performed per
substrate concentration.
FIGURE 2: Reaction of BLC with peroxides. (A) Hydrogen
peroxide-mediated reduction of BLC compound I preformed with
100 µM peroxoacetic acid. Spectral changes observed after addition
of 10 µM hydrogen peroxide to 3 µM BLC compound I. Spectrum
1 is the first detectable spectrum (1.3 ms after compound I is mixed
with H2O2). Spectrum 2 was taken 20 ms after mixing and
dominated during H2O2 degradation. Finally, compound I was reformed due to the excess of peroxoacetic acid in the reaction mixture
(spectrum 3, taken after 3 s). Conditions: 50 mM phosphate buffer
at pH 7.0 and 25 °C. (B) Time trace followed at 404 nm for the
reaction in panel A. Arrows indicate times of spectra selection.
The inset shows the time trace including the single-exponential fit
that reflects reduction of compound I by 10 µM hydrogen peroxide.
(C) Plot of the pseudo-first-order rate constants for compound I
reduction (kobs) vs the concentration of H2O2. (D) Time traces of
H2O2 degradation followed at 240 nm for the reaction of 2 µM
ferric BLC with 2 mM (gray line), 10 mM (thin line), and 20 mM
(thick line) hydrogen peroxide. Conditions: 50 mM phosphate
buffer at pH 7.0 and 25 °C. The inset shows the time trace and
single-exponential fit for the reaction of 2 µM BLC with 10 mM
hydrogen peroxide.
RESULTS
Mechanism of the Reaction of BLC with Peroxides. In
monofunctional catalases, the formation of an oxoiron(IV)
porphyryl radical compound I (Por•+FeIVdO) cannot be
followed spectroscopically using a moderate excess of
hydrogen peroxide, because reduction of compound I by
H2O2 (reaction 2) is faster than its formation (reaction 1).
k1
PorFeIII + H2O2 98 Por•+FeIVdO + H2O
k2
(1)
Por•+FeIVdO + H2O2 98 PorFeIII + O2 + H2O (2)
For comparison with KatGs, we revisited BLC to monitor
the spectral features of intermediates formed during the
reaction of the ferric enzyme with H2O2 and the kinetics of
H2O2 degradation at 240 nm in the stopped-flow apparatus.
As an example, a 2 µM solution of BLC depleted a 2 mM
solution of H2O2 within 300 ms (Figure 2D). The time trace
of H2O2 depletion can be exactly represented by a singleexponential equation (inset of Figure 2D). The spectral
features that predominated during the reaction strongly
resembled those of the ferric enzyme, consistent with a
catalatic cycle of only reactions 1 and 2 where k2 > k1, as
described in the literature (1). No red shift of the Soret band
33
1186 Biochemistry, Vol. 46, No. 5, 2007
Jakopitsch et al.
or other features typical of a BLC compound I species were
observed over the pH range of 5-10 (not shown).
In BLC and monofunctional heme b catalases, generally
an oxoiron(IV) porphyryl radical compound I (characterized
by a hypochromicity at the Soret region of 40-45% and a
long wavelength band at 665-670 nm of an intensity almost
equal to that of the original CT band of the ferric enyzme)
is formed during reaction with peroxoacetic acid (1) (Figure
2A, spectrum 3). The reaction of preformed compound I with
hydrogen peroxide (reaction 2) was followed in sequentialmixing mode, revealing that BLC compound I readily reacts
with hydrogen peroxide directly to the ferric enzyme (Figure
2A), underlining the fact that the oxoiron(IV) porphyryl
radical species indeed participates in the catalatic cycle and
is responsible for the oxidation of H2O2 to O2. Spectra 1
and 2 in Figure 2 recorded 1.3 and 20 ms, respectively, after
mixing 10 µM H2O2 with 3 µM BLC compound I already
resemble (15-20% hypochromicity) that of the ferric enzyme
and did not change during H2O2 degradation. After complete
H2O2 dismutation, the compound I spectrum reappeared as
a result of reaction with excess peroxoacetic acid (spectrum
3, Figure 2). The time trace for reduction of compound I by
hydrogen peroxide was fitted to a single-exponential equation, and the resulting linear plot of the pseudo-first-order
rate constants versus H2O2 concentration (Figure 2C) produced a second-order rate constant (k2) of 3.2 × 107 M-1
s-1 at pH 7.0, in good agreement with the value estimated
by Chance (21).
Kinetics of Hydrogen Peroxide Degradation Mediated by
Catalase-Peroxidases. A significant difference in the turnover
rates of BLC and KatG is evident (compare Figures 2D and
3C), with 2 µM KatG from Synechocystis (SynKatG)
degrading 10 mM hydrogen peroxide in ∼1.4 s compared
to 0.5 s for 2 µM BLC. Furthermore, the kinetics of
degradation of H2O2 by KatG differed from those of BLC
in that they could not be fitted to a single-exponential
equation (Figure 3B-D). Similar results were obtained with
KatGs from M. tuberculosis and B. pseudomallei.
To understand the differences in the shape of the time
traces between BLC and KatG, it is advisable to describe
the H2O2 dismutation reaction by a bi-uni mechanism similar
to the reaction catalyzed by superoxide dismutase, SOD (i.e.,
dismutation of superoxide to hydrogen peroxide and oxygen).
According to the mechanism described for SOD by Fee and
Bull (22), the catalatic cycle could be described by two
irreversible (an oxidative and a reductive) reactions, with E
being the enzyme in its resting state and CI representing a
compound I species irrespective of its electronic structure:
k1
k2
E + H2O2 {\
} [E-H2O2] 98 CI + H2O
k
-1
k3
k4
} [CI-H2O2] 98 E + O2 + H2O
CI + H2O2 {\
k
-3
(3)
(4)
The steady-state equation can be obtained using the method
of King and Altman as follows (23):
-d[H2O2]
2(1/k2 + 1/k4)-1[catalase][H2O2]
)
dt
(1/k2 + 1/k4)-1
+ [H2O2]
-1
1
1
+
k1k2
k3k4
k-1 + k2 k-3 + k4
(
)
The productive binding rates for the oxidative (ko) and
reductive (kr) half-reactions can be formulated as
ko ) k1[k2(k-1 + k2)]
kr ) k3[k4(k-3 + k4)]
Substitution of these binding rates into the steady-state
equation allows the expression of the turnover number (kcat)
and the apparent Michaelis-Menten constant, Km:
kcat ) (1/k2 + 1/k4)-1
Km )
(1/k2 + 1/k4)-1
(1/ko + 1/kr)-1
This leads to two limiting cases. (A) A Km much greater
than the H2O2 concentration corresponds to a second-order
process and represents the binding of hydrogen peroxide. A
pseudo-first-order degradation rate of H2O2 will be monitored
at 240 nm. (B) A Km much lower than the H2O2 concentration
represents saturation of the enzyme. A zero-order degradation
rate of H2O2 will be monitored at 240 nm.
An apparent Km value for BLC has been determined to
be 93 mM (24), but our studies are limited to concentrations
of hydrogen peroxide up to only 20 mM, because of the
increasing inaccuracy at absorbances in excess of 1 and the
vigorous oxygen evolution in the reaction cell at higher
concentrations. Therefore, in the case of BLC where Km .
[H2O2], a pseudo-first-order reaction is expected and observed (Figure 2D). By contrast, the Km values of KatGs are
much lower [4.2 mM for SynKatG, 2.5 mM for MtbKatG
(8), and 5.9 mM for BpKatG (25), all at pH 7]. As a
consequence, the shape of the time traces depends on the
H2O2 concentration, following almost pseudo-first-order
kinetics at H2O2 concentrations below 1 mM (Figure 3A)
and deviating from the single-exponential fit at 10 mM H2O2
(Figure 3B). At alkaline pH, however, the rate of depletion
of H2O2 was nearly linear at low concentrations of hydrogen
peroxide, and the apparent Km values at pH 8.5 were
determined to be 0.4 mM (SynKatG) at 30 °C and 0.22 mM
(BpKatG) at 37 °C, fully compatible with the observed
linearity of the time traces and the kinetic model proposed
above. At pH 8.5 and 10 mM H2O2, it follows that Km ,
[H2O2], and the model predicts a reaction that follows zeroorder kinetics, which is reflected by the experimental findings
(Figure 3C). At acidic pH, the apparent Km values are
comparable to those determined at pH 7.0 (4.7 mM for
SynKatG and 5.7 mM for BpKatG at pH 5.6) and the time
traces for 10 mM H2O2 degradation were nonlinear (Figure
3D) and did not fit well to a single-exponential equation.
The plot of the rate of H2O2 degradation determined by
stopped-flow against pH was similar in shape to a plot using
rates obtained from polarographic measurements of oxygen
34
Catalatic Intermediates of Catalase-Peroxidase
Biochemistry, Vol. 46, No. 5, 2007 1187
FIGURE 4: Reaction of wild-type Synechocystis KatG with hydrogen
peroxide at pH 7.0. Spectra were recorded 1.3 ms after mixing of
2 µM ferric KatG (gray line) with 2 mM (thin line) and 20 mM
(thick line) hydrogen peroxide. Conditions: 50 mM phosphate
buffer at pH 7.0 and 25 °C. The inset shows the comparison of the
time traces at 240 and 429 nm for the reaction of 2 µM ferric KatG
with 10 mM hydrogen peroxide. Conditions: 50 mM phosphate
buffer at pH 7.0 and 25 °C.
FIGURE 3: Kinetics of hydrogen peroxide degradation. (A) Time
trace at 240 nm (black line) and single-exponential fit (gray line)
for the reaction of 2 µM ferric wild-type KatG with 1 mM hydrogen
peroxide. Conditions: 50 mM phosphate buffer at pH 7.0 and 25
°C. (B) Time trace (black line) and single-exponential fit (gray line)
for the reaction of 2 µM ferric wild-type KatG with 10 mM
hydrogen peroxide. Conditions as in panel A. (C) Time traces at
240 nm for the reaction of 2 µM ferric KatG with 10 mM hydrogen
peroxide at pH 7.0, 7.5, 8, and 8.5. Conditions: 50 mM phosphate
buffer at 25 °C. (D) Time traces at 240 nm for the reaction of 2
µM ferric KatG with 10 mM hydrogen peroxide at pH 5.0, 5.6,
6.0, and 7.0. Conditions: 50 mM phosphate buffers and 50 mM
citrate/phosphate buffer at pH 5.0 and 25 °C. (E) pH dependence
of hydrogen peroxide degradation determined by following H2O2
degradation at 240 nm in the stopped-flow apparatus. (F) pH
dependence of oxygen evolution determined with a Clark-type
electrode. Conditions: 50 mM citrate/phosphate buffers (pH 4.07.0) and 50 mM Tris-HCl buffers (pH 7.5-9.0) at 25 °C.
evolution (Figure 3E,F), but with maximum activity slightly
shifted from pH 6.5 to 6.0. Similar results were obtained
with MtbKatG and BpKatG, although the decrease in
stopped-flow-determined activity in the acidic region was
much more pronounced than in the polarographically determined rates. These small differences notwithstanding, the
pH profiles of all KatGs exhibit a sharp optimum at pH
6-6.5, whereas BLC, and catalases in general, exhibit a
broad pH optimum extending from pH 5.6 to 8.5 (not
shown).
Reaction of Ferric Synechocystis KatG with Hydrogen
Peroxide. A recently constructed variant of SynKatG, E253Q
situated in the substrate entrance channel, has been found to
slow the catalatic turnover sufficiently to allow the identification of spectral features of reaction intermediates in the
catalase cycle (26). With a rate too rapid to be measured by
stopped-flow techniques, a 50-fold excess of H2O2 generated
an intermediate exhibiting the spectral features of a lowspin species, including a red-shifted Soret band, hyperchromicity in the Q-band region (502 and 542 nm), formation
of a broad shoulder around 520 nm, and disappearance of
the high-spin CT band at 637 nm. Carrying out the same
experiment with native SynKatG produced an intermediate
with the same spectral features, but only at a much higher
excess (at least 1000-fold) of H2O2, necessitated by the higher
catalatic activity of native KatG compared to E253Q (Figure
4). Generally, increasing amounts of H2O2 caused an
increasingly more pronounced red shift of the Soret band
within 1.3 ms of mixing, and the reaction intermediate was
evident only until all of the H2O2 was exhausted (inset of
Figure 4). For example, after 1 s, the time needed to
completely deplete 10 mM H2O2 with 2 µM SynKatG, the
spectrum of ferric protein was recovered.
The pH dependence of the appearance of the reaction
intermediate was directly related to the pH dependence of
the catalatic reaction. At pH 8.5, the reaction is only 15%
of the rate at pH 7, resulting in less H2O2 being necessary
for the appearance of the spectral signatures of the intermediate (Figure 5A). No other spectral changes were evident even
with a 1000-fold excess of H2O2 at pH 8.5 (Figure 5A). A
different situation was observed at acidic pH (Figure 5B)
where the first intermediate that could be trapped had spectral
features completely different from those observed at pH 7.0
and 8.5. A 1000-fold excess of H2O2 within 1.3 ms generated
a spectrum with a slightly decreased and red-shifted Soret
band, and addition of even greater excesses (up to 100000fold) caused a shift of the Soret band to 417 nm and the
appearance of two distinct peaks at 545 and 578 nm (Figure
5B), reminiscent of the spectra of compound III of SynKatG
(27), of the intermediate from the SynKatG variant Y249F
mixed with a moderate excess of hydrogen peroxide (9), and
of the alkaline forms of plant-type peroxidases (“FeIII-OH”)
(28).
At intermediate pH 6.5, the first trapped spectrum exhibits
a broad shoulder around 520 nm as well as a shoulder around
580 nm suggesting a mixture of the spectral signatures
observed at pH 7.0 and 5.6 (Figure 5C). As the insets of
panels A-C of Figure 5 indicate, there was a pH-dependent
correlation between H2O2 degradation and the intermediate
present. Analysis of spectra at pH 8.5 and 5.6 (Figure 6A,B)
35
1188 Biochemistry, Vol. 46, No. 5, 2007
FIGURE 5: pH dependence of the reaction of ferric wild-type
Synechocystis KatG with hydrogen peroxide. (A) Spectra recorded
1.3 ms after 2 µM ferric KatG (gray line) was mixed with 200 µM
(thin line) and 2 mM (thick line) hydrogen peroxide at pH 8.5.
Conditions: 50 mM phosphate buffer at pH 8.5 and 25 °C. The
inset shows time traces at 240 nm (black line) and 521 nm (gray
line) for the reaction of 2 µM wild-type KatG with 2 mM hydrogen
peroxide at pH 8.5. (B) Spectra recorded 1.3 ms after 2 µM ferric
KatG (gray line) was mixed with 2 mM (thin line) and 200 mM
(thick line) hydrogen peroxide at pH 5.6. Conditions: 50 mM
phosphate buffer at pH 5.6 and 25 °C. The inset shows time traces
at 240 nm (black line) and 429 nm (gray line) for the reaction of
2 µM wild-type KatG with 10 mM hydrogen peroxide at pH 5.6.
Conditions as in panel B. (C) Spectra recorded 1.3 ms after 3 µM
ferric KatG (gray line) was mixed with 10 mM (thick line) hydrogen
peroxide at pH 6.5. Conditions: 50 mM phosphate buffer at pH
6.5 and 25 °C. The inset shows time traces at 240 nm (black line)
and 429 nm (gray line) for the reaction of 2 µM wild-type KatG
with 10 mM hydrogen peroxide at pH 6.5. Conditions as in panel
C.
in conjunction with the rates of H2O2 degradation (Figure
6C,D) reveals a single spectral signature (418 and 520 nm)
at pH 8.5 throughout the H2O2 degradation phase and
continuously changing spectra during advanced H2O2 depletion at pH 5.6 starting with a 1.3 ms spectrum with peaks at
415, 545, and 580 nm and ending with the Soret band at
407 nm and reappearance of the CT band at 637 nm.
Reaction of SynKatG Sequentially with Peroxoacetic Acid
and H2O2. The oxidation of KatGs by peroxoacetic acid
has been shown to give rise to a calculated reaction rate of
2-6 × 104 M-1 s-1, depending on the source of KatG (8,
11, 29). The spectral signatures of the SynKatG compound
I intermediate include a 40-50% hypochromicity of the Soret
band and two distinct bands at 604 and 643 nm which were
attributed to an oxoiron(IV) porphyrin radical cation species
Jakopitsch et al.
FIGURE 6: pH dependence of spectral transitions during hydrogen
peroxide turnover of Synechocystis KatG. (A) Reaction of 2 µM
KatG with 2 mM H2O2 at pH 8.5. The thick line is the spectrum
1.3 ms after mixing. This spectral signature persisted for 1.5 s.
Subsequent spectra were taken after 1.8 s and after 2.2 s. For
orientation, the spectrum of ferric KatG at pH 8.5 is colored gray.
(B) Reaction of 2 µM KatG with 20 mM H2O2 at pH 5.6. The
thick line is the spectrum observed 1.3 ms after mixing. Subsequent
spectra were taken at 430 ms and 1.4 s. The gray line is the spectrum
for ferric KatG at pH 5.6. (C) Hydrogen peroxide depletion recorded
at 240 nm for the reaction in panel A. (D) Hydrogen peroxide
depletion recorded at 240 nm for the reaction in panel B.
(20). No pH-dependent differences in the kinetics of its
formation or its spectral properties were observed in the pH
range of 5.6-8.5. EPR has demonstrated that upon reaction
of SynKatG with peroxoacetic acid in the absence of electron
donors, a Por•+ and, subsequently, two protein-based radicals,
a Trp• and a Tyr•, are formed (10).
In contrast to BLC, the compound I form of SynKatG
produced by PAA did not appear to react readily with
moderate levels of H2O2 back to the resting state. However,
the addition of a large excess of hydrogen peroxide did cause
a spectral transition, suggesting formation of the same
intermediates that were formed in the direct reaction of ferric
SynKatG with H2O2 (Figure 7A). For example, spectra
recorded during the reaction of 3 µM SynKatG compound I
with 6 mM hydrogen peroxide at pH 7.0 reveal a reaction
intermediate with a Soret band at 418 nm, a broad shoulder
at 520 nm, and no absorbance in the CT region while H2O2
degradation was taking place. Upon depletion of H2O2, the
PAA-generated compound I reappeared, a result of the excess
36
Catalatic Intermediates of Catalase-Peroxidase
FIGURE 7: Reaction of hydrogen peroxide with Synechocystis KatG
compound I preformed with peroxoacetic acid. (A) Spectral changes
observed in the reaction of 3 µM WT KatG compound I with 6
mM hydrogen peroxide. Spectra were taken immediately after
mixing (1.3 ms), after 15 ms, and after 50 ms. Conditions: 50 mM
phosphate buffer at pH 7.0 and 25 °C. (B) Spectral changes upon
mixing 3 µM WT KatG compound I with 10 mM hydrogen
peroxide. The first spectrum is the compound I spectrum, and
subsequent spectra were taken after 1.3 ms and after 20 ms.
Conditions: 50 mM phosphate buffer at pH 6.0 and 25 °C. (C)
Time trace at 420 nm and single-exponential fit for the reaction in
panel A. The inset shows the plot of the pseudo-first-order rate
constant vs the concentration of hydrogen peroxide.
PAA being in the reaction mixture (not shown). The time
course of this reaction as reflected by hyperchromicity and
a shift to 418 nm of the Soret band during the first 50 ms
was monophasic (black line in Figure 7C) and could be fit
to a single-exponential equation (gray line). Plotting these
pseudo-first-order rate constants versus H2O2 concentration
(inset of Figure 7C) yielded a second-order rate constant of
1.3 × 104 M-1 s-1. The intercept of the plot was relatively
high (38 s-1) which reflected the fact that the formed
intermediate was not a stable end product but subjected to
permanent turnover during H2O2 degradation. At pH 8.5, the
spectral transitions were similar to those at pH 7.0 (not
shown), whereas at pH 6.0, the Soret band shifted to 416
nm which was accompanied by absorbance decreases at 604
and 643 nm and the appearance of peaks at 545 and 580 nm
all following monophasic kinetics with a calculated rate
constant of 1.5 × 104 M-1 s-1 (intercept of 12 s-1) (Figure
7B).
Reaction of MtbKatG and BpKatG with Peroxoacetic Acid
and H2O2. To compare the properties of SynKatG with those
Biochemistry, Vol. 46, No. 5, 2007 1189
FIGURE 8: Reaction of M. tuberculosis and B. pseudomallei KatG
with PAA. (A) Spectral changes observed upon mixing of 2 µM
ferric M. tuberculosis KatG with 200 µM PAA. The thick gray
line is for the ferric enzyme; subsequent spectra were taken after
1.3 ms (gray line), 220 ms (black line), and 1.1 s (thick black line).
Conditions: 50 mM phosphate buffer at pH 7.0 and 25 °C. (B)
Spectral changes observed upon mixing of 2 µM ferric B.
pseudomallei KatG with 100 µM PAA. The dashed line is for the
ferric enzyme; subsequent spectra were taken after 1.3 ms (gray
line), 200 ms (black line), 1.9 s (thick black line), and 10 s (thick
gray line). Conditions: 50 mM phosphate buffer at pH 7.0 and 25
°C. (C) pH dependence of compound I formation in M. tuberculosis.
The time traces for the reaction of 2 µM ferric M. tuberculosis
KatG with 200 µM PAA observed at 408 nm at various pH values
are shown. (D) pH dependence of compound I formation in B.
pseudomallei. The time traces for the reaction of 2 µM ferric B.
pseudomallei KatG with 200 µM PAA observed at 407 nm at
different pH values are shown.
of other KatGs, MtbKatG and BpKatG were first treated with
peroxoacetic acid, revealing spectral changes quite different
from those observed in SynKatG, including a red-shifted
Soret band (to 415 nm) and two new maxima around 549
and 590 nm (Figure 8A,B) that were stable for at least 10 s.
At pH 7, the spectral changes of MtbKatG exhibit isosbestic
points with a monophasic transition (Figure 8A), whereas
the spectral changes of BpKatG were clearly not monophasic
(Figure 8B). These spectra are similar to those previously
reported for MtbKatG (29, 30), but with a more red-shifted
Soret band compared to the reported band at 411 nm (29,
30), more prominent maxima at 549 and 590 nm, and a less
intense shoulder at 655 nm (29). These spectral features
varied only slightly over the pH range of 4.5-8.5 with a
small increase in the magnitude of the 655 nm shoulder at
higher pH.
37
1190 Biochemistry, Vol. 46, No. 5, 2007
Jakopitsch et al.
FIGURE 10: Spectral features of redox intermediates in BpKatG at
pH 6.5. The black line is for 3 µM ferric BpKatG, the bold gray
line for compound I obtained upon mixing of 3 µM ferric BpKatG
with 200 µM PAA (final concentrations), and the thick black line
for the intermediate formed upon addition of 50 mM hydrogen
peroxide to compound I preformed with PAA. Conditions: 50 mM
phosphate buffer at pH 6.5 and 25 °C.
FIGURE 9: Reaction of M. tuberculosis KatG with hydrogen
peroxide. (A) Spectral changes observed upon reaction of 2 µM
M. tuberculosis KatG with 10 mM hydrogen peroxide at pH 7.0.
Times at which spectra were taken are indicated (dashed line for
the ferric enzyme). The inset shows the corresponding time trace
for hydrogen peroxide degradation at 240 nm. Conditions: 50 mM
phosphate buffer at pH 7.0 and 25 °C. (B) Spectral changes
observed upon reaction of 2 µM M. tuberculosis KatG with 10
mM hydrogen peroxide at pH 5.6. The dashed line is for the ferric
enzyme. The inset shows the corresponding time trace at 240 nm.
Conditions: 50 mM phosphate buffer at pH 5.6 and 25 °C.
In contrast to SynKatG, the reactions of both MtbKatG
and BpKatG with PAA were pH-dependent and the kinetics
were not monophasic at all pH values. At both acidic and
alkaline pHs, the reaction of MtbKatG with PAA was
biphasic and slower than at pH 7 (Figure 8C). The kinetics
of oxidation of BpKatG by PAA were slower than for
MtbKatG, revealing greater complexity, including an initial
hypochromicity of the Soret band at 407 nm followed by an
increase in absorbance and red shift to 415 nm (Figure 8D).
The time trace of the changes at 407 nm in combination with
the other spectral transitions suggests the existence of a
transient intermediate in the reaction pathway leading to the
compound I species with the observed spectral features.
Mixing of MtbKatG and BpKatG with H2O2 at pH 8.5
produced intermediates with the same spectral features that
were observed for SynKatG at pH 8.5 and 7.0 (418 and 520
nm) for as long as H2O2 was being degraded (data not
shown). At pH 7.0, mixing of both MtbKatG and BpKatG
with H2O2 gave rise to spectra very similar to that of
SynKatG at pH 5.6 (Figure 9A). For example, the reaction
of 2 µM MtbKatG with 10 mM hydrogen peroxide at pH
7.0 caused, within 1.3 ms and lasting for 600 ms, a red shift
of the Soret band (416 nm), the appearance of two peaks
around 545 and 580 nm, and a diminished CT1 band.
Ultimately, the absorbance increased generally without a
change in the position of the maxima at 416, 545, and 580
nm, and even after depletion of 10 mM H2O2 (1.8 s; see the
inset of Figure 9A), these spectral features persisted (Figure
9A, spectra taken after 2.5 s). The reaction of ferric MtbKatG
with H2O2 at pH 5.6 led to spectral changes similar to those
observed at pH 7.0 (Figure 9B), except that the general
increase in absorbance occurred faster (within 100 ms). After
complete dismutation of H2O2, the spectrum remained almost
identical to that resulting from PAA-mediated oxidation of
MtbKatG (415, 549, and 590 nm). Similar results were
obtained with BpKatG.
Compared to that of SynKatG, the degradation of hydrogen
peroxide by both MtbKatG and BpKatG was slower at pH
5.6, taking 10 s to dismutate 10 mM H2O2 compared to 2 s
for SynKatG. The reaction of BpKatG with H2O2 was also
tested at pH 4.5, made possible by its greater resistance to
acidic conditions compared to SynKatG. A pH of 4.5 is the
pH optimum for the peroxidase activity in BpKatG, and
catalase activity is decreased to 30-40% of maximum levels
at pH 6.5 but still higher than at pH 8.5 (31). The resulting
spectral features of the intermediate were similar to those
observed at pH 5.6 and 7.0 (415, 548, and 580 nm), but the
reaction was slower with the spectrum at 1.3 ms, suggesting
a mixture of the ferric enzyme and the intermediate. The
reaction followed pseudo-first-order kinetics, and the bimolecular rate constant determined at a single concentration of
hydrogen peroxide (1 mM) was 1.3 × 104 M-1 s-1.
The final step was to investigate the reaction of compound
I of BpKatG, generated using PAA, with H2O2. No spectral
changes were observed using a moderate excess of H2O2,
but a large excess (1000-fold range) resulted in the formation
of intermediates with the same spectral features and pH
dependence (Figure 10) as those formed in the direct mixing
of the ferric protein with hydrogen peroxide (Figure 9). The
transition between the spectral features of PAA-generated
compound I of BpKatG and the spectral features of the
intermediate dominating in the presence of H2O2 could be
fitted to a single-exponential equation which yielded a
second-order rate constant of 1.7 × 104 M-1 s-1 at both pH
6.5 and 8.5.
DISCUSSION
Despite catalyzing the same reaction (2H2O2 f 2H2O +
O2), heme-containing monofunctional catalases and bifunctional catalase-peroxidases do not share sequence or structural
38
Catalatic Intermediates of Catalase-Peroxidase
similarities, raising the question of whether the reaction
pathways are similar or different. Whereas catalases have
been the subject of study for decades, the interest in catalaseperoxidases has developed more recently, in part because
of its role in mediating isoniazid resistance in M. tuberculosis,
but also from the standpoint of determining how an enzyme
which so closely resembles a class I peroxidase can dismutate
H2O2 at reasonable rates. The determination of crystal
structures of KatGs, now from four different organisms, has
led to the identification of several catalase-specific residues,
subsequently confirmed by site-directed mutagenesis studies.
Such unique features have suggested a number of unusual
mechanisms controlling the catalase reaction, but a clear
picture of the reaction pathway has remained elusive. To
address this question, a comprehensive kinetic and spectral
investigation of the reaction of three different catalaseperoxidases and one monofunctional catalase with H2O2 and
PAA using stopped-flow techniques has been carried out and
correlated with the kinetics of H2O2 degradation also
monitored by stopped-flow spectroscopy.
The rate of H2O2 degradation by BLC is much faster than
the rate exhibited by KatGs, but in both cases, the kinetics
can be explained by a simple reaction pathway involving
reactions 3 and 4. Changing the H2O2 concentration in the
assay from below to higher than the apparent Km of KatGs
caused a change in the kinetic pattern of H2O2 degradation
consistent with the change from a substrate-unsaturated to
substrate-saturated state. Unfortunately, it was technically
not possible to raise the H2O2 concentration above the
apparent Km for monofunctional catalases in the stoppedflow system to determine if they responded similarly.
However, the overall reaction pathway involving H2O2
binding provides a reasonable explanation for the kinetic
responses of the two enzymes. On the other hand, the very
different pH profiles of the two classes of enzymes suggest
some fundamental differences.
These differences are also evident in the spectral features
of reaction intermediates formed during H2O2 dismutation
by the two classes of enzymes which differ significantly.
Two oxidized products have been observed after reaction of
BLC with peroxoacetic acid, an oxoferryl prophyrin radical
species and a presumed hydroxoferryl protein radical species
(1, 16). Neither of these species is evident in the reaction
with H2O2 because the rate of reaction 2 is so much faster
than the rate of reaction 1 that compound I does not
accumulate. However, compound I preformed with peroxoacetic acid is reduced to the ferric state by H2O2 at a
rapid rate, confirming that it is an intermediate in the catalatic
pathway.
The much slower turnover rate for the catalatic reaction
in KatGs compared to monofunctional catalases suggested
that it might be possible to capture and characterize reaction
intermediates of KatG with H2O2. This proved to be the case
with the rapid appearance, within 1.3 ms of mixing H2O2
with enzyme, of the spectral signatures of two low-spin
species that are different from the spectra of the compound
I species of both BLC and KatG generated by peroxoacetic
acid. Specifically, at pH 8.5, the spectra exhibited maxima
at 418 and 520 nm, while at pH 5.6, the spectra exhibited
maxima at 415, 545, and 580 nm; both spectra had lost the
CT1 band at 640 nm. The relative proportions of the two
intermediates varied at intermediate pHs, depending on the
Biochemistry, Vol. 46, No. 5, 2007 1191
pH and the KatG. Maximal conversion of the enzyme into
the reaction intermediate required saturation of the enzyme
with substrate ([H2O2] > apparent Km).
The spectrum with features at 418 and 520 nm does not
resemble the spectrum of any reaction intermediate previously reported in peroxidases or catalases. On the other hand,
the spectrum with features at 415, 545, and 580 nm of the
intermediate predominating at acidic pH resembles the
spectra of compound III of plant peroxidases (32), the ferrous
form of WT and the Y249F variant of SynKatG treated with
O2 (27), the ferric form of MtbKatG treated with superoxide
(33), compound II of MtbKatG treated with excess H2O2
(13), the Y238F variant of BpKatG treated with peroxoacetic
acid (unpublished data), several KatG variants treated with
a small excess of H2O2 (8, 9, 13), HRP (28) [but not MtKatG
(34)] at alkaline pH with a hydroxyl ion distal ligand,
Arthromyces ramosus peroxidase with NH2OH bound (35),
and finally, to some extent, the inactive N-terminal domain
of KatG of E. coli (416, 536, and 568 nm) (36). In the latter
case, incubation of the N-terminal domain with a separately
expressed C-terminal domain resulted in a partial restoration
of both catalase and peroxidase activities as well as highspin spectral features of wild-type KatG (6). In the case of
the NH2OH complex with the peroxidase, the crystal structure
of the complex suggests coordination of the nitrogen atom
to the heme iron and hydrogen bonding of the hydroxyl group
with the distal histidine possibly representative of compound
“0” or the H2O2-peroxidase complex prior to the reaction
(35). The conclusion here should be that compound III-like
spectra are not uncommon and may be exhibited by different
six-coordinate low-spin structures.
The crystal structures of a number of catalase, peroxidase,
and catalase-peroxidase peroxoacetic acid-generated reaction
intermediates have recently been reported that reveal Fe-O
bond lengths longer than the value of 1.65 Å expected for a
classical compound I (Por•+FeIVdO) structure. Micrococcus
lysodeikticus catalase (37), Helicobacter pylori catalase
(2IQF, manuscript in review), CCP (38), and BpKatG (39)
were converted to intermediates with Fe-O bond lengths
of 1.82, 1.85, 1.87, and 1.93 Å, respectively. In all cases,
the longer length was explained in terms of a PorFeIV-OH
species formed by a transfer of an electron from the protein
to the porphyryl radical, and this has been corroborated in
recent QM/MM calculations (40). Unfortunately, the situation
has been complicated somewhat by QM and QM/MM
calculations that postulate the existence of six isomers of
horseradish peroxidase compound II, of which the two
experimentally observed reaction intermediates, FeIVdO and
FeIV-OH, are the least stable while the singlet and triplet
states of the Por•+FeIII-OH and Por•+FeIII-OH2 complexes
are more stable (41).
EPR studies have revealed a number of radical-based
reaction intermediates that are formed during the treatment
of catalases, peroxidases, and catalase-peroxidases with
peroxoacetic acid or H2O2. The classic compound I species
has a radical on the porphyrin, a result of a one-electron
transfer to the iron (PorFeVdO f Por•+FeIVdO). This has
been observed in virtually all heme peroxidases and catalases.
However, specific enzymes in all classes also support the
migration of an electron from the protein into the heme,
quenching the porphyrin radical and producing a protein
radical based on either a Trp or Tyr residue (10, 16-19,
39
1192 Biochemistry, Vol. 46, No. 5, 2007
32). In such cases, protonation of the oxoferryl to form a
hydroxoferryl species occurs rapidly (PorFeIVdO + H+ f
PorFeIV-OH). In SynKatG, W106 has been identified as the
site of a Trp•+ radical while the location of the Tyr• remains
unidentified. The diversity of radical sites even in the same
class of enzymes is evident in the identification of the
surface-situated Y353 as a radical site in MtKatG (42).
Specific tyrosines have been identified as radical sites in CCP
and lignin peroxidase (32).
The proximity of the KatG-specific adduct (MYW) to the
reaction center stacked just 3.4 Å above the heme has led to
conjecture about its role in the reaction. One proposal is that
it forms one component of a molecular switch inductively
controlling the catalase reaction (39). Arg426 can adopt two
conformations depending on the pH and the oxidation state
of the heme (39). In the Y conformation favored at pH >6.5,
Arg429 is in ionic association with the adduct. Conformation
Y is in equilibrium with conformation R, which is favored
at pH <6.5 and predominates in KatG oxidized by PAA.
Formation of an adduct radical as an intermediate during
MYW formation has been proposed (30), but no radical has
so far been experimentally associated with the adduct in EPR
studies. This suggests that if an adduct radical is formed
during catalysis, it is transient or short-lived, and freezequench EPR techniques will be required for its identification.
Furthermore, the fact that the adduct is required for the
catalase but not the peroxidase reaction suggests that if a
transient radical on the adduct (MYW•+) is formed, it has a
role only in the second stage of the catalase reaction, i.e.,
H2O2 oxidation, and not in the formation of compound I or
the associated electron transfer pathway leading to protein
radical formation (10, 13). Alternative radical sites close to
the heme, such as the proximal tryptophan observed in CCP
(43), might also be considered.
In all of this, Arg426 plays a key role. On the one hand,
its association with the tyrosinate ion on the adduct is
required for optimal catalatic rates (10, 39). The role of the
putative adduct radical becomes complicated as a result,
because removal of an electron from the adduct during radical
formation would reduce the negative charge on the adduct,
thereby weakening its association with Arg426. Therefore,
quenching of the adduct radical to re-form the tyrosinate ion
followed by its reassociation with Arg426 must either be a
concerted part of or precede compound I reduction. Second,
the change between the 415, 545, and 580 nm (low pH) and
418 and 520 nm (high pH) species with a midpoint around
pH 6.5 correlates well with the 50:50 mixture of R and Y
conformations of the Arg426 side chain at pH 6.5 (44). The
influence of Arg426 on the predominant reaction intermediate
could be a result of inductive stabilization of a particular
intermediate or even the selection of alternate reaction
pathways.
In light of the foregoing, a number of schemes can be
envisioned that provide possible identities to the intermediates responsible for the previously unknown 418 and 520
nm and compound III-like 415, 545, and 580 nm intermediates. They range from the relatively simple compound
I-substrate complex (Por•+FeIVdO-H2O2) to alternative
compound I species that are active in H2O2 oxidation
(reaction 4) and have the porphyryl radical quenched by an
electron from the adduct (MYW•+PorFeIV-OH), an alternative amino acid near the heme (AA•+PorFeIV-OH), or both
Jakopitsch et al.
(MYW•+PorFeIII-OH-AA•+). Alternatively, because KatG
is a class I peroxidase, it might utilize the modified catalatic
mechanism of monofunctional peroxidases that includes
formation of compound III (PorFeII-O2 / PorFeIII-O2•-)
(32). In peroxidases, the slow decay of compound III is
responsible for the low rate of catalatic turnover, but in KatG,
a nearby radical located on the adduct (or an alternative site)
[MYW•+(AA•+)PorFeII-O2 / MYW•+(AA•+)PorFeIII-O2•-]
guarantees both a high turnover rate and the release of
molecular oxygen. In any case, the pH-dependent change in
the reaction intermediate can be explained by simple protonation and/or deprotonation of several intermediates or,
alternatively, by different intermediates being stabilized at
different pHs.
In summary, spectra suggestive of two different reaction
intermediates, depending on pH, appear after KatGs are
mixed with H2O2. One of the intermediates has not been
reported previously, and it is clearly different from the
signatures of intermediates generated after peroxoacetic acid
treatment. Alternative techniques, including possibly freezequench EPR and Mössbauer spectroscopy, will have to be
applied to differentiate among the possibilities.
REFERENCES
1. Nicholls, P., Fita, I., and Loewen, P. C. (2001) Enzymology and
structure of catalases, AdV. Inorg. Chem. 51, 51-106.
2. Klotz, M. G., and Loewen, P. C. (2003) The molecular evolution
of catalatic hydroperoxidases: Evidence for multiple lateral
transfer of genes between prokaryota and from bacteria into
eukaryota, Mol. Biol. EVol. 20, 1098-1112.
3. Ko, T. P., Day, J., Malkin, A. J., and McPherson, A. (1999)
Structure of orthorhombic crystals of beef liver catalase, Acta
Crystallogr. D55, 1383-1394.
4. Yamada, Y., Fujiwara, T., Sato, T., Igarashi, N., and Tanaka, N.
(2002) The 2.0 Å crystal structure of catalase-peroxidase from
Haloarcula marismortui, Nat. Struct. Biol. 9, 691-695.
5. Carpena, X., Loprasert, S., Mongkolsuk, S., Switala, J., Loewen,
P. C., and Fita, I. (2003) Catalase-peroxidase KatG of Burkholderia pseudomallei at 1.7 Å resolution, J. Mol. Biol. 327, 475489.
6. Baker, R. D., Cook, C. O., and Goodwin, D. C. (2006) Catalaseperoxidase active site restructuring by a distant and “inactive”
domain, Biochemistry 45, 7113-7121.
7. Zamocky, M., Regelsberger, G., Jakopitsch, C., and Obinger, C.
(2001) The molecular peculiarities of catalase-peroxidases, FEBS
Lett. 492, 177-182.
8. Ghiladi, R. A., Knudsen, G. M., Medzihradszky, K. F., and de
Montellano, P. R. (2005) The Met-Tyr-Trp cross-link in Mycobacterium tuberculosis catalase-peroxidase (KatG): Autocatalytic
formation and effect on enzyme catalysis and spectroscopic
properties, J. Biol. Chem. 280, 22651-22663.
9. Jakopitsch, C., Auer, M., Ivancich, A., Ruker, F., Furtmuller, P.
G., and Obinger, C. (2003) Total conversion of bifunctional
catalase-peroxidase (KatG) to monofunctional peroxidase by
exchange of a conserved distal side tyrosine, J. Biol. Chem. 278,
20185-20191.
10. Jakopitsch, C., Ivancich, A., Schmuckenschlager, F., Wanasinghe,
A., Poltl, G., Furtmuller, P. G., Ruker, F., and Obinger, C. (2004)
Influence of the unusual covalent adduct on the kinetics and
formation of radical intermediates in Synechocystis catalase
peroxidase: A stopped-flow and EPR characterization of the
MET275, TYR249, and ARG439 variants, J. Biol. Chem. 279,
46082-46095.
11. Regelsberger, G., Jakopitsch, C., Ruker, F., Krois, D., Peschek,
G. A., and Obinger, C. (2000) Effect of distal cavity mutations
on the formation of compound I in catalase-peroxidases, J. Biol.
Chem. 275, 22854-22861.
12. Hillar, A., Peters, B., Pauls, R., Loboda, A., Zhang, H., Mauk, A.
G., and Loewen, P. C. (2000) Modulation of the activities of
catalase-peroxidase HPI of Escherichia coli by site-directed
mutagenesis, Biochemistry 39, 5868-5875.
40
Catalatic Intermediates of Catalase-Peroxidase
13. Yu, S., Girotto, S., Zhao, X., and Magliozzo, R. S. (2003) Rapid
formation of compound II and a tyrosyl radical in the Y229F
mutant of Mycobacterium tuberculosis catalase-peroxidase disrupts
catalase but not peroxidase function, J. Biol. Chem. 278, 4412144127.
14. Schonbaum, G. R., and Chance, B. (1976) in The Enzymes, Vol.
13, Academic Press, New York.
15. Fita, I., and Rossmann, M. G. (1985) The active center of catalase,
J. Mol. Biol. 185, 21-37.
16. Ivancich, A., Jouve, H. M., Sartor, B., and Gaillard, J. (1997) EPR
investigation of compound I in Proteus mirabilis and bovine liver
catalases: Formation of porphyrin and tyrosyl radical intermediates, Biochemistry 36, 9356-9364.
17. Ivancich, A., Jakopitsch, C., Auer, M., Un, S., and Obinger, C.
(2003) Protein-based radicals in the catalase-peroxidase of Synechocystis PCC6803: A multifrequency EPR investigation of wildtype and variants on the environment of the heme active site, J.
Am. Chem. Soc. 125, 14093-14102.
18. Jakopitsch, C., Obinger, C., Un, S., and Ivancich, A. (2006)
Identification of Trp106 as the tryptophanyl radical intermediate
in Synechocystis PCC6803 catalase-peroxidase by multifrequency
electron paramagnetic resonance spectroscopy, J. Inorg. Biochem.
100, 1091-1099.
19. Chouchane, S., Girotto, S., Yu, S., and Magliozzo, R. S. (2002)
Identification and characterization of tyrosyl radical formation in
Mycobacterium tuberculosis catalase-peroxidase (KatG), J. Biol.
Chem. 277, 42633-42638.
20. Jakopitsch, C., Ruker, F., Regelsberger, G., Dockal, M., Peschek,
G. A., and Obinger, C. (1999) Catalase-peroxidase from the
cyanobacterium Synechocystis PCC 6803: Cloning, overexpression in Escherichia coli, and kinetic characterization, Biol. Chem.
380, 1087-1096.
21. Chance, B. (1949) The Primary and Secondary Compounds of
Catalase and Methyl or Ethyl Hydrogen Peroxide. IV. Reactions
with Hydrogen Peroxide, J. Biol. Chem. 180, 947-959.
22. Fee, J. A., and Bull, C. (1986) Steady-state kinetic studies of
superoxide dismutases. Saturative behavior of the copper- and
zinc-containing protein, J. Biol. Chem. 261, 13000-13005.
23. Segel, I. H. (1975) Enyzme Kinetics, Wiley-Interscience, New
York.
24. Switala, J., and Loewen, P. C. (2002) Diversity of properties among
catalases, Arch. Biochem. Biophys. 401, 145-154.
25. Deemagarn, T., Carpena, X., Singh, R., Wiseman, B., Fita, I., and
Loewen, P. C. (2005) Structural characterization of the Ser324Thr
variant of the catalase-peroxidase (KatG) from Burkholderia
pseudomallei, J. Mol. Biol. 345, 21-28.
26. Jakopitsch, C., Droghetti, E., Schmuckenschlager, F., Furtmuller,
P. G., Smulevich, G., and Obinger, C. (2005) Role of the main
access channel of catalase-peroxidase in catalysis, J. Biol. Chem.
280, 42411-42422.
27. Jakopitsch, C., Wanasinghe, A., Jantschko, W., Furtmuller, P. G.,
and Obinger, C. (2005) Kinetics of interconversion of ferrous
enzymes, compound II and compound III, of wild-type Synechocystis catalase-peroxidase and Y249F: Proposal for the
catalatic mechanism, J. Biol. Chem. 280, 9037-9042.
28. Howes, B. D., Rodriguez-Lopez, J. N., Smith, A. T., and
Smulevich, G. (1997) Mutation of distal residues of horseradish
peroxidase: Influence on substrate binding and cavity properties,
Biochemistry 36, 1532-1543.
29. Chouchane, S., Lippai, I., and Magliozzo, R. S. (2000) Catalaseperoxidase (Mycobacterium tuberculosis KatG) catalysis and
isoniazid activation, Biochemistry 39, 9975-9983.
Biochemistry, Vol. 46, No. 5, 2007 1193
30. Ghiladi, R. A., Medzihradszky, K. F., and Ortiz de Montellano,
P. R. (2005) Role of the Met-Tyr-Trp cross-link in Mycobacterium
tuberculosis catalase-peroxidase (KatG) as revealed by KatG(M255I), Biochemistry 44, 15093-15105.
31. Singh, R., Wiseman, B., Deemagarn, T., Donald, L. J., Duckworth,
H. W., Carpena, X., Fita, I., and Loewen, P. C. (2004) Catalaseperoxidases (KatG) exhibit NADH oxidase activity, J. Biol. Chem.
279, 43098-43106.
32. Dunford, H. B. (1999) Heme Peroxidases, Wiley-VCH, New York.
33. Ghiladi, R. A., Cabelli, D. E., and Ortiz de Montellano, P. R.
(2004) Superoxide reactivity of KatG: Insights into isoniazid
resistance pathways in TB, J. Am. Chem. Soc. 126, 4772-4773.
34. Chouchane, S., Girotto, S., Kapetanaki, S., Schelvis, J. P., Yu, S.,
and Magliozzo, R. S. (2003) Analysis of Heme Structural
Heterogeneity in Mycobacterium tuberculosis Catalase-Peroxidase
(KatG), J. Biol. Chem. 278, 8154-8162.
35. Wariishi, H., Nonaka, D., Johjima, T., Nakamura, N., Naruta, Y.,
Kubo, S., and Fukuyama, K. (2000) Direct binding of hydroxylamine to the heme iron of Arthromyces ramosus peroxidase.
Substrate analogue that inhibits compound I formation in a
competetive manner, J. Biol. Chem. 275, 32919-32924.
36. Baker, R. D., Cook, C. O., and Goodwin, D. C. (2004) Properties
of catalase-peroxidase lacking its C-terminal domain, Biochem.
Biophys. Res. Commun. 320, 833-839.
37. Murshudov, G. N., Grebenko, A. I., Brannigan, J. A., Antson, A.
A., Barynin, V. V., Dodson, G. G., Dauter, Z., Wilson, K. S.,
and Melik-Adamyan, W. R. (2002) The structures of Micrococcus
lysodeikticus catalase, its ferryl intermediate (compound II) and
NADPH complex, Acta Crystallogr. D58, 1972-1982.
38. Bonagura, C. A., Bhaskar, B., Shimizu, H., Li, H., Sundaramoorthy, M., McRee, D. E., Goodin, D. B., and Poulos, T. L. (2003)
High-resolution crystal structures and spectroscopy of native and
compound I cytochrome c peroxidase, Biochemistry 42, 56005608.
39. Carpena, X., Wiseman, B., Deemagarn, T., Singh, R., Switala, J.,
Ivancich, A., Fita, I., and Loewen, P. C. (2005) A molecular switch
and electronic circuit modulate catalase activity in catalaseperoxidases, EMBO Rep. 6, 1156-1162.
40. Hersleth, H. P., Ryde, U., Rydberg, P., Gorbitz, C. H., and
Andersson, K. K. (2006) Structures of the high-valent metal-ion
haem-oxygen intermediates in peroxidases, oxygenases and catalases, J. Inorg. Biochem. 100, 460-476.
41. Derat, E., and Shaik, S. (2006) Two-state reactivity, electromerism,
tautomerism, and “surprise” isomers in the formation of compound
II of the enzyme horseradish peroxidase from the principal species,
compound I, J. Am. Chem. Soc. 128, 8185-8198.
42. Zhao, X., Girotto, S., Yu, S., and Magliozzo, R. S. (2004) Evidence
for radical formation at Tyr-353 in Mycobacterium tuberculosis
catalase-peroxidase (KatG), J. Biol. Chem. 279, 7606-7612.
43. Sivaraja, M., Goodin, D. B., Smith, M., and Hoffman, B. M. (1989)
Identification by ENDOR of Trp191 as the free-radical site in
cytochrome c peroxidase compound ES, Science 245, 738-740.
44. Carpena, X., Wiseman, B., Deemagarn, T., Herguedas, B.,
Ivancich, A., Singh, R., Loewen, P. C., and Fita, I. (2006) Roles
for Arg426 and Trp111 in the modulation of NADH oxidase
activity of the catalase-peroxidase KatG from Burkholderia
pseudomallei inferred from pH-induced structural changes, Biochemistry 45, 5171-5179.
BI062266+
41
chapter three
Probing hydrogen peroxide oxidation kinetics of wild-type
Synechocystis catalase-peroxidase (KatG) and selected variants
Jutta Vlasits, Paul G. Furtmüller, Christa Jakopitsch, Marcel Zamocky
and Christian Obinger
Submitted to Biochimica et Biophysica Acta
Probing hydrogen peroxide oxidation kinetics of wild-type Synechocystis
catalase-peroxidase (KatG) and selected variants
Jutta Vlasits, Paul G. Furtmüller, Christa Jakopitsch, Marcel Zamocky, and
Christian Obinger*
1
Department of Chemistry, Division of Biochemistry, BOKU – University of Natural
Resources and Applied Life Sciences, Muthgasse 18, A-1190 Vienna, Austria
*corresponding author: phone: +43-1-36006-6073; fax: +43-1-36006-6059; email:
[email protected]
Running title: Hydrogen peroxide oxidation by catalase-peroxidase
43
Abbreviations:
KatG, catalase-peroxidase; BLC, bovine liver catalase; PAA,
peroxoacetic acid; RFQ-EPR, rapid freeze-quench electron
paramagnetic resonance; HS, high-spin, LS, low-spin; 5-c, five
coordinated; 6-c, six coordinated; kon, apparent bimolecular rate
constant of cyanide binding to ferric heme protein; k1, apparent
bimolecular rate constant of compound I formation; k2, apparent
bimolecular rate constant of the transition of compound I to ferric
KatG.
Key words:
Catalase-peroxidase;
hydrogen
peroxide
reduction;
hydrogen
peroxide oxidation; catalase activity; compound I; cyanide complex;
stopped-flow spectroscopy.
44
Abstract
Catalase-peroxidases (KatGs) are unique bifunctional heme peroxidases that exhibit
peroxidase and substantial catalase activities. However, the reaction pathway of hydrogen
peroxide dismutation, including the electronic structure of the redox intermediate that
actually oxidizes H2O2, is not clearly defined. Several mutant proteins with diminished
overall catalase but wild-type-like peroxidase activity have been described in the last
years. However, understanding of decrease in overall catalatic activity needs
discrimination between reduction and oxidation reactions of hydrogen peroxide. Here, by
using sequential-mixing stopped-flow spectroscopy, we have investigated the kinetics of
the transition of KatG compound I (produced by peroxoacetic acid) to its ferric state by
trapping the latter as cyanide complex. Apparent bimolecular rate constants (pH 6.5, 20°C)
for wild-type KatG and the variants Trp122Phe (lacks KatG-typical distal adduct),
Asp152Ser (controls substrate access to the heme cavity) and Glu253Gln (channel
entrance) are reported to be 1.2  104 M-1 s-1, 30 M-1 s-1, 3.4  103 M-1 s-1, and 8.6  103 M-1
s-1, respectively. These findings are discussed with respect to steady-state kinetic data and
proposed reaction mechanism(s) for KatG. Assets and drawbacks of the presented method
are discussed.
45
Introduction
Typical (monofunctional) catalases (EC 1.11.1.6) and catalase-peroxidases (KatGs,
EC 1.11.1.7) have neither sequence nor structural homology, but both catalyze the
dismutation of hydrogen peroxide: 2 H2O2  2 H2O + O2. In typical catalases the catalatic
mechanism is well characterized with conventional compound I [oxoiron(IV) porphyrin cation radical intermediate] being responsible for hydrogen peroxide oxidation [1]. The
reaction pathway in KatGs is not as clearly defined. Like in monofunctional catalases,
reaction of KatGs with alkyl peroxides generates an oxoiron(IV) porphyrin -cation
radical species, which rapidly transforms to a protein radical species [2-6]. Recently, it has
been demonstrated by stopped-flow spectroscopy in combination with rapid-freeze-quench
electron paramagnetic resonance (RFQ-EPR) spectroscopy that the protein radical is
located, at least transiently, at the KatG-typical covalent distal side adduct (Trp-Tyr-Met)
[6, 7]. Reaction of this redox intermediate with hydrogen peroxide constitutes the second
phase of the dismutation in which H2O2 acts as two-electron reductant and dioxygen is
released. In detail, it has been proposed that the oxoferryl heme rapidly reacts with
hydrogen peroxide to produce oxyferrous [Fe(II)-O2 ↔ ] heme that decomposes to ferric
KatG and superoxide, the latter being oxidized to O2 by the Trp-Tyr-Met adduct radical [5,
6]. In sum, this mechanism is non-scrambling (i.e. both oxygen atoms derive from the
same molecule of hydrogen peroxide) as has been demonstrated recently [8].
In order to understand the impact of mutations on the catalatic activity of KatG it is
necessary to distinguish between the role of H2O2 as oxidant and as reductant. The
heterolytic cleavage of H2O2 (i.e. compound I formation) is common to all heme
peroxidases and catalases and is known to be very fast (~1-5  107 M-1 s-1) [1, 9, 10],
whereas efficient H2O2 oxidation is only catalyzed by catalases and KatGs [1]. Typically,
for dismutating enzymes (counts also for superoxide dismutases) it is difficult to measure
the kinetics of the two half reactions since adding of substrate to either redox intermediate
immediately promotes the enzyme to cycle. Here, we have adapted the sequential stoppedflow instrument in order to measure the kinetics of the two-electron reduction of preformed compound I with hydrogen peroxide. In detail, wild-type KatG and selected
variants with great variability of catalatic activity were oxidized by peroxoacetic acid, to
compound I, which was immediately reduced by hydrogen peroxide in the presence of
cyanide in order to trap formed ferric KatG as low-spin (LS) complex. This allowed for the
first time extraction of apparent H2O2 oxidation rates by KatG compound I. The
46
corresponding bimolecular rates are presented and compared with published overall kinetic
parameters of KatG. The effect of mutations on the H2O2 oxidation reaction is discussed
with respect to proposed reaction mechanism(s).
Materials and Methods
Reagents and recombinant protein production. Standard chemicals and
biochemicals were obtained from Sigma at the highest grade available. Hydrogen peroxide
was obtained from Sigma and its concentration was determined using an extinction
coefficient at 240 nm of 39.4 M-1 cm-1. Peroxoacetic acid (PAA) was a 39% solution
supplied by Fluka. Its concentration was determined iodometrically. Bovine liver catalase
was obtained from Sigma and used without further purification. Its concentration was
determined using a molar extinction coefficient 404 nm of 105,000 M-1 cm-1 per heme [5].
Heterologous expression in E. coli, purification and characterization of recombinant wildtype KatG and variants was reported recently: wild-type [11], Trp122Phe [12], Asp152Ser
[13] and Glu253Gln [14], respectively.
Stopped-flow spectroscopy. Transient-state measurements were performed using a
sequential-mixing SX.18MV stopped-flow spectrophotometer (Applied Photophysics Ltd.)
equipped with a 1 cm observation cell thermostatted at 20°C and pH 6.5, the maximum of
the catalatic activity. Calculation of pseudo-first-order rate constants (kobs) from
experimental time traces was performed with the SpectraKinetic work station (Version
4.38) interfaced to the instrument. The substrate or ligand concentrations were at least five
times that of the enzyme to allow determination of pseudo first-order rate constants.
Second-order rate constants were calculated from the slope of the linear plot of the pseudo
first-order rate constants versus substrate concentration. To follow spectral transitions, a
PD.1 photodiode array accessory (Applied Photophysics) connected to the stopped-flow
machine together with Xscan diode array scanning software (Version 1.07) were utilized.
The conventional stopped-flow technique was used to measure the kinetics of
cyanide binding to the ferric heme proteins. Final concentrations: 1 µM KatG and 10–200
µM cyanide in 50 mM phosphate buffer, pH 6.5. The first data point was recorded 1.5 ms
after mixing, and 2000 data points were accumulated. The calculated kobs values (followed
at a wavelength with largest absorption difference between high-spin and low-spin species)
were fitted to the expression kobs = kon[CN-] + koff, where kon is the rate of cyanide binding
47
and koff the rate of dissociation of cyanide from the Fe(III) species. Values for the
equilibrium dissociation constants were calculated using KD = koff / kon.
The sequential-mixing stopped-flow technique was used to measure compound I
reduction by hydrogen peroxide in the presence of sodium cyanide. In the first step native
KatG and peroxoacetic acid (PAA) were premixed in the aging loop and after a defined
delay time (determined by maximum hypochromicity at the Soret maximum) H2O2 and
cyanide were added. In the beginning, 6 µM enzyme were mixed with an excess of PAA
(20 to 66-fold excess) to elucidate maximum hypochromicity and optimum delay time.
Finally, pre-formed compound I was mixed with various concentrations of hydrogen
peroxide ranging from 10 µM to 200 µM (Trp122Phe: 20 µM to 10 mM) in the presence
of 1 mM cyanide (50 mM phosphate buffer, pH 6.5).
Results
Kinetics of cyanide binding to ferric KatG. Since cyanide was intended to be used
for trapping ferric KatG formed by compound I reduction, it was necessary to re-evaluate
the kinetics of cyanide complex formation of native wild-type and mutant KatG. Ferric 5coordinated (5-c) high-spin (HS) KatG from Synechocystis has its Soret band at 406 nm,
Q-bands at 502 and 545 nm and a CT1 band (i.e. a porphyrin-to-metal charge transfer
band) at 635 nm [10]. Upon cyanide binding the Soret peak shifted to 422 nm (isosbestic
point at 414 nm) and a new prominent peak around 540 nm was formed. Formation of the
low-spin (LS) complex was followed at 427 nm (corresponding to the largest difference
between the high-spin and the low-spin species) and was monophasic. The apparent
bimolecular rate constant, kon, was calculated to be (5.1 ± 0.2)  105 M-1 s-1 at pH 6.5 and
20°C. With Glu253Gln both the spectral characteristics of the high- to low-spin transition
as well as the kinetics of complex formation were similar to wild-type KatG (Table 1).
Substitution of Trp122 by Phe not only inhibits formation of the KatG-typical
covalent adduct [15] but also decreases formation of 6-c HS and LS species [16].
Compared to wild-type KatG binding of cyanide (Soret band of LS complex at 419 nm)
was significantly slower, namely (3.9 ± 0.4) × 104 M-1 s-1. In contrast to the other proteins
investigated here, cyanide binding to Asp152Ser was biphasic with a rapid first phase
responsible for at least 85% of absorbance increase. This confirms already published data
[13]. By fitting the first rapid phase with a single-exponential equation pseudo first-order
rate constants, kobs, were obtained that linearly increased with the concentration of cyanide.
48
The calculated apparent bimolecular rate constant was (1.9 ± 0.2) × 106 M-1 s-1, about five
times faster than that of wild-type KatG.
Table 1. Kinetic constants for compound I formation (k1) and reduction (k2) as well as cyanide complex
formation of recombinant wild-type KatG from Synechocystis PCC 6803, selected variants (Glu253Gln,
Asp152Ser, Trp122Phe) and bovine liver catalase, BLC (pH 6.5, 20°C). n.d., not determinable.
Wild-type
Glu253Gln
Asp152Ser
Trp122Phe
BLC
( 104 M-1 s-1)
Compound I formation
Hydrogen peroxide
n.d.
n.d.
750 [13]
8.2 [12]
n.d.
3.9 [12]
3.0 [14]
26 [13]
18 [12]
-
kon
51
82 [14]
190
3.9
110
-1
koff (s )
7.6
10.7
7.3
2.9
3.7
KD (µM)
15
13
3.8
74
3.4
1.2
0.86
0.34
0.003
1100
3500 [12]
890 [14]
200 [13]
n.d.
212 000 [20]
Peroxoacetic acid
Cyanide binding
Compound I* reduction
Hydrogen peroxide
(in presence of cyanide)
Catalatic activity
kcat (s-1)
* Formed with peracetic acid
In addition to the kon values Table 1 depicts also koff and KD values for the
corresponding cyanide complexes. Cyanide complexes of KatG are relatively stable with
dissociation constants in the micromolar region. For comparative purposes we have also
investigated cyanide binding to ferric bovine liver catalase (shift of Soret band from 404
nm to 420 nm) and calculated kon to be (1.1 ± 0.03) × 106 M-1 s-1 (Table 1).
Kinetics of transition of compound I to ferric KatG. In the following sequential
stopped-flow experiments cyanide was present in order to suppress cycling of catalaseperoxidase with excess H2O2. Principally, ferric KatG (formed by two-electron reduction
of compound I) has two possibilities to react, namely either with H2O2 to compound I or
with cyanide to the corresponding LS complex at rates summarized in Table 1. Assuming
that the (unknown) rate of compound I formation by H2O2 in KatG is of the same order of
magnitude as in other heme peroxidases (k1 ~1  107 M-1 s-1 [10]), the cyanide
49
concentration had to be at least 50-times higher than that of added H2O2. Typically, with
wild-type KatG concentrations of H2O2 <100 µM and 1 mM NaCN were used.
Figure 1A shows the spectral transition of 3 µM compound I mixed with 100 µM
hydrogen peroxide in the presence of 1 mM cyanide. Oxidation of ferric KatG with
peroxoacetic acid to compound I has been shown to give rise to a calculated reaction rate
of 2-7  104 M-1 s-1 depending on the source of KatG [2, 12, 17, 18]. The spectral
signatures of the resulting Synechocystis intermediate included a 40-50 % hypochromicity
of the Soret band and two distinct bands at around 604 and 643 nm, which were attributed
to an oxoiron(IV) porphyrin cation radical species [5, 11]. Upon its mixing with hydrogen
peroxide in the presence of cyanide the Soret band shifted directly to 420 nm (apparent
isosbestic points at 403 and 464 nm) and a new peak at 540 nm appeared, closely
resembling the spectrum of the cyanide complex (Figure 1A).
The reaction was monophasic and could be fitted using a single-exponential
equation. A typical time trace is shown in Figure 1B. Plotting these pseudo-first-order rate
constants versus H2O2 concentration (Figure 1C) yielded a linear correlation that allowed
calculation of the apparent second-order rate, k2, from the slope: (1.2 ± 0.2)  104 M-1 s-1 at
pH 6.5 and 20°C. This clearly suggests that the rate limiting step in the consecutive
Reactions 1 & 2 is compound I reduction by H2O2 (i.e. release of dioxygen). Figure 1D
depicts the calculated concentration changes of the redox intermediates involved in this
reaction sequence. Since kon > k2, ferric KatG did not accumulate but was rapidly
transferred to its LS complex.
compound I + H2O2  Fe(III) + H2O + O2
Reaction 1
Fe(III) + CN¯  Fe(III)-CN¯
Reaction 2
Decrease in the concentration of compound I is accompanied by an increase of the KatGcyanide complex (Figure 1D), whereas the concentration of ferric KatG is highest at
around 6 nM (<< 1% of total KatG protein) in the first phase of reaction (Figure 1E). This
has been calculated based on the fact that (i) kon > k2, and (ii) ferric KatG reaches a steadystate level:
k2[compound I][H2O2] = kon[Fe(III)][CN¯]
d[Fe(III)]/dt = 0 = k2[compound I][H2O2] – kon[Fe(III)][CN¯]
[Fe(III)]steady-state = {k2[compound I][H2O2]} / {kon[CN¯]}
50
Under assumption that in the initial phase of reaction the concentrations of compound I,
H2O2 (100 µM) and CN¯ (1 mM) are constant and using k2 = (1.2 ± 0.2)  104 M-1 s-1 and
kon = 5.1  105 M-1 s-1, [Fe(III)] steady-state was calculated to be ~6 nM (out of 1.5 µM KatG).
Thus, under these conditions ferric KatG does not contribute to the observed spectral
transitions. As a consequence the compound I spectrum seemed to be transformed directly
into that of the cyanide complex and the corresponding time traces could be fitted with a
single-exponential equation. These calculations support our experimental setup for
determination of H2O2 oxidation rates in catalase-peroxidases.
Similar spectral transitions were observed for the variant Glu253Gln and k2 was
calculated to be (8.6 ± 0.1)  103 M-1 s-1 at pH 6.5 and 20°C. This corresponds to ~72% of
the wild-type rate. In case of Trp122Phe high concentrations of H2O2 had to be added to
compound I in order to induce a spectral shift of the Soret maximum from 409 nm to 417
nm and formation of peak at 540 nm (data not shown). The rate of transition was
extremely slow (30 M-1 s-1, <0.3% of wild-type rate) thereby confirming previous data that
clearly showed the importance of Trp122 (and of the intact KatG-typical Trp122-Tyr249Met275 adduct) in catalatic activity, i.e. oxidation of H2O2 [10].
51
(A)
0.24
0.06
420
0.2
406
0.05
604
Absorbance
bance
643
0.16
0.04
540
0 12
0.12
0 03
0.03
0.08
0.02
0.04
0.01
0
0
300
(C)
0.11
2.5
0.09
2.0
0.08
1.0
0.06
0.5
05
0.05
0.0
10
0
(E)
1.6E-06
40
80
120
H2O2 (µM)
160
0.05
0.1
1.E-08
8.E-09
8
E 09
Conc. (M)
1.2E-06
Conc. (M)
700
1.5
0.07
5
Time (s)
600
3.0
0.1
0
(D)
500
Wavelength (nm)
kobs (s-1)
Abs (427 nm)
(B)
400
8.0E-07
4.0E-07
6.E-09
4.E-09
2.E-09
0 0E+00
0.0E+00
0 E+00
0.E+00
0
2
Time (s)
4
0
Time (s)
Figure 1 Vlasits et al.
52
Figure 1. Reaction between compound I of wild-type Synechocystis KatG with hydrogen peroxide in the
presence of cyanide. (A) Spectral changes upon mixing 3 µM compound I (pre-formed with 200 µM PAA &
delay time of 5 s) with 100 µM H2O2 and 1 mM NaCN in 50 mM phosphate buffer, pH 6.5 and 20°C.
Spectra were recorded 1.3 ms, 100 ms, 280 ms, 430 ms, 880 ms, and 2.2 s after mixing. (B) Typical time
trace at 427 nm and fitted using a single-exponential equation. Conditions: 1.5 µM KatG, 100 µM H2O2, 1
mM NaCN. (C) Plot of the pseudo-first order rate constants versus the concentration of hydrogen peroxide.
(D) & (E) Calculated changes in concentration of redox intermediates involved the consecutive reaction
compound I  ferric KatG  cyanide complex. Conditions as in (B).
In case of Asp152Ser, formation of the LS complex by addition of hydrogen
peroxide and cyanide to compound I was biphasic (Supplemental Figure). Supplemental
Figure B shows the time trace at 427 nm fitted with a single- (thin line) and a doubleexponential (thick line) equation. In the latter case the second term was independent of the
concentration of H2O2. With both fits very similar k2 values could be obtained: (3.4 ± 0.1)
× 103 M-1 s-1 (~28% of wild-type rate).
Additionally, we have probed the reaction of BLC compound I pre-formed with
PAA and hydrogen peroxide in the presence of cyanide. Within milliseconds the BLC
cyanide complex was formed with typical peaks at 420 nm, 550 and 585 nm, respectively
(Figure 2A). Similar to KatG the rate of complex formation strongly depended on the H2O2
concentration and the time traces could be fitted to a single-exponential equation (Figure
2B). This allowed to calculate apparent k2 to be 1.1  107 M-1 s-1 (Figure 2C) which
compares with a published rate calculated from steady-state data (3.2  107 M-1 s-1 [19]).
Despite the fact that in case of BLC k2 > kon, compound I reduction was the rate-limiting
step under the actual conditions. Since in BLC k2 is more than two orders of magnitude
faster than in KatG, the used H2O2 concentrations (< 15 µM) were low (at least 70times
lower than cyanide concentration of 1 mM). The experimental approach is supported by
calculations of the concentration of the redox intermediates in the time course of reaction
(Figures 2D & E). Ferric BLC did not accumulate but was effectively trapped in its LS
cyanide complex. Its highest concentration in the initial phase of reaction was calculated to
be ~0.03 µM (out of 1.5 µM KatG protein) (Figures 2D & E).
53
(A)
0.03
0.14
420
0.12
0.025
407
550
Absorbance
bance
0.1
0.02
585
0.08
0.015
0 015
0.06
0.01
0.04
0.005
0.02
0
0
300
400
500
Wavelength (nm)
210
0.07
180
150
0.062
0.054
0.046
90
0.038
30
0.03
0
0.1
Time (s)
0
0.2
(E)
1.6E-06
1.2E-06
Conc. (M)
Conc. (M)
120
60
0.0
(D)
700
(C)
0.078
kobs (s-1)
Abs (425 nm)
(B)
600
8.0E-07
3
6
9 12
H2O2 (µM)
15
18
6.0E-08
4.0E-08
2.0E-08
4.0E-07
0 0E+00
0.0E+00
0 0E+00
0.0E+00
0
0.5
Time (s)
1
0
0.02
0.04
Time (s)
Figure 2 Vlasits et al.
54
Figure 2. Reaction between compound I of bovine liver catalase (BLC) with hydrogen peroxide in the
presence of cyanide. (A) Spectral changes upon mixing 1.5 µM BLC compound I (pre-formed with 100 µM
PAA & delay time of 4 s) with 2 µM H2O2 and 1 mM NaCN in 50 mM phosphate buffer, pH 6.5 and 20°C.
Spectra were recorded immediately after mixing (1.3 ms), 22 ms, 50 ms, 73 ms, 110 ms, and after 4 s. (B)
Typical time trace at 425 nm including fit using a single-exponential equation. Conditions: 1.5 µM KatG, 2
µM H2O2, 1 mM NaCN. (C) Plot of the pseudo-first order rate constants versus the concentration of
hydrogen peroxide (2-15 µM). (D) & (E) Calculated changes in concentration of redox intermediates
involved the consecutive reaction compound I  ferric BLC  cyanide complex. Conditions as in (A).
Discussion
It has been demonstrated that mixing of compound I (prepared with PAA) of wildtype Synechocystis KatG with moderate levels of hydrogen peroxide (<100 µM) did not
induce spectral transitions simply because (i) the enzyme immediately entered the catalatic
cycle and (ii) the rate of compound I formation (k1) was apparently higher than that of
compound I reduction (k2). However, mixing millimolar concentrations of H2O2 with preformed compound I or directly with ferric KatG in the stopped-flow instrument led to the
formation of discrete LS species (absorbance maxima at 415 nm, 545 nm and 580 nm at
low pH and 418 nm and 520 nm at high pH) that represented the dominating redox species
during catalase turnover [5]. These kinetic and spectral investigations clearly demonstrated
differences in the mechanism of hydrogen peroxide dismutation mediated by catalaseperoxidases and monofunctional catalases and might concern mainly the oxidation reaction
of H2O2. However, the so far performed experiments did not allow to measure this
individual reaction since addition of hydrogen peroxide immediately prompted both heme
enzymes to cycle.
Here, based on the fact that native KatG and BLC form relatively stable cyanide
complexes with KD values in the low micromolar region and that the rates of complex
formation can easily be followed by convential stopped-flow spectroscopy, we have
decided to follow compound I reduction mediated by H2O2 in the presence of excess
cyanide. This guaranteed that (i) ferric KatG did not accumulate and was immediately
trapped in its LS complex, and (ii) the kinetics of cyanide complex formation was fully
dependent on peroxide concentration. The findings demonstrate that the kinetics of
transition of compound I to ferric KatG is about 2-3 orders of magnitude slower than the
corresponding reaction in monofunctional catalases, reflecting the differences in the
published kcat values [20]. Thus, monofunctional catalases are able to catalyze the twoelectron oxidation of H2O2 more efficiently than KatGs. Another discrepance between the
two catalatically active enzymes concerns the relative rates of k1 and k2. In KatG k2 = 1.2 
104 M-1 s-1 << k1, whereas in BLC k2 > k1.
55
Nevertheless, KatGs are the only heme enzymes with a peroxidase-typical active
site architecture that can dismutate hydrogen peroxide at reasonable rates. In close
correlation with the ability to oxidize H2O2 is the KatG-typical distal covalent adduct
Trp122-Tyr249-Met275. Its disruption, that occurs upon exchange of Trp122 or Tyr249,
converts a bifunctional KatG into a monofunctional peroxidase [12, 21]. In Trp122Phe
very high amounts of hydrogen peroxide were necessary in order to reduce compound I
and, finally, induce formation of the cyanide complex. The slow rate (30 M-1 s-1)
underlines the importance of the adduct in the catalase turnover of KatG that has been
proposed [5] and recently demonstrated by rapid freeze-quench electron paramagnetic
resonance experiments [6]. Shortly, in KatG classical compound I [oxoiron(IV) porphyrin
cation radical] is rapidly converted to an isoelectronic species by formation of the adduct
radical [i.e. oxoiron(IV) adduct radical]. The latter rapidly reacts with the second H2O2
molecule to produce oxyferrous heme [Fe(II)-O2 ↔ Fe(III)-O2●-], which decomposes to
ferric enzyme and superoxide, the latter being quantitatively oxidized at the adduct radical
closing the adduct shell and releasing dioxygen.
The chosen experimental setup does not distinguish between the individual reaction
steps during transition of compound I to ferric KatG and release of O2. But based on the
observation that (i) formation of the oxyferrous form is faster than release of dioxygen [6]
and (ii) the fact that its spectral signatures could be trapped in stopped-flow experiments
upon addition of high amounts of H2O2 to either ferric KatG or pre-formed compound I
[5], it is reasonable to assume that the calculated rates in this work reflect the kinetics of
the transition of oxyferrous KatG to the ferric form. In conventional (monofunctional)
peroxidases the decay of the oxyferrous (i.e. compound III) form is very slow. In KatG the
unique function of the adduct radical is to promote the oxidation of superoxide radical to
3
O 2.
Another structural feature that enables KatG to be catalatically active is the
architecture of the substrate channel. In contrast to typical peroxidases KatGs have a more
deeply buried active site, and the proposed access route for hydrogen peroxide is provided
by a channel that is similar but longer and more restricted than that in other heme
peroxidases, very similar to monofunctional catalases [1, 14, 20]. Aspartate 152 is fully
conserved in KatG and is situated in the heme distal cavity at the main access channel
entrance. It has been suggested that it might modulate the H2O2 entry into the active site
and its substitution by serine has significantly reduced the overall catalatic activity. This is
56
confirmed by the present study that showed a diminished (28% of wild-type) rate of
transition of compound I of Asp152Ser to its native state.
The last variant chosen for this study was Glu253Gln because – despite its location
far away from the active site at the entrance of the access channel – it has been
demonstrated that it also exhibits decreased overall dismutation activity compared to the
wild-type protein [14]. Interestingly, this was also reflected by the present study. Exchange
of Glu253 had an effect on the rate of reduction of compound I to ferric Glu253Gln most
probably by partially impaired substrate (electron donor) delivery. The acidic residue is
conserved in KatGs and helps in maintenance of a (rigid) H-bonding network. The latter is
a precondition for catalysis of H2O2 dismutation [14].
Table 1 and Figure 3 demonstrate that both the apparent rate constants for
compound I reduction obtained in this transient kinetics study as well as the turnover
numbers (kcat) calculated from steady-state experiments follow the same order: wild-type
KatG > Glu253Gln > Asp152Ser >> Trp122Phe. The discrepancies in absolute percentage
values related to wild-type KatG (100%), i.e. the fact that the impact of the mutations on
the overall turnover numbers is always more pronounced than on k2 values, suggest that
effects of mutation(s) on substrate delivery might concern also compound I formation (k1).
In any case, the suggested method for the first time allows direct monitoring of the second
half of the dismutation reaction of a catalase and is a valuable tool to study structurefunction relationships in these unique bifunctional enzymes but also in monofunctional
catalases.
57
58
koff
(H2O2)
k2
0%
~0
200 s-1
D152S
W122F
3500 s-11
E253Q
WT
kcat
890 s-1
100 %
(B)
0.003 × 104 M-1 s-1
0.34 × 104 M-1 s-1
0 86 × 104 M-1 s-1
0.86
1 2 × 104 M-11 s-11
1.2
k2
W122F
D152S
E253Q
WT
0%
100 %
Figure 3. (A) Reaction sequence that provides the basis for the experimental setup and (B) schematic comparison of catalatic turnover
numbers and apparent bimolecular oxidation rates of H2O2 of wild-type KatG (100%: kcat = 3500 s-1 and k2 = 1.2  104 M-1 s-1) from
S
Synechocystis
h
andd selected
l
d variants.
i
Compound I
(PAA)
k1
KatG[Fe(III)]
kon
KatG[Fe(III)−CN¯]
(A)
Achnowledgements
This work was supported by the Austrian Science Fund FWF Projects No. 18751 & 20996.
References
[1]
M. Zamocky, P.G. Furtmuller, C. Obinger, Evolution of catalases from bacteria to
humans, Antioxid. Redox. Signal. 10 (2008) 1527-1548.
[2]
S. Chouchane, I. Lippai, R.S. Magliozzo, Catalase-peroxidase (Mycobacterium
tuberculosis KatG) catalysis and isoniazid activation, Biochemistry 39 (2000)
9975-9983.
[3]
R.A. Ghiladi, K.F. Medzihradszky, P.R. Ortiz de Montellano, Role of the Met-TyrTrp cross-link in Mycobacterium tuberculosis catalase-peroxidase (KatG) as
revealed by KatG(M255I), Biochemistry 44 (2005) 15093-15105.
[4]
A. Ivancich, C. Jakopitsch, M. Auer, S. Un, C. Obinger, Protein-based radicals in
the catalase-peroxidase of Synechocystis PCC6803: a multifrequency EPR
investigation of wild-type and variants on the environment of the heme active site,
J. Am. Chem. Soc. 125 (2003) 14093-14102.
[5]
C. Jakopitsch, J. Vlasits, B. Wiseman, P.C. Loewen, C. Obinger, Redox
intermediates in the catalase cycle of catalase-peroxidases from Synechocystis PCC
6803, Burkholderia pseudomallei, and Mycobacterium tuberculosis, Biochemistry
46 (2007) 1183-1193.
[6]
J. Suarez, K. Ranguelova, A.A. Jarzecki, J. Manzerova, V. Krymov, X. Zhao, S.
Yu, L. Metlitsky, G.J. Gerfen, R.S. Magliozzo, An oxyferrous heme/protein-based
radical intermediate is catalytically competent in the catalase reaction of
Mycobacterium tuberculosis catalase-peroxidase (KatG), J. Biol. Chem. 284 (2009)
7017-7029.
[7]
X. Zhao, S. Yu, K. Ranguelova, J. Suarez, L. Metlitsky, J.P. Schelvis, R.S.
Magliozzo, Role of the oxyferrous heme intermediate and distal side adduct radical
in the catalase activity of Mycobacterium tuberculosis KatG revealed by the
W107F mutant, J. Biol. Chem. 284 (2009) 7030-7037.
[8]
J. Vlasits, C. Jakopitsch, M. Schwanninger, P. Holubar, C. Obinger, Hydrogen
peroxide oxidation by catalase-peroxidase follows a non-scrambling mechanism,
FEBS Lett 581 (2007) 320-324.
[9]
B.H. Dunford, Heme peroxidases, Wiley-VCH, New York, 1999.
59
[10]
G. Smulevich, C. Jakopitsch, E. Droghetti, C. Obinger, Probing the structure and
bifunctionality of catalase-peroxidase (KatG), J. Inorg. Biochem. 100 (2006) 568585.
[11]
C. Jakopitsch, F. Ruker, G. Regelsberger, M. Dockal, G.A. Peschek, C. Obinger,
Catalase-peroxidase from the cyanobacterium Synechocystis PCC 6803: cloning,
overexpression in Escherichia coli, and kinetic characterization, Biol. Chem. 380
(1999) 1087-1096.
[12]
G. Regelsberger, C. Jakopitsch, F. Ruker, D. Krois, G.A. Peschek, C. Obinger,
Effect of distal cavity mutations on the formation of compound I in catalaseperoxidases, J. Biol. Chem. 275 (2000) 22854-22861.
[13]
C. Jakopitsch, M. Auer, G. Regelsberger, W. Jantschko, P.G. Furtmuller, F. Ruker,
C. Obinger, Distal site aspartate is essential in the catalase activity of catalaseperoxidases, Biochemistry 42 (2003) 5292-5300.
[14]
C. Jakopitsch, E. Droghetti, F. Schmuckenschlager, P.G. Furtmuller, G. Smulevich,
C. Obinger, Role of the main access channel of catalase-peroxidase in catalysis, J.
Biol. Chem. 280 (2005) 42411-42422.
[15]
C. Jakopitsch, D. Kolarich, G. Petutschnig, P.G. Furtmuller, C. Obinger, Distal side
tryptophan, tyrosine and methionine in catalase-peroxidases are covalently linked
in solution, FEBS Lett. 552 (2003) 135-140.
[16]
H.A. Heering, C. Indiani, G. Regelsberger, C. Jakopitsch, C. Obinger, G.
Smulevich, New insights into the heme cavity structure of catalase-peroxidase: a
spectroscopic approach to the recombinant synechocystis enzyme and selected
distal cavity mutants, Biochemistry 41 (2002) 9237-9247.
[17]
R.A. Ghiladi, G.M. Knudsen, K.F. Medzihradszky, P.R. Ortiz de Montellano, The
Met-Tyr-Trp cross-link in Mycobacterium tuberculosis catalase-peroxidase (KatG):
autocatalytic formation and effect on enzyme catalysis and spectroscopic
properties, J. Biol. Chem. 280 (2005) 22651-22663.
[18]
B. Wiseman, J. Colin, A.T. Smith, A. Ivancich, P.C. Loewen, Mechanistic insight
into the initiation step of the reaction of Burkholderia pseudomallei catalaseperoxidase with peroxyacetic acid, J. Biol. Inorg. Chem. 14 (2009) 801-811.
[19]
B. Chance, The primary and secondary compounds of catalase and methyl or ethyl
hydrogen peroxide; reactions with hydrogen peroxide, J. Biol. Chem. 180 (1949)
947-959.
60
[20]
J. Switala, P.C. Loewen, Diversity of properties among catalases, Arch. Biochem.
Biophys. 401 (2002) 145-154.
[21]
C. Jakopitsch, M. Auer, A. Ivancich, F. Ruker, P.G. Furtmuller, C. Obinger, Total
conversion of bifunctional catalase-peroxidase (KatG) to monofunctional
peroxidase by exchange of a conserved distal side tyrosine, J. Biol. Chem. 278
(2003) 20185-20191.
61
(A)
0.03
410
0.12
406
415
0.025
0.1
0.02
600
Absorbance
ce
0.08
640
0.015
540
0.06
0.01
0.04
0.005
0.02
0
0
-0.005
0 005
300
(B)
400
500
600
Wavelength (nm)
(C)
0.09
25.0
20.0
0.08
kobs (s-1)
Abs (427 nm)
0.085
700
0.075
0.07
15.0
10.0
0.065
5.0
0.06
0.0
0.055
0
0.2
0.4
Time (s)
0.6
0.8
0
2
4
6
H2O2 (mM)
Supplemental Figure. Reaction of Synechocystis KatG variant Asp152Ser compound I (prepared using
peracetic acid) with hydrogen peroxide and cyanide. (A) Spectral changes upon mixing 1.5 µM compound I
with 10 mM H2O2 in 1 mM NaCN.
NaCN Spectra were recorded immediately after mixing (1.3
(1 3 ms,
ms bold black
line), 32 ms (thin black line), 820 ms (dotted line), 3.1 s (thin grey line), and after 10 s (thick grey line).
Conditions: 50 mM phosphate buffer at pH 6.5 and 20°C. (B) Time trace of the reaction in A, monitored at
427 nm and fitted using a single-exponential (thin line) and a double-exponential (bold line) equation. (C)
Plot of the pseudo-first order rate constants, obtained from the double-exponential fit in B, versus the
62
chapter four
Disrupting the H-bond network in the main access channel of
catalase-peroxidase: effect on the redox thermodynamics of the
Fe(III)-Fe(II) couple
Jutta Vlasits, Marzia Bellei, Christa Jakopitsch, Paul G. Furtmüller, Marcel Zamocky,
Marco Sola, Gianantonio Battistuzzi and Christian Obinger
Submitted to FEBS Journal
Disrupting the H-bond network in the main access channel of catalaseperoxidase: effect on the redox thermodynamics of the Fe(III)-Fe(II) couple
Jutta Vlasits,1 Marzia Bellei,2 Christa Jakopitsch,1 Francesca De Rienzo,2,3 Paul G.
Furtmüller,1 Marcel Zamocky,1 Marco Sola,2,3 Gianantonio Battistuzzi,2 and Christian
Obinger1*
1
Department of Chemistry, Division of Biochemistry, BOKU – University of Natural
Resources and Applied Life Sciences, Muthgasse 18, A-1190 Vienna, Austria
2
Department of Chemistry, University of Modena and Reggio Emilia, via Campi 183,
41100 Modena, Italy
3
INFM-CNR National Center on nanoStructures and bioSystems at Surfaces (S3), Via
Campi 213/A, 41100 Modena, Italy
*corresponding author: phone: +43-1-36006-6073; fax: +43-1-36006-6059; email:
[email protected]
64
Abbreviations: KatG, catalase-peroxidase; CcP, cytochrome c peroxidase; APX,
ascorbate peroxidase; MPO, myeloperoxidase; HRP-C, horseradish peroxidase isoform C;
ARP, Athromyces ramosus peroxidase; E°', standard reduction potential (pH 7, 25°C); SHE,
standard hydrogen electrode; H°'rc, enthalpy change for the reaction center upon reduction of
the oxidized protein; S°'rc, entropy change for the reaction center upon reduction of the
oxidized form; RR, resonance Raman; Fe-Im, iron-imidazole; ASA, accessible surface area.
Keywords:
catalase-peroxidase;
catalase
activity;
reduction
potential;
redox
thermodynamics; enthalpy; entropy; heme cavity architecture; access channel; protein
modeling.
65
Abstract
Catalase-peroxidases are the only heme peroxidases with substantial hydrogen peroxide
dismutation activity. In order to understand the role of the redox chemistry in the bifunctional
activity of these metalloproteins, catalatically active and inactive mutant proteins have been
probed in spectroelectrochemical experiments. In detail, in a comparative study wild-type
KatG from Synechocystis has been compared with variants with (i) disrupted KatG-typical
adduct (Trp122-Tyr249-Met275), (ii) mutation of the catalytic distal His123-Arg119 pair, (iii)
altered channel accessibility to the heme cavity (Asp152, Ser335) and modified charge at the
channel entrance (Glu253). A valuable insight into the mechanism of reduction potential (E°')
modulation in KatG has been obtained from the factorization of the corresponding enthalpic
and entropic components, determined from the analysis of the temperature dependence of E°'.
Moreover, the model structures of the Fe(III) and Fe(II) forms of wild-type KatG from
Synechocystis were computed using the available X-ray structures of KatGs from different
origins as templates. Obtained results, discussed together for the published resonance Raman
data on the strength of the proximal iron-imidazole bond as well as with catalytic properties of
the corresponding mutants, clearly demonstrate that the absolute value of the midpoint
potential of the Fe(III)/Fe(II) couple has no impact on catalase and peroxidase activity.
Besides the role of the Trp-Tyr-Met adduct as radical site in catalysis, it is the architecture of
the heme cavity and of the long and constricted substrate channel that distinguishes KatGs
from monofunctional peroxidases. The role of maintenance of an ordered matrix of oriented
water dipoles for H2O2 oxidation and the effect of its disruption is clearly seen by
modification of enthalpic and entropic contributions to E°' that reflect changes in polarity,
electrostatics, continuity and accessibility of solvent to the metal center as well as solvent
reorganization effects upon reduction.
66
Introduction
Catalase-peroxidases (KatGs) have raised considerable interest, since, despite having
striking sequence homologies with heme peroxidases like ascorbate peroxidase (APX) or
cytochrome c peroxidase (CcP) [1], they exhibit substantial catalase activity similar to
monofunctional catalases. Thus, they are ideal model oxidoreductases to study structural
modifications that enable a peroxidase to efficiently dismutate hydrogen peroxide. The
available crystal structures of KatGs from Haloarcula marismortui (pdb code 1ITK),
Burkholderia pseudomallei (pdb code 1MWV), Mycobacterium tuberculosis (pdb code 1SJ2),
and Synechococcus PCC 7942 (pdb code 1UB2) [2-5] revealed that the principal organization
of their active site is very similar to those of CcP [6] and APX [7, 8], including the proximal
(His290-Trp341-Asp402; Synechocystis PCC 6803 numbering) and the distal (Arg119Trp122-His123) triad (Figure 1). Unique to KatG and essential for catalatic activity (that is
completely absent in both APX and CcP) is a highly conserved peculiar distal site adduct that
includes Trp122-Tyr249-Met275 [2-5, 9] and plays a role as transient radical site in the
catalatic turnover [10, 11]. In addition, at variance with typical peroxidases, but similar to
monofunctional catalases, KatGs have a longer and more restricted access channel to the
deeply buried heme b. KatG-specific large loop insertions build up this channel and also
connect the distal and proximal domains. Large loop 1 (LL1) is of particular interest, since it
contains a number of highly conserved residues that are essential for efficient H2O2 oxidation.
These include Tyr249 (part of the covalent adduct), Ile248 and Glu253, the latter creating an
acidic entrance to the channel. Isoleucine 248 is hydrogen-bonded to KatG-specific Asp152
that, together with Ser335, forms the narrowest part of the channel and controls access to the
distal heme cavity (Figure 1). Resonance Raman data and structural analysis demonstrated the
existence of a very rigid and ordered structure built up by the interaction of these residues
with distal and (via LL1) proximal amino acids, with the heme itself, and with the solute
matrix in the channel. Disruption of these interactions typically affected the catalase but not
the peroxidase activity of KatG [12, 13]. Here, we address the question to which extent the
protein matrix and the ordered matrix of oriented water dipoles contribute to the reduction
potential of the heme iron and in consequence to the catalatic activity of KatG. We have
investigated the thermodynamics of the one-electron reduction, E°', of the ferric heme in wildtype KatG from Synechocystis and in variants that lack (i) the KatG-typical adduct
67
(Trp122Phe, Tyr249Phe), (ii) the peroxidase-typical catalytic residues (Arg119Asn,
His123Gln), as well as (iii) in variants involved in stabilization of the solute matrix of the
access channel (Asp152Ser, Ser335Gly and Glu253Gln). Analysis of the temperature
dependence of the corresponding reduction potentials, allowed factorization of the
corresponding enthalpic and entropic components and in consequence of the protein- and
solvent-based contributions to E°'. Moreover, we have computed the model structures of the
Fe(III) and Fe(II) forms of wild-type KatG from Synechocystis using the available X-ray
structures of KatGs as templates [2-5]. Obtained findings are compared with data from
enzyme kinetics of the corresponding mutants as well as with those known from
monofunctional peroxidases and, finally, discussed with respect to the unique H2O2
dismutation activity of KatG.
Results
The final model of catalase-peroxidase from Synechocystis PCC 6803 includes residues
49-754. Residues 1−48 are excluded since, in the multiple sequence alignment used
(Supplemental Figure S1), they correspond to residues which are not resolved in the X-ray
structures of the catalase-peroxidases used as templates. This N-terminal segment is
presumably very flexible and is not conserved in the catalase-peroxidases family [14], as
shown by the sequence alignment (see Supplemental Figure S1). Omitting this portion is
unlikely to be relevant for the protein function.
As expected from both the high sequence identities (55%-72%) and sequence
similarities (78%-88%) between target and templates, the fold of the modelled structure of
Synechocystis catalase-peroxidase is very similar to that of the templates (rmsd range: 5.6 –
10.2 Å) and shows two structurally similar domains (Figure 1A), which are mainly formed by
α-helices: the N-terminal domain (residues 49−461) and the C-terminal domain (residues
462−754). Although having only 19% sequence identity (and about 30% sequence similarity),
the two domains share a similar fold (the secondary structure elements of the N-terminal
segment 88-439 superimpose to those in the C-terminal segment 475-746 with a 1.2 Å rmsd),
which is common in the members of the bacterial and plant peroxidase families, including
cytochrome c peroxidase [6], ascorbate peroxidase [7, 8], and horseradish peroxidase C [15].
The topological arrangement of the α-helices and β-sheets is largely conserved. In particular,
68
the conformations of both KatG-specific large loops LL1 and LL2 are fully conserved [2]. The
main secondary structure elements found in the modelled structure are almost consistent with
those predicted by the PredictProtein web service (see Supplemental Figure S1).
Figure 1. (A) Cartoon representation of the 3D modeled structure of KatG from Synechocystis PCC 6803. (B)
Focus on the oxidized (left) and reduced (right) active sites of the Synechocystis PCC 6803 KatG models. The
most relevant residues and the water molecules in the active site cavity are highlighted and labeled. Code:
arrows: β-strands; barrels: α-helices; red liquorice: heme group; bold liquorice: protein site residues; thin
liquorice: water molecules; cyan: C atoms; blue: N atoms; red: O atoms; yellow: S atoms. This picture was
drawn with the VMD software [73].
The single heme b group of the protein is buried inside the N-terminal domain and can
be reached by the substrate through a narrow hydrophilic channel that prevents access of large
substrates. In the C-terminal domain, the cleft corresponding to the heme site is occupied by
the C-terminal loop 589-619. The extension of the active site cavity and the architecture of the
heme site are similar to those of the templates and other class I peroxidases [2]. The conserved
69
key residues (R119, W122, and H123 in the distal pocket, H290, W341, and D402 in the
proximal pocket, and S335) show an arrangement closely similar to that observed in the X-ray
structures of catalase-peroxidases and class I peroxidases (Figure 1). The same holds for the
distal covalent adduct formed by W122, Y249 and M275, which is peculiar to the catalaseperoxidase family [2]. The H-bond network and the arrangement of the water molecules
within the active site cavity in the modelled structure of ferric Synechocystis KatG are
consistent with those observed in the catalase-peroxidase templates (Figure 2).
The ferric and ferrous forms are highly similar: the overall rmsd is 0.10 Å, while the
rmsd computed on their Cα chain is 0.04 Å. The computed rmsd values - taking into account
only the heme and the conserved residues in the active site is 0.14 Å - clearly indicate that the
most relevant structural differences between the models of the two redox states of the protein
are found in the active site region (Figure 1B). These small local structural modifications do
not significantly alter the solvent accessibility of the heme group, whose water accessible
surface area (ASA) is similar in the reduced and in the oxidized protein (31Å2 versus 30Å2,
respectively), as is its subdivision into hydrophobic and hydrophilic components (80% and
20% for the oxidized protein; 77% and 23% for the reduced protein). On the other hand, the
water accessible surface area of the whole distal active site (including the heme and the
residues R119, W122, H123, D152, Y249, which constrict the active site cavity) slightly
increases upon Fe(III) reduction, passing from 77 Å2 to 81 Å2. Also the distribution of the
hydrophilic and hydrophobic components is slightly altered (passing from 48% hydrophobic
and 52% hydrophilic in the ferric protein to 43% hydrophobic and 57% hydrophilic in the
ferrous enzyme), as a consequence of the small reduction-induced displacement of the side
chains of residues H123, R119 and D152 (Figure 1B). Although the ASAs of the distal cavity
in the two redox forms are comparable, the water molecule distribution in the distal cavity
slightly changes upon Fe(III) reduction. In particular, the water molecules closest to the heme
shift slightly away from the metal center (Figure 1B), in agreement with a decreased
electrostatic interaction with the ferrous ion, thereby slightly increasing their H-bonding
interactions with the aminoacids inside the distal cavity (Figure 2).
70
(A)
His123
Asp152
Arg119
Trp122
Met275
W13
W15
Tyr249
W11
W12
W17
W9
W10
W14
W16
W18
Ser335
(B)
W13
Asp152
W8
W7
W12
W15
W3
W10
W9
W14
W6
W5
Ser335
W4
W1
W2
Asn251
Pro252
Glu253
(C)
W13
W12
W8
W9
W11
W7
W5
W6
W10
W4
W3
W2
W1
Figure 2, Vlasits et al.
71
Figure 2. (A) Detailed view of the active site of KatG from Burkholderia pseudomallei (Synechocystis numbers
in parentheses) with the interacting water (W) molecules starting at the constriction of the channel. Water
molecules are numbered 9 – 18 and correspond to the following structural water molecules of Burkholderia. W9,
position 3086; W10, position 3180; W11, position 2653; W12, position 2511; W13, position 3138; W14, position
3071; W15, position 2764; W16, position 3897; W17, position 3085; W18, position 2229. (B) Detailed view of
the main channel showing the residues investigated in this study, including the amino acids asparagine and
proline, and the interacting water molecules. W9, W10 and W12 – W15 correspond to the positions in A. W1,
position 3558; W2, position 2472; W3, position 2860; W4, position 3149; W5, position 3620; W6, position 4107;
W7, position 3898; W8, position 3770. (C) Surface structure of the main cannel of KatG. Green, heme; blue,
Ser324 (Ser335); cyan, Asp141 (Asp152); yellow, Glu242 (Glu253); red, water molecules; W1 – W8 correspond
to the positions in B and W9 – W13 to the positions in A. The figures were constructed using the coordinates
deposited in the Protein Data Bank (accession code 1MWV).
The spectroscopic properties of ferric wild-type recombinant KatG from Synechocystis
are indicative of predominant five-coordinate high-spin heme with the Soret band at 407 nm,
Q-bands at 502 and 545 nm and a CT1 band (i.e. a porphyrin-to-metal charge transfer band) at
635 nm, very similar to those reported in the literature [13]. For comparison and in order to
guarantee identical conditions, the standard reduction potential, E°', of the Fe(III)/Fe(II)
couple of wild-type protein has been re-measured. Figure 3A depicts a representative family
of spectra of ferric KatG at different applied potentials in the OTTLE cell. Ferric KatG is
directly reduced to its ferrous form (spectral bands at 438 nm and 557 nm) with a clear
isosbestic point at 420 nm. The calculated midpoint reduction potential for the Fe(III)/Fe(II)
couple, determined from the corresponding Nernst plot (inset to Figure 3A), was identical to
the value published recently (i.e. –0.226 ± 0.005 V, 25°C and pH 7.0) [14]. Similarly, the
ferric mutant proteins were exposed to different potentials and the one-electron reduction was
followed spectroscopically. As a representative example, the spectral transition of the variant
Asp152Ser that mimics CcP and exhibits significantly lower catalase activity compared to the
wild-type protein [15] is shown in Figure 3B. Ferric Asp152Ser (Soret: 406 nm, CT1: 640 nm)
is directly converted (isosbestic point at 417 nm) to its ferrous form (spectral bands at 438 nm
and 557 nm). Similar to the wild-type protein and all other mutants investigated in this work,
the slope of the linear Nernst plot was close to the theoretical value of RT/F = 0.059 V at 25°C
and thus consistent with a one-electron redox reaction. The determined E°' of the Fe(III)/Fe(II)
couple of Asp152Ser was -0.208 ± 0.010 V (25°C and pH 7.0), 20 mV more positive than the
wild-type protein. For details of spectral transitions, Nernst plots and calculation of E°' see
Supplemental Figure S2. Table 1 summarizes the E°' values of the other variants.
72
73
-0.226
-0.222
-0.319
-0.305
-0.208
-0.226
-0.226
-0.306
-0.183
+0.005
KatG W122F
KatG Y249F [14]
KatG H123Q
KatG R119N
KatG D152S
KatG S335G
KatG E253Q
HRP-C [16]
ARP [17]
MPO [18]
*data from this work
-0.226
25°C
(= -ΔH°'rc,int/F)
KatG WT [14]
Protein
E°' (V)
+3
-18
+91
+14
+4
+8
+104
+81
+9
+6
+10
-120
+210
-27
-61
-40
+247
+166
-44
-51
-18
(J mol-1 K-1)
(KJ mol-1)
+17
ΔS°'rc
ΔH°'rc
-0.031
+0.187
-0.943
-0.145
-0.041
-0.083
-1.078
-0.840
-0.093
-0.062
-0.176
(V)
- ΔH°'rc/F
+0.031
-0.371
+0.648
-0.083
-0.188
-0.124
+0.763
+0.513
-0.136
-0.158
-0.056
(V)
TΔS°'rc/F
246 [42]
[40, 41]
211, 230
243 [39]
203, 251 [12]
--------------------
[24]
203, 237, 250
202, 249 [23]
202, 249 [23]
203, 251 [13]
202, 251 [23]
203, 251 [23]
(cm-1)
ν(Fe-Im)
-
-
-
0.8 [12]
2.1*
2.4 [15]
60 [26]
16 [26]
5 [9]
n.d. [26]
4.1 [26]
(mM)
KM
-
-
-
890 [12]
2500*
200 [15]
17 [26]
0.83 [26]
6 [9]
n.d. [26]
3500 [26]
(s-1)
kcat
Table 1. Thermodynamic parameters for Fe(III)Fe(II) reduction in wild-type Synechocystis KatG and several variants. For comparison purposes, the
corresponding thermodynamic parameters of horseradish peroxidase isoform C (HRP-C), Arthromyces ramosus peroxidase (ARP), and myeloperoxidase (MPO)
are presented. Average errors on E°', H°'rc, and S°'rc values are ±0.005 V, ±5 kJ mol-1, and ±8 kJ mol-1. In addition resonance Raman data in the lowfrequency region as well as kinetic parameters of the catalase reaction are presented.
Principally, the reduction potentials of Trp122Phe, Tyr249Phe, Ser335Gly and
Glu253Gln were found to be more or less identical (-0.222 to -0.226 V) with that of the wildtype protein.
E (V)
-0.14
-0.24
0.246
-0.29
-1.05 -0.45 0.15 0.75
Log X
0.126
-0.13
0.286
0.086
Absorbance
Absorbance
0.166
(B)
-0.19
E (V)
(A) 0.206
-0.17
-0.21
-0.25
0.206
-0.3
0.166
0.1
0.5
Log X
0.9
0.126
0.086
0.046
0.046
0.006
0.006
322
342
362
382
402
422
442
Wavelength (nm)
462
482
322
342
362
382
402
422
442
462
482
Wavelength (nm)
Figure 3. (A) Electronic spectra of wild-type catalase-peroxidase from Synechocystis PCC 6803 obtained at
various applied potentials at 25 °C. (B) Electronic spectra of the variant, KatG(D152S) measured at 25°C. The
insets depict the corresponding Nernst plots, were X represents [(Aλredmax – Aλred)/(Aλoxmax – Aλox)] with λox = 406
nm and λred = 438 nm for wild-type KatG, and λox = 406 nm and λred = 436 nm for KatG(D152S).
By contrast, E°' of both His123Gln (-0.319 ± 0.005 V) and Arg119Asn (-0.305 ± 0.005 V)
exhibited a significantly more negative reduction potential than wild-type KatG.
To gain a deeper insight into the mechanism of E°' modulation, the temperature
dependence of the reduction potential was investigated (Figure 4). This allows factorization of
the corresponding enthalpic (H°'rc) and entropic (S°' rc) components of the reduction
reaction. In all investigated proteins, the oxidized state was enthalpically stabilized over the
reduced state, although the relative contributions of H°'rc were different. Without
consideration of His123Gln and Arg119Asn, the enthalpic contribution to the negative E°'
value was in the range of -41 mV to -176 mV following the hierarchy Ser335Gly <
Trp122Phe < Asp152Ser < Glu253Gln < wild-type KatG (Table 1). In both His123Gln and
Arg119Asn the enthalpic contribution to the negative E°' value was very high (-840 mV and 1078 mV, respectively).
With the exception of His123Gln and Arg119Asn, reduction of wild-type KatG and the
other mutant proteins led to a decrease in entropy, with S°'rc values ranging from -18 to -44 J
mol-1 K-1, thus contributing -56 to -188 mV (Table 1) to the negative E°' value of these
metalloproteins and following the opposite hierarchy compared to enthalpic factors:
74
Ser335Gly > Trp122Phe > Asp152Ser > Glu253Gln > wild-type KatG (Table 1). By contrast,
reduction of ferric His123Gln and Arg119Asn is entropically favored. The corresponding
entropic contributions (+513 mV and +763 mV, respectively) partially compensate the
enthalpic stabilization of the ferric state in these two mutant proteins.
-0.20
E°' (V)
-0.25
-0.30
A
-0.35
10
20
30
40
T (°C )
E°'/T x 10-3 (V K-1)
-0.6
Figure 4. (A) Temperature dependence of the
reduction potential and (B) E°'/T versus 1/T plots
for wild-type Synechocystis PCC 6803 KatG ()
and for its Y249F (), W122F (), D152S (),
S335G (), E253G (), H123Q () and R119N
() variants. The slope of the plots yields the
S°'rc/F and – H°'rc/F values, respectively,
respectively. Solid lines are least square fits to the
data points. Experimental conditions: 0.02 mM and
0.03 mM protein, respectively, in 50 mM phosphate
buffer and 10 mM NaCl at pH 7.0.
-0.8
-1.0
B
-1.2
3 .2
3.3
3.4
1/T x 10
-3
3 .5
3.6
-1
(K )
75
Discussion
The modelled structure of Synechocystis KatG is very similar to those of the
catalase-peroxidases used as templates, as expected from the high sequence identities
(55%-72%) and similarities (78%-88%) existing between the proteins. The structural
similarity extends to the residues belonging to the heme cavity (R119, W122, H123 in the
distal zone; H290, W341, D402 in the proximal pocket; S335 and W122, Y249, M275, the
latter forming the distal covalent adduct) and to the local H-bond network, which involves
several conserved water molecules, as well as to the KatG-specific large loop insertions,
which build up the access channel to the deeply buried heme b and connect the distal and
proximal domains.
For the purpose of this work, the comparison of the modelled structures of the ferric
and ferrous forms of Synechocystis KatG is particularly relevant, since it clearly suggests
that electron uptake by the ferric heme causes very limited structural changes, which are
mainly limited to a slight rearrangement of the residues and of the water molecules in the
distal cavity, inducing a slight increase of its solvent accessibility and of the hydrophilic
component of the calculated ASAs (from 52% to 57%). This effect is opposite to that
observed in the native form and the six-coordinate adducts of HRP, in which Fe(III)
reduction induces a sensible decrease in ASA of the heme group and of the distal residues
[18]. These differences are reflected by the opposite sign of the corresponding S°'rc
values (Table 1). The positive reduction entropy of HRP has been attributed to a decreased
solvent ordering upon reduction, in agreement with a reduced electrostatic interaction of
the ferrous heme with the water molecules in the distal cavity as compared to the ferric
enzyme and the lower solvent accessibility of the heme group and of the distal residues in
the ferrous form is consistent with this picture [18, 19]. On the contrary, the negative
S°'rc, featured by wild-type Synechocystisis KatG indicates a reduction-induced increase
in the ordering of the water molecules in heme cavity (see below), which fits with the
enhanced polarity of the distal site indicated by ASA calculation and with the increased Hbonding interactions with the distal residues.
Table 1 compares the reduction thermodynamics of wild-type Synechocystis KatG
and of the seven mutant proteins with recent data obtained from other (catalaticallyinactive) heme peroxidases, namely horseradish peroxidase isoform C (HRP-C) [16],
Athromyces ramosus peroxidase (ARP) [17] and myeloperoxidase (MPO) [18]. In
76
addition, it correlates these data with resonance Raman investigations of the ironimidazolate bond strength and kinetic paramaters of the catalase activity.
Since reduction thermodynamics for metalloproteins in aqueous solution are deeply
affected by solvent reorganization phenomena [21-28], the thermodynamic data listed in
Table 1, as such, cannot be used to extract information on mutation-induced protein-based
effects on the thermodynamics of Fe(III) → Fe(II) reduction. In particular, it is worthy of
note that for all the present KatG variants the mutation-induced changes in ΔH°'rc and
ΔS°'rc are much larger than the corresponding changes in ΔG°'rc, and therefore in E°' (Table
1). This is typical of the presence of enthalpy-entropy compensation phenomena occurring
within a series of homologous species subjected to the same reaction at fixed temperature
[21-25, 29-32], according to equation 1 [29-32]:
Hi = a+ b (TiSi)
(1)
where a and b are constants [21-25, 29-32]. The compensation plot in which the
entropic contributions to ∆G°'rc at 298 K (TΔS°'rc) are plotted against the corresponding
enthalpic terms (-ΔH°'rc) for wild-type KatG and all the mutants investigated is linear
(Figure 5), with a regression coefficient (r2) of 0.999 and a slope close to unit (0.9),
indicative of an high degree of compensation. Such H-S compensation indicates that the
reaction enthalpies and entropies for the Fe(III) → Fe(II) reduction within the series are
dominated by solvent reorganization effects due to reduction-induced changes in the
hydrogen bonding network connecting the water molecules within the hydration sphere of
the protein [21-25, 29-32]. Since the above solvent reorganization effects are fully
compensatory at any temperature (H°'rc(solv) = TΔS°'rc(solv)) [29-32], it turns out that they
do not affect the free energy of Fe(III) → Fe(II) reduction. Therefore ΔG°'rc (-nFE°') in
native and mutated KatG is totally determined by protein-based effects.
ΔG°'rc (= -nFE°') = ΔH°'rc(int) + TΔS°'rc(int)
(2)
If we assume, to a first approximation, that the protein-based entropy changes related
to differences in conformational degrees of freedom are negligible, then the free energy of
Fe(III) to Fe(II) reduction and the reduction potential of the Fe(III)/Fe(II) couple in native
and mutated KatGs are determined by the protein-based selective enthalpic stabilization of
one of two redox states of the heme:
ΔG°'rc = H°'rc(int)
(4)
E°' = -H°'rc(int)/F
(5)
77
Hence, any deviation from the perfect H-S compensation, namely any mutationinduced change in E°', must arise from protein-based molecular factors, such as alterations
in the coordination features of the heme iron and changes in the electrostatics at the
interface between the heme and the protein and the solvent (the latter as a bulk effect).
The negligible role of protein-based entropic factors in defining the E°' values of the
present series of KatG mutants is confirmed by comparison of the 3D structures calculated
for the ferric and ferrous forms of the wild-type Synechocystis enzyme, which
demonstrates that Fe(III) reduction induces very limited changes in the overall protein
structure, as indicated by the small rmsd values between the two redox states (see above).
The same behavior has been observed in other heme peroxidases, i.e. horseradish
peroxidase isoform C (HRP-C) [16], Athromyces ramosus peroxidase (ARP) [17] and
myeloperoxidase (MPO) [18].
1.0
0.8
T298S°'rc/F (V)
0.6
0.4
0.2
0.0
-0.2
-1.2
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
-H°'rc/F (V)
Figure 5. Enthalpy/entropy compensation plot at 298 K for the reduction thermodynamics of wild-type
Synechocystis PCC 6803 KatG and its Y249F, W122F, D152S, S335G, E253G, H123Q and R119N variants.
Solid line is the least-squares fit to the data points.
KatG, HRP-C and ARP belong to the same peroxidase superfamily [19], that
comprises three classes of similar active site architecture. Besides the fully conserved
distal side catalytic residues His and Arg (His123 and Arg119, see Figure 2A), all
members have the proximal histidine (His259) hydrogen-bonded with an aspartate
(Asp402). The latter deprotonates the Nδ position of the proximal His thereby providing a
strong axial ligand with imidazolate character. It increases the electron density at the heme
iron and guarantees negative E°' values of the couple Fe(III)/Fe(II) that are a precondition
for efficient reaction with H2O2 and for prevention of binding of 3O2. Moreover, it helps to
78
maintain the iron-imidazole bond in optimized strength and geometry. In Class I
peroxidases (APXs, CcPs and KatGs) a H-bond between Asp402 and a conserved
tryptophan (Trp341) further restricts the position of the proximal His. The importance of
this architecture for the redox chemistry of these metalloproteins has been demonstrated by
CcP mutants with disrupted His-Asp interaction. Whereas wild-type CcP has been reported
to exhibit E°' values of the Fe(III)/Fe(II) couple of -194 mV [20] and -182 mV [21],
variants with exchanged Asp (Asp235Asn, Asp235Ala, Asp235Glu) had significantly
increased (more positive) reduction potentials (-79 mV, -78 mV, -113 mV, respectively)
[21]. The corresponding value of native soybean ascorbate peroxidase was reported to be 160 mV [22]. Thus, in Class I peroxidases E°'[Fe(III)/Fe(II)] values are placed within a
relatively small potential region of 66 mV ranging from -226 mV (KatG) to -160 mV
(APX). This excludes an important impact of the proximal heme architecture on the
mechanism of H2O2 oxidation since both CcP and APX are catalatically inactive. This
hypothesis is underlined by the fact that the iron-imidazole bond strength probed by
resonance Raman (RR) spectroscopy in the low-frequency region was very similar in wildtype and mutant KatG proteins [23, 24] although the catalytic properties of the investigated
variants had a broad variability ranging from wild-type activity to catalatically more or
less inactive enzymes (Trp122Phe or Tyr249Phe) (Table 1) [9, 12, 13, 15, 25, 26]. The RR
spectrum of wild-type KatG displays two bands at 205 and 253 cm-1 that were assigned to
two (Fe-Im) stretching modes [23], with the 203 cm-1 band corresponding to a species
whose proton resides on the imidazole, while the band at 251 cm-1 corresponds to a form
where the proton is transferred to the carboxylate. Since all investigated mutants - with the
exception of Asp152Ser - showed very similar spectral bands in this region (Table 1) [24],
it can be concluded that (i) the mutations had no impact on the proximal heme architecture
and the Fe-Im bond strength, and that (ii) observed differences in E°’ and reduction
thermodynamics are related to mutational effects on the distal site and/or substrate channel
architecture. In the case of Asp152Ser, the altered low-frequency RR spectrum has been
explained by the important role of Asp152 in stabilizing the KatG-typical large loop 1
(LL1) that connects the distal side with the proximal helices E and F [13, 23, 24].
One KatG-typical structural feature, that is absent in CcP and APX, is the distal side
covalent adduct Trp122-Tyr249-Met275. Its disruption, that occurs upon exchange of
Trp122 (Trp122Phe) or Tyr249 (Tyr249Phe) [13], totally converts a bifunctional KatG
into a monofunctional peroxidase [9, 21]. However, it does not modify the standard
reduction potential of the heme iron (Table 1), which depends on the protein-based
79
enthalpic selective stabilization of the ferric form of the enzyme (see above), namely the
coordination features of the heme iron (in agreement with the RR data) and the
electrostatics at the interface between the heme and the protein and the solvent.
Since changes in conformational degrees of freedom of the polypeptide chain upon
reduction can be neglected (see above) [14, 16-18, 27-29] and mainly solvent
reorganization effects contribute to S°'rc, the increase of order of solvent upon formation
of ferrous KatG can be related to its channel architecture and H-bond network. In fact,
KatGs have a more deeply buried active site compared to monofunctional (Class III) HRPC and the proposed access route for H2O2 is provided by a channel that is similar but
longer and more constricted than that in other heme peroxidases, very similar to
monofunctional catalases [12, 33, 41]. This channel architecture is typical for
catalatically-active enzymes and its manipulation usually decreases the catalase but
increases the peroxidase activity [12, 19, 30-32]. The entropic stabilization of the ferric
form in wild-type KatG fits with a diminished solvent accessibility in bifunctional KatG
compared to monofunctional HRP, which on the contrary features a positive S°’rc value
(Table 1) [19]. The increased loss of entropy upon Fe(III) reduction in both Trp122Phe and
Tyr249Phe mutants compared to the wild-type protein clearly indicates that disruption of
the distal covalent adduct Trp122-Tyr249-Met275 sensibly alters reduction-induced
solvent reorganization effects [33]. This agrees with the role played by Trp122 in the
extended network of hydrogen bonds existing in the heme cavity, which includes water
molecules, His123, Arg119 and KatG-specific Asp152 and Glu253 (Figure 2). On the
other hand, Tyr249 belongs to the KatG-typical LL1 loop that stabilizes the heme cavity
architecture. Disruption of the adduct might readjust the LL1 loop and increase the 6coordinate low-spin state of Fe(III) [23], decreasing the electrostatic interaction of the
metal ion with the other water molecules in the heme cavity. In any case, these data clearly
show that the competence of a peroxidase to oxidize hydrogen peroxide is not related
directly to the midpoint potential of the heme iron. The role of the unique Trp-Tyr-Met
adduct in catalase activity has been demonstrated recently [10, 11]: it is transiently
oxidized during turnover and supports the rapid oxidation reaction of the second H2O2
molecule as well as the release of dioxygen [10, 11]
Asparate 152 is also fully conserved in KatG and its substitution significantly
reduced the catalase activity with little effect on peroxidase activity (Table 1) [15, 34]. It is
situated in the heme distal cavity at the main access channel entrance. The positioning of
80
its carboxylate, that is hydrogen-bonded to several water molecules, less than 6 Å from the
guanidinium side chain of Arg119 suggests that it might modulate the H2O2 entry into the
active site, working in concert with Arg119 [34]. Removal of the negative charge made
Asp152Ser structurally and spectroscopically similar to CcP [23] and modifies the
proximal Fe-Im bond strength most probably due to disabled interaction with Ile248 that is
part of LL1 [24]. Simultaneously, H-bonded water molecules might be released upon
mutation [23].
In this context, the increased E°' of the Asp152Ser mutant compared to the wild-type
KatG [E°’ (E°’mut-E°’nat) = +18 mV, see Table 1] agrees with the decreased basicity of
the proximal His as well as with the deletion of a negative charge near the heme and with a
decreased polarity of the distal heme cavity, which reduces enthalpic stabilization of ferric
form of the enzyme due to ligand-binding and electrostatic interactions, respectively
(Table 1).
Asp152Ser features a S°'rc even more negative than wild-type KatG. Upon
exchange of the carboxylate group H2O molecules are released [24] and the continuous
solvent matrix is disturbed. As a consequence, reduction-induced solvent reorganization is
more restricted than in wild-type KatG. The fact that in a corresponding Burkholderia
mutant the KM for H2O2 oxidation was significantly increased [34] underlines the
importance of a stable continuum of water molecules in the substrate channel for efficient
delivery and binding of hydrogen peroxide.
Mutants of serine 335 (Ser315 in Mycobacterium tuberculosis) play a key role in
isoniazide-resistant strains in treatment of tuberculosis. As observed with the Trp122Phe
and Tyr249Phe variants, the Ser335Gly mutation does not influence the E°' value, but
induces a more negative S°'rc compared to wild-type KatG (Table 1), clearly
demonstrating it does not modify the protein-based enthalpic stabilization of the ferric
enzyme, although it sensibly alters the reduction induced solvent-reorganization. Together
with Asp152, Ser335 is located at the narrowest part of the substrate channel close to the
heme edge [35]. Its exchange by glycine should open the access to the heme active site for
the substrate molecules. Indeed, binding of O2 to ferrous Ser335Gly (as well as to ferrous
Arg119Asn and His123Gln, see below) has been seen in the initial electrochemical studies
and could only be hindered by addition of small amounts of laccase and catalase into the
protein solution. On the other hand Ser335 is not involved in maintenance of the solute
matrix [35] and thus its exchange should not directly have an impact on its rigidity. This is
81
reflected by its almost intact catalatic properties (Table 1). However, disruption of the
hydrogen bond between the side chain hydroxyl group of Ser335 and the carboxyl group
of one of the pyrrole propionates in observed in Mycobacterium KatG(Ser315Gly) [36],
might diminish the electrostatic interaction of the metal ion with the solvent matrix thus
being responsible for the observed negative S°'rc values.
Although loss of the carboxylate group at the entrance of the access channel in
Glu253Gln significantly alters the bifunctionality of KatG [12], its E°' and reduction
thermodynamics were nearly coincident with that of the wild-type protein (Table 1). This
fits with recently published data that demonstrated neither a change in the spin state nor in
the Fe-Im bond strength in Glu253Gln [12] and with the small influence that deletion of
surface charges generally have on E°' in redox metallo-proteins [51-53]. These data clearly
suggest that exchange of the carboxylate group has neither an impact on the protein-based
molecular factors which define the E°’ of the iron ion nor on solvent reorganization effects
at least in the main (inner) part of the access channel. Its role in catalysis (Table 1) might
be in attracting and orienting the incoming H2O and H2O2 dipoles [12].
Compared to the so far discussed KatG variants, exchange of the fully conserved
distal pair His-Arg significantly modified the redox behavior of the heme iron. In both
Arg119Asn and His123Gln mutants the midpoint potential shifted 79 mV and 93 mV,
respectively, to more negative (HRP-like) values (Table 1). In contrast to the other mutants
investigated in this work, exchange of both His123 and Arg119 affected both the catalatic
and the peroxidatic reaction underling the importance of His and Arg in the heterolytic
cleavage of hydrogen peroxide [37].
Interestingly, the behavior of the His123Gln KatG mutant coincides with that of
His52Gln and His52Asn CcP mutants [55], which feature E°' values 35 and 70 mV more
negative than the wild-type enzyme, respectively, and stabilize the ferric form of CcP more
effectively than the His52Glu and His52Asp mutations, in which a negative charge has
been inserted in the distal heme cavity [55]. Ion pairing of the carboxylate sidechains with
the distal arginine, mutation-induced alterations in the hydrogen bonding network and in
the heme legation have been indicated as the possible causes of the above effect [55].
RR spectroscopy indicates that in the His123Gln and Arg119Asn KatG variants the
proximal heme architecture is only marginally weakened (Table 1) [23], clearly
demonstrating that their lower reduction potentials is not the consequence of an increased
interaction with the axial histidine. On the other hand, the above mutations open the heme
82
distal cavity, inducing a deep reorganization of the hydrogen bonding network and
increasing the accessibility of solvent molecules to the metal ion. Opening of the heme
cavity in His123Gln and Arg119Asn is evident by the fact that both proteins were more
pronounced to dioxygen binding to their ferrous forms than wild-type KatG. A similar
effect has been described for HRP and CcP, where mutation of distal His and Arg
significantly enhanced the binding rate for O2 [38]. So special care had to be taken in the
spectroelectrochemical experiments in order to maintain anaerobic conditions and prevent
O2 binding (i.e. compound III formation).
Opening the heme cavity increases the solvent accessibility to the metal center and,
consequently, the polarity of the distal heme cavity, electrostatically stabilizing the ferric
form of the enzyme compared to the wild-type enzyme. Moreover, an increased polarity of
the heme distal site would partially quench the effect of the exchange of the positively
charge arginine with the neutral asparagine on E°', explaining, at least partially, the smaller
effect of the Arg119Asn mutation has on the reduction potential.
An increased solvent accessibility to the metal center, induced by the mutationinduced opening the heme cavity agrees with the positive ΔS°'rc values featured by both the
His123Gln and Arg119Asn mutants, which are typical for metalloproteins with solvent
exposed central ions [16, 27-29]. Exchange of both His and Arg furthermore promoted
reduction induced solvent reorganization effects since both are essential components of the
H-bond network. This is also obvious by an increased fraction of 6-coordinated high-spin
heme in both mutant proteins [23].
Summing up, this work describes the redox thermodynamics of bifunctional
catalase-peroxidase variants that carry mutations at sites known to be important in
catalysis as well as in stabilization of the KatG-typical H-bonding network. Exchange of
the catalytic distal residues His and Arg completely reorganizes the heme cavity
architecture and accessibility and dramatically modifies the redox properties of the
corresponding mutants. In other (catalatically active and inactive) mutants the midpoint
potential was very similar to that of wild-type KatG though the relative contributions of
H°'rc and S°'rc to E°' were different and compensated each other. This includes also
variants without the KatG-specific Trp-Tyr-Met adduct that is essential for H2O2
oxidation. Enthalpic and entropic contributions to E°' reflect mutation-mediated alterations
of charge and polarity of the distal heme cavity as well as of the rigidity and continuity of
83
the solvent molecules in the active site and substrate channel. Disruption of the solvent
matrix in the constricted access channel of KatG retards oxidation of H2O2 but (with
exception of distal His-Arg) not its reduction, i.e. heterolytic cleavage.
Materials and Methods
Mutagenesis, expression and purification of wild-type KatG and KatG variants form
Synechocystis PCC 6803 were described previously [9, 12, 15, 25, 26]. All chemicals were
reagent-grade.
Spectroelectrochemistry. All experiments were carried out in a homemade OTTLE
cell [16, 27, 28]. The three-electrode configuration consisted of a gold minigrid working
electrode (Buckbee-Mears, Chicago, IL), a homemade Ag/AgCl/KClsat microreference
electrode, separated from the working solution by a Vycor set, and a platinum wire as the
counter electrode. The reference electrode was calibrated against a saturated calomel
electrode before each set of measurements. All potentials are referenced to the SHE.
Potentials were applied across the OTTLE cell with an Amel model 2053
potentiostat/galvanostat. The function of the OTTLE cell was checked by measuring the
reduction potential of yeast iso-1-cytochrome c in conditions similar to those used in the
present work. The E°' value corresponded to that determined by cyclic voltammetry (+260
mV). Constant temperature was maintained by a circulating water bath, and the OTTLE
cell temperature was monitored with a Cu-costan microthermocouple. UV-vis spectra were
recorded using a Varian Cary C50 spectrophotometer. Since some of the mutants were
more prone to O2 binding to their ferrous form than wild-type KatG, besides flushing with
argon, small amounts of laccase and catalase were present in the cuvettes in these cases.
Variable temperature experiments were performed using a non-isothermal cell
configuration. The temperature of the reference electrode and the counter electrode was
kept constant, whereas that of the working electrode was varied. Factorization of enthalpic
and entropic components, was possible by calculating ΔS°'rc from the slope of the plot E°'
versus temperature, whereas ΔH°'rc could be obtained from the Gibbs-Helmholtz equation,
ΔG°'rc = ΔH°'rc - TΔS°'rc = -nFE°', thus from the slope of the plot of E°'/T versus 1/T [29].
The Faraday constant F is 96485.31 C mol-1, and n represents the number of electrons
transferred by the redox couple. All experiments were carried out under Argon over the
10-30°C (Ser335Gly), 10-35°C (Arg119Asn), 15-35°C (Trp122Phe, His123Gln), 15-40°C
(Asp152Ser, Glu253Gln) range using 1 mL samples containing protein (concentrations see
below) in 50 mM phosphate buffer, pH 7.0, containing 10 mM NaCl, in the presence of
84
various mediators: methyl viologen (mediator A) and a solution of lumiflavine-3-acetate,
indigo disulfonate, phenazine methosulfate, and methylene blue (mediators B). In detail,
protein and mediator concentrations were as follows: Asp152Ser (0.02 mM), 0.7 mM
mediator A and 3 µM mediators B; Glu253Gln (0.035 mM), 0.7 mM mediator A and 3.5
µM mediators B; Trp122Phe (0.016 mM), 0.28 mM mediator A and 2 µM mediators B;
His123Gln (0.025 mM), 0.2 mM mediator A and 2.5 µM mediators B; Arg119Asn (0.02
mM), 0.2 mM mediator A and 2.5 µM mediators B. In order to prevent binding of oxygen
to the Fe(II) form catalase from bovine liver and laccase from Rhus vernicifera, 0.5 µM
each, were added to the sample containing 0.024 mM Ser335Gly, 0.55 mM mediator A
and 2 µM mediators B.
Nernst plots consisted of at least six points and were invariably linear, with a slope
consistent with a one-electron reduction process.
Molecular modelling. Model building. The sequence of the catalase-peroxidase from
Synechocystis sp PCC 6803 (UniProtKB entry P73911) was extracted from the
UniProtKB/Swiss-Prot database [61] (www.expasy.org), with the accession code P73911.
A FASTA [62] search performed against the Protein Data Bank [63] identified the proteins
with the highest sequence identity (id) and similarity (ss) to the target sequence: the
catalase-peroxidases from Burkholderia Pseudomallei (PDB entry: 2FXH, chain A;
id=60%, ss=81%), Mycobacterium Tuberculosis (PDB entry: 2CCA, chain B; id=55%,
ss=78%), Haloarcula Marismortui (PDB entry: 1ITK, chain B; id=55%, ss=80%) and
Synechococcus sp. (PDB entry: 1UB2; id=72%, ss=86%). In the cases when more than one
3D structure was available, a few structural criteria (i.e. best resolution, pH around 7,
structural completeness, highest sequence identity to the target) were used to select the
best representative protein structure to be used as a template.
The target sequence was then aligned to the sequences of the four templates. The
multiple alignment used for modelling (see Supplemental Figure S1) was performed with
the software CLUSTALW [64] (www.ebi.ac.uk).
Modelling of the wt catalase-peroxidase from Synechocystis sp. PCC 6803 was
performed with the software MODELLER v. 8.2 [65]. Five model structures were
generated and subsequently i) assessed against the available structural experimental
information and ii) evaluated from a stereochemical and geometrical perspective with the
aid of the tools implemented in the software packages: MODELLER [65], PROCHECK
[66], QUANTA [67]. The values of the objective function output by MODELLER [65]
together with the values of the global G-factors and of the Ramachandran residue
85
distribution from PROCHECK [66], calculated for the models and the template structures,
are reported in Supplemental Table S1 (in the Supplementary materials). The modeled
structure with the best fitting to the experimental data and the best technical evaluation
was then selected and used for further work. Secondary structure prediction was performed
with the PredictProtein service (http://www.predictprotein.org/) [68].
Model refinement. The main structural problems are related to the difficulty of
reproducing in the 3D model the correct geometry of the adduct built by residues W122Y249-M275, which are covalently connected. The geometric distortion is due to the local
structural differences found in the four 3D structures used as templates. The correct
covalent geometry of the adduct was reproduced by manually rotating the side chains of
the three residues involved on the basis of the local structure of the catalase-peroxidase
from Synechococcus sp. (1UB2), which is the protein with the highest sequence identity to
the target protein.
The selected modelled structure was then optimized with the GROMACS suite [69],
using the GROMOS’96 force field [70]. The crystallographic water molecules in the cell
unit of the catalase-peroxidase from Synechococcus sp. (1UB2) were translated to the
Synechocystis sp PCC 6803 model, to correctly reproduce the active site local
environment. The system was then solvated in a cubic water cell with sides extending 15Å
from the protein boundary. All water molecules were treated explicitly with the SPCE
model. Short range interactions were treated with the Lennard-Jones potential using a cutoff of 14 Å, while long range interactions were calculated using the PME algorithm within
a radius of 10 Å and an interpolation order of 6. All covalent distances were restrained
with the LINCS algorithm [71]. 31 and 32 Na+ ions were added respectively to the water
bulk of the ferric and the ferrous proteins, respectively, to assure system neutrality. The
optimization protocol consisted of a preliminary Steepest Descent minimization until
energy converged to 1000 kJ/mol, followed by Conjugate Gradients until energy
convergence to 1000 kJ/mol was reached. During the calculations, the correct geometries
of both the covalent adduct and the heme-H190 system were preserved by adding a few
harmonic restraints to their bond length.
Calculation of protein structural parameters. Accessible surface areas were
computed for both the reduced and the oxidized 3D models of the catalase-peroxidase
from Synechocystis sp PCC 6803. ASAs were computed on the heme group and on the
redox active site (heme plus residues R119, W122, H123, Y249 and D152) using the
software package QUANTA [67]. The calculations, based on the Richards method [72],
86
used an all-atoms parameterization for the protein and a conventional sphere of 1.4 Å as
the probe simulating a solvent (water) molecule.
Acknowledgements
This work was supported by the Ministero dell’Universitá e della Ricerca Scientifica
(MUR) of Italy (Programmi di Ricerca Scientifica di Rilevante Interesse Nazionale 2007
Prot. N. 2007KAWXCL and n. 20079Y9578), by the University of Modena and Reggio
Emilia (Fondo di Ateneo per la Ricerca, 2007), and the Austrian Science Fund FWF
project P18751 & P19793.
References
1.
Welinder KG (1992) Superfamily of plant, fungal and bacterial peroxidases. Curr
Opin Struct Biol 2, 388-393.
2.
Bertrand T, Eady NA, Jones JN, Jesmin, Nagy JM, Jamart-Gregoire B, Raven EL &
Brown KA (2004) Crystal structure of Mycobacterium tuberculosis catalaseperoxidase. J Biol Chem 279, 38991-38999.
3.
Carpena X, Loprasert S, Mongkolsuk S, Switala J, Loewen PC & Fita I (2003)
Catalase-peroxidase KatG of Burkholderia pseudomallei at 1.7A resolution. J Mol
Biol 327, 475-489.
4.
Wada K, Tada T, Nakamura Y, Kinoshita T, Tamoi M, Shigeoka S & Nishimura K
(2002) Crystallization and preliminary X-ray diffraction studies of catalaseperoxidase from Synechococcus PCC 7942. Acta Crystallogr D Biol Crystallogr 58,
157-159.
5.
Yamada Y, Fujiwara T, Sato T, Igarashi N & Tanaka N (2002) The 2.0 A crystal
structure of catalase-peroxidase from Haloarcula marismortui. Nat Struct Biol 9,
691-695.
6.
Finzel BC, Poulos TL & Kraut J (1984) Crystal structure of yeast cytochrome c
peroxidase refined at 1.7-A resolution. J Biol Chem 259, 13027-13036.
7.
Patterson WR & Poulos TL (1995) Crystal structure of recombinant pea cytosolic
ascorbate peroxidase. Biochemistry 34, 4331-4341.
8.
Sharp KH, Mewies M, Moody PC & Raven EL (2003) Crystal structure of the
ascorbate peroxidase-ascorbate complex. Nat Struct Biol 10, 303-307.
9.
Jakopitsch C, Auer M, Ivancich A, Ruker F, Furtmuller PG & Obinger C (2003)
Total conversion of bifunctional catalase-peroxidase (KatG) to monofunctional
87
peroxidase by exchange of a conserved distal side tyrosine. J Biol Chem 278, 2018520191.
10.
Jakopitsch C, Vlasits J, Wiseman B, Loewen PC & Obinger C (2007) Redox
intermediates in the catalase cycle of catalase-peroxidases from Synechocystis PCC
6803, Burkholderia pseudomallei, and Mycobacterium tuberculosis. Biochemistry
46, 1183-1193.
11.
Suarez J, Ranguelova K, Jarzecki AA, Manzerova J, Krymov V, Zhao X, Yu S,
Metlitsky L, Gerfen GJ & Magliozzo RS (2009) An oxyferrous heme/protein-based
radical intermediate is catalytically competent in the catalase reaction of
Mycobacterium tuberculosis catalase-peroxidase (KatG). J Biol Chem 284, 70177029.
12.
Jakopitsch C, Droghetti E, Schmuckenschlager F, Furtmuller PG, Smulevich G &
Obinger C (2005) Role of the main access channel of catalase-peroxidase in
catalysis. J Biol Chem 280, 42411-42422.
13.
Smulevich G, Jakopitsch C, Droghetti E & Obinger C (2006) Probing the structure
and bifunctionality of catalase-peroxidase (KatG). J Inorg Biochem 100, 568-585.
14.
Bellei M, Jakopitsch C, Battistuzzi G, Sola M & Obinger C (2006) Redox
thermodynamics of the ferric-ferrous couple of wild-type Synechocystis KatG and
KatG(Y249F). Biochemistry 45, 4768-4774.
15.
Jakopitsch C, Auer M, Regelsberger G, Jantschko W, Furtmuller PG, Ruker F &
Obinger C (2003) Distal site aspartate is essential in the catalase activity of catalaseperoxidases. Biochemistry 42, 5292-5300.
16.
Battistuzzi G, Borsari M, Ranieri A & Sola M (2002) Redox thermodynamics of the
Fe(3+)/Fe(2+) couple in horseradish peroxidase and its cyanide complex. J Am
Chem Soc 124, 26-27.
17.
Battistuzzi G, Bellei M, De Rienzo F & Sola M (2006) Redox properties of the
Fe3+/Fe2+ couple in Arthromyces ramosus class II peroxidase and its cyanide adduct.
J Biol Inorg Chem 11, 586-592.
18.
Battistuzzi G, Bellei M, Zederbauer M, Furtmuller PG, Sola M & Obinger C (2006)
Redox thermodynamics of the Fe(III)/Fe(II) couple of human myeloperoxidase in its
high-spin and low-spin forms. Biochemistry 45, 12750-12755.
19.
Zamocky M, Furtmuller PG & Obinger C (2008) Evolution of catalases from
bacteria to humans. Antioxid Redox Signal 10, 1527-1548.
88
20.
Conroy CW, Tyma P, Daum PH & Erman JE (1978) Oxidation-reduction potential
measurements of cytochrome c peroxidase and pH dependent spectral transitions in
the ferrous enzyme. Biochim Biophys Acta 537, 62-69.
21.
Goodin DB & McRee DE (1993) The Asp-His-Fe triad of cytochrome c peroxidase
controls the reduction potential, electronic structure, and coupling of the tryptophan
free radical to the heme. Biochemistry 32, 3313-3324.
22.
Jones DK, Dalton DA, Rosell FI & Raven EL (1998) Class I heme peroxidases:
characterization of soybean ascorbate peroxidase. Arch Biochem Biophys 360, 173178.
23.
Heering HA, Indiani C, Regelsberger G, Jakopitsch C, Obinger C & Smulevich G
(2002) New insights into the heme cavity structure of catalase-peroxidase: a
spectroscopic approach to the recombinant Synechocystis enzyme and selected distal
cavity mutants. Biochemistry 41, 9237-9247.
24.
Santoni E, Jakopitsch C, Obinger C & Smulevich G (2004) Comparison between
catalase-peroxidase and cytochrome c peroxidase. The role of the hydrogen-bond
networks for protein stability and catalysis. Biochemistry 43, 5792-5802.
25.
Jakopitsch C, Ruker F, Regelsberger G, Dockal M, Peschek GA & Obinger C (1999)
Catalase-peroxidase from the cyanobacterium Synechocystis PCC 6803: cloning,
overexpression in Escherichia coli, and kinetic characterization. Biol Chem 380,
1087-1096.
26.
Regelsberger G, Jakopitsch C, Ruker F, Krois D, Peschek GA & Obinger C (2000)
Effect of distal cavity mutations on the formation of compound I in catalaseperoxidases. J Biol Chem 275, 22854-22861.
27.
Battistuzzi G, Bellei M, Borsari M, Di Rocco G, Ranieri A & Sola M (2005) Axial
ligation and polypeptide matrix effects on the reduction potential of heme proteins
probed on their cyanide adducts. J Biol Inorg Chem 10, 643-651.
28.
Battistuzzi G, Bellei M, Leonardi A, Pierattelli R, De Candia A, Vila AJ & Sola M
(2005) Reduction thermodynamics of the T1 Cu site in plant and fungal laccases. J
Biol Inorg Chem 10, 867-873.
29.
Yee EL, Cave RJ, Guyer KL, Tyma PD & Weaver MJ (1979) A Survey of Ligand
Effects upon the Reaction Entropies of Some Transition Metal Redox Couples. J Am
Chem Soc 101, 1131-1137.
89
30.
Mate MJ, Sevinc MS, Hu B, Bujons J, Bravo J, Switala J, Ens W, Loewen PC & Fita
I (1999) Mutants that alter the covalent structure of catalase hydroperoxidase II from
Escherichia coli. J Biol Chem 274, 27717-27725.
31.
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.
32.
Nicholls P, Fita I & Loewen PC (2001) Enzymology and structure of catalases. Adv
Inorg Chem 51, 51-106.
33.
Santoni E, Jakopitsch C, Obinger C & Smulevich G (2004) Manipulating the
covalent link between distal side tryptophan, tyrosine, and methionine in catalaseperoxidases: an electronic absorption and resonance Raman study. Biopolymers 74,
46-50.
34.
Deemagarn T, Wiseman B, Carpena X, Ivancich A, Fita I & Loewen PC (2007) Two
alternative substrate paths for compound I formation and reduction in catalaseperoxidase KatG from Burkholderia pseudomallei. Proteins 66, 219-228.
35.
Zhao X, Yu H, Yu S, Wang F, Sacchettini JC & Magliozzo RS (2006) Hydrogen
peroxide-mediated isoniazid activation catalyzed by Mycobacterium tuberculosis
catalase-peroxidase (KatG) and its S315T mutant. Biochemistry 45, 4131-4140.
36.
Suarez J, Ranguelova K, Schelvis JP & Magliozzo RS (2009) Antibiotic resistance in
Mycobacterium tuberculosis: peroxidase intermediate bypass causes poor isoniazid
activation by the S315G mutant of M. tuberculosis catalase-peroxidase (KatG). J
Biol Chem 284, 16146-16155.
37.
Poulos TL & Kraut J (1980) The stereochemistry of peroxidase catalysis. J Biol
Chem 255, 8199-8205.
38.
Rodriguez-Lopez JN, Smith AT & Thorneley RN (1997) Effect of distal cavity
mutations on the binding and activation of oxygen by ferrous horseradish peroxidase.
J Biol Chem 272, 389-395.
39.
Smulevich G, Paoli M, Burke JF, Sanders SA, Thorneley RN & Smith AT (1994)
Characterization of recombinant horseradish peroxidase C and three site-directed
mutants, F41V, F41W, and R38K, by resonance Raman spectroscopy. Biochemistry
33, 7398-7407.
40.
Kjalke M, Andersen MB, Schneider P, Christensen B, Schülein M & Welinder KG
(1992) Fungal CIP has been shown to be identical to the commercial Coprinus
90
macrorhizus peroxidase (CMP) and Arthromyces ramosus peroxidase (ARP) both in
its covalent structure and enzymic properties. Biochim Biophys Acta 1120, 248-256.
41.
Smulevich G, Feis A, Focardi C, Tams J & Welinder KG (1994) Resonance Raman
study of the active site of Coprinus cinereus peroxidase. Biochemistry 33, 1542515432.
42.
Brogioni S, Stampler J, Furtmuller PG, Feis A, Obinger C & Smulevich G (2008)
The role of the sulfonium linkage in the stabilization of the ferrous form of
myeloperoxidase: a comparison with lactoperoxidase. Biochim Biophys Acta 1784,
843-849.
91
Supplementary Materials
Supplemental Table S1. Technical checks report. *Parameters of the Ramachendran residue distribution: percentage of
residues with correct (core), allowed (all), generously allowed (gener) or disallowed (disall) φ/ψ conformation.
**Geometric (G) factors: evaluation of the dihedral conformations (dihedrals) and of the covalent bonds (cov) of all the
protein residues and global evaluation (overall) of the protein geometry (values below -0.5: unusual; values below -1.0:
- highly unusual). ***MOLPDF: value of the MODELLER objective function, which estimate the content of structural
restraint violation in a modelled structure.
MODEL
Synechocystis sp PCC 6803
B. Pseudomallei (2fxh)
M. Tuberculosis (2cca)
H. Marismortui (1itk)
Synechococcus sp. (1ub2)
RAMACHANDRAN PLOT*
G FACTORs**
MOLPDF***
Core
Allow
Gener
Disall
Dihed
Cov
P73911_ 1
92,20
6,80
0,50
0,50
0,11
-0,15
Overall
0,01
25740,24023
P73911_2
91,80
7,50
0,70
0,00
0,14
-0,13
0,04
25242,64844
P73911_3
90,80
7,50
1,00
0,70
0,09
-0,18
-0,01
25632,85938
P73911_4
91,20
7,80
0,80
0,20
0,13
-0,14
0,03
26779,82813
P73911_5
90,80
91,5
8,80
8,3
0,20
0,2
0,20
0,1
0,12
-0,06
-0,13
0,43
0,03
0,14
25295,49609
91,3
8,5
0,2
0,0
-0,07
0,56
0,18
0,18
10,0
0,1
0,0
0,27
0,58
0,39
88,6
11,4
0,0
0,0
0,09
0,18
0,14
92
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
-------------MPGSDAGPRRRGVHEQRRNRMSNEAKCPFHQAAGN-G
-------------MP-----EQHPPITETTTGAASNGCPVVGHMKYPVEG
-------------------------MAETPNSDMS--GATGGRSKRPK--------------------------MTATQG---KCPVMHGGATTVNI-MGTQPARKLRNRVFPHPHNHRKEKPMANDQVPASKCPVMHGANTTGQN-:
.
36
32
21
20
48
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
TSNRDWWPNQLDLSILHRHSSLSDPMGKDFNYAQAFEKLDLAAVKRDLHA
GGNQDWWPNRLNLKVLHQNPAVADPMGAAFDYAAEVATIDVDALTRDIEE
-SNQDWWPSKLNLEILDQNARDVGPVEDDFDYAEEFQKLDLEAVKSDLEE
-STAEWWPKALNLDILSQHDRKTNPMGPDFNYQEEVQKLD-AALKQDLQA
-GNLNWWPNALNLDILHQHDRKTNPMDDGFNYAEAFQQLDLAAVKQDLHH
.. :***. *:*.:* ::
.*:
*:*
. :* *:. *:.
HHHHHH
HHHHHHH HHHHHHHHHH
HHHH
HHHH
HHHHHHH HHHHHHHHHH
86
82
70
68
97
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
LMTTSQDWWPADFGHYGGLFIRMAWHSAGTYRTADGRGGAGEGQQRFAPL
VMTTSQPWWPADYGHYGPLFIRMAWHAAGTYRIHDGRGGAGGGMQRFAPL
LMTSSQDWWPADYGHYGPLFIRMAWHSAGTYRTADGRGGAAGGRQRFAPI
LMTDSQDWWPADWGHYGGLMIRLTWHAAGTYRIADGRGGAGTGNQRFAPL
LMTDSQSWWPADWGHYGGLMIRMAWHAAGTYRIADGRGGAATGNQRFAPL
:** ** *****:**** *:**::**:***** ******. * *****:
HH
HHHHHHHHHH EEE
HHH
HHHHHHHHHHHHH
136
132
120
118
147
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
NSWPDNANLDKARRLLWPIKQKYGRAISWADLLILTGNVALESMGFKTFG
NSWPDNASLDKARRLLWPVKKKYGKKLSWADLIVFAGNCALESMGFKTFG
NSWPDNANLDKARRLLLPIKQKYGQKISWADLMILAGNVAIESMGFKTFG
NSWPDNTNLDKARRLLWPIKQKYGNKLSWADLIAYAGTIAYESMGLKTFG
NSWPDNVNLDKARRLLWPIKKKYGNKLSWGDLIILAGTMAYESMGLKVYG
******..******** *:*:***. :**.**: :*. * ****:*.:*
HHHHHH
HHHHHH
HHHHHHHH HHHHH
HHHHHHH HHHHHHH
HHHHHHHHHHHHHHHH
186
182
170
168
197
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
FAGGRADTWEPE-DVYWGSEKIWLELSGGPNSRYSGDRQ-LENPLAAVQM
FGFGRVDQWEPD-EVYWGKEATWLG-----DERYSGKRD-LENPLAAVQM
YAGGREDAFEEDKAVNWGPEDEFET-----QERFDEPGE-IQEGLGASVM
FAFGREDIWHPEKDIYWGPEKEWFPPSTNPNSRYTGDRE-LENPLAAVTM
FAGGREDIWHPEKDIYWGAEKEWLASSDHRYG--SEDRESLENPLAAVQM
:. ** * :. : : ** * :
: ::: *.* *
HHH
HHHHHH
33
333
234
225
214
217
245
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
GLIYVNPEGPDGNPDPVAAARDIRDTFARMAMNDEETVALIAGGHTFGKT
GLIYVNPEGPNGNPDPMAAAVDIRETFRRMAMNDVETAALIVGGHTFGKT
GLIYVNPEGPDGNPDPEASAKNIRQTFDRMAMNDKETAALIAGGHTFGKV
GLIYVNPEGVDGNPDPLKTAHDVRVTFARMAMNDEETVALTAGGHTVGKC
GLIYVNPEGVDGHPDPLCTAQDVRTTFARMAMNDEETVALTAGGHTVGKC
********* :*:*** :* ::* ** ****** **.** .****.**
EEEE
HHHHHHHHHHHHHHH
HHHHHEEE
33
333
HHHHHHHHHHHHHH
HHHHHHHHHHHH E
284
275
264
267
295
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
HGAG-PASNVGAEPEAAGIEAQGLGWKSAYRTGKGADAITSGLEVTWTTT
HGAG-PADLVGPEPEAAPLEQMGLGWKSSYGTGTGKDAITSGIEVVWTNT
HGADDPEENLGPEPEAAPIEQQGLGWQNKNGNSKGGEMITSGIEGPWTQS
HGNG-NAALLGPEPEGADVEDQGLGWINKTQSGIGRNAVTSGLEGAWTPH
HGNS-KAELIGPEPEGADVVEQGLGWHNQNGKGVGRETMSSGIEGAWTTH
** .
:*.***.* :
**** .
.. * : ::**:* **
HHH
EE
E
333E
4444 33333
E
333 E
EE
333
324
314
316
344
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
PTQWSHNFFENL-FGYEWELTKSPAGAHQWVAKG--ADAVIPDAFDPSKK
PTKWDNSFLEIL-YGYEWELTKSPAGAWQYTAKDGAGAGTIPDPFGGPGR
PTEWDMGYINNL-LDYEWEPEKGPGGAWQWAPKSEELKNSVPDAHDPDEK
PTQWDNGYFAVCSLNYDWELKKNPAGAWQWEPINPREEDLPVDVEDPSIR
PTQWDNGYFYML-FNHEWELKKSPAGAWQWEPVNIKEEDKPVDVEDPNIR
**:*. .::
.::** *.*.** *: . .
* .
:
HHHHHHH HH
EEE
HHHHHH H
EEEEE
EEEEE
380
373
363
366
393
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
HRPTMLTTDLSLRFDPAYEKISRRFHENPEQFADAFARAWFKLTHRDMGP
S-PTMLATDLSLRVDPIYERITRRWLEHPEELADEFAKAWYKLIHRDMGP
QTPMMLTTDIALKRDPDYREVMETFQENPMEFGMNFAKAWYKLTHRDMGP
RNLVMTDADMAMKMDPEYRKISERFYQDPAYFADVFARAWFKLTHRDMGP
HNPIMTDADMAMIKDPIYRQISERFYREPDYFAEVFAKAWFKLTHRDLGP
* :*::: ** *..: . : ..* :. **:**:** ***:**
430
422
413
416
443
93
predSS
model secstr
HHH
HHHHHHHHHH HHHHHHHHHHHHHHHHHHH
HHHHHHHH HHHHHHHHHHHH HHHHHHHHHHHHHHHHH
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
RARYLGPEVPAEVLLWQDPIPAVDHPLIDAADAAELKAKVLASGLTVSQL
VARYLGPLVPKQTLLWQDPVPAVSHDLVGEAEIASLKSQIRASGLTVSQL
PERFLGPEVPDEEMIWQDPLPDADYDLIGDEEIAELKEEILDSDLSVSQL
KARYIGPDVPQEDLIWQDPIPAGNRNY----DVQAVKDRIAASGLSISEL
KSRYLGPDVPQEDLIWQDPIPPVDYTLS-EGEIKELEQQILASGLTVSEL
*::** ** : ::****:* .
:
:: .: *.*::*:*
HHH
HHHHHHHHHHHHHH
HHHH
333
HHHHHHHHHHHH
HHHH
480
472
463
462
492
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
VSTAWAAASTFRGSDKRGGANGARIRLAPQKDWEANQPE-QLAAVLETLE
VSTAWAAASSFRGSDKRGGANGGRIRLQPQVGWEVNDPDGDLRKVIRTLE
VKTAWASASTYRDSDKRGGANGARLRLEPQKNWEVNEPE-QLETVLGTLE
VSTAWDSARTYRNSDKRGGANGARIRLAPQKDWEGNEPD-RLPKVLAVLE
VCTAWDSARTFRSSDYRGGANGARIRLEPQKNWPGNEPT-RLAKVLAVLE
* *** :* ::*.** ******.*:** ** .* *:*
* *: .**
HHHHHHHHHH
EEEEE
HH HHHHHHHHHH
HHHHHHHH
4444
333 HHHHH
HH HHHHHHHHHH
529
522
512
511
541
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
AIRTAFNGAQRGGKQVSLADLIVLAGCAGVEQAAKNAGHAVTVPFAPGRA
EIQESFNSAAPGNIKVSFADLVVLGGCAAIEKAAKAAGHNITVPFTPGRT
NIQTEFNDSRSDGTQVSLADLIVLGGNAAVEQAAANAGYDVEIPFEPGRV
GISA--------ATGATVADVIVLAGNVGVEQKARAAGVEIVLPFAPGRG
NIQANF------AKPVSLADLIVLGGGAAIAKAALDGGIEVNVPFLPGRG
*
.:.**::**.* ..: : * .* : :** ***
HHHHHH
HHHHHHHHH
EEE
HHHHH
HHHHHHHHHHHHHHHHHHHH
579
572
562
553
585
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
DASQEQTDVESMAVLEPVADGFRNYLKGKYRVPAEVLLVDKAQLLTLSAP
DASQEQTDVESFAVLEPKADGFRNYLGKGNPLPAEYMLLDKANLLTLSAP
DAGPEHTDAPSFDALKPKVDGVRNYIQDDITRPAEEVLVDNADLLNLTAS
DATAEQTDTESFAVLEPIHDAIATGSSRTMRQRLKNCCLIATQLLGLTAP
DATQAMTDAESFTPLEPIHDGYRNWLKQDYAVSPEELLLERTQLMGLTAP
**
**. *: *:* *. .
:
: ::*: *:*.
HHHHH
HHH
HHHHHHHHHHHH
HHHH333
EE4444EE
HHHHHHHHHHHH
HH
629
622
612
603
635
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
EMTVLLGGLRVLGANVGQSRHGVFTAREQALTNDFFVNLLDMGTEWKPTA
EMTVLVGGLRVLGANYKRLPLGVFTEASESLTNDFFVNLLDMGITWEPSP
ELTALIGGMRSIGANYQDTDLGVFTDEPETLTNDFFVNLLDMGTEWEPAA
EMTVLIGGLRVLGTNHGGTKHVVFTDREGVLTNDFFVNLTDMNYLWKP-EMTVLIGGMRVLGTNHGGTKHGVFTDRVGVLSNDFFVNLTDMAYQWRP-*:*.*:**:* :*:*
***
*:******* **
*.*
HHHEEEE EEEEE
EEE
EE HHHHHHH
EE
HHHHHHHHHHHH
HHHHHHH
EEEEE
679
672
662
651
683
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
ADADVFEGRDRATGELKWTGTRVDLVFGSHSQLRALAEVYGSADAQEKFV
ADDGTYQGKD-GSGKVKWTGSRVDLVFGSNSELRALVEVYGADDAQPKFV
DSEHRYKGLDRDTGEVKWEATRIDLIFGSNDRLRAISEVYGSADAEKKLV
AGKNLYEICDRKTNQVKWTATRVDLVFGSNSILRAYSELYAQDDNKEKFV
AGNNLYEIGDRQTGEVKWTATKVDLVFGSNSILRSYAEVYAQDDNREKFV
.
:: * :.::** .:::**:***:. **: *:*. * . *:*
EEEEEE
EEEEEEEEEEE
HHHHHHHHHHH
HHHHHH
EEEEE
EEEEE HHHHHHHH HHHHHHH
HHHHH
729
721
712
701
733
2FXH_A|PDBID|CHAIN|SEQUENCE
2CCA_B|PDBID|CHAIN|SEQUENCE
1ITK_B|PDBID|CHAIN|SEQUENCE
1UB2 |PDBID|CHAIN|SEQUENCE
P73911|P73911_SYNY3
conservation
predSS
model secstr
RDFVAVWNKVMNLDRFDLA-QDFVAAWDKVMNLDRFDVR-HDFVDTWSKVMKLDRFDLE-RDFVAAWTKVMNADRFDLD-RDFVAAWTKVMNADRFDLPRG
:*** .* ***: ****:
HHHHHHHHHHH
HHHHHHHHHHHHHHHHHH
748
740
731
720
754
Supplemental Figure S1. Multiple sequence alignment of the target sequence (P73911) to the catalase-peroxidase
sequences of Burkholderia pseudomallei (2FXH_A), Mycobacterium tuberculosis (2CCA_B), Haloarcula marismortui
(1ITK_B) and Synechococcus sp. (1UB2). The sequence conservation (conservation), the secondary structure prediction
(predSS) and the secondary structure of the model (model secstr) are also reported: “*”= identical aminoacids; “:” =
conserved aminoacids; “.”= low residue conservation; H = α-helix; E = β-strand; 3 = 3-turn; 4 = 4-turn. Color code: red
= non resolved aminoacids; white = residues in the distal sites; pink = covalent adduct; yellow = residues identifying the
access channel to the active site; cyan = Fe coordinating histidine; grey: residues in the proximal site; blu = distal
histidine; green = distal arginine.
94
(A)
(B)
-0.18
E (V)
-0.21
0.205
-0.24
-0.5
-0.2
0.07
0.1
Log X
0.4
0.05
-0.35
-0.45 -0.15 0.15 0.45
Log X
0.155
0.105
0.055
0.03
0.01
0.005
322
342
362
382
402
422
442
462
482
322
342
362
Wavelength (nm)
(C)
-0.26
(D)
0.2
382 402 422 442
Wavelength (nm)
462
-0.21
-0.32
-0.35
0.16
482
-0.18
0.21
E (V)
E (V)
-0.29
-0.24
-0.27
0.16
-0.38
-0.9 -0.6 -0.3 0
Log X
0.12
0.08
0.3
Absorbance
Absorbance
-0.32
-0.38
-0.27
Absorbance
E (V)
0.09
Absorbance
-0.26
0.255
-0.29
0.11
-0.3
-0.7 -0.4 -0.1 0.2
Log X
0.11
0.06
0.04
0
0.01
322
342
362
382 402 422 442
Wavelength (nm)
462
482
322
342
362
382 402 422 442
Wavelength (nm)
462
482
Supplemental Figure S2. (A) Electronic absorption spectra of the Synechocystis KatG variant KatG(W122F) obtained
at various applied potentials at 25°C. The inset depicts the Nernst plot, were X represents [(Aλredmax – Aλred)/(Aλoxmax –
Aλox)] with λox = 409 nm and λred = 430 nm. The absorbance at 430 nm was used for calculations, while the absorbance
maximum of the reduced species is at 426 nm. Conditions: 0.016 mM in 50 mM phosphate buffer and 10 mM NaCl at
pH 7.0. (B) Electronic absorption spectra (at 25°C) and Nernst plot for the variant KatG(H123Q). X represents [(Aλredmax
– Aλred)/(Aλoxmax – Aλox)] with λox = 406 nm and λred = 439 nm. The absorbance maximum of the reduced species is at 429
nm. Conditions: 0.025 mM in 50 mM phosphate buffer and 10 mM NaCl at pH 7.0. (C) Electronic absorption spectra
(at 25°C) and Nernst plot for the variant KatG(R119N). X represents [(Aλredmax – Aλred)/(Aλoxmax – Aλox)] with λox = 407
nm and λred = 439 nm. The absorbance maximum of the reduced species is at 424 nm. Conditions: 0.02 mM in 50 mM
phosphate buffer and 10 mM NaCl at pH 7.0. (D) Electronic absorption spectra (at 25°C) and Nernst plot for the variant
KatG(S335G). X represents [(Aλredmax – Aλred)/(Aλoxmax – Aλox)] with λox = 409 nm and λred = 431 nm. The absorbance
maximum of the reduced species is at 425 nm. Conditions: 0.024 mM in 50 mM phosphate buffer and 10 mM NaCl at
pH 7.0. (E) Electronic absorption spectra (at 25°C) and Nernst plot for the variant KatG(E253Q). X represents [(Aλredmax
– Aλred)/(Aλoxmax – Aλox)] with λox = 406 nm and λred = 437 nm, where the reduced species exhibits its absorbance
maximum. Conditions: 0.035 mM in 50 mM phosphate buffer and 10 mM NaCl at pH 7.0
95
chapter five
Putative substrate binding site(s) in bifunctional
catalase-peroxidases
Jutta Vlasits, Marcel Zamocky, Christa Jakopitsch and Christian Obinger
Unpublished Data
Putative substrate binding site(s) in bifunctional catalase-peroxidases
Jutta Vlasits, Marcel Zamocky, Christa Jakopitsch and Christian Obinger
Department of Chemistry, Division of Biochemistry, Metalloprotein Research Group,
BOKU – University of Natural Resources and Applied Life Sciences, Muthgasse 18,
A-1190 Vienna, Austria
Introduction
Catalase-peroxidases (KatGs), present in prokaryotes and fungi, are unique
bifunctional enzymes since they exhibit a predominant catalase activity together with a
substantial peroxidatic activity with broad specificity. The latter becomes significant in the
presence of a suitable one-electron donor and at low (i.e. submillimolar) levels of
hydrogen peroxide. However, neither the in vivo peroxidatic substrate nor its binding site
have been identified, leaving the actual role of the peroxidatic reaction undefined. Though
being homologous to cytochrome c peroxidase (CcP) and ascorbate peroxidase (APX),
KatG is the only heme peroxidase that is able to dismutate hydrogen peroxide (2 H2O2 →
2 O2 + 2 H2O) by a non-scrambling mechanism (1) with rates similar to monofunctional
catalases. Together with ascorbate peroxidases, present in chloroplastic organisms, and
cytochrome c peroxidase, found mainly in mitochondrial organisms, KatGs constitute class
I of the plant, fungal, and bacterial heme peroxidase superfamily (2-4). The four available
crystal structures of KatG enzymes from Haloarcula marismortui (Protein Data Bank code
1ITK), Burkholderia pseudomallei (code 1MWV), Mycobacterium tuberculosis (code
1SJ2), and Synechococcus PCC 7942 (code 1UB2) (5-8) revealed that the heme pocket
contains catalytic residues virtually identical to those of other peroxidases belonging to
class I. In particular, the distal and proximal heme pockets contain the amino acid triads
His-Arg-Trp and His-Trp-Asp, respectively. However, features unique to catalaseperoxidases were also found; like the unusual covalent adduct consisting of the distal side
residues Trp122, Tyr249, and Met275 (in Synechocysits numbering) (5-8), and an arginine
(Arg439) that seems to act as a molecular switch with the guanidinium group being either
in association with the tyrosinate ion of the adduct (Y-conformation) or being shifted to a
region containing two other arginine residues (R-conformation) (Figure 1A) (3, 6).
The catalatic activity is closely related with the funnel-shaped main access channel
that leads to heme b, which is more deeply buried than in (monofunctional) peroxidases.
97
Moreover, the channel is longer and restricted near Asp152 allowing only small molecules
to enter the heme cavity. An ordered continuum of water molecules and rigid H-bonding
network is important for the H2O2 oxidation reaction. Whenever residues in the active site
that are part of this network were mutated and hence the hydrogen-bonds disrupted, the
catalatic activity was heavily reduced, while the peroxidase activity remained unaffected.
As mentioned above the binding and oxidation site of conventional peroxidase
substrates like guaiacol and ABTS as well as the nature of the endogenous one electron
(AH2) donor of catalase-peroxidases is unknown (H2O2 + 2 AH2 → 2 H2O + 2 AH). An
access route approximately in the plane of the heme, normally found in (monofunctional)
peroxidases is blocked by loops in KatG. However, from structural analysis of
Burkholderia KatG Carpena et al. (6) described another access route providing direct
access to the core of the protein. It is located in a region that encompasses two structural
features with possible functional significance. One is the presence of a cleft with a Ushaped region aligned with polar residues (Figure 1B). The second feature is related with a
putative role of the “flipping” Arg439 (Figure 1A). In one conformation Arg439 is located
on the surface at the bottom of the “U”, whereas in the other conformation it increases the
depth of the cleft and reduces the positively charges on its surface. Such a striking and
well-defined cavity, combined with the potential for functional changes through the simple
movement of a side-chain (Arg439), begs the suggestion that this cavity might be the
binding site for a peroxidase (i.e one-electron donating) substrate.
In order to test whether the U-shaped cleft functions as peroxidase substrate
binding site, we have performed a sequence alignment of 156 genes encoding both
prokaryotic and eukaryotic KatGs (Figure 2) in order to see conservation of relevant amino
acids. Finally, five residues were found to be conserved in almost all katG genes available
so far in public databases, namely Asn278, Glu281, Lys435, Trp459 and Arg509
(Synechocystis
numbering).
The
variants
Asn278Asp,
Glu281Asp,
Glu281Gln,
Lys435Gln, Lys435Arg, Trp459Phe, and Arg509Lys were recombinantly expressed in E.
coli and the overall catalase and peroxidase (guaiacol and ABTS) activity was
investigated. Furthermore, we have tested the reaction of the ferric proteins with both
hydrogen peroxide and peroxoacetic acid in order to see whether the spectral and kinetic
features follow the known pattern of wild-type KatGs (9). Data obtained for the variant
Arg439Ala are also presented for comparative studies.
98
(A)
(B)
Arg497
(R509)
Arg426
(R439)
R
Trp446
(W459)
Y
Lys422
(K435)
Asn267
(N278)
Glu270
(E281)
(C)
Figure 1. Detailed view of conserved amino acid residues in the U-shaped region of Burkholderia
pseudomallei KatG. Figures were constructed using PyMol. (A) KatG-specific residues in the vicinity of the
heme including the distal site covalent adduct Trp-Tyr-Met and the „flipping“ arginine, which can adopt two
conformations (Y and R) (corresponding Synechocystis residues are given in brackets). Additionally, the
position of the residues Asn267(278), Glu270(281), Lys422(435), Trp446(459), and Arg497(509) that are
highly conserved residues in U-shaped region. (B) Slab view showing the U-shaped region with the amino
acids Asn267 (red), Glu270 (yellow), Lys422 (blue), Trp446 (orange), Arg497 (cyan), and Arg426
(magenta), the covalent adduct (grey), and the heme (green). (B) U-shaped region in Burkholderia
pseudomallei KatG constructed using the PyMol plugin CASTp with the same coloured residues as in (A)
and (B).
99
(A)SSPPCCP01
SeCP
SsPCP01
BpCP01_K96
AmeCP01
DhaCP01
PprCP01_DS
PcaCP01
GsuCP01_PC
DaCP01
PphCP01
CphCP01-_D
PaeCP01
RnCP01
SpoCP01_DS
SIsCP
DarCP01_RC
GkaCP01
GstCP01
BhaCP01
SseeCP01
SsNASCP01
LvCP01
HsCP01
NpCP01
HmaCP01
AfCP01-Arc
TpsCP01
YpCP01
YpsCP01
E_ColiPCP_
BamCP02
BviCP02_G4
FtCP01
LpnCP01_Le
LpnCP01_su
LpnCP01_Pa
SdCP01
SonCP01
NeCP01
NarCP01
IlCP01
IbCP01
PatCP01
CpsCP01
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
--DHRYG-SEDRESLENPLAAVQMGLIYVNPEGVDGHP-DPLCTAQDVRTTFARMA----------------MNDEETVALTAGGHTVGKCHGNSKAE---NSRYTGD---RELENPLAAVTMGLIYVNPEGVDGNP-DPLKTAHDVRVTFARMA----------------MNDEETVALTAGGHTVGKCHGNGNAA--------RHTTEGALDQPLAAVEMGLVYVNPEGPHGHP-DPVASGPDVRDTFARMG----------------MTMEETVALVAGGHTFGKCHGAAPVS-----------GDRQLENPLAAVQMGLIYVNPEGPDGNP-DPVAAARDIRDTFARMA----------------MNDEETVALIAGGHTFGKTHGAGPAS--------RYSGDRELENPLAAVQMGLIYVNPEGPNGQP-SALASGKDIRDTFARMA----------------MNDEETVALVAGGHTFGKCHGAGSAT--------RYSGDRDLENPLAAVQMGLIYVNPEGPNGQP-SVLASGRDVRDTFKRMA----------------MNDEETVALVAGGHTFGKCHGAGPAS-------SRYSGDRDLENPLAAVQMGLIYVNPEGPDGNP-DPVASGRDVRETFGRMA----------------MNDEETVALVAGGHTFGKCHGAGPAT-------SRYSGDRDLENPLAAVQMGLIYVNPEGPDGNP-DPVASGRDVRETFARMA----------------MNDEETVALVAGGHTFGKCHGAGPAT--------RYSGDRDLENPLAAVQMGLIYVNPEGPNGEP-DPVASGRDVRETFARMA----------------MNDEETVALVAGGHTFGKCHGAGPAT-------SRYSGERDLDNPLAAVQMGLIYVNPEGPDGNP-DPVASGKDVRETFARMA----------------MNDEETVALVAGGHTFGKCHGAGDPS--------RYSGERDLENPLAAVQMGLIYVNPEGPDGNP-DPVAAGRDIRETFARMA----------------MNDEETVALVAGGHTFGKCHGVGDPK--------RYSGERDLENPLAAVQMGLIYVNPEGPDGNP-DPVAAGRDIRETFARMA----------------MNDEETVALVAGGHTFGKCHGVGDPN--------RYSGERDLENPLAAVQMGLIYVNPEGPDGKP-DPVAAGRDIRETFARMA----------------MNDEETVALVAGGHTFGKCHGVGDPK-------SRYSGERQLDNPLAAVQMGLIYVNPEGPDGNP-DPVASGRDVRDTFTRMG----------------MTDEETVALVAGGHTFGKAHGAGDPI-------SRYSGARDLENPLAAVQMGLIYVNPEGPDGNP-DIVASGHDVIETFGRMA----------------MDEAETVALVAGGHTFGKAHGNGPAS-------ARYSGDRVLEDPLAAVQMGLIYVNPEGPDGNP-DPIASGRDIRETFARMA----------------MNDEETVALTAGGHTFGKAHGNGPVD-------SRYSGERNLDNPLAAVQMGLIYVNPEGPDGNP-DPVASGRDIRETFARMA----------------MNDEETVALTAGGHTFGKAHGAGDPA-------ERYSGDRELENPLAAVQMGLIYVNPEGPDGKP-DPVAAARDIRETFRRMG----------------MNDEETVALIAGGHTFGKAHGAGPAS-------ERYSGDRELENPLAAVQMGLIYVNPEGPDGKP-DPKAAARDIRETFRRMG----------------MNDEETVALIAGGHTFGKAHGAGPAT-------KRYSGDRELENPLAAVEMGLIYVNPEGPDGKP-DPIKAAHDIRETFGRMG----------------MNDEETVALIAGGHTFGKAHGAGNPD-------KRYSQERALEDPLAAVQMGLIYVNPEGPDGKP-DLQGSANDIRDTFGRMA----------------MNDYETVALIAGGHTFGKAHGAVGGE-------KRYSQERALEDPLAAVQMGLIYVNPEGPDGKP-DLQGSANDIRDTFGRMA----------------MNDYETVALIAGGHTFGKAHGAVGGE-------NRYRGERELENPLAAVQMGLIYVNPEGPNANP-DPLASAHDIRETFARMA----------------MNDEETVALVAGGHTFGKSHGAGDAA-------DRFDADGSLKWPLGNTVMGLIYVNPEGPNGEP-DLEGSAKNIRESFGKMA----------------MNDKETVALIAGGHTFGKVHGADDPE-------DRYDEAGELPEPLGATVMGLIYVNPEGPDGEP-DLEGSAANIRESFGRMA----------------MNDEETVALIAGGHTFGKVHGADDPD-------ERFDEPGEIQEGLGASVMGLIYVNPEGPDGNP-DPEASAKNIRQTFDRMA----------------MNDKETAALIAGGHTFGKVHGADDPE--------KRGEKEELERPFAATEMGLIYVNPEGPGGNP-DPLGSAQEIRVAFRRMG----------------MNDEETVALIAGGHAFGKCHGAG-PA------RSAEDDSTGLEKPLGAVQMGLIYVNPEGPGGNP-DILASAKDIRETFSRMG----------------MSDFETVALIAGGHTFGKAHGSADPS-------NR-DKNGKLPKPLAATQMGLIYVNPEGPNGKP-DPVAAAKDIREAFARMA----------------MNDEETVALIAGGHTFGKAHGAASPE-------NR-DKNGKLPKPLAATQMGLIYVNPEGPNGKP-DPVAAAKDIREAFARMA----------------MNDEETVALIAGGHTFGKAHGAASPE-------NR-DKNGKLQKPLAATQMGLIYVNPEGPGGKP-DPLASAKDIREAFSRMA----------------MDDEETVALIAGGHTFGKAHGAASPE-------NR-DKNGKLEKPLAATQMGLIYVNPEGPNGNP-DPVAAAKDIRDAFGRMA----------------MNDEETLALIAGGHTFGKAHGAASPD-------NR-DKNGKLEKPLAATQMGLIYVNPEGPNGNP-DPVAAAQDIREAFGRMA----------------MNDEETLALIAGGHTFGKAHGAASPD-------SN-VRDGKLAPAYAATQMGLIYVNPEGPDGKP-DIKGAASEIRQAFRAMG----------------MTDKETVALIAGGHTFGKTHGAVPEDKV
------KRQDKVGKLEKPLAATVMGLIYVNPEGPNGVP-DPLAAAEKIRETFGRMA----------------MNDEETVALIAGGHAFGKTHGAASGK-------KRQDKDGKLEKPLAATVMGLIYVNPEGPNGVP-DPLAAAEKIRETFGRMA----------------MNDEETVALIAGGHAFGKTHGAASGK-------KRQDKDGKLEKPLAATVMGLIYVNPEGPNGVP-DPLAAAEKIRETFGRMA----------------MNDEETVALIAGGHAFGKTHGAASGK-------RRHSGDRKLDKPFGATEMGLIYVNPEGPHGNP-DPIAAAHDIRQAFGRMG----------------MSDEETVALIAGGHTFGKAHGAHKPS-------KRHNGKGELAKPMGATQMGLIYVNPEGPNGVP-DPLASAKEIRDTFGRMA----------------MNDEETVALIAGGHTFGKAHGAHDPA-------ERHDKRGELAKPLAAVQMGLIYVNPEGPGGNP-DPLAAARHIRESFGRMA----------------MNDEETVALIAGGHTFGKAHGAHKPE-------QRYHGDRKLQNPLAAVQMGLIYVNPEGPNGNP-DPLLAAKDIRETFGRMA----------------MNDEETVALIAGGHTFGKAHGARKPE-------ERYSGDRELENPLAAVQMGLIYVNPEGPNGKP-DPLLAAKDIRDTFGRMA----------------MNDEETVALIAGGHTFGKAHGAHKPE-------KRYKGERKLDNPLAAVQMGLIYVNPEGPNGNP-DPLLAAKDIRDTFGRMA----------------MNDEETVALIAGGHTFGKAHGAHKPE-------ERYSGERKLEEPLAAVQMGLIYVNPEGPNGNH-DPISAAADIRDVFARMA----------------MNDEETVALIAGGHTFGKAHGAHKPD-------KRRDKKGKLKGPLAAVEMGLIYVNPEGPHGKP-DPLLAANDIRMSFGRMA----------------MNDEEIVALLAGGHTLGKAHGAKKPN--
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
302
274
271
291
271
271
277
280
268
277
273
273
273
270
279
279
247
276
276
275
277
277
274
260
253
271
260
295
276
276
276
281
281
278
282
282
282
279
282
281
291
291
204
294
281
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
484
454
454
472
454
454
460
463
451
460
456
456
456
453
460
462
430
458
458
457
460
460
457
444
437
455
442
483
458
458
458
463
463
463
465
465
465
463
466
465
473
475
388
478
465
(B)SSPPCCP01
SeCP
SsPCP01
BpCP01_K96
AmeCP01
DhaCP01
PprCP01_DS
PcaCP01
GsuCP01_PC
DaCP01
PphCP01
CphCP01-_D
PaeCP01
RnCP01
SpoCP01_DS
SIsCP
DarCP01_RC
GkaCP01
GstCP01
BhaCP01
SseeCP01
SsNASCP01
LvCP01
HsCP01
NpCP01
HmaCP01
AfCP01-Arc
TpsCP01
YpCP01
YpsCP01
E_ColiPCP_
BamCP02
BviCP02_G4
FtCP01
LpnCP01_Le
LpnCP01_su
LpnCP01_Pa
SdCP01
SonCP01
NeCP01
NarCP01
IlCP01
IbCP01
PatCP01
CpsCP01
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
RHNPIMTDADMAMIKDP---IYRQISERFYREPDYFAEVFAKAWFKLTHRDLGPKSRYLGPDVPQEDLIWQDPIP-PVDYTLSEGEIKELE----QQILA
RRNLVMTDADMAMKMDP---EYRKISERFYQDPAYFADVFARAWFKLTHRDMGPKARYIGPDVPQEDLIWQDPIP-AGNR---NYDVQAVK----DRIAA
ASAPIMTTADLSLRHD---PLMEPVARRFHQDQDAFADAFARAWFKLTHRDLGPRALYLGADVPEEIQIWQDPVPALDHPLIGAVEIKALK----QKLLA
KHRPTMLTTDLSLRFD---PAYEKISRRFHENPEQFADAFARAWFKLTHRDMGPRARYLGPEVPAEVLLWQDPIPAVDHPLIDAADAAELK----AKVLA
RHAPMMTTADLALRMD---PIYGPIAKGFRENPEAFADAFGRAWFKLTHRDMGPRTRFLGPEVPTEELIWQDPVPAVDFELIDEDDITGLK----GKIID
RHAPMMTTADLSLRMD---PIFGPIAKRFRDNPEEFADAFARAWFKLTHRDMGPRSRYLGPEVPEEELIWQDPVPPVDHELIDEQDIADLK----AKLLA
KHRPMMTTADLSLRFD---PIYEPIARDYQQNPEKFADAFARAWFKLTHRDMGPRSRYLGAEVPAEELIWQDPVPTVDHQLIDGQDIAALK----DTILA
KFRPMMTTADLSLRFD---PIYEPIARRYLENPEEFADAFARAWFKLTHRDMGPRSRYLGSEVPAEELIWQDPVPAVDHELIDAADIADLK----VKILA
KRRPMMTTADLSLRFD---PIYEPIARRYLANPEEFADAFARAWFKLTHRDMGPRSRYLGPDVPAEELIWQDPVPAVTHQLIDRQDIAALK----GTILA
KVRPMMTTADLSLKFD---PIYEPIARDYLANPEKFADAFARAWFKLTHRDMGPRSRYLGPEVPEEDLIWQDPVPAVDHPLIDDADIASLK----AAILS
RVPTMMTTADLAMRMD---PLYGPIARRYYEHPEQFADAFARAWFKLTHRDMGPRSRYLGAEVPAEELIWQDPVPAVDHELIGEGKIKELK----KRILA
RVPTMMTTADLAMRMD---PVYGPISRRYYEHPDQFADAFARAWFKLTHRDMGPKSRYLGAEVPAEDLIWQDPVPAVDHELIGEGEIAELK----KRLLA
RVPTMMTTADLAMRMD---PIYGPISQRYYEHPDQFADAFARAWFKLTHRDMGPHSRYLGAEVPAEELIWQDPVPALDHDLIDAEEIAELK----KRLLA
KHAPMMTTADMSLRMD---EGFEKISRDFHANPDKFADAFARAWFKLTHRDMGPKSRYLGSDVPAEDLIWQDPLPARDYDLIDASDIAQLK----KDIAA
KHRPMMTTADLSLRYH---PKLLPHAKRFAEDPAAFADAFARAWFKLTHRDMGPRARYLGKEVPAEELIWQDPIPAG--TQIDADDAAALK----AQILA
KVPIMMTTADMAMKMD---PIYGPISKRFHENPEEFAEAFKRAWFKLTHRDMGPKACYLGADVPDEDLIWQDPLPAVDHALIEEADIASLK----ADILA
KHPPMMTTADLAMRFD---PIYGPISRRFHQDPAAFADAFARAWFKLTHRDLGPKARYLGPEVPAEDLVWQDPIPAVDHPLIEVTDVASLK----AKLLA
KVPPMMMTTDLALRFD---PEYEKIARRFYENPEEFADAFARAWFKLTHRDMGPKTRYLGPEVPKEDFIWQDPIPTVDYELSD-AEIEEIK----AKILN
KVPTMMMTTDLALRFD---PEYEKIARRFHQNPEEFAEAFARAWFKLTHRDMGPKTRYLGPEVPKEDFIWQDPIPEVDYELTE-AEIEEIK----AKILN
KVPTVMLTTDLALRHD---PEYEKISRRFHKNPDEFADAFARAWFKLLHRDMGPKARYLGPEVPAEDFIWQDPVPTVDYELTD-AEVEELK----AKILD
RHAPIMFTSDLALRED---PEYFKISTRFHENPEEFADAFARAWFKLTHRDMGPIARYLGKDVPSEELIWQDPIPAVTHTLVDDADVASLK----AKVLE
RHAPIMFTSDLALRED---PEYFKIATRFHENPEEFADAFARAWFKLTHRDMGPIARYLGKDVPSEELIWQDPIPAVTHTLVDDADVASLK----AKVLE
RHAPMMTTADLALKLD---PAYAKISKHYHENPEIFADAYARAWFKLTHRDMGPKALYLGNEVPAEDLIWQDPIPAVDHDLISSADVTALK----AQIMA
TEDVMMLTTDVALKDD---PDYREVLETFQENPREFQQSFSKAWYKLIHRDMGPSERFLGPEVPEETMIWQDPLPDADYDLVDDEAVAALK----SELLE
KEDVMMLTTDVALKRD---PDYREVLERFQENPDEFQEAFAKAWYKLIHRDMGPPERFLGPEVPEETLIWQDPLPDADYDSIGDEEVAELK----EALLD
KQTPMMLTTDIALKRD---PDYREVMETFQENPMEFGMNFAKAWYKLTHRDMGPPERFLGPEVPDEEMIWQDPLPDADYDLIGDEEIAELK----EEILD
KHRPRMLTADLALRFD---PEFSKIARRFLENPEEFEKAFAIAWYKLTHRDMGPKDCYIGKYVPEETFVWQDPLPRRDYELVDEKDVEELK----RRILA
KHLPIMFTTDLALRYD---PIYGPISQRFHLNPHEFTDAFKRAWYKLCHRDMGPLQRHLGQWLPTEDLIWLDPIPSSNGNTINVNDVSILKSKISDLINS
FHPLMMFTTDIALKVD---PEYKKITTRFLENPEEFKMAFARAWFKLTHRDMGPAARYLGDEVPKETFIWQDPLPAANYKMIDSADISELK----DKILK
FHPLMMFTTDIALKVD---PEYKKITTRFLENPEEFKMAFARAWFKLTHRDMGPAARYLGDEVPKETFIWQDPLPAANYKMIDSADISELK----DKILK
LHPLMMFTTDIALKVD---PEYKKITTRFLNDPKAFEQAFARAWFKLTHRDMGPAARYLGNEVPAESFIWQDPLPAADYTMIDGKDIKSLK----EQVMD
RHPLMMFTTDIALKVD---PAYSAIAKRFHAHPEEFKLAFAKAWFKLTHRDLGPKARYLGKDVPKVDLIWQDPLPVAGYQMIGDADIAELK----RRILA
LHPLMMFTTDIALKVD---PAYSKIAKRFEEHPDEFKLAFAKAWFKLTHRDLGPKARYLGKDVPKVDLIWQDPLPVAGYQTIGAADIADLK----SRILA
KHKPIMFTTDLALKED---DGFNKYTQEFYNNPEEFKEEFAKAWFKLTHRDMGPKSRYIGPWIPEQNFIWQDPVPAADYKQVSTQDIAQLE----QDIIN
RHAPVMLTTDLALKFD---PVYSKIAKRFLDNPKEFDDAFARAWFKLIHRDMGPRSRYLGSLVPKEIMIWQDPVPPVDYKLVDANDIANLK----GKILN
RHAPVMLTTDLALKFD---PVYSKIAKRFLDNPKEFDDAFARAWFKLIHRDMGPRSRYLGSLVPKEAMIWQDPVPPVDYKLVDANDIANLK----GKILN
RHAPVMLTTDLALKFD---PVYSKIAKRFLDNPKEFDDAFARAWFKLIHRDMGPRSRYLGSLVPKEIMIWQDPVPPVDYKLVDANDIANLK----GKILN
RHAPIMLTTDLALKED---PAYRKITQRWLEDPEEFTRAFARAWFKLTHRDMGPVSRYKGELVPSDTFVWQDPVPVADYKQIGERDVKKLK----AAILD
RHAPIMFTSDIALKAD---PIYREITTRFLKNPQEFELAFAKAWFKLTHRDLGPKARYLGADVPAEALIWQDPIPALDHPLIDNADIKALG----NKILA
RHAPIMFTTDIALKVD---PEYQKISKRFLDNPKEFGIAFAKAWFKLTHRDMGPKARYVGAEIPAEDFIWQDPIPKVDHKLIDAQDIARLK----SAILA
LHKPIMFTTDLALKTD---PAYRKITTKFRQNPDAFADAFARAWFKLTHRDMGPRWRYLGAMVPAEELIWQDPVPKATYAMIDAADVSALK----GRILA
RHAPIMFTTDLSLKED---PEYRKIAKRFHEDPKEFELAFAKAWFKLTHRDMGPKQSYLGDMAPQEDLLWQDPIPAVDFELINENDVEQLK----VAILD
RHAPIMFTTDLALKFD---PEYRKIAKRFHENPEEFELAFAKAWFKLTHRDMGPKARYLGSMAPKEDLIWQDPIPKVDYTLINESEISELK----KQILA
RHAPIMFTTDLALKED---PQYRKIAERFHANPKEFELAFSKAWFKLTHRDMGPRARYVGDESPTDDFLWQDPIPSVDYSLIDKRDIQHLK----TKLLD
RHAPIMFTTDIALKED---PQFRKIVERFRADPTQFDLAFAKAWFKLTHRDMGPRARYVGAEVPSEVLMWQDPIPAINYQLITDKDIKQLK----KQITN
100
(C)SSPPCCP01
SeCP
SsPCP01
BpCP01_K96
AmeCP01
DhaCP01
PprCP01_DS
PcaCP01
GsuCP01_PC
DaCP01
PphCP01
CphCP01-_D
PaeCP01
RnCP01
SpoCP01_DS
SIsCP
DarCP01_RC
GkaCP01
GstCP01
BhaCP01
SseeCP01
SsNASCP01
LvCP01
HsCP01
NpCP01
HmaCP01
AfCP01-Arc
TpsCP01
YpCP01
YpsCP01
E_ColiPCP_
BamCP02
BviCP02_G4
FtCP01
LpnCP01_Le
LpnCP01_su
LpnCP01_Pa
SdCP01
SonCP01
NeCP01
NarCP01
IlCP01
IbCP01
PatCP01
CpsCP01
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
S-GLTVSELVCTAWDSARTFRSSDYRGGANGARIRLEPQKNWPGNEPT-RLAKVLAVLENIQANFAKP-------VSLADLIVLGGGAAIAKAALDGG-I
S-GLSISELVSTAWDSARTYRNSDKRGGANGARIRLAPQKDWEGNEPD-RLPKVLAVLEGISAATG---------ATVADVIVLAGNVGVEQKARAAG-V
T-GCSVGALVATAWGAASTFRGSDRRGGANGGRIRFQPQNTWEVNDPE-QLRSVLQTLETVQQQFNAEATGGQR-VSMADLIVLAGSAAVEQAAAAGG-H
S-GLTVSQLVSTAWAAASTFRGSDKRGGANGARIRLAPQKDWEANQPE-QLAAVLETLEAIRTAFNGAQRGGKQ-VSLADLIVLAGCAGVEQAAKNAG-H
S-GLSLSELVSTAWASASSFRGSDKRGGANGARIRLAPQKDWEVNQPE-QLQTVLQALEKIQDTFNSAQSGKKR-ISLADLIVLAGCAGVEQAAKNAG-F
S-GLSVSQLVSTAWASASTFRGSDKRGGANGARIRLAPQKDWEINQPA-QLAEVLAALEGIQTEFNSSQSGTKK-VSLADLIVLGGAAAIEQAARNAG-Q
S-GLSVSQLVSTAWASASTFRGSDKRGGANGGRIRLAPQKDWEVNQPA-QLKTVLQTLERIQKEYNDAQSGGKR-VSLADLIVLGGCAGIEQAAKNAG-Q
S-GLSIPQLVSTAWASASTFRGSDKRGGANGARIRLAPQKDWEVNQPA-QLKTVLQTLEGIQQAFNGAQAGGKK-VSLADLIVLGGCAAVEQAAGNAG-H
S-GLSIADLVATAWASASTFRGSDKRGGANGARIRLAPQKDWAVNQPA-RLATVLQALEGIRQEFNNAQSGGKQ-VSLADLIVLGGCAAVEQAAKKAG-H
T-GLSVAELVSTAWASASTFRGSDKRGGANGARIRLAPQKDWEVNQPK-ILDAVLQKLGSVQQQFNAAQAGNKK-VSLADLIVLGGCAAVEQAARVAG-C
S-GLSIPELLSTAWASASTFRSSDKRGGANGSRIRLSPQKEWEVNQPE-QLQRVLGKLEEIRNAFNGEQSGGKQ-VSLADLIVLGGCAAVEEAARRAG-N
S-GLPIPELVSTTWASASTFRGSDKRGGANGSRIRLAPQKDWEVNQPE-QLQRVLEKLEEIRHAFNGEQSGGKR-VSLADLIVLGGCAAVEEAARRAG-N
S-GLSIPELVSTAWASASTFRGSDKRGGANGARIRLAPQKDWEVNQPE-QLQRVLHKLEEIRNTFNGEQSGNKQ-VSLADMIVLGGCAAVEEAAGKAG-T
T-GLSVADMVSTAWASASTYRGSDHRGGANGARIRLAPQKDWEANEPA-KLAKVLDTLEGVQTAFNSAQS-GKA-VSLADLIVLAGVVGVEKAAADAG-H
S-GLSVSDMVSTAWASASTFRGSDMRGGANGARIRLAPQKDWEVNEPA-KLARVLGVLEGIQAGFSSG---GKS-VSMADLIVLAGCAGVEQAAKAGG-H
S-GLSVQDLVYVAWSSASTFRGSDKRGGANGARIRLAPQKDWEVNEPA-KLEKVLNALEGIQSSFNAAASGGKK-VSLADLIVLAGSAAVEKAAKDAG-F
S-GLSTAELVSTAWASASTFRGSDKRGGANGARIRLAPQKDWAANQPA-QLAKVLGVLEGIQQAFNSAQTGGKK-VSLADLIVLGGCAAVEAAAKAAG-F
S-GLTVSELVKTAWASASTFRNSDKRGGANGARIRLAPQKDWEVNEPE-RLAKVLSVYEDIQRELP------KK-VSIADLIVLGGSAAVEKAARDAG-F
S-GLTVSELVKTAWASASTFRNSDKRGGANGARIRLAPQKDWEVNEPE-RLAKVLSVYEDIQRELP------KK-VSIADLIVLGGSAAVEKAARDAG-F
S-GLTVSELVTTAWASASTFRNSDKRGGANGARIRLAPQKDWEVNQPE-QLEKVLSVLENIQSQLD------KK-VSIADLIVLGGSAAVEKAAKEAG-F
S-GLSVQELVATAWGSAATFRGSDKRGGANGARIRLAPQKDWPVNEPD-QLAKVLGVLETIRADFNNAQSGDKE-ISVADLIVLAGAAGIEKAAKDAG-H
S-GLSVQELVATAWGSAATFRGSDKRGGANGARIRLAPQKDWPVNEPD-QLAKVLGVLETIRADFNNAQSGDKE-ISVADLIVLAGAAGIEKAAKDAG-H
T-GLSVQELVRTAWASASTFRGSDNRGGANGARIRLAPQKDWTVNAPA-ELAKVLAALEGVQSAFNAAQSGNKR-VSLADLIVLGGSAAIEAAAKAGG-H
S-ELSIPQLVKTAWASASTYRDSDKRGGANGARIRLEPQRSWEVNEPE-QLEAALSTYEDIQAEFND-ARSDDMRVSLADLIVLGGNAAIEQAAADAGYS-ELSVAQLVKTAWASASTYRDSDKRGGANGARIRLEPQRSWEVNEPA-ALADALETYEAIQEEFNS-ARSDAVRVSLADLIVLGGNAAVEQAAADAGYS-DLSVSQLVKTAWASASTYRDSDKRGGANGARLRLEPQKNWEVNEPE-QLETVLGTLENIQTEFND-SRSDGTQVSLADLIVLGGNAAVEQAAANAGYS-GLSLSQLVYFAWASASTYRNSDRRGGANGARIRLKPMSVWEVNHPE-ELKKVIAAYEKIQQEFNEGAKGSEKRISIADLIVLGGIAAVEEAARRAGFS-TLSVSDLVKAAWASASTYRCTDHRGGANGGRIRLNPQKSWDVNDPS-SLGKVIVTLESIQQNFNAMN--SNQ-VSFADLVVLGGNVAIEEAARRAGHY
T-GLSDTKLIKTAWASASTFRGTDFRGGDNGARIRLAPQKDWPVNDPA-ELHSVLAALMEVQNNFNKDRSDGKK-VSLSDLIVLGGNAAIEDAAKKAGYT-GLSDTKLIKTAWASASTFRGTDFRGGDNGARIRLAPQKDWPVNDPA-ELHSVLAALMEVQNNFNKDRSDGKK-VSLSDLIVLGGNAAIEDAAKKAGYL-GIPASELIKTAWASASTFRVTDYRGGNNGARIRLQPEINWEVNEPE-KLKKVLASLTSLQREFNKKQSDGKK-VSLADLIVLSGNAAIEDAARKAGVS-GVPKAELIKTAWASAASFRATDYRGGANGARIRLAPENDWAVNDPA-SLSKVLKSLEGIQSGFNRNRTDGKQ-VSLADLIVLGGSAAVEDAARKAGYS-GLPKSELIKTAWASAASFRGTDYRGGANGARIRLAPENGWAVNDPA-SVAKVLKSLEAIQSGFNKGRTDGKQ-VSLADLIVLGGSAAVEDAARKAGYS-GLTNQQLIKTAWDSASTYRKTDYRGGSNGARIALAPEKDWQMNEPA-KLEVVLTKLKEIQTNFNNSKTDGTK-VSLADLIVLGGNVGVEQAAKQAGYS-GLTTPELVKTAWASASTFRGTDMRGGANGARIRLAPQKDWPANDPQ-ELAKVLKTLESIQNNFNNAQADGKK-ISLADLIVLGGNAAIEQAAKQAGYS-GLTTSELVKTAWASASTFRGTDMRGGANGARIRLAPQKDWPANDPQ-ELAKVLKTLESIQNNFNNAQADGKK-ISLADLIVLGGNAAIEQAAKQAGYS-GLSTSELVKTAWASASTFRGTDMRGGANGARIRLSPQKDWPANDPQ-ELAKVLKTLESIQNNFNNAQADGKK-ISLADLIVLGGNAAIEQAAKQAGYS-GLSTSDLVKTAWASAASFRTTDMRGGANGARIRLAPQKDWAVNQPQ-DLQRVLKVLEGVQREFNK-KSRKTK-VSLADVIVLGGAAAIEQAAKKAGHS-GLTVPELVRTAWASASSFRGTDMRGGANGARIRLEPMMNWQANNPK-ELAKVLAKLEKVQKDFNGSLKGSKK-VSLADVIVLGGSVAVEKAAKEAGVS-GLTIPELVRTAWASAATFRGTDMRGGANGARIRLAPQKDWEVNDPA-ELAKVLTRLESIRNNFNRTLKGGKK-VSFADVIVLGGSAAVEEAARKAGYT-GLTGPELVRAAWASAASFRGTDMRGGTDGGRIRLAPQKDWAANNPA-ELARVLKALEGVATEFNRAAKDGKK-VSVANLVVLGGNAAIEQAAAKAGVS-GLSVPQLVRTAWASASSFRGTDMRGGANGARIALEPQMNWEANNPA-ELKKVLDTLKGVQEDFNDELSGDKY-VSLADVIVLGGAAAIEKAGKDAGYS-KLTVPELVRTAWAAASSFRATDMRGGANGARIALEPQINWEANNPE-ELRRVLNELKRIRAEFNQSLSGNKQ-ISLADTIVLAGAAAIEKAAEDAGYG-DVTPAQLVKTAWAAAASFRATDMRGGVNGARIQLAPQKNWEVNNPD-ELNTVLMFLNKVKKDFNDGQSGNKQ-VSLADLIVLGGAAAIEHAASKNDFS-GLTTSELVRTAWAAASSHRVTDMRGGANGARINLEPQNSWAVNNPK-ELGKVLAKLEGIQARFNK-KSAKTK-VSLADVIVLGGATAIENAAAKAGN-
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
574
542
550
568
550
550
556
559
547
556
552
552
552
548
553
558
526
548
548
547
556
556
553
540
533
551
539
578
554
554
554
559
559
559
561
561
561
558
562
561
569
571
484
574
560
Figure 2. Selected parts of the multiple sequence alignment of 156 KatGs. 45 complete bacterial sequences
including residues that contribute to the structural composition of the putative substrate binding site.
Abbreviations of analyzed sequences correspond to PeroxiBase nomenclature (http://peroxibase.isb-sib.ch/ for all
details)*. Arrows indicate the residues mutated in the present study. (A) Asn278, Glu281. (B) Lys435, Arg439,
Trp459. (C) Arg509.
* Passardi, F., Theiler, G., Zamocky, M., Cosio, C., Rouhier, N., Teixera, F., Margis-Pinheiro, M., Ioannidis, V.,
Penel, C., Falquet, L., and Dunand, C. (2007) PeroxiBase: the peroxidase database, Phytochemistry 68, 16051611.
101
Materials and Methods
Reagents – Standard chemicals and biochemicals were obtained from Sigma at the
highest grade available.
Mutagenesis
–
A pET-3a expression vector containing the cloned catalase-
peroxidase gene from the cyanobacterium Synechocystis PCC 6803 (10) was used as the
template for PCR. Oligonucleotide site-directed mutagenesis was performed using PCRmediated introduction of silent mutations as described previously (11). Unique restriction
sites flanking the region to be mutated were selected. The flanking primers for the
Synechocystis KatG variants Asn278Asp, Glu281Asp, and Glu281Gln were 5′-AAT GAT
CAG GTA CCG GCC AGT AAA TG-3′ (containing a KpnI restriction site) and 5′-AGT
GCA GAC TAG TTC GGA AAC G-3′ (containing a SpeI restriction site). The mutant
primer with the desired mutation changing Asn278 to Asp and a silent mutation
introducing a PauI restriction site was 5′-TTT GCG CGC ATG GCC ATG GAT GAC
GAG GAA AC-3′ and the internal 3′-primer with the same introduced restriction site was
5′-CAT GCG CGC AAA GGT AGT G-3′ (with point mutations in italics and introduced
restriction sites underlined). To substitute Glu281 with Asp and Gln, respectively, the used
internal 5′-primer with a PstI restriction site introduced by silent mutation was 5′-CTT
ACT GCA GGG GGG CAC AC-3′ and the mutant primers with the same introduced
restriction site were 5′-CC CCC TGC AGT AAG GGC AAC GGT ATC CTC GTC ATT
C-3′ and 5′-CC CCC TGC AGT AAG GGC AAC GGT TTG CTC GTC ATT C-3′,
respectively.
For the variants Lys435Gln, Lys435Arg, and Trp459Phe the flanking primers were
5′-GG CAC CCG GAT CCT TTA TG-3′ (containing a BamHI restriction site) and 5′-AGT
GCA GAC TAG TTC GGA AAC G-3′ (containing a SpeI restriction site). Following
internal primers were used to substitute Lys435 with Gln and Arg, respectively: 5′-GCC
AAA GCT TGG TTT CAA CTA ACT CAC-3′ changed Lys435 to Gln, 5′-CC AAA GCT
TGG TTT AGA CTA ACT CAC-3′ changed Lys435 to Arg, and 5′-A CCA AGC TTT
GGC AAA TAC-3′ a 3′-primer used for both KatG variants (underlined the introduced
HindIII restriction site). To substitute Trp459 with Phe the following internal primers
containing an AccIII restriction site were used: 5′-T GGT CCG GAT GTG CCC CAG
GAA GAT TTA ATT TTT CAA GAC CCC-3′ and 5′-AC ATC CGG ACC AAG GTA
AC-3′. The flanking primers 5′-GG CAC CCG GAT CCT TTA TG-3′ (containing a
BamHI restriction site) and 5′-GCC CCT AGG GAG ATC AAA CCG G-3′ (containing an
AvrII restriction site) and the internal primers, both with an introduced BbeI restriction
102
site, 5′-GGG GGC GCC AAT GGA G-3′ and 5′-ATT GGC GCC CCC TTT ATA GTC
GGA G-3′, the latter containing the desired mutation, were used to create the KatG variant
Arg509Lys.
Expression and Purification – The mutant recombinant catalase-peroxidases were
expressed in Escherichia coli BL21(DE3)pLysS and purified as described previously (10,
12) using the same conditions as for the wild-type enzyme.
Circular dichroism – CD studies were carried out using a PiStar-180 spectrometer
from Applied Photophysics. Far-UV (190–250 nm) experiments were performed using a
protein concentration of 2 µM and a cuvette with 1 mm path length. A good signal-tonoise ratio in the CD spectra was obtained by averaging 3 scans (step, 1 nm; bandwidth, 3
nm; default sampling size, 150 000; maximum sampling size, 500 000). The protein
concentration was calculated from the known amino acid composition and absorption at
280 nm according to Gill and Hippel (13).
Steady-state kinetics – Catalase activity was determined polarographically in 50
mM phosphate buffer using a Clark-type electrode (YSI 5331 oxygen probe) inserted into
a stirred water bath (YSI 5301B). All reactions were performed at 30°C and started by
addition of KatG. One unit of catalase is defined as the amount that decomposes 1 µmol of
H2O2/min at pH 7.0 and 30°C. To cover the pH range 3.0–9.0, 50 mM citrate/phosphate
and 50 mM Tris-HCl buffer were used.
Peroxidase activity was monitored spectrophotometrically using 1 mM H2O2 and 5
mM guaiacol (ε470 = 26.6 mM-1 cm-1) or 2.5 mM H2O2 and 0.3 mM ABTS (ε414 = 36.8
mM-1 cm-1). Alternatively, 1 mM peroxoacetic acid (PAA) and 5 mM guaiacol were used.
One unit of peroxidase is defined as the amount that oxidizes 1 µmol of electron
donor/min at pH 5.5 and 25°C. To cover the pH range 3.0–9.0 same buffers as for catalatic
activity were used.
Transient-state kinetics – Transient-state measurements were made using a
SX.18MV stopped-flow spectrometer (Applied Photophysics Ltd.) equipped with a 1 cm
observation cell. Calculation of pseudo first-order rate constants (kobs) from experimental
traces at the Soret maximum was performed with a SpectraKinetic work station (Version
4.38) interfaced to the instrument. The substrate concentrations were at least five times that
of the enzyme to allow determination of pseudo first-order rate constants. Observed
pseudo-first-order rates were plotted as a function of substrate concentration to obtain
apparent second-order rate constants for compound I formation. To follow spectral
transitions, a PD.1 photodiode array accessory (Applied Photophysics Ltd.) connected to
103
the stopped-flow device together with XScan diode array scanning software (Version 1.07)
was utilized. The kinetics of oxidation of ferric catalase-peroxidase to compound I by
peroxoacetic acid were followed in the single mixing mode. Catalase-peroxidase and
peroxoacetic acid were mixed to give a final concentration of 2 µM enzyme and 20–200
µM PAA. The first data point was recorded 1.5 ms after mixing, and 2000 data points were
accumulated. Sequential mixing stopped-flow analysis was used to monitor compound I
reduction by one-electron donors. In the first step, the enzyme was mixed with
peroxoacetic acid, and after a defined delay time, the formed compound I was mixed with
ascorbate as the electron donor.
All stopped-flow measurements were conducted at 25°C in 50 mM phosphate
buffer, pH 7.0, and at least three determinations were performed per substrate
concentration.
Results
Spectral properties – The UV-Vis absorption spectra of wild-type KatG and
variants exhibited the typical bands of a heme b containing peroxidase. The Soret band
varied between 406 and 407 nm and the two bands in the visible region were at 510 and
637 nm (CT).
The far-UV (190-250 nm) CD spectrum is a sensitive probe of protein secondary
structure. The CD spectra of all variants (data not shown) were similar to that of wild-type
KatG and showed typical features of an α-helical protein structure with the 222 and 208
nm (negative) dichroic bands. Therefore mutations did not induce changes in the overall
secondary structure and if conformational changes did occur, they must be very localized
and minimal.
Overall bifunctional activity – The kinetic parameters for the catalase and
peroxidase activities of wild-type KatG and the investigated variants are summarized in
Table 1. Additionally, data obtained upon exchange of the “flipping” arginine with alanine
(Arg439Ala) (14) are presented. Wild-type KatG exhibits an overwhelming catalase
activity with an apparent kcat of 3500 s-1 (11). Significant changes in the catalatic activities
were only observed for the variants Lys435Arg and Lys435Gln. The kcat values of these
mutants were 70 and 49%, respectively, of that of wild-type KatG. Similar turnover
numbers, compared to wild-type, were obtained for the variants Asn278Asp, Glu281Asp,
and Glu281Gln, whereas Trp459Phe and Arg509Lys showed slightly higher turnover
numbers. Regarding the apparent KM value in catalase activity, Lys435Gln is the only
104
variant analyzed within this work, that exhibits diminished affinity with H2O2 (6.1 mM)
compared to the wild-type protein (4.1 mM).
In WT KatG, the pH profile of the catalase activity has a sharp maximum at pH 6.5
(Figure 3, black line) (15). The same behaviour was shown for the other variants (data not
shown), except of the Lys435Gln variant. The pH dependence of the catalatic activity in
Lys453Gln shows a plateau between pH 5-6 and a maximum at pH 7.5 (Figure 3, gray
line).
1800
1400
1600
1200
µmol O2 min-1mg-1
1400
1000
1200
1000
800
800
600
600
400
400
200
200
0
0
3.5
4.5
5.5
6.5
7.5
8.5
9.5
pH
Figure 3. pH-dependence of the catalatic activity in wild-type KatG
(black line) and Lys435Gln (gray line). The catalatic activity was
determined polarographically. Conditions were as follows: 5 mM
hydrogen peroxide, 50 mM citrate/phosphate or 50 mM Tris-HCl
buffer at 30°C.
The effect of the performed mutations on the peroxidase activity (Table 1) was small. In
some variants the peroxidase activity was even slightly enhanced. Two substrates, namely
guaiacol and ABTS, were used as artificial one-electron donors. No significant differences
between wild-type and most variants using guaiacol in combination with hydrogen
peroxide as the peroxidase substrate were observed. Only Lys435Arg showed a decrease in
activity of about 80%, whereas the variants Arg509Lys and Arg439Ala (14) exhibited a
1.5 and 2-fold increase in peroxidase activity. With ABTS most of the variants exihibited
slightly reduced or wild-type like activities. Lys435Gln and Arg439Ala had an increased
capacity to oxidize ABTS.
In order to exclude the interference of the overwhelming catalatic activity with the
peroxidase activity, the guaiacol assays were also performed using peroxoacetic acid
instead of hydrogen peroxide. Under these conditions not only Lys435Arg but also
Glu281Asp, Lys435Gln, and Arg439Ala (14) exhibited decreased peroxidase activity
(Table 1). However, the effect was not very significant thereby excluding an important role
of the mutated residues in binding and oxidation of either ABTS or guaiacol. This is also
105
underlined by the fact that the pH maximum of peroxidase activity determined with ABTS
as one-electron donor was at pH 5.5 for wild-type KatG as well as for all mutants
investigated in the present study (data not shown).
Compound I formation with peroxoacetic acid – Because of the overwhelming
catalatic activity in catalase-peroxidases the formation of compound I mediated by
hydrogen peroxide cannot be followed spectroscopically. As a consequence, organic
peroxides (e.g. peroxoacetic acid) have to be used to monitor the formation of an
oxoiron(IV) porphyryl radical compound I (Por+FeIV=O), which (at least in most
monofunctional peroxidases) is typically characterized by a 40-50% hypochromicity in the
Soret region. Compound I formation from ferric Synechocystis wild-type KatG with
peroxoacetic acid is characterized by a 45% hypochromicity at 407 nm and the appearance
of two prominent peaks around 600 and 650 nm (11). Similar results were obtained for the
variants investigated in the present work. However, the observed hypochromicity at the
Soret region was less pronounced (27–36%) compared to the wild-type (data not shown).
The reaction of wild-type KatG with peroxoacetic acid followed at 407 nm was
monophasic and could be fitted using a single exponential equation. Observed pseudo firstorder rates (kobs) plotted as a function of peroxoacetic acid concentration resulted in an
apparent second-order rate constant (kapp) of (3.9 ± 0.3) × 104 M-1s-1 at pH 7.0 (11). No
significant differences were seen in the reactions of the variants with peroxoacetic acid.
The oberved reactions were monophasic and could be fitted using a single exponential
equation. The resulting kobs values plotted against peroxoacetic acid concentration yielded
apparent second-order rate constants ranging from 1.9 × 104 M-1 s-1 to 5.1 × 104 M-1 s-1 at
pH 7.0 (Table 1). Only in Arg439Ala the reaction was significantly increased by a factor
of 18 (2).
Reaction of ferric KatG with hydrogen peroxide – Because of the overwhelming
catalase activity only the ferric form of wild-type KatG can be monitored when mixed with
moderate levels of hydrogen peroxide. Addition of a large excess of hydrogen peroxide,
however, immediately forms a redox intermediate with a red-shifted Soret band (414 nm)
and two new peaks at higher wavelength (520 and 580 nm) at intermediate pH values (9).
This pH dependent species was formed immediately after mixing (1.3 ms) and was the
dominant redox intermediate as long as hydrogen peroxide was available.
106
107
0.7
0.8
2.8
1.0
5.1
n.d.
0.3
2.0
9.3
1.9
17.9
0.8
3.1
n.d.
0.5
2.5
10.5
1.6
21.3
0.9
3.7
n.d.
× 104 M-1 s-1
0.5
1.8
12.7
6.1
2.8
0.8
3.2
n.d.
0.1
1.8
7.8
2.9
10.1
0.4
2.7
n.d.
0.7
2.1
7.7
2.2
17.7
0.4
1.9
n.d.
0.9
2.7
8.0
2.2
17.2
1.1
71.1
230
1.3
1.8
12.8
2.1
0.8
1. Regelsberger, G., Jakopitsch, C., Ruker, F., Krois, D., Peschek, G. A., and Obinger, C. (2000) Effect of distal cavity mutations on the formation
of compound I in catalase-peroxidases, J. Biol. Chem. 275, 22854-22861.
2. Jakopitsch, C., Ivancich, A., Schmuckenschlager, F., Wanasinghe, A., Poltl, G., Furtmuller, P. G., Ruker, F., and Obinger, C. (2004) Influence of
the unusual covalent adduct on the kinetics and formation of radical intermediates in Synechocystis catalase peroxidase: a stopped-flow and EPR
characterization of the MET275, TYR249, and ARG439 variants, J. Biol. Chem. 279, 46082-46095.
*Kinetic data were extracted from spectral transitions using ProK software from Applied Photophysics.
Compound I reduction
Ascorbate*
3.9
Peroxoacetic acid
n.d.
0.7
2.2
0.6
2.8
n.d.
7.5
1.3
24.6
10.6
4.1
8.5
Hydrogen peroxide
Compound I formation
Peroxidase activity
ABTS (units mg-1)
Guaiacol (units mg-1)
Hydrogen peroxide
Peroxoacetic acid
KM (mM H2O2)
kcat/KM (× 105 M-1 s-1)
Table 1. Steady-state and pre-steady-state kinetic parameters of catalase and peroxidase activity of wild-type KatG from Synechocystis and selected
variants. For comparative purposes published data from Arg439Ala are also presented.
Wild-type
N278D
E281D
E281Q
K435Q
K435R
W459F
R509K
R439A
(1)
(2)
Catalase activity
kcat (s-1)
3500
3200
3400
3400
1700
2450
3900
3800
170
Like in wild-type KatG all the variants did not allow monitoring of compound I
formation or did not show any spectral transitions upon using moderate excess of hydrogen
peroxide. However, high excess of hydrogen peroxide led to similar spectral transitions as
observed in the wild-type enzyme at pH 8.5 (9). As a representative example the spectral
transitions of the ferric Arg509Lys mixed with millimolar H2O2 at pH 7.0 is shown in
Figure 4. The reaction started with a 1.3 ms spectrum with peaks at 414 nm and 520 nm
and ended with the Soret band at 407 nm and reappearance of the CT band at 637 nm after
depletion of hydrogen peroxide. The variants Glu281Asp/Gln, Lys435Arg, Trp459Phe,
and Arg509Lys showed similar spectral transitions upon mixing of 2 µM enzyme with
2500-fold excess of hydrogen peroxide (data not shown). Different from the other variants
but similar to wild-type KatG at pH 5.6 (9) were the spectral transitions observed in
Lys435Gln. Because of the low catalase activity only a 100-fold excess of hydrogen
peroxide was necessary to get a reaction intermediate with peaks at 410, 545, and 578 nm
at pH 7.0 (Figure 5A, black line). Upon using a higher excess of hydrogen peroxide (500fold) the Soret maximum was red-shifted to 414 nm (Figure 5A, bold line).
Interestingly and different from wild-type KatG, the spectrum observed
immediately after mixing (1.3 ms, Figure 5B, bold line) was followed by transition to a
spectrum that is usually observed by mixing PAA with ferric KatG (Figure 5B, black line).
The same transition was observed in Arg439Ala (14). Because of the low catalatic activity
of Arg439Ala it was possible to monitor compound I formation with hydrogen peroxide.
Based on these findings the same approach was used for the variant Lys435Gln, which
exhibits also diminished (~50%) catalatic activity. Nevertheless, it was not possible to
monitor compound I formation using hydrogen peroxide, most probably because the
catalatic activity of Arg439Ala was significantly lower (4.9% of the wild-type acitivity)
than that of Lys435Gln (14).
108
0.2
520
1.3 ms
Absorbance
414
0.15
470 ms
0.1
700 ms
0.05
0
300
400
500
600
700
Wavelength (nm)
Figure 4. Reaction of 2 µM KatG variant Arg509Lys with 5 mM
hydrogen peroxide. Spectra were recorded immediately after mixing
(1.3 ms, bold line) and after 470 ms, 700 ms and 10 s (gray line).
Conditions: 50 mM phosphate buffer, pH 7.0, and 25°C.
(A)
(B)
0.24
0.24
0.21
0.21
414
410
0.18
0.15
0.15
Absorbance
Absorbance
0.18
545
0.12
0.09
578
0.06
0.03
0.12
545
0.09
578
0.06
0.03
0
0
300
400
500
Wavelength (nm)
600
700
300
400
500
600
700
Wavelength (nm)
Figure 5. Reaction of KatG variant Lys435Gln with hydrogen peroxide. (A) Reaction intermediates
monitored 1.3 ms after mixing of 2 µM Lys435Gln (gray line) with 200 µM H2O2 (black line) or 2 mM H2O2
(bold line). Conditions: 50 mM phosphate buffer, pH 7.0, and 25°C. (B) Reaction of 2 µM Lys435Gln (gray
line) with 200 µM H2O2. Bold line depicts the spectrum immediately after mixing (1.3 ms) followed by
compound I (black line) formed after 150 ms. Conditions as in (A).
Compound I reduction with ascorbate – Addition of one-electron donors to
peroxidase compound I leads to the formation of compound II. A typical plant peroxidase
type compound II, as observed in the KatG variant Tyr249Phe (16), exhibits a Soret band
red-shifted to 418 nm and two additional peaks at 535 and 560 nm. This is in contrast to
wild-type KatG (10) and all KatG variants analyzed within this work, which show
different behavior. The time trace at 407 nm, extracted from the spectra, shows that
compound I reduction with ascorbate follows a biphasic curve with a fast initial phase
followed by a slower phase. The intermediate formed in wild-type KatG after full
depletion of ascorbate exhibits a Soret maximum at 407 nm and a distinct peak at 626 nm
(10). The spectral transitions in the variant Arg509Lys are shown in Figure 6, with the first
spectrum 1.3 ms after mixing, resembling compound I (prepared with PAA) in grey, and
109
the spectrum of the intermediate after the fast phase (500 ms) with the Soret band at 409
nm and a new peak at 626 nm in black. The last spectrum (Figure 6, bold line), recorded
after 10 s, exhibits the same absorption maxima like the 500 ms spectrum but more
pronounced.
0.14
409
0.12
Absorbance
0.1
0.08
0.06
626
0.04
0.02
0
310
390
470
550
630
Wavelength (nm)
Figure 6. Reduction of Arg509Lys compound I with ascorbate. The first spectrum
was recorded 1.3 ms after addition of ascorbate (gray line) still resembles that of
compound I (formed with 50 µM PAA , delay time of 3.5 s). The following spectra
were recorded 500 ms (black line) and 10 s (bold line) after mixing of compound I
with ascorbate. Conditions: 1 µM enzyme, 2 mM ascorbate, 50 mM phosphate buffer,
pH 7.0, and 25°C.
Discussion
Based on structural analyses (6) and sequence alignment a putative peroxidase
substrate binding site was investigated by mutational analysis together with kinetic
analysis. The U-shaped cleft on the KatG surface comprises some highly conserved
residues, from which five were mutated within this work. Exchange of the residues
Asn278, Glu281, Lys435, Trp459, and Arg509 did not alter the overall secondary structure
of the mutant proteins, which is mainly α-helical. Both Lys435 variants are, besides
Arg439Ala, the only recombinant oxidoreductases with decreased catalatic activity
compared to wild-type KatG. In Lys435Gln kcat was 49% and in Lys435Arg 70% of wildtype activity. Unfortunately, the peroxidase activity of the variants was even less affected
upon mutation. The only variant with reduced oxidation capacity of both substrates was
Lys435Arg, although the effect was relatively small. Exchange of Lys435 with glutamine
led (similar to Arg439Ala) to a slightly increased peroxidase activity using ABTS in
combination with hydrogen peroxide. This increase in the peroxidase activity using ABTS
in combination with hydrogen peroxide correlates roughly with the decrease in catalatic
activity since both pathways compete for hydrogen peroxide. Interestingly, this effect was
not observed upon using guaiacol in combination with hydrogen peroxide (Table 1).
110
The reduced catalatic activity in Lys435Gln fit with the findings that the spectral
features and kinetic transitions upon mixing with hydrogen peroxide are significantly
different to wild-type KatG (9). In detail, the first spectrum obtained immediately after
mixing with hydrogen peroxide showed a red-shifted Soret band and two additional peaks
at 545 and 678 nm followed by transition to a spectrum that is obtained usually by mixing
with PAA. The same transitions were observed in the Arg439Ala variant (14) but unlike in
this variant compound I formation mediated by hydrogen peroxide could not be monitored
in Lys435Gln. In contrast to hydrogen peroxide mediated spectral transitions there was no
difference in oxidation kinetics of all seven ferric proteins by PAA.
Compound I reduction with ascorbate in the variants followed the same kinetic and
spectral transitions as observed in the wild-type enzyme (10). The reaction starts with a
fast phase followed by a slower phase. The spectrum obtained after the initial fast phase
did not change until all ascorbate was consumed and exhibits features different from a
typical plant type compound II (observed also in KatG variant Trp249Phe), namely a lowspin oxoiron(IV) species. The Soret band in wild-type and mutant KatG compound II is
only slightly (1-2 nm) shifted compared to the ferric protein and showed a new band at 626
nm. The spectrum is indicative of a HS ferric form
To sum up, the region around the “flipping” arginine (Arg439) that forms an Ushaped cleft consists of highly conserved residues. This peculiar cleft on the protein
surface combined with the arginine 439 on the surface at the bottom of the “U” was probed
as putative one-electron donor binding site that bridges substrate binding and oxidation
with the heme cavity. However, neither the catalase nor the peroxidase activity was altered
significantly by mutation. Regarding the catalase activity this was as expected since H2O2
delivery, reduction and oxidation needs the intactness of the main access channel. And this
architecture was apparently not affected in the investigated proteins. However, the only
small effects on peroxidase activity, though using one-electron donors of different size and
chemistry (guaiacol and ABTS) also suggest that this region is not involved in binding and
oxidation of at least small aromatic substrates. Consequently, both the substrate binding
site and the nature of the endogenous one-electron donor of both prokaryotic and
eukaryotic KatG remain unknown.
111
References
1.
Vlasits, J., Jakopitsch, C., Schwanninger, M., Holubar, P., and Obinger, C. (2007)
Hydrogen peroxide oxidation by catalase-peroxidase follows a non-scrambling
mechanism, FEBS Lett 581, 320-324.
2.
Passardi, F., Bakalovic, N., Teixeira, F. K., Margis-Pinheiro, M., Penel, C., and
Dunand, C. (2007) Prokaryotic origins of the non-animal peroxidase superfamily
and organelle-mediated transmission to eukaryotes, Genomics 89, 567-579.
3.
Smulevich, G., Jakopitsch, C., Droghetti, E., and Obinger, C. (2006) Probing the
structure and bifunctionality of catalase-peroxidase (KatG), J. Inorg. Biochem. 100,
568-585.
4.
Welinder, K. G. (1992) Superfamily of plant, fungal and bacterial peroxidases,
Curr. Opin. Struct. Biol. 2, 388-393.
5.
Bertrand, T., Eady, N. A., Jones, J. N., Jesmin, Nagy, J. M., Jamart-Gregoire, B.,
Raven, E. L., and Brown, K. A. (2004) Crystal structure of Mycobacterium
tuberculosis catalase-peroxidase, J. Biol. Chem. 279, 38991-38999.
6.
Carpena, X., Loprasert, S., Mongkolsuk, S., Switala, J., Loewen, P. C., and Fita, I.
(2003) Catalase-peroxidase KatG of Burkholderia pseudomallei at 1.7A resolution,
J. Mol. Biol. 327, 475-489.
7.
Wada, K., Tada, T., Nakamura, Y., Kinoshita, T., Tamoi, M., Shigeoka, S., and
Nishimura, K. (2002) Crystallization and preliminary X-ray diffraction studies of
catalase-peroxidase from Synechococcus PCC 7942, Acta. Crystallogr. D Biol.
Crystallogr. 58, 157-159.
8.
Yamada, Y., Fujiwara, T., Sato, T., Igarashi, N., and Tanaka, N. (2002) The 2.0 A
crystal structure of catalase-peroxidase from Haloarcula marismortui, Nat. Struct.
Biol. 9, 691-695.
9.
Jakopitsch, C., Vlasits, J., Wiseman, B., Loewen, P. C., and Obinger, C. (2007)
Redox intermediates in the catalase cycle of catalase-peroxidases from
Synechocystis PCC 6803, Burkholderia pseudomallei, and Mycobacterium
tuberculosis, Biochemistry 46, 1183-1193.
10.
Jakopitsch, C., Ruker, F., Regelsberger, G., Dockal, M., Peschek, G. A., and
Obinger, C. (1999) Catalase-peroxidase from the cyanobacterium Synechocystis
PCC
6803:
cloning,
overexpression
in
Escherichia
coli,
and
kinetic
characterization, Biol. Chem. 380, 1087-1096.
112
11.
Regelsberger, G., Jakopitsch, C., Ruker, F., Krois, D., Peschek, G. A., and Obinger,
C. (2000) Effect of distal cavity mutations on the formation of compound I in
catalase-peroxidases, J. Biol. Chem. 275, 22854-22861.
12.
Regelsberger, G., Jakopitsch, C., Engleder, M., Ruker, F., Peschek, G. A., and
Obinger, C. (1999) Spectral and kinetic studies of the oxidation of monosubstituted
phenols and anilines by recombinant Synechocystis catalase-peroxidase compound
I, Biochemistry 38, 10480-10488.
13.
Gill, S. C., and von Hippel, P. H. (1989) Calculation of protein extinction
coefficients from amino acid sequence data, Anal. Biochem. 182, 319-326.
14.
Jakopitsch, C., Ivancich, A., Schmuckenschlager, F., Wanasinghe, A., Poltl, G.,
Furtmuller, P. G., Ruker, F., and Obinger, C. (2004) Influence of the unusual
covalent adduct on the kinetics and formation of radical intermediates in
Synechocystis catalase-peroxidase: a stopped-flow and EPR characterization of the
MET275, TYR249, and ARG439 variants, J. Biol. Chem. 279, 46082-46095.
15.
Jakopitsch, C., Droghetti, E., Schmuckenschlager, F., Furtmuller, P. G., Smulevich,
G., and Obinger, C. (2005) Role of the main access channel of catalase-peroxidase
in catalysis, J. Biol. Chem. 280, 42411-42422.
16.
Jakopitsch, C., Auer, M., Ivancich, A., Ruker, F., Furtmuller, P. G., and Obinger,
C. (2003) Total conversion of bifunctional catalase-peroxidase (KatG) to
monofunctional peroxidase by exchange of a conserved distal side tyrosine, J. Biol.
Chem. 278, 20185-20191.
113
appendix one
Intracellular catalase/peroxidase from the phytopathogenic fungus
Magnaporthe grisea: expression analysis and biochemical
characterisation of the recombinant protein
Marcel Zamocky, Paul G. Furtmüller, Marzia Bellei, Gianantonio Battistuzzi,
Johannes Stadlmann, Jutta Vlasits and Christian Obinger
Biochemical Journal 418 (2009), 443 – 541
www.biochemj.org
Biochem. J. (2009) 418, 443–451 (Printed in Great Britain)
443
doi:10.1042/BJ20081478
Intracellular catalase/peroxidase from the phytopathogenic rice blast
fungus Magnaporthe grisea : expression analysis and biochemical
characterization of the recombinant protein
*Metalloprotein Research Group, Division of Biochemistry, Department of Chemistry, University of Natural Resources and Applied Life Sciences, Muthgasse 18 A-1190 Vienna, Austria,
†Department of Chemistry, University of Modena and Reggio Emilia, Modena, Italy, ‡Glycobiology Research Group, Division of Biochemistry, Department of Chemistry, University of
Natural Resources and Applied Life Sciences, Muthgasse 18 A-1190 Vienna, Austria, and §Institute of Molecular Biology, Slovak Academy of Sciences, Bratislava, Slovakia
Phytopathogenic fungi such as the rice blast fungus Magnaporthe grisea are unique in having two catalase/peroxidase (KatG)
paralogues located either intracellularly (KatG1) or extracellularly (KatG2). The coding genes have recently been shown
to derive from a lateral gene transfer from a (proteo)bacterial
genome followed by gene duplication and diversification. Here we
demonstrate that KatG1 is expressed constitutively in M. grisea. It
is the first eukaryotic catalase/peroxidase to be expressed heterologously in Escherichia coli in high amounts, with high purity and
with almost 100 % haem occupancy. Recombinant MagKatG1
is an acidic, mainly homodimeric, oxidoreductase with a predominant five-co-ordinated high-spin haem b. At 25 ◦C and
pH 7.0, the E 0 (standard reduction potential) of the Fe(III)/Fe(II)
couple was found to be − 186 +
− 10 mV. It bound cyanide
monophasically with an apparent bimolecular rate constant of
5
−1
−1
◦
(9.0 +
− 0.4) × 10 M · s at pH 7.0 and at 25 C and with a
K d value of 1.5 μM. Its predominantly catalase activity was
characterized by a pH optimum at 6.0 and kcat and K m values
of 7010 s−1 and 4.8 mM respectively. In addition, it acts as a
versatile peroxidase with a pH optimum in the range 5.0–5.5
using both one-electron [2,2 -azinobis-(3-ethylbenzothiazoline6-sulfonic acid) o-dianisidine, pyrogallol or guaiacol] and
two-electron (Br− , I− or ethanol) donors. Structure–function
relationships are discussed with respect to data reported for
prokaryotic KatGs, as is the physiological role of MagKatG1.
Phylogenetic analysis suggests that (intracellular) MagKatG1 can
be regarded as a typical representative for catalase/peroxidase of
both phytopathogenic and saprotrophic fungi.
INTRODUCTION
reported [11–14], mainly due to lack of successfully produced
recombinant protein.
Most interestingly, the plant pathogenic fungus Magnaporthe
grisea (anamorph Pyricularia grisea) has two katG genes. This
ascomycete, known as rice blast fungus, is classified among the
most dangerous phytopathogens worldwide, possibly attacking
not only most rice (Oryza sativa) variants, but also numerous other
grass species, including economically important crops such as
wheat (Triticum aestivum), barley (Hordeum vulgare) and millet
(Panicum miliaceum) [15]. In addition to genes encoding KatGs,
its genome [16] revealed numerous virulence-associated genes
and those coding for enzymes involved in secondary metabolism.
From the two distinct KatG paralogues of M. grisea, one, namely
KatG1, seems to be located intracellularly, and the other, namely
KatG2, might be secreted from the fungal cells, since it contains
a (predicted) signal sequence. Localization could be related to
distinct functions, i.e. intracellular degradation of H2 O2 deriving
from fungal catabolism [17] and/or contribution to the enzymatic
armoury of phytopathogenic fungi against reactive oxygen species
produced during plant–pathogen interaction (oxidative burst)
[18,19]. Even a possible signalling function for KatGs has been
discussed [2]. As is obvious from an inspection of corresponding
sequences and phylogenetic analyses [6], both KatG1 and KatG2
Catalase/peroxidases (KatGs) are unique bifunctional oxidoreductases accomplishing efficiently both peroxidatic and
catalatic activity within a single active site [1,2]. Despite their
striking sequence similarities and close phylogenetic relationship
with ascorbate peroxidases and cytochrome c peroxidases [3],
they are the sole representatives within the whole superfamily of
non-animal haem (i.e. plant, fungal and bacterial) peroxidases [4]
capable of both the reduction and efficient oxidation of H2 O2 [5].
In this way they are similar to typical catalases, which represent
a quite different evolutionary line. A recent phylogenetic analysis
[6] has demonstrated that most of the 347 so-far-found katG
sequences are encoded within eubacterial and archaebacterial
genomes. About 40 % of all sequenced prokaryotes possess
katG gene(s) and, so far, comprehensive functional and structural
investigations have only been performed with eubacterial and
archaebacterial catalase/peroxidases [7–10]. However, KatGs can
be found also in Ascomycota (sac fungi) and Basidiomycota
(club fungi), as well as in a few protists [3,6]. In contrast
with prokaryotic KatGs, the corresponding eukaryotic proteins
have hardly been described, and only a few scattered data on
localization and gene expression and enzyme activities have been
Key words: catalase, KatG, Magnaporthe grisea, oxidative stress,
peroxidase, phytopathogenic fungus.
Abbreviations used: ABTS, 2,2 -azinobis-(3-ethylbenzothiazoline-6-sulfonic acid); E 0 , standard reduction potential; ECD, electronic CD; His6 ,
hexahistidine; HS, high-spin; 3D, three-dimensional; KatG, catalase/peroxidase; katG1, gene encoding MagKatG1; katG2 , gene encoding MagKatG2;
LGT, lateral gene transfer; LS, low-spin; MagKatG1, intracellular KatG from Magnaporthe grisea ; MagKatG2, secreted KatG from M. grisea ; MCAC,
metal chelate affinity chromatography; NcKatG2, KatG from Neurospora crassa ; ORF, open reading frame; OTTLE, optically transparent thin-layer
(spectro)electrochemical; RT–PCR, reverse-transcription PCR; SynKatG, KatG from Synechocystis PCC6803.
1
To whom correspondence should be sent (email [email protected]).
115
c The Authors Journal compilation c 2009 Biochemical Society
Biochemical Journal
Marcel ZAMOCKY*§1 , Paul G. FURTMÜLLER*, Marzia BELLEI†, Gianantonio BATTISTUZZI†, Johannes STADLMANN‡,
Jutta VLASITS* and Christian OBINGER*
444
M. Zamocky and others
seem to represent two distinct and abundant fungal KatG groups
with different structure–function patterns.
Here, we focus on the expression analysis, recombinant production and biochemical characterization of intracellular catalase/
peroxidase of M. grisea (MagKatG1). The complete ORF
(open reading frame) of katG1 (2217 bp) has been successfully
cloned and heterologously expressed, enabling a comprehensive
investigation of its biophysical and biochemical properties. The
spectral, redox and kinetic data presented are compared with
those reported for the corresponding prokaryotic enzymes and are
discussed with respect to structure–function relationships and
physiological role(s) of eukaryotic catalase/peroxidases.
MATERIALS AND METHODS
Organism
The ascomycete M. grisea, strain MA 829 (Cooke), was obtained
from the internal collection of BOKU (the University of Natural
Resources and Applied Life Sciences), Vienna, Austria. It was
first grown on MPG-agar plates (containing 20 g of malt extract,
1 g of peptone, 20 g of glucose and 20 g of agar in 1 litre) for
4–6 days at room temperature (25 ◦C). For the isolation of RNA
and activity monitoring, it was grown in liquid medium with the
same composition (without agar) at 25 ◦C in a shaking incubator
at 130 rev./min for 3 days.
Sequence-similarity searches: PeroxiBase
All available sequences of peroxidase genes and proteins were
systematically annotated, compared and analysed in PeroxiBase
(http://peroxibase.isb-sib.ch/; see [20] for details). All sequence
names used in the present paper correspond to the nomenclature of
PeroxiBase, where links to the corresponding genes and genomes
can also be found. The name and identification number of the
intracellular catalase/peroxidase from M. grisea are MagKatG1
and 2288 respectively; the corresponding UniProt accession
number is A4R5S9.
Multiple sequence alignment
Multiple sequence alignment of fungal and bacterial KatG proteins was performed with ClustalX, Version 2.0.5 [21]. The following optimized parameters were applied: Gonet protein weight
matrix with gap opening penalty, 8; gap extension penalty, 0.2; and
gap separation distance, 4. The aligned output was presented using
GeneDoc, a tool for editing and annotating multiple sequence
alignments developed by K.B. Nicholas and H.B. Nicholas in
1997 (http://www.genedoc.us/).
Promoter and subcellular targeting prediction
Prediction of intron and native promoter location within the genomic DNA of M. grisea strain 70-15 was performed in the
SoftBerry suite (http://www.softberry.com) using the ascomycetespecific parameters.
Cloning, heterologous expression and purification of recombinant
MagKatG1
The cloning strategy for the cDNA and the complete katG1
gene is described in Supplementary Figure S1 at http://www.
BiochemJ.org/bj/418/bj4180443add.htm. Briefly, the complete
gene was first maintained in TOPO vector (Invitrogen) and,
after modification, it was expressed within the pET21d vector
(Novagen), allowing a His6 (hexahistidine)-tag fusion. Hetero
c The Authors Journal compilation c 2009 Biochemical Society
logous expression was achieved from Escherichia coli BL21 DE3
Star cells (Invitrogen) at 16 ◦C and addition of haemin (10 μM
final concn.) concomitantly with isopropyl β-D-thiogalactoside
(1 mM final concn.) in the cultivation medium. Cells of 16-hold cultures were centrifuged for 15 min at 4500 g, and pellets
were resuspended in homogenization buffer containing 50 mM
sodium phosphate, pH 8.0, and protease inhibitors (1 mM PMSF,
1 μg/ml leupeptin and 1 μg/ml pepstatin). Homogenization was
performed by two ultrasonication cycles on a Vibra-Cell Model
CV17 (Sonics Materials Inc.) sonicator (length of each cycle,
45 s; pulser 50 %; intensive cooling between the cycles). The
crude homogenate was centrifuged for 20 min at 20 000 g and
the cleared supernatant was applied on to a 30 ml MCAC (metal
chelate affinity chromatography) Chelating Sepharose Fast Flow
(GE Healthcare) column loaded with Ni2+ ions. After loading,
the column was washed with 150 ml of buffer A (50 mM sodium phosphate, pH 8.0, containing 500 mM NaCl and protease
inhibitors). Bound His-tagged protein was eluted with a linear
gradient of buffer A to 100 % buffer B (50 mM sodium phosphate,
pH 6.5, containing 500 mM NaCl and 500 mM imidazole). For
final purification and determination of putative oligomeric structure of MagKatG1, gel-permeation chromatography (Superdex
200 prep grade; GE Healthcare) was performed. The column
volume was 250 ml and the sample volume was 1 ml. The buffer
was 5 mM sodium phosphate, pH 7.0, containing 0.15 M NaCl,
and was pumped at a flow rate of 0.5 ml/min and a maximal
pressure: 0.3 MPa.
The purity of fractions obtained was analysed by SDS/PAGE
under standard conditions using a Bio-Rad Mini-Protean system
employing a 12 % (w/v) polyacrylamide separating gel. The gels
were either stained with Coomassie Brilliant Blue or blotted
on to nitrocellulose membrane (Amersham Biosciences) for the
detection of MagKatG1 by immunoblotting using a polyclonal
antibody raised against catalase/peroxidase from Neurospora
crassa [23].
To reveal the expression pattern of MagKatG1, native PAGE
with a linear gradient of 4–18 % acrylamide was used. Native
electrophoresis was performed with 10 mM Tris/HCl buffer,
pH 8.3, containing 14.4 g/l glycine in the Mini-Protean apparatus
at 150 V for 30 min, followed by 65 V for 3 h. Catalase staining
was performed as described in [24], with catalatically active bands
occurring as white bands on a dark-green background. Peroxidase
activity staining was performed by using o-dianisidine as the
electron donor.
MS analysis
In order to analyse proteolytic degradation of MagKatG1,
bands in the SDS/PAGE gels were excised, destained, carbamidomethylated and digested with sequencing-grade bovine trypsin
(Sigma–Aldrich). The extraction from gel pieces was performed
in the same mode as described in [25]. Extracts were dried in a
SpeedVac® concentrator and reconstituted with water containing
0.1 % formic acid. Subsequent Matrix-assisted laser desorption
ionization–time-of-flight MS analysis was performed on a
Voyager-DE STR (Applied Biosystems) instrument with α-cyano4-hydroxycinnamic acid as matrix.
ECD (electronic CD) spectrometry and structure prediction
ECD spectra were recorded on a PiStar-180 Spectrometer
equipped with a thermostatically controlled cell holder (Applied
Photophysics). For recording far-UV spectra (260–190 nm), the
quartz cuvette had a path length of 1 mm. The spectral bandwidth
was 5 nm, the step size was 1 nm, the scan time was 624 s and the
116
Magnaporthe grisea catalase/peroxidase
protein (MagKatG1) concentration was 4 μM. Protein concentration was calculated from the known amino acid composition
and A280 , as described in [26]. All ECD measurements were
performed in 5 mM phosphate buffer, pH 7.0, at 25 ◦C. Each
spectrum was automatically corrected with the baseline to remove
birefringence of the cell. The instrument was flushed with nitrogen
with a flow rate of 5 litres/min.
In addition, the secondary structure of MagKatG1 was predicted
with PSIPRED [27]. The 3D (three-dimensional) structure of
MagKatG1 was modelled using EsyPred 3D [28] and KatG from
Burkholderia pseudomallei, the bacterium causing melioidosis
(1MWV). The reliability of the obtained model was verified using
What IF (http://swift.cmbi.ru.nl/servers/htmL/index.htmL).
Electronic UV–visible spectroscopy
The UV–visible spectrum of purified recombinant ferric
MagKatG1 in 5 mM phosphate buffer, pH 6.5, was routinely
recorded on a Zeiss Specord 10 diode-array photometer over the
range 200–800 nm at 25 ◦C. We determined the absorption coefficient at Soret band (ε408 110 650 M−1 · cm−1 ) for highly purified
MagKatG1 from the slope of the linear correlation between A408
and seven different protein concentrations (results not shown).
Ferrous MagKatG1 was produced by the addition of sodium
dithionite, either directly or from a freshly prepared anaerobic
stock solution in a glove box (Meca-Plex; Neugebauer) with a
positive pressure of nitrogen (2.5 kPa; 25 mbar). All solutions
were made anaerobic by flushing with nitrogen gas (oxygen
< 3 p.p.m.) and stored in the glove box. The spectrum of the lowspin cyanide complex of MagKatG1 was recorded after mixing
7 μM protein with 800 μM cyanide in 5 mM phosphate buffer,
pH 6.5.
445
instrument. For a total of 100 μl/shot into the optical observation
cell with a 1 cm light path, the fastest time for mixing two
solutions and recording the first data point was of the order of
1.3 ms. Formation of low-spin cyanide complex was monitored
at 426 nm, the Soret maximum difference between high-spin
MagKatG1 and the cyanide complex. In a typical experiment,
one syringe contained 2 μM MagKatG1 in 5 mM phosphate
buffer, pH 7.0, and the second syringe contained at least a 2.5fold excess of cyanide in 100 mM phosphate buffer, pH 7.0.
At high molar excess of cyanide over catalase/peroxidase, the
formation of cyanide complex occured too quickly, and the majority of absorbance change occurred within the dead time
of the instrument. As a consequence, pseudo-first-order
conditions could not be maintained over the whole concentration
range. Three determinations were performed for each ligand
concentration. The mean of the pseudo-first-order rate constants,
kobs , was used in the calculation of the second-order rate
constant obtained from the slope of a plot of kobs versus cyanide
concentration. All measurements were performed at 25 ◦C.
Equilibrium binding of cyanide to MagKatG1 was studied by
spectroscopic titration of 1 μM enzyme in 100 mM phosphate
buffer, pH 7.0, with increasing concentrations of cyanide (1–
800 μM). The enzyme spectrum was recorded before and after
the addition of constant volumes of cyanide solution, and the free
enzyme spectrum was substracted from the complex spectrum
after correction for volume dilution. Plotting the ratio of cyanide
concentration to absorbance difference between HS (high-spin)
and LS (low-spin) MagKatG1 (peak maximum minus valley)
against the cyanide concentration allowed determination of K d .
Overall kinetic parameters
Spectroelectrochemistry
0
Determination of the E of the Fe(III)/Fe(II) couple of MagKatG1
was carried out using a homemade OTTLE [optically transparent
thin-layer (spectro)electrochemical] cell [29]. The three-electrode configuration consisted of a gold mini-grid working
electrode (Buckbee-Mears), a homemade Ag/AgCl/KClsatd microreference electrode, separated from the working solution by a
Vycor set, and a platinum wire as counter-electrode. The reference
electrode was calibrated against a saturated calomel (HgCl) electrode before each set of measurements. All potentials are
referenced to the standard hydrogen electrode. Potentials were
applied across the OTTLE cell with an Amel model 2053 potentiostat/galvanostat. The constant temperature was maintained
by a circulating water bath, and the OTTLE cell temperature
was monitored with a Cu-costan microthermocouple. UV–visible
spectra were recorded using a Varian Cary C50 spectrophotometer. Spectra of 28 μM MagKatG1 equilibrated in 50 mM
sodium phosphate buffer, pH 7.0, containing 10 mM NaCl,
were recorded at various applied potentials. Paraquat (850 μM),
lumiflavine-3-acetate (3 μM), indigo disulfonate, phenazine
methosulfate and Methylene Blue were used as electron
mediators. Results are presented as Nernst plot, that allowed
calculation of the standard reduction potential, E0 , of the
Fe(III)/Fe(II) couple MagKatG1.
Kinetics and thermodynamics of formation of the
MagKatG1–cyanide complex
In order to probe haem accessibility, apparent bimolecular rate
constants of complex formation between cyanide and ferric
MagKatG1 were determined by convential stopped-flow spectroscopy using a model SX-18MV Applied Photophysics, U.K.
Catalase activity was determined both polarographically using
a Clark-type electrode (YSI 5331 oxygen probe) inserted into a
stirred water bath (YSI 5301B) or spectrophotometrically by
measuring the decomposition of H2 O2 at 240 nm as described in
[30]. For spectrophotometric measurements, one unit was defined
as the amount of enzyme catalysing the conversion of 1 μmol of
H2 O2 /min at an initial concentration of 15 mM H2 O2 at the pH
optimum and at 25 ◦C.
Peroxidase activity was monitored spectrophotometrically
using 1 mM H2 O2 and 5 mM ABTS (ε414 31.1 mM−1 · cm−1 ) [31]
or 5 mM guaiacol (ε470 26.6 mM−1 · cm−1 ) or 1 mM o-dianisidine
(ε460 11.3 mM−1 · cm−1 ) or 1 mM pyrogallol (ε430 2.5 mM−1 ·
cm−1 ). One unit of peroxidase was defined as the amount of
enzyme that decomposes 1 μmol of electron donor/min at pH
optimum and 25 ◦C.
Additionally, peroxidase activity with ethanol as a twoelectron donor was determined at 340 nm in a coupled reaction
using aldehyde dehydrogenase as described previously [32].
Chlorination and bromination activity of recombinant MagKatG1
was followed by halogenation of monochlorodimedone (100 μM)
dissolved in 100 mM phosphate buffer, pH 7.0, containing 10 mM
H2 O2 and either 10 mM bromide or 100 mM chloride. Rates
of halogenation were determined from the initial linear part of
the time traces using an absorption coefficient for monochlorodimedone at 290 nm of 19.9 mM−1 · cm−1 [33]. One unit was
defined as the amount that decomposes 1 μmol of monochlorodimedone/min at pH 7.0 and 25 ◦C. Iodide oxidation was
monitored at 353 nm (i.e. the absorbance maximum of tri-iodide
(I3 − ; ε353 25.5 mM−1 · cm−1 ) in 100 mM phosphate buffer, pH 7.0,
containing 10 mM H2 O2 and 10 mM iodide at pH 7.0 and 25 ◦C
[34]. One unit was defined as the amount that produces 1 μmol
of I3 − min−1 at pH 7.0 and 25 ◦C.
117
c The Authors Journal compilation c 2009 Biochemical Society
446
M. Zamocky and others
For all steady-state measurements at least three determinations
for each electron-donor concentration were performed, and
reactions were started by the addition of MagKatG1.
RESULTS AND DISCUSSION
Statistics from PeroxiBase clearly suggest that, in eukaryotic
organisms, KatGs are predominantly spread among the Fungal
Kingdom. A recent phylogenetic analysis [2] demonstrated the
existence of two distinct types of fungal katG genes. Mainly
phytopathogenic fungi (e.g. M. grisea or Gibberella zeae, the
latter causing head blight of wheat) contain both paralogues, with
katG1 most probably encoding an intracellular, and katG2 an
extracellular, catalase/peroxidase.
Analysis and comparison of the whole genome of M. grisea
[16] with katG1 and katG2 showed a significant difference in
the G + C contents (for katG1, 59.2 versus 51.6 %), suggesting
an LGT (lateral gene transfer) of katG genes from bacteria to
a fungal progenitor, as has been demonstrated for other fungi
[3,6]. Interestingly, the rather high G + C contents of MagkatG1
resembled that of various Burkholderia species (58–68 %),
thereby indicating a putative proteobacterial ancestor. In order to
answer the question of whether only the katG1 gene or a longer
genome portion had been transferred by LGT, we also analysed
the putative promoter region of MagkatG1 with GENSCANW
(http://genes.mit.edu/GENSCAN.htmL). A conserved 40-bplong promoter motif was identified at positions − 1838 to − 1799
(see PeroxiBase ID=2288 for further details). Moreover, a typical
CGGAGT box was found in positions − 657 to − 652, similar to
a promoter region in the catB gene of Aspergillus oryzae [35].
The G + C contents of this 1838-bp-long region is only 49.73 %,
thus even lower than the average value for the entire Magnaporthe
genome. This suggests that only the coding region was transferred
via LGT from a proteobacterial into the fungal genome.
In order to probe whether expression of MagKatG1 is induced
by oxidative stress, cultures of Magnaporthe grisea were grown
in MPG in the absence and presence of H2 O2 or peroxyacetic acid
or paraquat (final concentrations 0.5–2 mM) added in the middleexponential or early-stationary growth phase respectively. RT–
PCR analysis of total RNA isolated from filtered mycelia allowed
quantification of katG1-specific mRNA under various conditions. The RT-PCR signal intensities obtained from various
cDNA samples (Figure 1) clearly demonstrate that katG1 is
expressed constitutively. Only with paraquat was the expression
level enhanced to 120 % compared with that under standard
conditions. This might suggest a role of intracellular MagKatG1
in the continuous degradation of H2 O2 deriving from fungal metabolism and is in contrast with MagKatG2, which, owing to
the presence of a (predicted) signal sequence, is located extracellularly and is produced on demand; that is, its expression is
significantly enhanced under oxidative stress conditions that could
occur during plant attack [17].
Subsequently MagKatG1 was heterologously expressed and
purified. The entire ORF (2217 bp) was cloned in TOPO vector
using outer primers and modified by introducing NcoI, AgeI
and NotI restriction-endonuclease sites without affecting the
coding sequence (see Supplementary Figure S1 at http://www.
BiochemJ.org/bj/418/bj4180443add.htm). This allowed cloning
in pET21d vector (Novagen) in-frame with a C-terminal His6 tag fusion. Heterologous expression of the recombinant plasmid
pMZM1 was performed in E. coli strain BL21 DE3 Star
(Invitrogen) at 16 ◦C and, in the presence of haemin, avoided
formation of inclusion bodies and produced recombinant protein
with almost 100 % haem occupancy. On average about 30 mg
c The Authors Journal compilation c 2009 Biochemical Society
Figure 1 DNA electrophoresis (0.85 % agarose) of RT–PCR products
isolated from M. grisea
Lane 1, M. grisea grown under standard conditions; lanes 2 and 3, grown under oxidative
stress induced by addition of 0.1 mM paraquat (lane 2) or 0.3 mM peroxyacetic acid (lane 3);
lane 4, amplification of the whole gene after induction of oxidative stress with paraquat; lane 5,
molecular size standards with the size given in bp. In lanes 1–3 the same amount of RT–PCR
product was loaded.
of soluble KatG1 could be obtained by lysis from cell pellets of
1 litre of liquid culture. Protein purification included MCAC
followed by gel-permeation chromatography as described in the
Materials and methods section.
Analysis of purity by SDS/PAGE revealed the existence of one
major band at 85 kDa corresponding to monomeric MagKatG1
(theoretical molar mass of monomer with one haem b and a
His6 -tag is 83.17 kDa), but additional minor bands were always
present at 160, 51, 31 and 27 kDa respectively, irrespective
of variations in the purification protocol. This prompted us to
probe the protein pattern by immunoblotting using a polyclonal
antibody raised against NcKatG1 (PeroxiBase ID=2181; also
known as NcCat2; see Supplementary Figure S5 at http://www.
BiochemJ.org/bj/418/bj4180443add.htm for a sequence comparison). Figure 2(A) shows that all the bands at 51, 31, and
27 kDa gave positive signals after staining suggesting proteolytic
degradation of MagKatG1. This was proven by tryptic digestion
and peptide mass-mapping (see Supplementary Figure S2 at http://
www.BiochemJ.org/bj/418/bj4180443add.htm). The molecular
masses of the main peptides produced were fully in accordance
with those of peptides obtained by in silico trypsin-mediated
‘digestion’ of MagKatG1. The band at 160 kDa suggested the
presence of a covalently linked dimeric form of MagKatG1.
Native electrophoresis of recombinant MagKatG1 revealed
positive catalase and peroxidase stains at the position of the
dimeric protein, as well as a minor catalase activity at the position of tetrameric MagKatG1 (Figure 2B). In order to analyse the
oligomeric structure further, gel-permeation chromatography
was performed (Superdex 200 prep grade calibrated with
a Gel Filtration Calibration Kit; GE Healthcare). It clearly
demonstrated that about 95 % of the protein existed in
homodimeric form (180 +
− 4 kDa), whereas about 5 % existed
as homotetramer (352 +
− 8 kDa) (see Supplementary Figure S3
at http://www.BiochemJ.org/bj/418/bj4180443add.htm), thereby
reflecting the data obtained by native electrophoresis. The two
so far partially analysed fungal peroxidases isolated directly
from Penicillium simplicissimum [11] and N. crassa [12], which
both contain only one intracellular KatG, were reported to
118
Magnaporthe grisea catalase/peroxidase
Figure 2
447
(A) SDS/PAGE of recombinant MagKatG1 purified from E. coli BL21 DE3 Star homogenate and (B) active staining of catalase and peroxidase activity
(A) Lane 1, Coomassie Brilliant Blue-stained crude extract; lane 2, pool from MCAC affinity column, fractions 4–6; lane 3, pool from MCAC affinity column, fractions 7–10; lane 4, protein from main
elution peak of Superdex 200 chromatography (i.e. sample from lane 3 purified over a second column); lanes 5 and 6, Western-blot analysis using a polyclonal antibody raised against NcKatG1
[lane 5, MCAC fractions 4–6 (corresponding to Coomassie Brilliant Blue, lane 2); lane 6, MCAC fractions 7–10 (corresponding to Coomassie Brilliant Blue, lane 3)]. Bands at 51, 31 and 27 kDa
derive from proteolytic degradation of the 83 kDa protein. (B) Active staining of catalase (lanes 1 and 2) and peroxidase activity (lanes 3 and 4). Lane 1, pool from MCAC fractions 7–10; lane 2, pool
from main peak of Superdex 200 chromatography (sample from lane 1 further purified); lane 3, pool from MCAC fractions 7–10; lane 4, pool from main peak of Superdex 200 (sample from lane 3
further purified). Bovine liver catalase (Sigma) and Comamonas testosteroni KatG (PeroxiBase ID 5351) were used as native standards. The molecular masses of standard proteins are given in kDa.
be homodimeric non-covalently linked proteins. Concomitant
occurrence of homodimeric, with some homotetrameric, KatGs
has been reported from prokaryotic proteins [7].
The pI of recombinant MagKatG1 was determined to be
5.7 +
− 0.2, which is in good agreement with the calculated value
of 6.0 using the ExPaSy server (http://www.expasy.org/tools/
pi_tool.htmL).
In order to gain more insight into the structural features of
recombinant MagKatG1, both ECD spectroscopy as well as
structure modelling was performed. Figure 3(A) depicts the farUV ECD spectrum of recombinant MagKatG1 in comparison
with catalase/peroxidase from Synechocystis (SynKatG) at an
identical concentration, i.e. A280 . Both the prokaryotic and the
eukaryotic enzyme showed the typical features of a predominantly α-helical protein (dichroic bands at 222 and 208 nm).
However, the α-helical content of the eukaryotic enzyme is
more pronounced, which is also underlined by sequence analysis
and secondary-structure prediction using PSIPRED [27] (see
Supplementary Figure S4 at http://www.BiochemJ.org/bj/418/
bj4180443add.htm).
Figure 3(B) presents a 3D-model of monomeric MagKatG1 in
overlay with the known structure of KatG from B. pseudomallei
(1MWV). Probing the homology model with What IF yielded a
root-mean-square Z-score for bond lengths and angles of 0.911
and 1.268, respectively, and a root-mean-square deviation in
bond distances of 0.0018 nm (0.018 Å). All predicted bonds
were in agreement with standard bond lengths. The active site
(Figure 3B) comprises the typical conserved proximal (His279 Asp389 -Trp330 ) and distal (Arg108 -His112 -Trp111 ) triads found in all
so-far-investigated KatGs [1].
Recent mutational and kinetic analysis of bacterial KatGs
revealed peculiar structural features essential for the catalatic
activity. These include a unique covalent adduct between distal
Trp111 and KatG-typical Tyr238 and Met264 (Figure 3B) [1].
Additionally, it has been demonstrated that a distal aspartate
residue (Asp120 in MagKatG1) at the entrance of the channel
into the haem cavity contributes to maintenance of a defined
hydrogen-bonding network necessary for H2 O2 oxidation [1].
All these amino acids are strictly conserved in all prokaryotic
and eukaroytic KatGs (see Supplementary Figure S5). One
obvious difference between prokaryotic and eukaryotic catalase/
peroxidases concerns the KatG-specific loop 1 (LL1 in Supplementary Figure S5B) that is known to contribute to the architecture
of the main channel and that controls access to the haem cavity [8].
Owing to the insertion of 23–27 amino acids in fungal KatG1s,
LL1 is significantly longer.
More details about the nature of the haem and redox properties
were obtained by UV–visible spectroscopy and spectroelectrochemical studies. Figure 4 depicts the electronic UV–visible
spectra of HS ferric (Figure 4A), HS ferrous (Figure 4B) and LS
ferric (Figure 4C) recombinant MagKatG1 in comparison with
those of SynKatG. Ferric MagKatG1 exhibits the typical bands of
a haem b-containing peroxidase in the visible and near-UV region
with a Soret band at 408 nm, Q-bands at 504 nm and 546 nm
and a CT1 band (porphyrin-to-metal charge-transfer band) at
635 nm (Figure 4A). This compares with the corresponding
SynKatG absorption maxima at 406 nm (Soret), 502 and 542 nm
(Q-bands) and 637 nm (CT1) which have been demonstrated
to be representative of a five-co-ordinate HS haem co-existing
with a small portion of six-co-ordinate HS haem and a
119
c The Authors Journal compilation c 2009 Biochemical Society
448
M. Zamocky and others
Figure 4 Spectral characterization of recombinant MagKatG1 (black line)
in comparison with prokaryotic SynKatG (grey line)
Figure 3 (A) Electronic UV–visible and far-UV ECD spectra of purified
recombinant MagKatG1 and (B) a homology model of the 3D structure of
MagKatG1 monomer
(A) Electronic UV–visible and far-UV ECD spectrum (inset) of purified recombinant MagKatG1
(continuous line, 4.2 μM) in 5 mM phosphate buffer, pH 7.0. For comparison the corresponding
spectra of SynKatG (broken line) recorded under identical conditions are depicted. (B) Homology
model of the 3D structure of MagKatG1 monomer revealed by EsyPred [28]. An overlay with the
known structure of KatG from B. pseudomallei (PDB code 1MWV) is presented. The Figure was
constructed using the Swiss PDB Viewer.
six-coordinate LS haem, as has been demonstrated by resonance
Raman spectroscopy of the prokaryotic protein [1]. The purity
number (Reinheitszahl in German), i.e. A408 /A280nm , of purified recombinant MagKatG1 varied over the range 0.63–0.65
(Figure 3A), which is comparable with values for heterologously
expressed prokaryotic KatGs and indicates 100 % haem
occupancy [36]. From the two eukaryotic KatGs isolated from
P. simplicissimum [11] and N. crassa [12], a Soret absorption
at 407 nm and 405 nm respectively was reported. This clearly
suggests a very similar architecture of the active site of prokaryotic
and intracellular eukaryotic KatGs.
On reduction of ferric MagKatG1 with dithionite in a glove
box and recording the spectrum in a gas-tight cuvette, the Soret
band shifted to ≈ 432 nm. A good spectrum of ferrous SynKatG
under identical conditions had its peak maxima at 438 nm and
at 562 nm, with a shoulder at about 590 nm. Both the position
and shape of the Soret band indicate the presence of a mixture of
c The Authors Journal compilation c 2009 Biochemical Society
Arrows indicate peak maxima. (A) Ferric proteins in 5 mM sodium phosphate buffer, pH 7.0, at
25 ◦C. (B) Ferrous proteins in 5 mM sodium phosphate buffer, pH 7.0, at 25 ◦C. (C) Low-spin
cyanide complexes obtained by mixing of ferric proteins with 800 μM NaCN in 5 mM sodium
phosphate buffer, pH 7.0, at 25 ◦C.
dominating ferrous MagKatG1, with some dioxygen adduct (i.e.
compound III), which was underlined by analysis of the second
derivative spectrum (results not shown). Despite several trials,
we did not succeed in obtaining pure ferrous MagKatG1, which
might suggest a higher affinity for dioxygen of the eukaryotic
ferrous oxidoreductase compared with the prokaryotic enzyme.
Figure 4(C) depicts the LS cyanide complex of ferric
MagKatG1 compared with the cyanide complex of SynKatG.
Cyanide converts the HS (S = 5/2) iron state to the LS (S = 1/2)
state, thereby shifting the Soret peak of ferric MagKatG1 from 408
to 419 nm with a clear isosbestic point at 416 nm. An additional
peak of the cyanide complex of MagKatG1 was at 536 nm, with a
shoulder at 572 nm. This compares with the corresponding peaks
of LS SynKatG at 421 and 533 nm.
Determination of the E0 of the Fe(III)/Fe(II) couple of recombinant MagKatG1 was performed spectroelectrochemically.
Figure 5 shows a representative family of spectra of MagKatG1
at various applied potentials. The midpoint reduction potential
(Em ) was determined from the corresponding Nernst plot (inset
to Figure 5). At 25 ◦C and pH 7.0, E0 was found to be
(− 186 +
− 10) mV, which is 40 mV more positive than that of
SynKatG (− 226 mV) [37], but within the range reported for other
120
Magnaporthe grisea catalase/peroxidase
449
Figure 5 Electronic spectra of MagKatG obtained at various applied
potentials in spectroelectrochemical experiments
Spectra were recorded at 25 ◦C. The corresponding Nernst plot is reported in the inset, where
X stands for:
A max
λreduced − A λreduced
A max
λoxidized − A λoxidized
where λoxidized is 406 nm and λreduced is 435 nm. The conditions were as follows: sample volume,
1 ml; [protein], 28 μM in 50 mM phosphate buffer, pH 7.0, containing 10 mM NaCl; [paraquat],
850 μM; [lumiflavine-3-acetate], [indigo disulfonate], [phenazine methosulfate] and [Methylene
Blue], 3 μM.
class I peroxidases such as cytochrome c peroxidase (− 194 mV)
[38] and ascorbate peroxidase (− 160 mV) [39]. This suggests
conservation of the structure of the haem b moiety as well as
of the electrostatic interactions of the iron ion with the proximal
histidine residue (His279 ) and the immediate surrounding polypeptide matrix. This especially concerns the interaction between
the proximal histidine residue and conserved aspartate residue
(Asp389 ) that serves to deprotonate the N δ position of the
proximal histidine residue, providing a strong axial ligand with an
imidazolate character (Figure 3B). In class I peroxidases, despite
their different enzymatic properties, this interaction seems to be
carefully optimized in strength and geometry.
Finally, we investigated in addition the kinetics of cyanide
binding of one- and two-electron oxidation reactions catalysed
by recombinant MagKatG1. In haem protein chemistry, cyanide
is a useful ligand to probe accessibility and binding to the haem
iron. Figure 6(A) depicts a series of spectra obtained upon mixing
ferric MagKatG1 (2 μM) with 30 μM KCN. Under conditions of
excess of cyanide, pseudo-first-order rate constants, kobs , could
be obtained from single-exponential fits of the obtained time
traces (Figure 6B). The apparent second-order rate constant (kon ),
calculated from the slope of the linear plot of kobs versus cyanide
concentration:
kobs = kon · [HCN] + koff (Figure 6C)
5
−1
−1
◦
was (9.0 +
− 0.4) × 10 M · s at pH 7.0 and 25 C. This compares
with 4.8 × 105 M−1 · s−1 and 6.9 × 105 M−1 · s−1 determined with
KatGs from Synechocystis PCC 6803 KatG [36] and Anacystis
nidulans [40]. In both the prokaryotic [1] and eukaryotic
(results not shown) KatGs, the pH-dependence of kon clearly
demonstrates that HCN is the liganding species. From the
−1
intercept (1.4 +
− 0.4 s ) of the linear plot (Figure 6C), the dissociation rate constant (koff ) was obtained, allowing the calculation
of the dissociation constant (K d ) from the koff /kon ratio, i.e.
1.5 +
− 0.3 μM obtained
− 0.4 μM. This compares with a K d of 1.4 +
by thermodynamic binding studies (Figure 6D). This very small
K d value might indicate a more opened substrate channel compared with that of bacterial KatGs. This is also supported by the
Figure 6 Kinetics and thermodynamics of cyanide binding to recombinant
MagKatG1
(A) Spectral transition showing the formation of the low-spin cyanide complex. The reaction of
2 μM ferric MagKatG1 with 30 μM cyanide was measured in the conventional stopped-flow
mode. Arrows indicate changes in absorbance with time. Conditions were 50 mM phosphate
buffer, pH 7.0, and 25 ◦C. (B) Typical time trace at 423 nm with single-exponential fit.
(C) Dependence of k obs values on the cyanide concentration. The association rate constant
k on was calculated from the slope and the dissociation rate constant k off from the intercept.
(D) Hanes plot deduced from titration of MagKatG1 with increasing concentration of cyanide.
Conditions were 7 μM MagKatG1 in 50 mM phosphate buffer, pH 7.0, and 25 ◦C.
high affinity of O2 for ferrous MagKatG1 that hampered
the formation of pure ferrous protein, even when samples were
prepared in the glove box.
Prokaryotic catalase/peroxidases have been demonstrated to
be bifunctional enzymes with both a catalatic and peroxidatic
activity. The pH profile of both activities of MagKatG1 (Figure 7)
showed a sharp maximum, with the catalase activity being
highest at pH 6.0 and the peroxidase activity at pH 5.5 (substrate
guaiacol) and pH 5.0 (substrate ABTS). Recombinant MagKatG1
exhibited a specific catalase activity of 3430 +
− 280 units/mg
of protein at pH 6.0 and a specific peroxidatic activity for
guaiacol of 2.7 +
− 0.4 units/mg at pH 5.5 respectively (Table 1 and
Supplementary Figure S6 at http://www.BiochemJ.org/bj/418/
bj4180443add.htm). Guaiacol peroxidation was still present at
pH 3.0 (0.63 unit/mg of protein). The specific activity of ABTS
oxidation at pH 5.0 was determined to be 32.3 +
− 2.1 units/mg of
protein. In addition, the peroxidase activity towards o-dianisidine
and pyrogallol was determined (Table 1 and Supplementary
Figure S6) with the hierarchy of electron donors being:
ABTS > o-dianisidine > pyrogallol > guaiacol
Whether molecules similar to these compounds play a role in
metabolism of this rice-blast fungus and might affect expression
of MagkatG1 remains a topic for future investigation.
The apparent K m and kcat for H2 O2 dismutation (i.e. catalatic activity) were determined at the pH optimum, both
polarimetrically and spectrophotometrically at H2 O2 concentrations ranging from 0.5 to 10 mM. The kinetic parameters were
−1
determined to be 4.8 +
− 230 s (kcat ), as well
− 0.4 mM (K m ), 7010 +
121
c The Authors Journal compilation c 2009 Biochemical Society
450
M. Zamocky and others
Table 1
Selected kinetic parameters of recombinant MagKatG1
Average values for typical clones are presented. Kinetic traces are presented in Supplementary
Figure S6.
Sample (clone)
Activity parameter
Value
M133
K m (H2 O2 )
k cat (H2 O2 )
k cat / K m (H2 O2 )
4.8 +
− 0.4 mM−1
7010 +
− 230 s
6
−1
−1
(1.46 +
− 0.06) × 10 M · s
Catalase activity
Peroxidase activity with, as substrate:
o -Dianisidine
ABTS
Guaiacol
Pyrogallol
Chlorination of monochlorodimedone*
Bromination of monochlorodimedone*
Iodination of monochlorodimedone*
3430 +
− 280 units/mg
M133 + M134 +
M166 + M167
Figure 7 pH–activity profiles for catalase and peroxidase activities of
recombinant MagKatG1 (clone M133)
The concentration of MagKatG1 was 0.2 μM and all measurements were performed at 25 ◦C in
100 mM phosphate or citrate buffers in the range between pH 3.0 and 9.0.
6
−1
−1
as (1.46 +
− 0.06) × 10 M · s (kcat /K m ) respectively, the values
thus falling within the range reported for prokaryotic enzymes
(kcat 4900–15 900 s−1 and K m 3–5 mM) [7]. It is noteworthy that
the apparent K m values determined for the KatGs isolated directly
from P. simplicissimum and N. crassa were determined to be
higher, namely 10.8 mM [11] and 13.0 mM [12] respectively.
In addition, we analysed the capacity of MagKatG1 to oxidize
ethanol, halides and NADH. Ethanol oxidation was monitored
indirectly in a reaction system that continuously provided
H2 O2 . The observed specific activity for ethanol peroxidation
(0.3 unit/mg; Table 1) was at least 20-fold lower than the
average values reported for typical catalases [32]. Interestingly,
MagKatG1 can catalyse halogenation reactions using bromide
and iodide as two-electron donors (Table 1). The recombinant
oxidoreductase exhibited a relatively high bromination activity
compared with prokaryotic SynKatG (2.56 versus 0.7 unit/mg)
[34]. NADH oxidation was very low, even over the pH range
8.0–9.0 reported to be optimal for bacterial KatGs [7]. After
subtraction of unspecific background, the NADH oxidation
activity of the eukaryotic enzyme was very small (Table 1), namely
∼ 0.4 % and ∼ 15 % of the NADH oxidase activity reported for
B. pseudomallei [41] and E. coli KatG [7].
Thermostability tests on MagKatG1 were performed over the
range 25–65 ◦C by incubating the enzyme at the corresponding
temperature for defined time intervals, followed by measuring the
catalase and peroxidase activities at 25 ◦C (Table 2). Comparison
with equivalent data for bacterial counterparts [7] revealed that
MagKatG1 is a mesophilic enzyme that is more thermostable
than Synechocystis KatG or B. pseudomallei KatG [7]. Generally,
the catalase activity is more susceptible to temperature-induced
inactivation than is the peroxidase activity. Whether the observed
(limited) thermostability of the peroxidatic activity may be related
to the fact that M. grisea attacks predominantly various cultivars
of rice which grow optimally only in a subtropical climate remains
to be investigated.
Summing up, phytopathogenic fungi are unique in having two
KatG paralogues located either intracellularly (KatG1) or extracellularly (KatG2). The coding genes have been shown to derive
from an LGT from a (proteo)bacterial genome, followed by gene
duplication and diversification. MagKatG1 is the first eukaryotic
catalase/peroxidase to be heterologously expressed in E. coli in
high amounts, with high purity and with almost 100 % haem occupancy. The acidic, mainly homodimeric, oxidoreductase contains
c The Authors Journal compilation c 2009 Biochemical Society
14.2 +
− 4.6 units/mg
32.3 +
− 2.1 units/mg
2.7 +
− 0.4 units/mg
3.3 +
− 0.3 units/mg
0.36 +
− 0.06 unit/mg
2.56 +
− 0.11 units/mg
0.07 +
− 0.02 unit/mg
M133 + M166
(Per)oxidation of ethanol
NADH oxidase
0.3 +
− 0.1 unit/mg
2.2 +
− 0.5 units/μmol of haem
M133
K d,cyanide
k on,cyanide
k off,cyanide
1.4 +
− 0.3 μM 5 −1 −1
(9.0 +
× 10 M · s
− 0.4) −1
1.4 +
− 0.4 s
*Monochlorodimedone has been demonstrated not to function as an electron donor for
recombinant MagKatG1.
Table 2
M166)
Thermostability of recombinant MagKatG1 (clones M133 and
Abbreviation: n.d., not determined.
Temperature (◦C)
Loss of catalase activity
Loss of ABTS peroxidation activity
40
45
50
55
60
65
95 % after 22 h
95 % after 330 min
95 % after 48 min
95 % after 15 min
95 % after 8 min
95 % after 1 min
Stable; < 10 % loss after 48 h
Stable: < 10 % loss after 28 h
Stable: < 10 % loss after 28 h
n.d.
50 % after 7 h
50 % after 4 h
a predominantly five-co-ordinated high-spin haem b with spectral
and redox properties comparable with those of prokaryotic
enzymes. Its overwhelming catalase activity is characterized by
a pH optimum at pH 6.0 and kcat and K m values of 7010 s−1
and 4.8 mM respectively. The intracellular metalloenzyme is
expressed constitutively and might contribute to continuous
degradation of H2 O2 deriving from fungal metabolism in the
cytosol. In addition, the bifunctional enzyme acts as a versatile
peroxidase with pH optimum over the pH range 5.0–5.5, oxidizing
both one- (ABTS, o-dianisidine, pyrogallol and guaiacol) and twoelectron donors (Br− and I− ) at reasonable rates. However, similar
to prokaryotic KatGs, the biochemical and physiological role of
the peroxidase activity of fungal KatGs is unknown, as too is the
nature of the (putative) endogenous electron donor.
ACKNOWLEDGMENTS
We thank Dr Christian Würtz and Dr Hanspeter Rottensteiner (both of Abteilung
Systembiochemie, Institut für Physiologische Chemie, Ruhr-Universität Bochum, Bochum,
Germany) for providing us with the KatG-specific antibody.
122
Magnaporthe grisea catalase/peroxidase
FUNDING
This research was supported by the Fonds zur Förderung der wissenschaftlichen Forschung
[FWF (Austrian Science Fund)] [FWF project P19793].
REFERENCES
1 Smulevich, G., Jakopitsch, C., Droghetti, E. and Obinger, C. (2006) Probing the structure
and bifunctionality of catalase–peroxidase (KatG). J. Inorg. Biochem. 100, 568–585
2 Zamocky, M., Furtmüller, P. G. and Obinger, C. (2008) The evolution of catalases from
bacteria to man. Antioxid. Redox Signalling 10, 1527–1548
3 Zamocky, M. (2004) Phylogenetic relationships in class I of the superfamily of bacterial,
fungal, and plant peroxidases. Eur. J. Biochem. 271, 3297–3309
4 Passardi, F., Bakalovic, N., Teixeira, F. K., Margis-Pinheiro, M., Penel, C. and Dunand, C.
(2007) Prokaryotic origins of the non-animal peroxidase superfamily and organelle
mediated transmission to eukaryotes. Genomics 89, 567–579
5 Zamocky, M., Regelsberger, G., Jakopitsch, C. and Obinger, C. (2001) The molecular
peculiarities of catalase-peroxidases. FEBS Lett. 492, 177–182
6 Passardi, F., Zamocky, M., Favet, J., Jakopitsch, C., Penel, C., Obinger, C. and Dunand, C.
(2007) Phylogenetic distribution of catalase–peroxidases: are there patches of order in
chaos? Gene 397, 101–113
7 Singh, R., Wiseman, B., Deemagarn, T., Jha, V., Switala, J. and Loewen, P. C. (2008)
Comparative study of catalase-peroxidases (KatGs). Arch. Biochem. Biophys. 471,
207–214
8 Yamada, Y., Fujiwara, T., Sato, T., Igarashi, N. and Tanaka, N. (2002) The 2.0 Å crystal
structure of catalase–peroxidase from Haloarcula marismortui . Nat. Struct. Biol. 9,
691–695
9 Bertrand, T., Eady, N. A. J., Jones, J. N., Jesmin, J. M., Nagy, J. M., Jamart-Gregoire, B.,
Raven, E. L. and Brown, K. A. (2004) Crystal structure of Mycobacterium tuberculosis
catalase–peroxidase. J. Biol. Chem. 279, 38991–38999
10 Ten-I, T., Kumasaka, T., Higuchi, W., Tanaka, S., Yoshimatsu, K., Fujiwara, T. and Sato, T.
(2007) Expression, purification, crystallization and preliminary X-ray analysis of the
Met244Ala variant of catalase-peroxidase (KatG) from the haloarchaeon Haloarcula
marismortui . Acta Crystallogr. Sect. F Struct. Biol. Crystallogr. Commun. 63, 940–943
11 Fraaije, M. W., Roubroeks, H. P., Hagen, W. R. and van Berkel, W. J. H. (1996) Purification
and characterization of an intracellular catalase–peroxidase from Penicillium
simplicissimum . Eur. J. Biochem. 235, 192–198
12 Peraza, L. and Hansberg, W. (2002) Neurospora crassa catalases, singlet oxygen and cell
differentiation. Biol. Chem. 383, 569–575
13 Paris, S., Wysong, D., Debeaupuis, J.-P., Shibuya, K., Philippe, B., Diamond, R. D. and
Latgé, J.-P. (2003) Catalases of Aspergillus fumigatus . Infect. Immun. 71, 3551–3562
14 Scherer, M., Wei, H., Liese, R. and Fischer, R. (2002) Aspergillus nidulans
catalase–peroxidase gene (cpeA ) is transcriptionally induced during sexual development
through the transcription factor StuA. Eukaryotic Cell 1, 725–735
15 Valent, B. and Chumley, F. G. (1994) Avirulence genes and mechanisms of genetic
instability in the rice blast fungus. In Rice Blast Disease (Zeigle, R. S., Leong, S. A. and
Teng, P. S., eds.), pp. 111–134, International Rice Research Institute, Los Baños,
Philippines
16 Dean, R. A., Talbot, N. J., Ebbole, D. J., Farman, M. L., Mitchell, T. K., Orbach, M. J.,
Thon, M., Kulkarni, R., Xu, J. R., Pan, H. et al. (2005) The genome sequence of the rice
blast fungus Magnaporthe grisea . Nature 434, 980–986
17 Zamocky, M., Jakopitsch, C., Vlasits, J. and Obinger, C. (2007) Fungal catalase
peroxidases – a novel group of bifunctional oxidoreductases. J. Biol. Inorg. Chem. 12,
S97
18 Mittler, R. (2002) Oxidative stress, antioxidants and stress tolerance. Trends Plant. Sci. 7,
405–410
19 Xue, C., Li, L., Seong, K. and Xu, J.-R. (2004) Fungal pathogenesis in the rice blast
fungus Magnaporthe grisea . In Plant–Pathogen Interactions (Talbot, N. J., ed.),
pp. 138–165, CRC Press, Boca Raton, FL
451
20 Passardi, F., Theiler, G., Zamocky, M., Cosio, C., Rouhier, N., Teixera, F., Margis-Pinheiro,
M., Ioannidis, V., Penel, C., Falquet, L. and Dunand, C. (2007) PeroxiBase: The
peroxidase database. Phytochemistry 68, 534–539
21 Larkin, M. A., Blackshields, G., Brown, N. P., Chenna, R., McGettigan, P. A., McWilliam,
H., Valentin, F., Wallace, I. M., Wilm, A., Lopez, R. et al. (2007) Clustal W and Clustal X
version 2.0. Bioinformatics 23, 2947–2948
22 Reference deleted
23 Schliebs, W., Würtz, C., Kunau, W.-H., Veenhuis, M. and Rottensteiner, H.-P. (2006) A
eukaryote without catalase-containing microbodies: Neurospora crassa exhibits a unique
cellular distribution of its four catalases. Eukaryotic Cell 5, 1490–1502
24 Woodbury, W. A., Spencer, K. and Stahlmann, M. A. (1971) An improved procedure using
ferricyanide for detecting catalase isozymes. Anal. Biochem. 44, 301–305
25 Kolarich, D., Altmann, F. and Sunderasan, E. (2006) Structural analysis of the
glycoprotein allergen Hev b 4 from natural rubber latex by mass spectrometry.
Biochim. Biophys. Acta 1760, 715–720
26 Gill, S. C. and von Hippel, P. H. (1989) Calculation of protein extinction coefficients from
amino acid sequence data. Anal. Biochem. 182, 319–326
27 McGuffin, L. J., Bryson, K. and Jones, D. T. (2000) The PSIPRED protein structure
prediction server. Bioinformatics 16, 404–405
28 Lambert, C., Leonard, N., De Bolle, X. and Depiereux, E. (2002) EsyPred3D: prediction of
proteins 3D structures. Bioinformatics 18, 1250–1256
29 Battistuzzi, G., D’Onofrio, M., Borsari, M., Sola, M., Macedo, A. L., Moura, J. J. and
Rodrigues, P. (2000) Redox thermodynamics of low-potential iron–sulfur proteins.
J. Biol. Inorg. Chem. 5, 748–760
30 Roggenkamp, R., Sahm, W. and Wagner, F. (1974) Microbial assimilation of methanol
induction and function of catalase in Candida boidinii . FEBS Lett. 41, 283–286
31 Barr, D. P. and Aust, S. D. (1993) On the mechanism of peroxidase-catalyzed oxygen
production. Arch. Biochem. Biophys. 303, 377–382
32 Zamocky, M., Herzog, C., Nykyri, L. M. and 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
33 Kettle, A. J. and Winterbourn, C. C. (1994) Assays for the chlorination activity of
myeloperoxidase. Methods Enzymol. 233, 502–512
34 Jakopitsch, C., Regelsberger, G., Furtmüller, P. G., Rüker, F., Peschek, G. A. and Obinger,
C. (2001) Catalase–peroxidase from Synechocystis is capable of chlorination and
bromination reactions. Biochem. Biophys. Res. Commun. 287, 682–687
35 Hisada, H., Sano, M., Hata, Y., Abe, Y. and Machida, M. (2008) Deletion analysis of the
catalase-encoding gene (catB ) promoter from Aspergillus oryzae . Biosci. Biotechnol.
Biochem. 72, 48–53
36 Jakopitsch, C., Rüker, F., Regelsberger, G., Dockal, M., Peschek, G. A. and Obinger,
C. (1999) Catalase–peroxidase from the cyanobacterium Synechocystis PCC 6803:
cloning, overexpression in Escherichia coli , and kinetic characterization. Biol. Chem.
380, 1087–1096
37 Bellei, M., Jakopitsch, C., Battistuzzi, G., Sola, M. and Obinger, C. (2006) Redox
thermodynamics of the ferric–ferrous couple of wild-type Synechocystis KatG and
KatG(Y249F). Biochemistry 45, 4768–4774
38 Conroy, C. W., Tyma, P., Daum, P. H. and Erman, J. E. (1978) Oxidation–reduction
potential measurements of cytochrome c peroxidase and pH dependent spectral
transitions in the ferrous enzyme. Biochem. Biophys. Acta 537, 62–69
39 Jones, D. K., Dalton, D. A., Rosell, F. I. and Raven, E. L. (1998) Class I heme peroxidases:
characterization of soybean ascorbate peroxidase. Arch. Biochem. Biophys. 360,
173–178
40 Engleder, M., Regelsberger, G., Jakopitsch, C., Furtmüller, P. G., Rüker, F., Peschek, G. A.
and Obinger, C. (2000) Nucleotide sequence analysis, overexpression in Escherichia coli
and kinetic characterization of Anacystis nidulans catalase–peroxidase. Biochimie 82,
211–219
41 Singh, R., Wiseman, B., Deemagarn, T., Donald, L. J., Duckworth, H. W., Carpena, X., Fita,
I. and Loewen, P. C. (2004) Catalase–peroxidases (KatG) exhibit NADH oxidase activity.
J. Biol. Chem. 279, 43098–43106
Received 25 July 2008/30 October 2008; accepted 10 November 2008
Published as BJ Immediate Publication 10 November 2008, doi:10.1042/BJ20081478
123
c The Authors Journal compilation c 2009 Biochemical Society
Biochem. J. (2009) 418, 443–451 (Printed in Great Britain)
doi:10.1042/BJ20081478
SUPPLEMENTARY ONLINE DATA
Intracellular catalase/peroxidase from the phytopathogenic rice blast
fungus Magnaporthe grisea : expression analysis and biochemical
characterization of the recombinant protein
Marcel ZAMOCKY*§1 , Paul G. FURTMÜLLER*, Marzia BELLEI†, Gianantonio BATTISTUZZI†, Johannes STADLMANN‡,
Jutta VLASITS* and Christian OBINGER*
*Metalloprotein Research Group, Division of Biochemistry, Department of Chemistry, University of Natural Resources and Applied Life Sciences, Muthgasse 18 A-1190 Vienna, Austria,
†Department of Chemistry, University of Modena and Reggio Emilia, Modena, Italy, ‡Glycobiology Research Group, Division of Biochemistry, Department of Chemistry, University of
Natural Resources and Applied Life Sciences, Muthgasse 18 A-1190 Vienna, Austria, and §Institute of Molecular Biology, Slovak Academy of Sciences, Bratislava, Slovakia
Figure S1
Cloning methodology and heterologous expression of MagkatG1
(A) Oligonucleotide primers used for cloning and RT–PCR (introduced restriction sites are shown in bold type). (B) Recombinant plasmid used for the heterologous expression of the MagkatG1
gene (shown in red, ligated into pET21d vector). This plasmid construct was transformed in E. coli BL21 DE3 Star cells (Invitrogen).
1
To whom correspondence should be sent (email [email protected]).
124
c The Authors Journal compilation c 2009 Biochemical Society
M. Zamocky and others
Figure S2
Peptide pattern obtained by MS analysis of putative (proteolytic) degradation products of recombinant MagKatG1 separated by SDS/PAGE
Selected bands at 160 (A), 83 (B), 51 (C), 31 kDa (D) and 27 kDa (E) were digested with trypsin.
c The Authors Journal compilation c 2009 Biochemical Society
125
Magnaporthe grisea catalase/peroxidase
Figure S3
Analysis of oligomeric structure of recombinant MagKatG1 by gel-permeation chromatography on Superdex 200 prep grade
Recombinant M. grisea KatG1 was subjected to Ni2+ MCAC on a column loaded with Superdex 200 prep grade.
126
c The Authors Journal compilation c 2009 Biochemical Society
M. Zamocky and others
Figure S4
Secondary-structure prediction for MagKatG1 using PSIPRED
Abbreviations used: H, helix; E, strand; C, coil; confidence from 0 (=low) to 9 (=high). Colour
scheme for essential residues: magenta, distal side of the prosthetic haem group; cyan, proximal
side of haem; orange and yellow, distal and proximal residues respectively that are also found
in the (non-haem) C-terminal domain.
c The Authors Journal compilation c 2009 Biochemical Society
127
Magnaporthe grisea catalase/peroxidase
Figure S5
Selected parts of multiple sequence alignment (ClustalX) of ten bacterial and 33 fungal catalase/peroxidases
(A) Distal haem side, including Arg108 , Trp111 , His112 , Asp120 and Asn121 (see the arrows), the latter being the H-bonding partner of proximal His112 (MagKatG1 numbering). (B) part of large loop
1. (C) Continuation of large loop 1 (including Tyr238 ) and sequence region containing Met264 (part of the KatG-specific covalent adduct) and proximal His279 . (D) Sequence region containing the
H-bonding partner of proximal His279 , i.e. Asp389 and Trp330 . The colour scheme corresponds to the level of conservation: blue, highest (invariant); green, moderate; yellow, low. In the interests of
brevity, the individual enzymes are not identified.
128
c The Authors Journal compilation c 2009 Biochemical Society
M. Zamocky and others
Figure S6
Selected time traces of the reaction between recombinant MagKatG1 (clone M133) and various electron donors
(A) H2 O2 degradation monitored at 240 nm; (B) ABTS oxidation (414 nm); (C) guaiacol oxidation (470 nm); (D) o -dianisidine oxidation (460 nm); (E) pyrogallol oxidation (430 nm); and
(F) monochlorodimedone oxidation by released hypobromous acid. For detailed reaction conditions, see the Materials and methods section of the main paper.
Received 25 July 2008/30 October 2008; accepted 10 November 2008
Published as BJ Immediate Publication 10 November 2008, doi:10.1042/BJ20081478
c The Authors Journal compilation c 2009 Biochemical Society
129
appendix two
Protein- and solvent-based contributions to the redox
thermodynamics of lactoperoxidase and eosinophil peroxidase
Gianantonio Battistuzzi, Marzia Bellei, Jutta Vlasits, Srijib Banerjee, Paul G. Furtmüller,
Marco Sola and Christian Obinger
Archives of Biochemistry and Biophysics, in press
Protein- and solvent-based contributions to the redox
thermodynamics of lactoperoxidase and eosinophil peroxidase
Gianantonio Battistuzzi1*, Marzia Bellei1, Jutta Vlasits2, Srijib Banerjee2, Paul G.
Furtmüller2, Marco Sola1 and Christian Obinger2*
1
Department of Chemistry, University of Modena and Reggio Emilia, via Campi 183,
41100 Modena, Italy
2
Department of Chemistry, Division of Biochemistry, BOKU – University of Natural
Resources and Applied Life Sciences, Muthgasse 18, A-1190 Vienna, Austria
*corresponding authors:
1
E-mail: [email protected], Phone: +39-59-2055117;
Fax: +39-59-3735543
2
E-mail: [email protected], Phone: +43-1-36006-6073;
Fax: +43-1-36006-6059
131
Keywords: Lactoperoxidase, eosinophil peroxidase; reduction potential; redox
thermodynamics; enthalpy, entropy, heme cavity, channel architecture.
Abbreviations:
LPO,
lactoperoxidase;
EPO,
eosinophil
peroxidase;
MPO,
myeloperoxidase; TPO, thyroid peroxidase; HRP, horseradish peroxidase; ARP,
Arthromyces ramosus peroxidase; KatG, catalase-peroxidase; E°', standard reduction
potential (pH 7, 25°C); SHE, standard hydrogen electrode;
H°'rc, enthalpy change for the
reaction center upon reduction of the oxidized protein;
S°'rc, entropy change for the
reaction center upon reduction of the oxidized form; RR, resonance Raman; FT-IR,
Fourier transform – infrared.
132
Abstract
Eosinophil peroxidase (EPO) and lactoperoxidase (LPO) are important constituents of the
innate immune system of mammals. These heme enzymes belong to the peroxidasecyclooxygenase superfamily and catalyze the oxidation of thiocyanate, bromide and nitrite
to hypothiocyanate, hypobromous acid and nitrogen dioxide that are toxic for invading
pathogens. In order to gain a better understanding of the observed differences in substrate
specificity and oxidation capacity in relation to heme and protein structure, a
comprehensive spectroelectrochemical investigation was performed. The reduction
potential (E°') of the Fe(III)/Fe(II) couple of EPO and LPO was determined to be -126 mV
and -176 mV, respectively (25°C, pH 7.0). Variable temperature experiments show that
EPO and LPO feature different reduction thermodynamics. In particular, reduction of
ferric EPO is enthalpically and entropically disfavoured, whereas in LPO the entropic
term, which selectively stabilizes the oxidized form, prevails on the enthalpic term that
favours reduction of Fe(III). The data are discussed with respect to the architecture of the
heme cavity and the substrate channel. Comparison with published data for
myeloperoxidase demonstrates the effect of heme to protein linkages and heme distortion
on the redox chemistry of mammalian peroxidases and in consequence on the enzymatic
properties of these physiologically important oxidoreductases.
133
Introduction
Vertebrate peroxidases constitute a subfamily of the peroxidase-cyclooxygenase
superfamily [1]. The main clades of this subfamily are represented by myeloperoxidases
(MPOs) and eosinophil peroxidases (EPOs) that are well separated from the
lactoperoxidase (LPO) branch. These soluble heme peroxidases are important constituents
of the innate immune system and are released when pathogens and parasites invade the
organisms [2-4]. Closely related with their role in host defense is their enzymatic activity.
They oxidize small anionic molecules, such as halides (chloride and bromide), thiocyanate,
and nitrite to the corresponding oxidation products [2-5]. These involve e.g. hypohalous
acids or nitrogen dioxide, i.e. (strong) halogenating and nitrating antimicrobial oxidants.
Clearly segregated from these three clades and thus distantly related with MPO, EPO, LPO
are thyroid peroxidases (TPOs) that are crucial in hormone biosynthesis [6].
These four heme oxidoreductases are found in all mammals including man, and differ
in their ability to bind and oxidize their small inorganic substrate molecules [5]. X-ray
structures of mammalian peroxidases have been published for human myeloperoxidase [79], as well as for caprine lactoperoxidase [10, 11], whereas the three dimensional
structures of both EPO and TPO are still unknown.
In both EPO and LPO, as well as in all vertebrate peroxidases, the methyl substituents
of pyrrole rings A and C of heme b form two ester bonds with highly conserved distal
aspartate and glutamate residues [3, 7-11, 14] (Figures 1B & 1D). The existence of these
covalent bonds disturbs the symmetry and planarity of the prosthetic group thereby
modulating the spectroscopic and redox properties of the metalloproteins [14, 15]. This is
very pronounced with myeloperoxidases that are singular in having an additional
sulfonium ion linkage between the heme 2-vinyl group and a conserved methionine [7-9,
14] (Figures 1A & 1C). X-ray structures of both LPO and MPO clearly show that the heme
iron in its resting (ferric) high-spin state is positioned out of plane toward the proximal
histidine, that is hydrogen bonded to the NH2 group of a fully conserved asparagine [9]
(Figure 1D). This proximal heme site architecture is a typical feature for vertebrate
peroxidases. Furthermore, the distal site of both LPO and MPO accommodates a
conserved distal glutamine residue (besides the peroxidase-typical His-Arg couple) and a
complex hydrogen bonding network that includes several conserved water molecules
leading from the substrate channel to the heme iron, which supports halide delivery and
binding (Figures 1E & 1F). Moreover, an array of conserved water molecules might
134
A
B
C
D
R239
D94
Q91
H95
R225
Q105
H109
D108
Q259
M243
E258
E242
H351
H336
N421
N437
E
F
Q91
H95
H109
W1
W1
W2
R239
R225
W3
W4
Q105
W4
W2
W5
W3
W5
W6
Figure 1. Covalent binding pattern of heme in (A) myeloperoxidase (MPO) and (B) in lactoperoxidase
(LPO) and eosinophil peroxidase (EPO). In MPO the heme is covalently bonded via two ester bonds to D94
and E242, as well as via a sulfonium ion linkage to M243. In addition the conserved distal residues H95,
135
R239 and Q91, as well as proximal H336 and N421 are shown (C). With the exception of the MPO-typical
sulfonium ion linkage, the same residues are conserved in LPO (D) and – suggested by sequence alignment
[15] in EPO. Figures (E) and (F) depict the conserved water molecules in the distal heme side of MPO and
EPO. This panel was constructed using the following coordinates deposited in the Protein Data Bank. MPO:
1CXP; LPO: 2NQX (B) and 3GC1 (F).
facilitate proton release from hydrogen peroxide to the surface of the protein [7, 10, 11,
16].
The H2O2-mediated two-electron oxidation of peroxidases to compound I [oxoferryl
species, Fe(IV)=O, plus porphyryl or protein radical] requires stabilization of the ferric
state that rapidly reacts with H2O2. This is usually guaranteed by the reduction potentials
(E°') of the Fe(III)/Fe(II) couple, which range from -180 to -300 mV (vs. SHE) [14, 1721]. The E°' value of the Fe(III)/Fe(II) couple of LPO has been reported to be -170 mV
[17], whereas that for MPO is distinctively high (+ 5 mV) [18], being similar to those of
globins, due to the electron-withdrawing sulfonium ion linkage [14, 15]. No redox data
about EPO and TPO are available so far.
Here, the standard reduction potential E°' [(Fe(III)/Fe(II)] of EPO and the changes in
enthalpy, H°'rc, and entropy, S°'rc, accompanying Fe(III) to Fe(II) reduction in EPO and
LPO have been determined for the first time by spectroelectrochemical experiments
carried out at different temperatures. These thermodynamic data are compared with those
of MPO [18] and some members of the non-animal heme peroxidase superfamily [19-21]
and discussed with respect to structural and functional differences between the two
superfamilies as well as between the four clades of mammalian peroxidases, obtaining new
hints concerning the molecular determinants of E°' in this important class of
metalloenzymes.
Materials and Methods
Lactoperoxidase from bovine milk was purchased as a lyophilized powder (Sigma
Chemical Co. type L-8257, purity index A412/A280 > 0.9). Essentially salt-free protein was
dissolved in 100 mM phosphate buffer and 100 mM NaCl (pH 7.0), and concentration was
determined by using the extinction coefficient 112,000 M-1 cm-1 at 412 nm [22].
Lyophilized salt-free human eosinophil peroxidase with a purity index (A413/A280) > 1 was
obtained from Planta Natural Products (http://www.planta.at). Its concentration was
determined at Soret maximum at 413 nm (110,000 M-1 cm-1) [23].
All chemicals were reagent grade.
136
Spectroelectrochemistry. All experiments were carried out in a home-made OTTLE
cell [18-21, 24-26]. The three-electrode configuration was realized with a gold minigrid
working electrode (Buckbee-Mears, Chicago, IL), a Ag/AgCl/KClsat microreference
electrode, separated from the working solution by a Vycor set, and a platinum wire as the
counter electrode. The reference electrode was calibrated against a saturated calomel
electrode before and after each set of measurements. All potentials are referenced to the
SHE. Potentials were applied across the OTTLE cell with an Amel model 2053
potentiostat/galvanostat. The functioning of the OTTLE cell was checked by measuring
the reduction potential of yeast iso-1-cytochrome c in conditions similar to those used in
the present work [18-21, 24-26]. The E°' value was identical to that determined by cyclic
voltammetry (+260 mV) [26]. The temperature of the OTTLE cell, controlled with a
circulating water bath, was measured with a Cu-costan microthermocouple. UV-vis spectra
were recorded using a Varian Cary C50 spectrophotometer.
Variable temperature experiments were carried out using a “non-isothermal” cell
configuration [18-20, 26-29], namely, the temperature of the reference electrode and the
counter electrode was kept constant, whereas that of the working electrode was varied. For
this experimental configuration ΔS°'rc is calculated from the slope of the E°' versus
temperature plot, whereas ΔH°'rc is obtained from the Gibbs-Helmholtz equation, namely,
from the slope of the plot of E°'/T versus 1/T [30].
All experiments were carried out under Argon over the 15-35 °C range using 1 mL
samples containing 6 µM LPO and 4 µM EPO dissolved in 100 mM phosphate buffer and
100 mM NaCl at pH 7.0. Mediators were used as follows: 36 µM (LPO) or 20 µM (EPO)
methyl viologen and 2 µM (LPO) or 1 µM (EPO) lumiflavine-3-acetate, indigo
disulfonate, phenazine methosulfate, and methylene blue. Nernst plots consisted of at least
five points and were found to be linear, with a slope consistent with a one-electron
reduction process.
Results and Discussion
E°' [Fe(III)/Fe(II)]. The electronic spectra for bovine lactoperoxidase and human
eosinophil peroxidase at different applied potentials in the OTTLE cell (25°C, pH 7) are
shown in Figure 2, along with the corresponding Nernst plots (insets to Figure 2). The
electronic absorption spectrum of Fe(III) LPO shows a fairly sharp Soret band at 412 nm
and maxima in the visible range at 501 nm, 545 nm, 595 nm, and 631 nm, very similar to
those reported in the literature [13, 14, 31]. Reduction of ferric LPO proceeds through the
137
formation of a transient ferrous intermediate with band maxima at 444 nm (Soret band),
561 nm, and 593 nm, which was not stable but converted slowly within 45-60 minutes into
a new stable Fe(II) form of LPO, characterized by absorption maxima at 434 nm (Soret
band), 561 and 593 nm [31, 32]. This process is due to an equilibrium between a protein
form with a relatively open, unrestricted heme pocket, which transforms into a more
constrained conformer [31-33]. This phenomenon is typical for LPO, and has been
observed neither with MPO [31] nor with EPO (see below). Thus, to avoid any
interference from the transient Fe(II) intermediate, in this work LPO was first fully
reduced electrochemically, allowing its complete conversion into the stable ferrous form,
which was then subjected to (oxidative) spectroelectrochemical titrations. Therefore, the
E°' values and the corresponding reduction thermodynamics refer to the redox equilibrium
between the stable Fe(II) form of LPO and the ferric species.
The Nernst plots obtained from the spectroelectrochemical titrations of LPO are linear
(inset to Figure 2A) with a slope close to the theoretical value of RT/F=0.059 V at 25 °C,
as expected for the one-electron Fe(III)/Fe(II) redox process [18-20, 26-29, 34-36]. The
calculated standard reduction potential [E°' (Fe(III)/Fe(II)] of bovine LPO of -176 ± 5 mV
(25°C and pH 7.0) is slightly higher than a previous literature value (190 mV) [17],
probably as a consequence of differences in the ionic composition of the protein solutions
that were employed.
Figure 2B shows the electronic spectra of EPO measured at different applied
potentials [32]. From the linear Nernst plot (inset to Figure 2B) E°' was determined to be 126 ± 5 mV, 50 mV higher than in LPO. This difference is rather unexpected and
demonstrates that the similarity between the redox properties of the two peroxidases is less
strict than was previously supposed [5, 16].
The E°' values for the Fe(III)/Fe(II) couple in LPO and EPO are listed in table 1 along
with those for MPO and (non-animal) heme peroxidases with unmodified heme b. The E°'
values for LPO and EPO were 181 mV and 131 mV lower than in MPO [18],
unequivocally underlying the impact of the peculiar electron withdrawing sulfonium ion
bond and of the pronounced heme distortion on the redox chemistry of heme iron in
myeloperoxidase. Interestingly, the hierarchy of E°' [Fe(III)/Fe(II)], namely MPO > EPO >
LPO, follows the same trend that was observed for E°' of the compound I/Fe(III) couple,
i.e. MPO (1160 mV) > EPO (1100 mV) > LPO (1090 mV) [37, 38]. This suggests that
within a defined peroxidase the same molecular factors influence the redox properties of
the heme iron at different oxidation states. Additionally, it nicely reflects the capacity of
138
these oxidoreductases to oxidize halides. Only MPO is able to oxidize chloride [E°'
(HOCl, Cl-, H2O) = 1280 mV] at reasonable rates, and bromide oxidation [E°' (HOBr, Br-,
H2O) = 1130 mV] mediated by EPO compound I is faster than by LPO compound I [5,
16].
0,08
-0,1
A
E (V)
Absorbance
-0,14
0,06
-0,18
-0,22
-0,4
0,04
0,2
0,8
Log X
0,02
0
320
360
400
440
480
520
560
600
Wavelength (nm)
0,05
E (V)
Absorbance
0,04
-0,09
B
-0,14
0,03
-0,19
-0,8
0,02
-0,3 0,2
Log X
0,01
0
320
360
400
440
480
520
560
600
Wavelength (nm)
Figure 2. Electronic spectra of high-spin bovine LPO (A) and human EPO (B) obtained at various potentials.
Spectra were recorded at 25°C. The insets depict the corresponding Nernst plots, were X represents (AredmaxAre)/(Aoxmax-Aox). In case of LPO (A) oxmax = 412 nm and redmax = 434 nm, in case of EPO oxmax = 413
nm and redmax = 437 nm.
Due to the lack of the three dimensional structure of EPO, the observed differences
between LPO and EPO cannot be related to distinct structural features. In both proteins,
heme b is covalently bound by two ester bonds [10, 11, 39] resulting in similar spectral
properties. Compared with (non-animal, plant-type) peroxidases with unmodified heme b,,
both LPO and EPO have more positive reduction potentials of the Fe(III)/Fe(II) couple
(Table 1). This indicates that the two heme to protein ester bonds affects the redox
chemistry of the heme iron, in agreement with theoretical calculations [15]. Apparently
these effects are more pronounced in EPO than in LPO.
139
The electron density at the heme iron in peroxidases is sensibly influenced by the
imidazolate character of the proximal histidine, which depends on its H-bonding partner
that is a conserved aspartate and a glutamine in heme b (plant-type) peroxidases and in
vertebrate peroxidases, respectively. In the former enzymes, it is generally assumed that
the Asp-His-iron interaction at the proximal heme side mainly contributes to the (negative)
E°' by deprotonating the N position of the proximal histidine. This guarantees the stability
of the ferric form that is competent in binding and reducing of hydrogen peroxide. In
vertebrate peroxidases, the unprotonated N of the proximal histidine (anionic) is hydrogen
bonded with the NH2-group of a fully conserved asparagine (Figure 1 D), whose carbonyl
group interacts with the guanidinium group of a conserved arginine [9]. The pronounced
imidazolate character of the proximal histidine is confirmed by resonance Raman studies
in the low-frequency region [14] that demonstrated that the iron-imidazole stretching mode
of both LPO and MPO is at fairly high frequencies. These data suggest that – independent
of the nature of the H-bonding partner – in both heme peroxidase superfamilies the
proximal histidine has a pronounced imidazolate character, that strengthens the ironimidazole bond and exerts a comparable influence on the E°' of the heme iron, stabilizing
its higher oxidation states.
Table 1. Thermodynamic parameters for the Fe(III)  Fe(II) reduction in human eosinophil peroxidase
(EPO) and bovine lactoperoxidase (LPO)a.
Protein
E°’ (V)
b
ΔH°’rc
ΔS°’rc
(kJ mol-1)
(JK-1 mol-1)
ΔH°’rc(int)
(=G°'= -nFE°')
(kJ mol-1)
EPO
-0.126
+7
-18
+12
LPO
-0.176
-5
-73
+17
MPO [18]
+0.005
+3
+10
0
HRP [21]
-0.306
+91
+210
+29
ARP [20]
-0.183
-18
-120
+18
KatG [19]
-0.226
+17
-18
+22
a
For comparison purposes, the E°’ values of the Fe(III)/Fe(II) couple of human myeloperoxidase (MPO),
horseradish peroxidase isoform C (HRP), Arthromices ramosus peroxidase (ARP) and Synechocystis PC6803
catalase-peroxidase (KatG) are presented (ref. in parenthesis). Average errors for E°’, ΔH°’rc, and ΔS°’rc are
±0.005 V, ±4 kJ mol-1, and ±8 JK-1 mol-1, respectively.
b
At 298.15 K and referred to the SHE. Often the sum (-ΔH°’rc/F + T298ΔS°’rc/F) does not exactly match E°’
because the ΔH°’rc and ΔS°’rc values are rounded to the closest integer, as a result of experimental error.
140
Reduction thermodynamics. The temperature profiles of the E°' values for EPO and
LPO are shown in Figure 3, together with that of MPO [18]. In contrast to MPO, the
reduction potentials of both LPO and EPO decrease with increasing temperature. The
spectroelectrochemical response deteriorates above 35°C due to protein denaturation and
solvent evaporation. The obtained temperature profiles allowed factorization of the
enthalpy (H°'rc) and entropy (S°'rc) changes during the reduction reaction, which are
listed in Table 1 together with those of MPO [18] and non-animal (plant-type) heme
peroxidases [19-21].
EPO and LPO feature different reduction thermodynamics. Reduction entropies are
negative in both cases, but reduction enthalpy is positive for EPO and negative for LPO.
Hence, in EPO Fe(III) reduction is enthalpically and entropically disfavoured, whereas the
negative E°' value of LPO results from two opposing thermodynamic contributions: an
enthalpic term which favors Fe(III) reduction and a larger entropic term that disfavors it.
As discussed thoroughly elsewhere [40], changes in enthalpy and entropy contain
contributions arising from both the reorganization of solvent molecules within the
hydration sphere of the proteins and from protein-based factors. Thus the enthalpy (H°'rc)
and entropy (S°'rc) changes accompanying Fe(III) to Fe(II) reduction in heme proteins are
composed of intrinsic protein contributions (H°'rc,
int
and S°'rc,
int)
and solvent
reorganisation factors (H°’rc, solv and S°’rc, solv) [18-21, 27, 40] with H°'rc = H°'rc, int +
H°'rc, solv and S°'rc = S°'rc, int + S°'rc, solv.
H°'rc, int is mainly controlled by the donor properties of the axial heme ligands, the
polarity of the protein environment and the electrostatic interactions between the redox
center and polar and charged residues surrounding the heme, whereas S°'rc, int depends on
redox state-dependent differences in protein flexibility [18-21, 27, 40]. The available data
for the ferric and ferrous forms of heme proteins indicate that, in general, reductioninduced 3D structural changes are quite small, allowing, as a first approximation, to
attribute the measured S°'rc to reduction-induced solvent reorganization effects only, i.e.
S°'rc, int  0 [26, 27]. This approach proved to work with plant-type peroxidases (HRP [20,
21], ARP [20], KatG [19]), as well as with MPO [18]. In the latter case, in the absence of
structural information concerning the ferrous enzyme, this hypothesis is based on the small
local conformational changes observed upon substrate and/or ligand binding to the ferric
protein [7, 8]. Given the strict similarity between the structural effect of bromide,
thiocyanate and cyanide binding in MPO and LPO [7, 8, 10, 11], the above approach has
141
been extended to the interpretation of the reduction thermodynamics of lactoperoxidase
and eosinophil peroxidise. Hence, it follows that, to a first approximation, for LPO and
EPO S°'rc = S°'rc, solv = -73 JK-1mol-1 and -18 JK-1mol-1, respectively. This compares with
+10 JK-1mol-1 for MPO [18].
Since reduction-induced solvent reorganization effects (e.g. changes in the H-bonding
network involving the water molecules within the hydration sphere of the protein), have
been found to induce compensatory enthalpy and entropy changes [40-44], the
corresponding enthalpic contribution can be factorized out from the measured enthalpy
change, finally allowing estimation of the protein-based contribution to H°'rc:
H°'rc, int = H°'rc – H°'rc, solv = H°'rc – TS°'rc = G°'rc = -nFE°'
(n = 1)
Hence, to a first approximation, the measured E°' coincides with H°'rc,
int
and would
ultimately be determined by the selective enthalpic stabilization of one of the two redox
states by first coordination sphere and electrostatic effects. As a consequence H°'rc, int = nFE°' corresponds to +17 kJ/mol (LPO), +12 kJ/mol (EPO) and 0 kJ/mol (MPO) (Table
1).
142
0.00
A
E°' (V)
-0.05
-0.10
-0.15
-0.20
10
20
30
40
T (°C)
0.0
E°'/T x 10-3 (V K-1)
-0.1
B
-0.2
-0.3
-0.4
-0.5
-0.6
3.2
3.3
3.4
3.5
1/T x 10-3 (K-1)
Figure 3. Temperature dependence of the reduction potential (A) and E°’/T versus 1/T plots (B) for bovine
LPO (●), human EPO (■), and human MPO (▲) (17). The slope of the plot yields S°’rc/F and –H°’rc/F
values, respectively. Error bars have the same dimensions of the symbols.
Reduction enthalpy. Protein-based factors, such as metal-ligand binding interactions
and the electrostatics at the interface between the heme, the protein environment and the
solvent, selectively stabilize the ferric form of LPO and EPO much more efficiently than in
MPO. In plant-type peroxidases the large enthalpic stabilization of the ferric heme (Table
1) has been attributed to the basic character of the proximal histidine and to the polarity of
the distal heme cavity, which contains several water molecules involved in an extended
network of hydrogen bonds [19-21]. Both sequence analysis [16] and comparison of the
published X-ray structures [7-11] suggest that in mammalian peroxidases the heme cavity
architecture is conserved. In fact, the anionic character of the proximal histidine, ensuing
143
from its interaction with the adjacent asparagine [9], as well as the strength of the ironimidazolate bond and (Figure 1) [31] seem to be a common feature in this family of
peroxidases. Similarly, the interactions of the propionate sidechains on pyrrole rings C &
D are conserved [7-11, 16]. Also the catalytic amino acids His, Arg and Gln [16] in the
distal heme side are strictly conserved (Figures 1C & 1D) as is an extended H-bonding
network formed by distinct water molecules (Figures 1E & F), one of which is hydrogen
bonded to the N of the distal histidine and lies about midway between the Nof the latter
residue and Fe(III).
Based on these observations, it can be assumed that the contribution of these structural
features to H°'rc, int = [H°'rc, int(MPO) – H°'rc, int(LPO)] = -17 kJ/mol, and therefore to
E°' = [E°' (MPO) – E°' (LPO)] = 181 mV is quite small. Thus, it seems that the lower
enthalpic stabilization of ferric MPO and the resulting high E°' [Fe(III)/Fe(II)] is mainly a
consequence of the sulfonium linkage connecting the sulfur atom of Met243 with the carbon of the vinyl group on pyrrole ring A (Figure 1A) [7, 15, 18]. In fact, the positive
charge on the sulfur atom, which does not appear to be involved in any additional
electrostatic interactions with the protein [7, 8], would electrostatically stabilize the ferrous
form of the heme. On the other hand, its electron withdrawing effect reduces the basicity
of the four pyrrole nitrogens, thereby decreasing the electron density at the heme iron and
lowering the enthalpic stabilization of the ferric form of the enzyme due to ligand binding
effects. The latter effect is enhanced by the pronounced distortions imposed on the
porphyrin ring in MPO, which further reduces the contacts with the pyrrole nitrogens with
the iron ion [7, 8, 31].
The lower protein-based enthalpic stabilization of the ferric form in EPO compared to
LPO, H°'rc, int = [H°'rc, int(EPO) – H°'rc, int(LPO)] = -5 kJ/mol and the corresponding
E°' = E°' (EPO) – E°' (LPO)] = 50 mV, is difficult to explain due to the absence of
structural data [16]. Sequence [16] and mass spectrometric analysis [39] suggest a similar
heme cavity architecture and heme to protein binding mode. On the other hand, differences
in the electronic UV-vis spectra of the ferrous and ferric form of both enzymes [32] as well
in FT-IR data [31, 44] might reflect some differences in the interaction of the heme iron
with the proximal histidine and/or the distal water network.
144
Reduction entropy. Since reduction entropy in LPO and EPO is, to a first
approximation, mainly determined by reduction-induced solvent reorganization effects
within the hydration sphere of the proteins, the observed differences in S°'rc cannot be
easily correlated with structural features of the proteins unless significant differences exist
in protein-solvent interactions, especially at the heme moiety. Qualitatively, the negative
S°'rc values obtained for LPO and EPO indicate that reduction of Fe(III) increases the
ordering of the solvent molecules in the solvation sphere of both proteins. MPO, on the
contrary, features a small and positive reduction entropy [18] that has been ascribed to a
rigid hydrogen bond network involving conserved water molecules in the surrounding of
the heme, which would support the binding of its small anionic substrate and help to
transfer and incorporate the oxyferryl oxygen to fixed Cl- thereby forming hypochlorous
acid [16, 18].
Comparison of the 3D-structures of the iodide and bromide complexes of LPO and
MPO [8, 11] suggests that the overall rigidity of the hydrogen bond network involving the
water molecules in the surrounding of the heme is a common feature of mammalian
peroxidases and not limited to MPO. Therefore, the origin of the difference of the S°'rc
values between LPO and MPO (and EPO) must be searched outside the heme cavity. In
MPO the substrate channel is considerably shallow and has a funnel-like shape [7],
whereas in LPO the cylindrical access channel is narrower, longer and more hydrophobic
(Figure 4) [10, 11]. By contrast, the substrate channels of EPO and MPO feature an 80%
sequence homology [16], suggesting a similar channel architecture in both proteins.
Therefore, it appears that the heme distal site in LPO is less accessible and less solvent
exposed than in MPO and EPO (Figure 4). These differences are mirrored by the S°'rc
values reported in Table 1. In MPO and EPO Fe(III) reduction is associated with small
entropy changes (although of different sign), whereas in LPO reduction entropy is larger
and negative. These observations reflect a general feature in metalloprotein chemistry,
namely that the reduction entropy turns from positive to negative values with decreasing
the solvent accessibility of the metal center [19-21, 25-28, 30].
145
Acknowledgements
This work was supported by the Ministero dell’Universitá e della Ricerca Scientifica
(MUR) of Italy (Programmi di Ricerca Scientifica di Rilevante Interesse Nazionale 2007
Prot. N. 2007KAWXCL and n. 20079Y9578), by the University of Modena and Reggio
Emilia (Fondo di Ateneo per la Ricerca, 2007), and the Austrian Science Fund FWF
project P20664.
146
References
[1] M. Zamocky, C. Jakopitsch, P.G. Furtmüller, C. Dunand, C Obinger, The
peroxidase-cyclooxygenase superfamily: reconstructed evolution of critical
enzymes of the innate immune system, Proteins 71 (2008) 589-605.
[2] S.J. Klebanoff, Myeloperoxidase: friend and foe, J. Leukoc. Biol. 77 (2005) 598625
[3] J. Wang, A. Slungaard, Role of eosinophil peroxidase in host defense and disease
pathology, Arch. Biochem. Biophys. 445 (2006) 256-260.
[4] R. Ihalin, V. Loimaranta, J. Tenovuo, Origin, structure, and biological activities of
peroxidases in human saliva, Arch. Biochem. Biophys. 445 (2006) 261-268.
[5] J. Arnhold, E. Monzani, P.G. Furtmüller, M. Zederbauer, L. Casella, C. Obinger,
Kinetics and thermodynamics of halide and nitrite oxidation by heme peroxidases,
Eur. J. Inorg. Chem. 380 (2006) 3801-3811.
[6] J. Ruf, P. Carayon Structural and functional aspects of thyroid peroxidase, Arch.
Biochem. Biophys. 445 (2006) 269-277.
[7] T.J. Fiedler, C.A. Davey, R.E. Fenna, X-ray crystal structure and characterization
of halide-binding sites of human myeloperoxidase at 1.8 angstrom resolution, J.
Biol. Chem. 275 (2000) 11964-11971.
[8] M. Blair-Johnson, T. Fiedler, R. Fenna, Human myeloperoxidase: Structure of a
cyanide complex and its interaction with bromide and thiocyanate substrates at 1.9
angstrom resolution, Biochemistry 40 (2001) 13990-13997.
[9] X. Carpena, P. Vidossich, K. Schroettner, B. M. Calisto, S. Banerjee, J. Stampler,
M. Soudi, P.G. Furtmüller, C. Rovira, I. Fita, C. Obinger, Essential role of
proximal histidine-asparagine interaction in mammalian peroxidases, J. Biol.
Chem. 284 (2009) 25929–25937.
[10] A.K. Singh, N. Singh, S. Sharma, S. B. Singh, P. Kaur, A. Bhushan, A.
Srinivasan, T.P. Singh, Crystal structure of lactoperoxidase at 2.4 Å resolution, J.
Mol. Biol. 376 (2008) 1060-1075.
[11] I.A. Sheikh, A.K. Singh, N. Singh, M. Sinha, S.B. Singh, A. Bhushan, P. Kaur, A.
Srinivasan, S. Sharma, T.P. Singh, Structural evidence of substrate specificity in
mammalian
peroxidases:
structure
of
the
thiocyanate
complex
with
lactoperoxidase and its interactions at 2.4 Å resolution, J. Biol. Chem. 284 (2009)
14849-14856.
147
[12] I. Olsson, A.M. Persson, K. Strömberg, I. Winqvist, P.C. Tai, C.J. Spry,
Purification of eosinophil peroxidase and studies of biosynthesis and processing in
human marrow cells, Blood 66 (1985) 1143-1148.
[13] G. Sievers, Structure of milk lactoperoxidase. Evidence for a single polypeptide
chain, FEBS Lett. 127(1981) 253-256.
[14] M. Zederbauer, P. G. Furtmüller, S. Brogioni, C. Jakopitsch, G. Smulevich, C.
Obinger, Heme to protein linkages in mammalian peroxidases: Impact on
spectroscopic, redox and catalytic properties, Nat. Prod. Rep. 24 (2007) 571-584.
[15] A. Deverjan, A.E. Gaenko, U. Ryde, Effect of covalent links on the
structure,spectra and redox properties of myeloperoxidase – A density functional
study, J. Inorg. Biochem. 102 (2008) 1549-1557.
[16] P.G. Furtmüller, M. Zederbauer, W. Jantschko, J. Helm, M. Bogner, C. Jakopitsch,
C. Obinger, Active site structure and catalytic mechanisms of human peroxidases,
Arch. Biochim. Biophys. 445 (2006) 199-213.
[17] P.I. Ohlsson,, G. Paul The reduction potential of lactoperoxidase, Acta Chem.
Scand. 37 (1983) 917-921.
[18] G. Battistuzzi, M. Bellei, M. Zederbauer, P.G. Furtmüller, M. Sola, C. Obinger,
Redox thermodynamics of the Fe(III)/Fe(II) couple of human myeloperoxidase in
its high-spin and low-spin forms, Biochemistry 45 (2006) 12750-12755.
[19] M. Bellei, C. Jakopitsch, G. Battistuzzi, M. Sola, C. Obinger, Redox
thermodynamics of the ferric-ferrous couple of wild-type Synechocystis KatG and
KatG(Y249F), Biochemistry 45 (2006) 4768-4774.
[20] G. Battistuzzi, M. Bellei, L. Casella, C.A. Bortolotti, R. Roncone, E. Monzani, M.
Sola, Redox properties of the Fe3+/Fe2+ couple in Arthromyces ramosus class II
peroxidase and its cyanide adduct, J. Biol. Inorg. Chem. 11 (2006) 586-592.
[21] G. Battistuzzi, M. Borsari, A. Ranieri, M. Sola Redox thermodynamics of the
Fe3+/Fe2+ couple in horseradish peroxidase and its cyanide complex, J. Am. Chem.
Soc. 124 (2002) 26-27.
[22] K.G. Paul, P.I. Ohlsson in: K.M. Pruitt, J.O. Tenuovo (Eds.), The Lactoperoxidase
System, Chemistry and Biological Significance, Marcel Dekker, New York 1985,
pp. 15-30.
[23] B.G.J.N. Bolscher, H. Plat, R. Wever, Some properties of human eosinophil
peroxidase, a comparison with other peroxidases, Biochim. Biophys. Acta 784
(1984) 177-186.
148
[24] G. Battistuzzi, M. Borsari, F. Francia, M. Sola, Redox thermodynamics of the
native and alkaline forms of eukaryotic and bacterial class I cytochromes c,
Biochemistry 36 (1997) 16247-16258.
[25] G. Battistuzzi, M. Borsari, J.A. Cowan, C. Eicken, L. Loschi, M. Sola, Redox
chemistry and acid-base equilibria of mitochondrial plant cytochromes c,
Biochemistry 38 (1999) 5553-5562.
[26] G. Battistuzzi, M. Borsari, J.A. Cowan, A. Ranieri, M. Sola, Control of
cytochrome c redox potential: axial ligation and protein environment effects, J.
Am. Chem. Soc. 124 (2002) 5315-5324.
[27] G. Battistuzzi, M. Bellei, L. Casella, C.A. Bortolotti, R. Roncone, E. Monzani, M.
Sola, Redox reactivity of the heme Fe3+/Fe2+ couple in native myoglobins and
mutants with peroxidase-like activity, J. Biol. Inorg. Chem. 12 (2007) 951-958.
[28] V.T. Taniguchi, N. Sailasuta-Scott, F.C. Anson, H.B. Gray, Reduction potential
and thermodynamic parameters of adrenodoxin by the use of an anaerobic thinlayer electrode, Pure Appl. Chem. 52 (1980) 2275-2281.
[29] S. Dong, J. Niu, T.M. Cotton, Mediator-Assisted Continuous-Flow Column
Electrolytic Spectroelectrochemical Technique for the Measurement of Protein
Redox Potentials. Application to Peroxidase, Methods Enzymol. 246 (1995) 701735.
[30] E.L. Yee, R.J. Cave, K.L. Guyer, P.D. Tyma, M.J. Weaver, A survey of ligand
effects upon the reaction entropies of some transition metal redox couples, J. Am.
Chem. Soc. 101 (1979) 1131-1137.
[31] S. Brogioni, J. Stampler, P.G. Furtmüller, A. Feis, C. Obinger, G. Smulevich, The
role of the sulfonium linkage in the stabilization of the ferrous form of
myeloperoxidase: a comparison with lactoperoxidase, Biochim. Biophys. Acta
1784 (2008) 843-849.
[32] H.M. Abu-Soud, S.L Hazen, Interrogation of heme pocket environment of
mammalian peroxidases with diatomic ligands, Biochemistry 40 (2001) 1074710755.
[33] S. Galijasevic, G.M. Saed, M.P. Diamond, H.M. Abu-Soud, High dissociation rate
constant of ferrous-dioxy complex linked to the catalase-like activity in
lactoperoxidase, J. Biol. Chem. 279 (2004) 39470-39645.
149
[34] W.R. Heineman, M.L. Meckstroth, B.J. Norris, C.-H. Su, A small-volume thinlayer spectroelectrochemical cell for the study of biological components,
Bioelectrochemistry and Bioenergetics 6 (1979) 577-585.
[35] C.V. Millis, D. Cai, M.T. Stankovic, M. Tien, Oxidation-reduction potentials and
ionization states of extracellular peroxidases from the lignin-degrading fungus
Phanerochaete chrysosporium, Biochemistry 28 (1989) 8484-8489.
[36] Z.S. Farhangrazi, M.E. Fossett, L.S. Powers, W.R. Ellis, Variable-temperature
spectroelectrochemical study of horseradish peroxidase, Biochemistry 34 (1995)
2866-2871.
[37] J. Arnhold, P.G. Furtmüller, G. Regelsberger, C. Obinger, Redox properties of the
couple compound I/native enzyme of human myeloperoxidase and eosinophil
peroxidase, Eur. J. Biochem. 268 (2001) 5142-5148.
[38] P.G. Furtmüller, J. Arnhold, W. Jantschko, M. Zederbauer, C. Jakopitsch, C.
Obinger, Standard reduction potentials of all couples of the peroxidase cycle of
lactoperoxidase, J. Inorg. Biochem. 99 (2005) 1220-1229.
[39] C. Oxvig, A.R. Thomsen, M.T. Overgaard, E.S. Sorensen, P. Hojrup, J. Bjerrum,
G.J. Gleich, L. Sottrup-Jensen, Biochemical evidence for heme linkage through
esters with Asp-93 and Glu-241 in human eosinophil peroxidase. The ester with
Asp-93 is only partially formed in vivo, J. Biol. Chem. 274 (1999) 16953-16958.
[40] M. Sola, G. Battistuzzi, M. Borsari, Modulation of the free energy of reduction in
metalloproteins, Chemtracts: Inorg. Chem. 18 (2005) 73-86.
[41] G. Battistuzzi, M. Borsari, G. Di Rocco, A. Ranieri, M. Sola, Enthalpy/entropy
compensation phenomena in the reduction thermodynamics of electron transport
metalloproteins, J. Biol. Inorg. Chem. 9 (2004) 23-26.
[42] G.W. Canters, U. Kolczak, F.A. Armstrong, L.J. Jeuken, R. Camba, M. Sola The
effect of pH and ligand exchange on the redox properties of blue copper proteins,
Faraday Discuss. 116 (2000) 205-220.
[43] L. Liu, Q.-X. Guo, Isokinetic relationship, isoequilibrium relationship, and
enthalpy-entropy compensation, Chem. Rev. 101 (2001) 673-695.
[44] I.M. Kooter, A.J. Pierik, M. Merkx, B.A. Averill, N. Moguilevsky, A. Bollen, R.
Wever, Difference Fourier transform infrared evidence for ester bonds linking the
heme group in myeloperoxidase, lactoperoxidase, and eosinophil peroxidase, J.
Am. Chem. Soc. 119 (1997) 11542-11543.
150