1.1
Theoretische Modellierung aktiver Zentren von
molybdänabhängigen Enzymen
INAUGURAL – DISSERTATION
zur
Erlangung der Doktorwürde
der
Naturwissenschaftlich-Mathematischen Gesamtfakultät
der
Ruprecht-Karls-Universität Heidelberg
Vorgelegt von
M.Phil. Uzma Habib
aus: Wah Cantt, Pakistan
Thema
Theoretische Modellierung aktiver Zentren von
molybdänabhängigen Enzymen
Gutachter:
Tag der mündlichen Prüfung: . .2012
Erklärung
Erklärung gemäß § 8 (3) b) und c) der Promotionsordnung:
a) Ich erkläre hiermit an Eides statt, dass ich die vorgelegte Dissertation selbst verfasst und
mich keiner anderen als der von mir ausdrücklich bezeichneten Quellen und Hilfen bedient
habe.
b) Ich erkläre hiermit an Eides statt, dass ich an keiner anderen Stelle ein Prüfungsverfahren
beantragt bzw. die Dissertation in dieser oder anderer Form bereits anderweitig als
Prüfungsarbeit verwendet oder einer anderen Fakultät als Dissertation vorgelegt habe.
Heidelberg, den . . 2012
————————————
Uzma Habib
Dedication
To my family especially my parents,
the SKY of my life.
“‫ﺑوﮐﮯﻨﺎم‬١‫ﻣﻰ‬١‫”ﻣﯾرے ﭙﯾﺎرے‬
Message for life
Plan more than what you think you could do, then try
to accomplish it all. Build your castle in the clouds.
Then build a ship to take you there.
Aim high, even if you don’t make
it to the sun, you will at least
land among the
STARS.
ACKNOWLEDGEMENT
I like to thank ALLAH almighty for HIS blessings throughout our lives.
I am enormously indebted to PD. Dr. Matthias Hofmann, for offering me a
doctoral position under his supervision. I am grateful to him for providing me
the motivation and encouragement during my doctoral studies. I am indebted to
him for sharing his vast knowledge of chemistry and providing excellent training
of handling and presenting scientific projects. I am also thankful to him for
providing financial support and very nice company over the years.
I express my sincere gratitude to Prof. Dr. Roland Krämer for giving me place in
his group and for his kind formal supervision. I express my gratitude to all the
present and past members of AK Krämer during my PhD time, especially Dr.
Andriy Mokhir, Larisa, Helga, Katharina, Susanne, Annemarie, Birgit, Claudia,
Ute, Helen, Uli, Armin, Gerrit, Paul, Subrata, Michael, Ulrich and Volker.
I am grateful to Prof. Dr. Peter Comba for providing an excellent additional
education through the program of the Graduate College-850. I am also thankful
to all members of “Graduate College-850, Modeling of Molecular Properties”
especially Bodo Martin, for his help during seminars and conferences.
I am very much grateful to the administrative staff including Frau Jeannette
Grosse, Frau Claudia Aßfalg, Frau Silke Dussel, Frau Karin Stelzer and Frau
Marlies Schilli. I am also thankful to the official and technical staff of
Anorganisch Institüt for their help in completing my research.
My sincere gratitude to Dr. M. Rasul Jan and Dr. Jasmin Shah,
Chemistry Department, Peshawar University, Pakistan. A source of inspiration
that induce in me a thirst of acquiring knowledge, whose ever encouraging and
valuable guidance always persisted me to work hard and to come up with task
accomplished. For me they are a source of admiration Thanks to them for all they
did for me and for their kind consideration and care which is continued uptil
now and hope will forever, Inshallah.
I like to say especial thanks to Rabia, Zarghoona, Rashda, Saba, Faiza,
Shanthi, Sajjad Bhai, Farooq bhai and Awais. Words cannot express the nature
and extent of my deepest appreciation for them. They were always there in the
hour of need. I want to thank them for their kind behaviour, valuable advices,
timely suggestions and constant encouragement all the time.
My love and especial thanks to most precious Ischal, Shees and Shaeel, for their
unconditional love as during my stay in Germany I always get energy after
spending lovely time with them.
Acknowledgements are also due to my friends, Anashua, Kiran, Lubna,
Maryum, Tanuja and Avik, for their kindly cooperation, valuable suggestions
and encouragement throughout my stay.
Finally I wish to thank my parents for their moral support and their
encouragement to come so far from my home to achieve the goal. I would also
like to thank to my brothers and sisters and their beloved partners and their
precious kids for their love and support. Without their prayers I couldn’t have
put my best in the research and I am grateful for that.
Uzma
Table of Contents
Summary
Chapter 1: Introduction
1. Enzymes
1
2. Chemical properties of Molybdenum and Tungsten
1
3. Biological distribution of Molybdenum and Tungsten enzymes
2
3.1.Molybdenum enzymes
2
3.2.Tungsten enzymes
3
4. Pyranopterin cofactor
3
5. Classification and structures of the Molybdenum enzymes
5
5.1. The Molybdenum Hydroxylases
5
5.1.1. Aldehyde Oxidase
6
5.1.2. Xanthine Oxidase or Xanthine Dehydrogenase
6
5.2. The Eukaryotic Oxotransferases
7
5.2.1. Sulfite Oxidase
7
5.2.2. Assimilatory Nitrate Reductase
7
5.3. The Prokaryotic Oxotransferases
5.3.1.
8
DMSO Reductase
9
5.3.2. Trimethylamine N-Oxide Reductase
9
5.3.3. Nitrate Reductase
10
5.3.4. Formate Dehydrogenase
11
5.3.5. Pyrogallol-phloroglucinol Transhydroxylase
12
5.3.6. Arsenite Oxidase
12
6. Classification and structure of the Tungsten enzymes
6.1. The Aldehyde Ferredoxin Oxidoreductase Family
12
13
6.1.1. Aldehyde Ferredoxin Oxidoreductase
13
6.1.2. Formaldehyde Ferredoxin Oxidoreductase
14
6.1.3. Glyceraldehyde-3-Phosphate Ferredoxin Oxidoreductase
14
6.1.4. Carboxylic Acid Reductase
15
6.1.5. Aldehyde Dehydrogenase
15
6.2. The Formate/ Formyl Methanofuran Dehydrogenase Family
15
6.2.1. Formate Dehydrogenase
16
6.2.2. Formyl Methanofuran Dehydrogenase
16
6.3. The Acetylene Hydratase Family
16
Table of Contents
7. Tungsten-substituted Molybdenum enzymes
17
7.1. Xanthine Oxidase and Sulfite Oxidase
17
7.2. Trimethylamine Oxide Reductase
17
7.3. Dimethylsulfoxide Reductase
17
7.4. Nitrate Reductases
18
8. Molybdenum-substituted Tungsten enzyme
18
Chapter 2: Tungsten dependent Nitrate Reductase
19
1.
Introduction
19
2. Computational detail
21
3. Active site models
21
4. Results
23
5. Discussion
26
Chapter 3: Ethylbenzene Dehydrogenase
32
1. Introduction
32
2. Computational detail
37
3. Active site models
38
4. Results
39
5. Discussion
62
6. Conclusion
69
Chapter 4: Acetylene Hydratase
71
1. Introduction
71
2. Computational detail
76
3. Active site models
78
4. Results
82
5. Discussion
Chapter 5: Selenate Reductase
115
120
1. Introduction
120
2. Project I
124
2.1.
Computational detail
125
2.2.
Active site model
125
Table of Contents
2.3.
Results
126
2.4.
Discussion
145
3. Project II
149
3.1.
Computational detail
150
3.2.
Active site model
150
3.3.
Results
151
3.4.
Discussion
165
References
168
Summary
Summary
Molybdenum and tungsten active site model complexes, derived from the protein X-ray
crystal structure of the first W-containing nitrate reductase isolated from Pyrobaculum
aerophilum,
were
computed
for
nitrate
reduction
at
the
COSMO-
B3LYP/SDDp//B3LYP/Lanl2DZ(p) level of density functional theory (DFT). The
molybdenum containing active site model complex has a considerably larger activation
energy (34.4 kcal/mol) for the oxygen atom transfer from the nitrate to the metal center as
compared to the tungsten containing active site model complex (12.0 kcal/mol). Oxidation of
the educt complex is close to thermoneutral (-1.9 kcal/mol) for the Mo active site model
complex but strongly exothermic (-34.7 kcal/mol) for the W containing active site model
complex. The low relative energy for the oxidized W metal complex makes the regeneration
of the +IV oxidation state much more difficult as compared to the Mo metal complex. The
MVI to MIV reduction requires much more reductive power (more negative redox potential)
when the metal center M is a tungsten rather than a molybdenum atom. So, although the
reduction of nitrate is stimulated when W replaces Mo in the active site of Nar the catalytic
cycle breaks after the reduction of nitrate to nitrite when the biochemical reducer is not strong
enough to reduce the metal center.
Ethylbenzene dehydrogenase (EBDH) is an enzyme that catalyzes the oxygen-independent,
stereospecific hydroxylation of ethylbenzene to (S)-1-phenylethanol. EBDH active site
models, derived from protein X-ray crystal structure, were computed at the COSMOB3LYP/SDDp//B3LYP/Lanl2DZ(p) energy level of DFT in order to investigate most
probable mechanism, ionic or radical pathway. In addition, different protonation states and
participation of amino acid residues near to the Mo center were considered. Models with
protonation of His192, Lys450, Asp223 and model without protonation were investigated for
comparison. Computed relative energies indicate that the overall lowest energy barrier
pathway results when ionic and radical pathways are mixed. This mechanism of ethylbenzene
hydroxylation starts with a homolytic C1-Hs bond cleavage (TS1’) resulting in the formation
of a radical type intermediate (I’) and then in order to continue the reaction by the easier O1Hs
anion transfer, an electron needs to be transferred from the substrate to the Mo-OH moiety to
transform the di-radical to the zwitter ionic intermediate. Then the transfer of O1Hs anion
from the Mo to the cationic substrate (TS2) results in the formation of product bound
complex (P). Among those the protonated Lys site corresponds to the energetically best
pathway for the hydroxylation of ethylbenzene by EBDH.
Summary
Acetylene hydratase (AH) of Pelobacter acetylenicus is a tungsten (W) containing iron-sulfur
enzyme that catalyzes the transformation of acetylene to acetaldehyde. DFT studies were
performed on the model complexes derived from the native protein X-ray crystal structure of
AH. Based on the computational results we proposed the most likely nucleophilic mechanism
for the hydration of acetylene by the acetylene hydratase (AH) enzyme. In this mechanism,
the water (Wat1424) molecule is coordinated to the W center and Asp13 is assumed to be in
anionic form. The Wat1424 molecule is activated by W and then donates one of its proton to
the anionic Asp13 forming the W-bound hydroxide and protonated Asp13. The W-bound
hydroxide then attacks the C1 atom of acetylene together with the transfer of proton from the
Asp13 to its C2 atom, resulting in the formation of a vinyl alcohol intermediate complex. The
energy barrier associated with this step is 14.4 kcal/mol. The final, rate limiting, step
corresponds to the tautomerization of the vinyl alcohol intermediate to acetaldehyde via
intermolecular assistance of two water molecules, associated with the energy barrier of 18.9
kcal/mol. An alternative, electrophilc pathway, was also considered but the energy barriers are
found to be higher than for the nucleophilic pathway described here.
Sulfite oxidase (SO), selenate reductase (SeR) and nitrate reductases (NRs) are among the
mononuclear molybdenum enzymes involved in the catalysis of metabolic redox reactions.
The active site composition of SO has one molybdopterin (MPT) ligand and it oxidizes the
sulfite to sulfate, SeR has two MPT ligands and it reduces the selenate to selenite, while NRs
reduces nitrate to nitrite by either one or with two MPT’s at the active site. Is the active site
itself special in some way for the oxidation/reduction of one or the other substrate? Or do the
different active sites behave essentially the same way and it is the role of the protein to make
it specific. To clarify these, DFT studies were performed on the computational model
complex, [MoVIO2(S2C2Me2)SMe]- (A, derived from the X-ray crystal structure of native SO),
and on the experimental model complex [MoVIO2(mnt)2]2- (B, coordination mode similar to
the active site of SeR) for the oxidation of selenite and sulfite. For the oxidation of sulfite
model A which resembles the SO active site is clearly the best choice (lowest barrier, minor
exothermicity). For the reduction of selenate a smaller activation is computed for model A,
but the reaction is less exothermic with model B, which resembles the SeR active site.
DFT computations were also carried out on simple active site model complexes of SeR to
investigate different ways of binding the substrate and the OAT reaction. Unfortunately, the
results are little conclusive. Larger models might be needed to obtain more meaningful
computational results.
Zusammenfassung
Zusammenfassung
Ausgehend von der Protein-Röntgenbeugungs-Kristallstruktur der ersten wolframhaltigen
Nitratreduktase, die aus Pyrobaculum aerophilum isoliert wurde, wurden molybdän- und
wolframhaltige Modellkomplexe für das aktive Zentrum auf dem COSMO-B3LYP/SDDp//
B3LYP/Lanl2DZ(p) Dichtefunktionaltheorie(DFT)-Niveau hinsichtlich der Reduktion von
Nitrat berechnet. Der Mo-haltige Modellkomplex besitzt eine wesentlich größere
Aktivierungsenergie (34.4 kcal/mol) für den Sauerstoffatomtransfer (OAT) von Nitrat auf das
Metallzentrum als das W-Analogon (12.0 kcal/mol). Die Oxidation des Eduktkomplexes ist
nahezu thermoneutral (-1.9 kcal/mol) für den molybdänhaltigen Modellkomplexes, aber
deutlich exotherm (-34.7 kcal/mol) im Falle des Wolframkomplexes. Die vergleichsweise
niedrige relative Energie des oxidierten Wolframkomplexes erschwert die Regeneration der
Oxidationsstufe +IV deutlich im Vergleich zum Molybdänkomplex. Die Reduktion von MVI
zu MIV erfordert also eine höhere Reduktionskraft (negativeres Redoxpotenzial), wenn das
Metallzentrum M aus einem Wolfram- statt einem Molybdänatom besteht. Obwohl die
Reduktion von Nitrat gefördert wird, wenn W das Mo im aktiven Zentrum ersetzt, wird der
katalytische Zyklus nach der ersten Reduktion eines Nitrations unterbrochen, wenn das
biochemische Reduktionsmittel nicht stark genug ist, das Metallzentrum zu reduzieren.
Das Enzym Ethylbenzoldehydrogenase (EBDH) katalysiert die sauerstoffunabhängige
stereospezifische Hydroxylierung von Ethylbenzol zu (S)-1-Phenylethanol. Modellkomplexe
für das aktive Zentrum von EBDH, abgeleitet aus der Protein-RöntgenbeugungsKristallstruktur, wurden mit dem COSMO-B3LYP/SDDp//B3LYP/Lanl2DZ(p) DFT-Niveau
berechnet um den wahrscheinlichsten Mechanismus aufzufinden. Sowohl ionische als auch
radikalische Reaktionspfade wurden in Betracht gezogen und zwar für verschiedene
Protonierungszustände von Aminosäureresten, die wegen ihrer räumlichen Nähe zum
Molybdänzentrum an der Reaktion teilhaben könnten.
Modelle mit
protonierten
Aminosäureresten von His192, Lys450 und Asp223 sowie ein Modell ohne Protonierung wurden
vergleichend untersucht. Die berechneten relativen Energien weisen darauf hin, dass der
Reaktionspfad mit den insgesamt geringsten Barrieren aus einer Mischung von ionischen und
radikalischen Reaktionsschritten resultiert. Dabei wird die Hydroxylierung von Ethylbenzol
durch einen homolytischen C1-Hs Bindungsbruch (TS1‘) eingeleitet, was zu einem
diradikalischen Intermediat (I‘) führt. Um dann den leichteren Transfer eines OH Anions zu
ermöglichen, müsste zunächst ein Elektron vom Substrat auf die Mo-OH Einheit übertragen
werden, was aus dem diradikalischen das zwitterionische Intermediat werden lässt. Die
Zusammenfassung
Übertragung des O1Hs Anions vom Molybdän auf das kationische Substrat (TS2) lässt den
Produktkomplex (P) entstehen. Der energetisch günstigste Reaktionspfad für die
Hydroxylierung von Ethylbenzol durch EBDH ergibt sich dabei für die protonierte
Lysinseitenkette.
Acetylenhydratase (AH) aus Pelobacter acetylenicus ist ein wolframhaltiges Eisen-SchwefelEnzym, das die Umsetzung von Acetylen zu Acetaldehyd katalysiert. Modellkomplexe, die
aus der nativen AH-Proteinkristallstruktur abgeleitet wurden, wurden mit DFT-Methoden
untersucht. Basierend auf den Rechenergebnissen wird ein nukleophiler Mechanismus als der
wahrscheinlichste für die Hydratisierung von Acetylen durch AH vorgeschlagen. Gemäß
diesem Mechanismus ist ein Wassermolekül (Wat1424) an das W-Zentrum koordiniert und
Asp13 wird als ionisiert angenommen. Das Wassermolekül Wat1424 ist durch das W aktiviert
und gibt so ein Proton an die anionische Gruppe Asp13 ab, was zu einem W-gebundenen
Hydroxid und protoniertem Asp13 führt. Das W-gebundene Hydroxid greift das C1-Atom des
Acetylens an und damit gekoppelt erfolgt der Protonentransfer von Asp13 auf das C2-Atom.
Dadurch bildet sich ein intermediärer Vinylalkohol-Komplex. Die Energiebarriere für diesen
Schritt beträgt 14.4 kcal/mol. Den abschließenden, geschwindigkeitsbestimmenden Schritt
stellt die Tautomerisierung des Vinylalkohols zu Acetaldehyd dar. Dafür wird unter
Einbeziehung von zwei Wassermolekülen eine Barriere von 18.9 kcal/mol berechnet. Für
einen alternativen, elektrophilen Reaktionspfad wurden höhere Barrieren als für den hier
beschriebenen nukleophilen berechnet.
Sulfit Oxidase (SO), Selenat Reduktase (SeR) und Nitrat Reduktasen (NRs) gehören zu den
mononuklearen Molybdoenzymen, die metabolische Redoxreaktionen katalysieren. Das SO
aktive Zentrum besitzt einen Molybdopterinliganden (MPT) und oxidiert Sulfit zu Sulfat; SeR
mit zwei MPT reduziert Selenat zu Selenit; NRs schließlich können Nitrat zu Nitrit
reduzieren und besitzen entweder einen oder zweit MPT-Liganden im aktiven Zentrum. Ist
das aktive Zentrum in irgendeiner Art spezifisch für die Oxidation bzw. die Reduktion des
einen oder anderen Substrates? Oder verhalten sich die verschiedenartigen Aktiven Zentren
im Wesentlichen gleich und erst das Protein schafft Spezifität? Zur Klärung dieser Fragen
wurde die Selenit- und die Sulfitoxidation durch Modellkomplexe mit DFT rechnerisch
untersucht: [MoVIO2(S2C2Me2)SMe]- (A, abgeleitet von der Röntgenkristallstruktur der
nativen SO) und der auch experimentell untersuchte Modellkomplex [MoIVO(mnt)2]2- (B, mit
einem der SeR ähnlichen Koordinationsmuster). Für die Oxidation von Sulfit stellt Modell A
tatsächlich die beste Wahl dar (geringste Barriere, geringfügig exotherme Reaktionsenergie).
Zusammenfassung
Für die Reduktion von Selenat hingegen wird zwar mit Modell A eine geringere Barriere
berechnet, aber die Reaktion verläuft mit Modell B weniger stark exotherm.
Weitere DFT Rechnungen wurden an einfachen Modellkomplexen zum aktiven Zentrum der
SeR
unternommen,
um
verschiedene
Bindungsmodi
des
Substrats
und
die
Sauerstoffübertragung zu untersuchen. Die Ergebnisse sind leider wenig aufschlussreich.
Größere Modelle dürften nötig sein um sinnvollere Rechenergebnisse zu erhalten.
Chapter 1-Introduction
Introduction
1) Enzymes
Almost all processes in the biological cell need enzymes to occur at significant rates.
E+S
Substrate binding
ES
Catalytic Step
É + P (1.1)
Eq. 1.1: Simple mechanism for the enzyme (E) substrate (S) reaction, É = reduced enzyme,
P = product.
Enzymes are proteins that catalyze chemical reactions. Metals occur as natural constituents of
these proteins. Enzymes vary in their size and properties. Most enzymes are much larger than
the substrates they act on, and only a small portion of the enzyme (around 3–4 amino acids) is
directly involved in catalysis. Approximately one-third of all proteins and enzymes purified to
apparent homogeneity require metal ions as cofactors for biological function. They are called
metalloproteins. Metalloproteins when perform catalytic function are called metalloenzymes
and metal ions in these enzymes are usually part of the active sites. Active site is the region
that is involved in catalysis contains the catalytic residues, binds the substrate, and then
carries out the reaction.1,2
2) Chemical properties of Molybdenum and Tungsten
Molybdenum and tungsten are the only 4d (Mo) and 5d (W) transition metals (of group 6) that
are involved in the metabolism of biological systems. The atomic radii and electron affinity
for molybdenum and tungsten are almost the same because of the lanthanide contraction. The
coordination chemistry of both the elements is comparable. Both form complexes in oxidation
state ranges from –II to +VI, of which +IV, +V, and +VI oxidation states are relevant in
biological systems (Table 1.1).3
Despite of their chemical properties, their geochemistry in the Earth´s crust is quite different.
Molybdenum occurs as molybdenite, MoS2, whereas tungsten occurs as scheelite, CaWO4,
and wolframite, (Fe, Mn)WO4 in the Earth´s crust. The reduced tungstenite (WS2) is very rare
because WS2 is readily solubilized. 4
WS2 + 4H2O
WO42- + 2H2S + 4H+ + 2e- (1.2)
Primary mechanism for the uptake and transport of both molybdenum and tungsten involves
the binding of [MO4]2- tetrahedral anions in a cavity of suitable size by hydrogen bonds from
1
Chapter 1-Introduction
the polypeptide chain. Because of the same size of [MoO4]2- and [WO4]2-, the transport
proteins do not discriminate significantly between the two [MO4]2- anions. 5,6
Table 1.1: Physical and Chemical properties of molybdenum and tungsten.
Properties
Molybdenum (Mo)
Tungsten (W)
Atomic number
42
74
Average atomic weight
95.94
183.85
Electronic configuration of outer shell
4d5 5s1
4f14 5d4 6s2
Atomic radii (Ǻ)
1.40
1.40
Ionic radii for +IV oxidation state (Ǻ)
0.65
0.66
Ionic radii for +V oxidation state (Ǻ)
0.61
0.62
Ionic radii for +VI oxidation state (Ǻ)
0.59
0.60
Electronegativity
1.8
1.7
pKa of Oxo acid (MO42-/HMO4-)
3.87
4.60
Concentration in sea water
~100 nM
~ 1 pM
Concentration in fresh water
~ 5-50 nM
~ 500 pM
M=O bond length (Ǻ)
1.76
1.76
3) Biological distribution of Molybdenum and Tungsten enzymes
Enzymes containing molybdenum or tungsten at their active sites appear to be present in all
forms of life, from ancient archaea to man. These enzymes catalyze a wide range of reactions
in carbon, sulfur, and nitrogen metabolism. 7,8,9 In general, these enzymes utilize water as the
ultimate source or sink of oxygen in the overall catalytic reaction. The reactions are coupled
to electron transfer between substrate X/XO, a Fe-S center, heme, or flavin.
X + H2O
XO + 2H+ + 2e- (1.3)
The enzymes are referred to as oxotransferases when the substrate is transformed by primary
oxygen atom transfer and as hydroxylases when bound or unbound water or hydroxide is
directly involved in the substrate transformation reaction. 10
3.1 Molybdenum Enzymes
Molybdenum enzymes have been known for more than 75 years and are found in all types of
living systems.9,11,12 Molybdenum is widely available to biological systems due to the
2
Chapter 1-Introduction
solubility of its high-valent oxides in water. It is found in two basic forms: as an integral
component of the multinuclear M center of nitrogenases13,14 and as the mononuclear active
sites of a much more diverse group of enzymes that in general function catalytically to
transfer an oxygen atom either to or from a physiological acceptor/donor molecule. It is on the
basis of this commonly encountered aspect of catalysis that these enzymes are frequently
referred to as oxotransferases.10 These enzymes catalyze a variety of important reactions in
the metabolism of nitrogen and sulfur containing compounds and also of various carbonyl
compounds (e.g., aldehydes, formate, CO and CO2).
In plants, e.g., nitrate reductases
catalyze the first and rate-limiting step in the assimilation of nitrate from the soil, 15 and an
indole-3-acetaldehyde oxidase is responsible for the final step in the biosynthesis of the
hormone indole-3-acetate.16 In bacteria and archaea, these enzymes catalyze the
hydroxylation of a carbon center as the first step in the breakdown of these
compounds.17,18,19,20 In addition, fish bacterial enzymes, such as trimethylamine-N-oxide
(TMAO) reductase and dimethylsulfoxide (DMSO) reductase from algae catalyze terminal
oxidation reactions under a variety of conditions. 21 In humans, xanthine oxidoreductase
catalyzes the final two steps in purine metabolism: sequential hydroxylation of hypoxanthine
to xanthine and then uric acid.22 The closely related mammalian aldehyde oxidases are
important in the metabolism of a variety of aldehyde compounds and have implications
specifically in the biosynthesis of retinoic acid, both in the retina and in the developing
nervous system. 23 Finally, sulfite oxidase is responsible for the final step of sulfur metabolism
in humans: oxidation of sulfite to sulfate.24
3.2 Tungsten Enzymes
In contrast to molybdenum, evidence for the involvement of tungsten in biological systems
has been obtained not more than 25 years ago.25,26 Initial studies showed that the production
and activity of formate dehydrogenase, which catalyzes the first step in the reduction of CO2
to acetate, was significantly higher in acetogens if the bacteria were grown in the presence of
[WO4]2- rather than [MoO4]2-.27,28 The first tungsten containing enzyme was purified and
characterized, from an acetogen, a bacterium that grows by producing acetate.29 Tungsten is
important for the metabolism of hyperthermophilic archaea that live in volcanic vents on the
sea bed at > 100°C. 30 Although more than a dozen tungstoenzymes have been isolated and
characterized from bacteria and archaea, it has yet to be isolated from higher organism.
3
Chapter 1-Introduction
4) Pyranopterin cofactor
An organic pterin cofactor, commonly called Moco, is present in all mononuclear
molybdenum and tungsten enzymes. In the literature it is referred to as molybdopterin for
molybdenum enzymes and pyranopterin for tungsten enzymes. This cofactor does not
participate directly in catalysis but it appears to be involved in electron transfer from the
metal center once it has been reduced by a substrate.
The pterin structure of Moco is unique in nature and has probably been evolved in order to
control and maintain the special redox properties of Mo. The task of cofactor is to position the
catalytic metal, Mo, correctly within the active site, to control its redox behavior and to
participate with its pterin ring system in the electron transfer to or from the Mo atom. The
pterin with its several possible reduction states as well as different structural conformations
could also be important for channeling electrons to other prosthetic groups.31,32 X-ray
crystallographic analyses of Mo-enzymes revealed that the cofactor is not located on the
surface of the protein, but it is buried deeply within the interior of the enzyme and a tunnellike structure makes it accessible to the appropriate substrates.31,32,33
Moco containing enzymes can be distinguished according to their amino acid sequences,
spectroscopic properties, active site structures and catalyzed reactions. Eukaryotic
molybdenum enzymes usually contain only one pterin as cofactor, whereas prokaryotic
enzymes can contain different cofactors consisting of one or two pterins and a nucleotide
covalently linked to the pterin (Fig 1.1). It coordinates to the metal via the ene-di-thiolate
moiety.7
4
Chapter 1-Introduction
Fig. 1.1: Cofactors of molybdenum and tungsten enzymes, (A) Molybdenum cofactor called
molybdopterin (Moco)34, (B) Molybdopterin guanosine dinucleotide (MGD)35, (C) Extended
molybdenum cofactor (bis-MGD)36
5) Classification and structure of the Molybdenum enzymes
According to the molybdenum cofactors, molybdenum enzymes are classified into three
families (Fig. 1.2), named
5.1
The Molybdenum Hydroxylases (Xanthine Oxidase family)
5.2
The Eukaryotic Oxotransferases (Sulfite Oxidase family)
5.3
The Prokaryotic Oxotransferases (DMSO Reductase family)
Fig. 1.2: Active site composition of major families of mononuclear molybdenum enzymes
5
Chapter 1-Introduction
5.1 The Molybdenum hydroxylases
The molybdenum hydroxylases constitute the largest group of mononuclear molybdenum
enzymes. More than 20 enzymes have been characterized to varying degree ranging from the
xanthine and aldehyde oxidoreductases from higher organisms to bacterial enzymes
responsible for the hydroxylation of a diverse range of aromatic heterocycles.9
These enzymes contain the molybdenum cofactor, Moco, in which molybdenum is associated
with a pterin derivative, called molybdopterin. The active sites of these enzymes contain a
molybdenum atom coordinated with two dithiolene sulfur atoms of the molybdopterin unit,
terminal oxygen, sulfur atom and terminal hydroxide/water molecule (Fig. 1.2).37 These Moco
containing enzymes participate in hydroxylation and oxo-transfer reactions that are two
electron transfer processes occurring at the Moco site. Water is the ultimate source of the
oxygen atom incorporated into the substrate, and reducing equivalents are generated in this
process.38 Moco enzymes usually contain additional redox-active cofactors, such as ironsulphur clusters, flavins or haem centers, which mediate electron transfer from Moco to the
final electron acceptor.39
The properties of these enzymes are closely related.40,41,42 They play important roles in the
metabolism of exogenous compounds. These flavoenzymes generally catalyze nucleophilic
oxidation of N-heterocycles, resulting in metabolites different from those obtained via
electrophilic oxidation by the cytochrome P450 system. However, the substrate specification
among these enzymes differs.43
5.1.1. Aldehyde Oxidase
5.1.2. Xanthine Oxidase / Dehydrogenase
5.1.1. Aldehyde Oxidase
Aldehyde oxidase belongs to a large family of molybdenum enzymes whose members
catalyze the oxidative hydroxylation of a diverse range of aldehydes and nitrogen-containing
aromatic heterocycles, isoxazole, and isothiazole in reactions that necessarily involve the
cleavage of a C-H bond. These enzymes are properly considered hydroxylases, 44 although
product tautomerization in reactions involving heterocyclic substrates usually results in the
keto rather than enol form predominating in aqueous solution.
RCHO + H2O → RCOOH + 2H+ + 2e- (1.4)
6
Chapter 1-Introduction
The aldehyde oxidoreductase from Desulfovibrio gigas was the first mononuclear enzyme for
which an X-ray crystal structure was reported, at 2.25 Ǻ resolution45 (later refined to 1.8 Ǻ
resolution46). The pterin cofactor in this enzyme is a molybdopterin cytosine dinucleotide. It
has pair of 2Fe-2S centers but lacks the flavin domain. This crystal structure indicates the
presence of a glycine residue, Gly623, in the molybdenum-binding portion of the protein that is
conserved among the hydroxylase family of mononuclear molybdenum enzymes.45
5.1.2.
Xanthine Oxidase or Xanthine Dehydrogenase
Xanthine oxidase and xanthine dehydrogenase are two forms of the same enzyme, differing in
the co-substrate for the reaction: xanthine oxidase utilizes molecular oxygen whereas xanthine
dehydrogenase utilizes NAD+. These enzymes play an important role in the catabolism of
purines in some species, including humans.42, 47
Xanthine oxidase is a homodimer that possesses FAD (flavin adenine dinucleotide) in
addition to the molybdenum cofactor, Moco and as additional redox center a pair of 2Fe-2S
centers. Members of the xanthine oxidase family generally catalyze hydroxylation reactions
of the type
RH + H2O → ROH + 2H+ + 2e- (1.5)
This stoichiometry is unique among biological systems catalyzing hydroxylation reactions as
reducing equivalents are generated rather than consumed in the course of the reaction, and
water48 rather than dioxygen is utilized as the ultimate source of the oxygen atom
incorporated into the substrate. The overall reaction mechanism of these and related enzymes
is typically broken down into reductive and oxidative half-reactions of the catalytic cycle,
defined from the standpoint of the enzyme. It is the reductive half-reaction in which the
molybdenum center participates, with the metal becoming reduced from Mo (VI) to Mo (IV).9
Re-oxidation of the molybdenum center takes place via simple electron transfer to the other
redox-active centers of the enzyme, ultimately to the FAD where electrons are removed from
the enzyme by reaction with dioxygen.49,50,51,52
5.2 The Eukaryotic Oxotransferases
These enzymes also contain the molybdenum cofactor, Moco. The active sites of these
enzymes are composed of a molybdenum atom coordinated with two dithiolene sulfur atoms
of the molybdopterin unit, a sulfur atom of cysteinate and two terminal oxygen atoms. These
7
Chapter 1-Introduction
enzymes catalyze oxygen atom transfer reactions to or from an available electron lone pair of
substrate and can itself be subdivided into two families.
5.2.1.
Sulfite Oxidase
5.2.2.
Assimilatory nitrate reductase
5.2.1.
Sulfite Oxidase
Sulfite oxidase possesses another redox-active center in addition to the molybdenum: a b-type
cytochrome. It catalyzes the oxidation of sulfite to sulfate53,54 which is the terminal reaction in
the oxidative degradation of the sulfur-containing amino acids cysteine and methionine.
SO32- + H2O → SO42- + 2H+ + 2e- (1.6)
Oxidation of sulfite by the enzyme is efficient even at very low concentration of substrate and
occurs in the presence of electron acceptors other than oxygen. 55 The conversion of sulfite to
sulfate is a two-electron oxidation, whereas cytochrome c, the physiological electron acceptor
for sulfite oxidase, is a one electron acceptor. Sulfite oxidation occurs at the molybdenum
center and cytochrome c reduction at the heme site.
The enzyme also plays an important role in detoxifying exogenously supplied sulfite and
sulfur dioxide. In humans, the deficiency of sulfite oxidase may lead to neurological
problems, mental retardation, and dislocation of the ocular lens.
5.2.2. Assimilatory Nitrate Reductase
Assimilatory nitrate reductases also possess redox-active centers in addition to the
molybdenum: a b-type cytochrome and FAD. These enzymes have been isolated from fungi,
algae, and higher plants. They catalyzes a redox reaction involving an electron transport chain
at the two active sites, i) the cytochrome b reductase fragment contains the active site where
NAD(P)H transfer electrons to FAD and these electrons are transferred to the Mo-MPT by the
internal cytochrome b, ii) the nitrate reductase, Mo-MPT active site, to reduce nitrate to
nitrite.56,57 The overall reaction is
NO3- + NADH → NO2- + NAD+ + OH- (1.7)
This is the first and the rate-limiting step in nitrogen assimilation in higher plants, the uptake
of nitrate, a light dependent phenomenon. This metabolic role is different from that of the
dissimilatory nitrate reductase of bacteria. 58
8
Chapter 1-Introduction
5.3 The Prokaryotic Oxotransferases
The prokaryotic oxo-transferases family, also called DMSO reductase family, is a diverse
group of enzymes that catalyzes either oxygen atom transfer or other redox reactions. The
active site of these enzymes consists of a molybdenum atom coordinated by a pair of
dithiolene ligands, a terminal oxygen atom and one amino acid ligand (oxygen atom of
serinate or oxygen atom of aspartate or sulfur atom of cysteine or selenium atom of cysteine
or hydroxide, depending on the surrounding polypeptide chains). These enzymes are found as
integral components of multisubunit membrane-bound proteins that possess additional redoxactive centers in other subunits. They are particularly important in the anaerobic respiration
including the dissimilatory reduction of certain toxic oxo-anions.59
Members of this family so far have been divided into:
5.3.1. DMSO Reductase (DMSOR)
5.3.2. Trimethylamine N-Oxide Reductase (TMAOR)
5.3.3. Nitrate Reductase (NR)
5.3.4. Formate Dehydrogenase (FDH)
5.3.5. Pyrogallol-phloroglucinol transhydroxylase (TH)
5.3.6. Arsenite Oxidase (AO)
Fig. 1.3: Active site composition of the members of prokaryotic oxotransferases.
5.3.1. DMSO Reductase (DMSOR)
DMSO reductase, isolated from Rhodobacter species, is linked to the respiratory chain of the
organism. The sixth coordination site of molybdenum is occupied by the oxygen of a serinate
from the polypeptide (Fig. 1.3). This enzyme has no other redox-active center other than
molybdenum while the DMSO reductase from E.coli is a multi-subunit protein.21 Other than
molybdenum subunit it possesses four 4Fe/4S iron-sulfur clusters, and a membrane anchor
which lacks redox-active centers but has a quinol binding site.60
9
Chapter 1-Introduction
DMSO reductase is an oxotransferase that catalyzes the reductive deoxygenation of dimethyl
sulfoxide (DMSO) to dimethyl sulfide (DMS) in a two stage reaction. In the first stage,
DMSO binds to reduced (MoIV) enzyme, which is oxidized to MoVI with an extra oxygen
ligand and DMS is released. In the second stage, the reduced enzyme is regenerated by
transfer of two electrons from a specific cytochrome, resulting in the release of oxygen as a
water molecule.61 The overall reaction is:
Me2SO + 2e- + 2H+
Me2S + H2O (1.8)
5.3.2. Trimethylamine N-Oxide Reductase (TMAO)
Trimethylamine N-oxide, from E.coli, is a periplasmic molybdoenzyme that catalyzes the
reduction of trimethylamine N-oxide (TMAO) to trimethylamine (TMA) during an anaerobic
respiratory process.
(CH3)3NO + 2H+ + 2e- → (CH3)3N + H2O (1.9)
The properties of TMAO reductases have been studied in several organisms. The active site
structure is very similar to the DMSO reductase. Based on their substrate specificity, these
enzymes can be divided into two groups, the TMAO reductases which have high substrate
specificity and DMSO/TMAO reductases which can reduce a broad range of N and S-oxide
substrates. The TMAO reductase from E.coli, Shewanelly putrefaciens and Roseobacter
denitrificans are unable to reduce S-oxide compounds and belongs to the first group. In the
second group, the constitutive DMSO from E.coli and Proteus vulgaris and the DMSO
reductases from Rhodobacter capsulatus or R. sphaeroides can reduce TMAO as well as other
N and S-oxides. Except the constitutive DMSO reductases from E. coli and P. vulgaris which
are membrane bound all these molybdoenzymes are located in the periplasm and induced by
TMAO.62
TMAOR, from Escherichia coli, Shewanelly massilia, Salmonella typhimurium and
Roseobacter denitrificans, possess oxygen atom of serine at the sixth coordination site of
molybdenum (Fig. 1.3). It also possesses b-type cytochrome as a redox-active center.62
5.3.3. Nitrate Reductase (NR)
Nitrate reductases, molybdo-iron-flavoproteins, are playing key roles in the first step of
ammonification, denitrification and dissimilatory ammonification. They catalyze the
reduction of nitrate to nitrite, by transferring an oxygen atom from nitrate to molybdenum.
Two electrons and two protons are consumed in the process, and in the end a water molecule
10
Chapter 1-Introduction
is produced. Electrons are transferred to the active site of NR by the b-type cytochrome and
an iron-sulfur containing subunit of the nitrate reductase.
NO3- + 2H+ + 2e- → NO2- + H2O (1.10)
This reduction of nitrate is performed by the organisms for three principle reasons: to
incorporate nitrogen into biomolecules, to generate energy for cellular function, and to
dissipate extra energy by respiration.63,64 Although the chemical reaction is always the
reduction of nitrate to nitrite, the prokaryotic nitrate reductases have been classified into three
groups, assimilatory nitrate reductase (NAS), respiratory nitrate reductase (NAR) and
periplasmic nitrate reductase (NAP). NAR and NAP are linked to respiratory electron
transport systems and are located in the membrane and periplasm, respectively. NAS is
located in the cytoplasm. 65
These nitrate reductases from a variety of bacterial and archaeal sources are more closely
related to DMSO reductase. In the NAP, dissimilatory nitrate reductase from Desulfovibrio
desulfuricans (DdNapA), the sixth coordination site is occupied by a sulfur atom of cysteine
(Fig 1.4). It possesses a 4Fe/4S iron-sulfur cluster as a redox-active center in addition to
molybdenum in the same polypeptide chain. For NAR, respiratory nitrate reductase from
E.coli (Ec NarGHI, Ec NarGH), two structures were reported (Fig 1.4), where NarG
represents the molybdenum catalytic active site of the protein. The sixth coordination site of
molybdenum is occupied by an oxygen atom of aspartate (Asp222). In the NarGHI, Asp222 acts
as a bidentate ligand, but in NarGH, where an additional oxo group is attached with
molybdenum, it acts as a monodentate ligand.66 The two E.coli nitrate reductases (NAR) also
possess further redox-active sites, three 4Fe-4S, one 3Fe-4S and the membrane anchor subunit
a pair of b-type cytochromes. 67 Unlike eukaryotic nitrate reductase, no heme or flavin adenine
dinucleotide (FAD) is present in the active subunit of NAS.68
Fig. 1.4: Active site composition of nitrate reductases.
11
Chapter 1-Introduction
5.3.4. Formate Dehydrogenase (FDH)
Formate dehydrogenase, a membrane bound enzyme, catalyzes the oxidation of formate to
carbon dioxide as the first step in acetogenic glucose fermentation.69
HCOO- → CO2 + H+ + 2e- (1.11)
There are two molybdenum containing FDH from Escherichia coli, the FDH-H which is one
of the components of the formate hydrogen lyase complex in E.coli70 and FDH-N, a
membrane bound protein, which is supposed to participate in the redox loop of the proton
motive force generation across the membrane cell. 71 FDH from Desulfovibrio desulfuricans is
also a molybdenum containing enzyme.72,73 Desulfovibrio alaskensis has been shown to
produce two active isoforms, W-FDH and Mo-FDH, each containing either tungsten or
molybdenum as an active metal. The two isoforms have similar molecular properties and
cannot be chromatographically separated, which suggests they have the same isoelectronic
point.74
Selenocysteinate is present as the sixth ligand attached to the molybdenum in most of the
FDHs but there is an exception where, instead of selenocysteinate, one hydroxyl or sulfide
ligand is present (Fig 1.3).69
Instead of oxygen atom transfer from the molybdenum to the substrate, the current proposed
mechanism for these enzymes suggests a replacement of the hydroxyl group bound to the
molybdenum by one of the oxygen atoms from formate, resulting in the reduction of MoVI to
MoIV and the cleavage of C-H bond.70
5.3.5. Pyrogallol-phloroglucinol Transhydroxylase (TH)
The Mo containing transhydroxylase from the anaerobic microorganism Pelobacter
acidigallici is a cytoplasmic enzyme. It catalyzes the conversion of pyrogallol to
phloroglucinol by the transfer of hydroxyl group in the absence of oxygen.
C6H6O3 (1, 2, 3-trihydroxybenzene) → C6H6O3 (1, 3, 5-trihydroxybenzene) (1.12)
It is a non-redox reaction. The active site structure is similar to the DMSO reductase (Fig 1.3).
Although there are additional three iron-sulfur redox-active centers, [4Fe-4S], their role is still
not clear.75
5.3.6. Arsenite Oxidase
Arsenite Oxidase from Alcaligenes faecalis, a molybdo-iron enzyme, is involved in the
detoxification of arsenic. It catalyzes the oxidation of arsenite to arsenate.
AsIIIO2 - + 2H2O → AsVO43- + 4H+ + 2e- (1.13)
12
Chapter 1-Introduction
Like for the other subgroups of the prokaryotic oxidoreductase family, the molybdenum atom
in A. faecalis arsenite oxidase is not coordinated by the side chain of a serine, cysteine, or
selenocysteine residue (Fig 1.3). The corresponding residue is Alanine199 (Ala199) but there is
no covalent linkage between the molybdenum atom and the polypeptide chain. As a
consequence, the loop containing this residue is folded away from the Mo site and the active
site is significantly more exposed than in the other subgroups. It also possesses two redox
active sites, iron-sulfur containing [3Fe-4S] and [2Fe-2S] and azurin or cytochrome c.76
6) Classification and structure of Tungsten enzymes
Tungstoproteins are usually found in thermophilic organisms that grow in extreme
environments.77,78 Tungstoenymes are involved in the low redox potential reactions. They can
be classified into three functionally and phylogenetically distinct families:
6.1
The Aldehyde Ferredoxin Oxidoreductase Family (AOR)
6.2
The Formate/ Formyl Methanofuran Dehydrogenase Family (F(M)DH)
6.3
The Acetylene Hydratase Family (AH)
Fig. 1.5: Active site composition of tungsten containing enzymes.
6.1 The Aldehyde Ferredoxin Oxidoreductase Family (AOR)
The majority of the known tungstoenzymes belong to this family. The enzymes in this family
are related phylogenetically and display high sequence similarity at the amino acid level.
However, none of the enzyme of this family shows any sequence similarity to any known
molybdoenzyme. These enzymes catalyze the oxidation of various types of aldehydes to the
corresponding acids and use ferredoxin as redox active site.
RCHO + H2O + 2Fd(ox) → RCOOH + 2H+ + 2Fd(red) (1.14)
Where, RCHO represents the substrate with the aldehyde functional group and Fd represents
the electron acceptor ferredoxin.
13
Chapter 1-Introduction
Enzymes of this family are generally oxygen sensitive and have broad range of substrate
specificities. In these enzymes the tungsten atom coordinates to bis-pterin cofactors, to a
terminal oxygen atom and an amino acid ligand depending on the surrounding polypeptide
chain. It also possesses an electron transferring [4Fe-4S] cluster.79,80
Members of this family have been divided into:
6.1.1. Aldehyde Ferredoxin Oxidoreductase (AOR)
6.1.2. Formaldehyde Ferredoxin Oxidoreductase (FOR)
6.1.3. Glyceraldehyde-3-Phosphate Ferredoxin Oxidoreductase (GAPOR)
6.1.4. Carboxylic Acid Reductase (CAR)
6.1.5. Aldehyde Dehydrogenase (ADH)
6.1.1. Aldehyde Ferredoxin Oxidoreductase (AOR)
Aldehyde ferredoxin oxidoreductase (AOR) is purified from hyperthermophilic archaeon
Pyrococcus furiosus81, Pyrococcus strain ES482 and Thermococcus strain ES1.83 The enzyme
is extremely sensitive to oxygen. It oxidizes a broad range of both aliphatic and aromatic
aldehydes to their corresponding acids. It catalyzes the reversible oxidation of aldehydes to
their corresponding carboxylic acids with the accompanying reduction of the redox protein
ferredoxin, the proposed physiological electron acceptor, Fd (eq. 1.14).
The Fd of P. furiosus contains a single [4Fe-4S] cluster and it undergoes a one-electron redox
reaction. So one catalytic turnover per subunit requires the reduction of two molecules of Fd.
In the active site tungsten is coordinated by two diothiolene side chains from two pterin
cofactors, linked to each other via Mg2+ ion through their terminal phosphate groups. No
protein ligand is coordinated with the tungsten. The [4Fe-4S] cluster is only at a distance of
10Ǻ from the tungsten. It is coordinated by the protein via four cysteine ligands. One of these
cysteine forms a hydrogen bond with a pterin ring nitrogen atom, which indicates that pterin
might have an active role in the redox chemistry of this enzyme. 84,85
6.1.2. Formaldehyde Ferredoxin Oxidoreductase (FOR)
Formate ferredoxin oxidoreductase (FOR) is purified from Thermococcus litoralis86 and
Pyrococcus furious.87 It not only oxidizes formaldehyde (as the name indicates) but it can also
use other short chain (C1-C4) aldehydes as substrates. Although, its activity is lost with the
long chain aldehydes but it can oxidize C4-C6 acid substituted aldehydes and dialdehydes.88
CH2O + H2O + 2Fd(ox) → HCOOH + 2H+ + 2Fd(red) (1.15)
14
Chapter 1-Introduction
Structurally FOR is similar to AOR, i.e. it contains a tungsto-bis-pterin cofactor, an [4Fe-4S]
cluster and a Mg2+ ion bridging the two pterins but it also contains a Ca atom per subunit. The
function of Ca, however, is still not known.87
6.1.3. Glyceraldehyde-3-Phosphate Ferredoxin Oxidoreductase (GAPOR)
Glyceraldehyde-3-Phosphate Ferredoxin Oxidoreductase (GAPOR) is purified from
Pyrococcus furiosus89. It catalyzes the oxidation of glyceraldehyde-3-phosphate (GAP) to 3phosphoglycerate. It is absolutely specific for its substrate and so far GAP is its only
substrate. It cannot oxidize non-phosphorylated aldehydes such as formaldehyde,
acetaldehyde, benzaldehyde, nor related compounds like glucose-6-phosphate or glycerol 3phosphate3.
Glyceraldehyde-3-phosphate + H2O + 2Fd(ox) → 3-phosphoglycerate + 2H+ + 2Fd(red) (1.16)
This enzyme plays a key role in the glycolytic pathway during carbohydrate (maltose)
metabolism in P. furiosus, where it converts the GAP dehydrogenase and phosphoglycerate
kinase (PGK) to 3-phosphoglycerate. Like AOR and FOR, it contains a tungsto-bis-pterin
cofactor, an [4Fe-4S] cluster and two Zn atoms per subunit. The function of Zn is still not
known90.
6.1.4. Carboxylic Acid Reductase (CAR)
Carboxylic acid reductase (CAR) is isolated from moderately thermophilic acetogenic
bacteria, Clostridium formicoaceticum. It catalyzes the reduction of a wide range of aliphatic
and aromatic non-activated carboxylic acids to aldehydes at the expense of reduced viologens
(toxic bipyridinium derivatives of 4, 4'-bipyridyl).
RCOOH + 2H+ + 2“e-ˮ → RCHO + H2O (1.17)
However, neither its physiological electron carrier nor its function inside the cell is known.
CAR in Clostridium thermoaceticum seems to be present in two forms. One of them contains
a flavin (FAD) group. Compared with the other enzymes in the AOR family these two CAR
enzymes are more complex, with multiple subunits and a much higher Fe content.91,92
6.1.5. Aldehyde Dehydrogenase (ADH)
Aldehyde dehydrogenase (ADH) is purified from sulfate reducing Desulfovibrio gigas93and
Desulfovibrio simplex.94 ADH from D.gigas catalyzes the oxidation of wide variety of
15
Chapter 1-Introduction
aldehyde substrates. It sufficiently oxidizes the C3-C4 aldehydes like acetaldehyde and
propionaldehyde. It contains a tungsto-bis-pterin cofactor and one [4Fe-4S] cluster. The
physiological electron carrier inside the cell is not known. ADH from D. simplex oxidizes
both aliphatic and aromatic aldehydes and can use flavins (FMN or FAD) as electron carriers.
When D. gigas is grown in the absence of tungsten it produces a Mo-containing aldehydeoxidizing enzyme termed aldehyde oxidoreductase (AOX).92
6.2 The Formate/ Formyl Methanofuran Dehydrogenase Family (F(M)DH)
The enzymes of this family are extremely sensitive to oxygen inactivation, which makes them
difficult to isolate and characterize. They use carbon dioxide as a substrate. Members of this
family are:
6.2.1. Formate Dehydrogenase (FDH)
6.2.2. Formyl Methanofuran Dehydrogenase (FMDH)
6.2.1. Formate Dehydrogenase (FDH)
Formate dehydrogenase (FDH), a tungsten-selenium-iron-sulfur protein, is purified from
thermophilic acetogenic bacterium Clostridium thermaceticum95 and Desulfovibrio gigas96. It
catalyzes the reduction of CO2 to formate and is the first reaction of an autotroph CO2fixation pathway occurring in many anaerobic bacteria.95
CO2 + NADPH → HCOO- + NADP+ (1.18)
FDH from C. thermaceticum has tungsten atom coordinated to the two pterin cofactors. It also
possesses FeS centers and unusual amino acid selenocysteine. It is extremely oxygen sensitive
and uses NADP as the electron acceptor3,79. FDH from D. gigas was isolated in the presence
of oxygen, although the activity could be only measured under strictly anoxic conditions. This
FDH has mononuclear tungsten atom coordinated by two pterin cofactors, which are of the
dinucleotide form (MGD). It also possesses two [4Fe-4S] clusters as redox active sites. No
selenium atom is found in this enzyme. FDH from D. gigas is more related to the Mo
containing FDHs as compared to AOR.96
6.2.2. Formyl Methanofuran Dehydrogenase (FMDH)
N-Formylmethanofuran (formyl-MFR) is purified from hyperthermophilic sulfate reducing
archaeon, Archaeoglobus fulgidus.97 FMDH is involved in the first step of the conversion of
CO2 to methane. It catalyzes the reductive addition of CO2 to the organic cofactor
16
Chapter 1-Introduction
methanofuran (MFR) in a two-electron reaction. The physiological electron donor for this
reaction is not known.
CO2 + MFR + 2H+ + 2“e-ˮ → CHO-MFR + H2O (1.19)
FDMH has a mononuclear tungsten atom coordinated by two pterin cofactors (MGD). It also
possesses Fe/S clusters but no selenium atom is found.84
6.3 The Acetylene Hydratase Family (AH)
This family has only one member characterized so far named Acelylene hydratase (AH). It is
a tungsten-iron-sulfur protein and is purified from mesophilic anaerobe Pelobacter
acetylenicus. 98,99 It is extremely oxygen sensitive and its activity is lost upon exposure to air.
It catalyzes the hydration of acetylene to acetaldehyde, a net hydration reaction rather than a
redox reaction.
C2H2 + H2O → CH3CHO (1.20)
AH contains a mononuclear tungsten atom coordinated by pterin cofactors (MGD), sulfur
atom of cysteine and a terminal oxygen atom. It also possesses a [4Fe-4S] cluster which is not
far from the tungsten center. It is a distinct class besides other tungstoenzymes as its reaction
does not involve electron transfer. Also, the oxidation state of tungsten does not change
during the reaction.84,100
7) Tungsten-substituted Molybdenum enzymes
Historically, Tungsten (W) has been regarded as an antagonist of the biological function of
molybdenum. Early attempts to substitute W into active sites of molybdoenzymes resulted in
inactive metal-free enzymes or W-substituted enzymes with little or no activity.30 This failure
could be because of the organism incapable of growing on the tungstate-containing medium
or due to the pathway used for growth. Once the alternative growth medium and suitable
pathway were established, growth was achieved and the effect of tungsten on the activity of
enzyme could be examined. The successful tungsten substituted molybdenum enzymes are:
7.1 Xanthine Oxidase and Sulfite Oxidase
7.2 Trimethyamine Oxide Reductase
7.3 Dimethylsulfoxide Reductase
7.4 Nitrate Reductase
17
Chapter 1-Introduction
7.1 Xanthine Oxidase and Sulfite Oxidase
The activity of the molybdenum enzymes, xanthine oxidase and sulfite oxidase, in rats has
been shown to be lowered by addition of tungstate [WO4]2- to the drinking water of rats on a
molybdenum dependent diet.101
7.2 Trimethylamine Oxide Reductase (TMAOR)
Tungsten substituted periplasmic TMAO reductase from E.Coli has been purified,
characterized and shown to be enzymatically active. W-TMAOR is capable of catalyzing the
reduction of TMAO to TMA as well as the reduction of DMSO to DMS at high redox
potential while Mo-TMAOR can catalyze the reduction of various N-oxides but not S-oxides.
W-TMAOR is more stable to high temperatures than Mo-TMAOR. However, the specific
activity of the W-substituted enzyme is half than that of Mo-TMAOR. 102
7.3 Dimethylsulfoxide Reductase (DMSOR)
Tungsten was substituted and characterized successfully for DMSO reductase from
R.capsulatus. W-DMSOR reduces DMSO about 17 times faster than Mo-DMSOR103. This
result is consistent with the previous observation that oxygen atom transfer from the substrate
to the reduced metal center is faster for tungsten than molybdenum. 104 However, in contrast to
Mo-DMSOR, W-DMSOR does not catalyze the oxidation of DMS.
7.4 Nitrate Reductases (NRs)
Tungsten substituted membrane bound respiratory nitrate reductase (Nar) was purified from
Pyrobaculum aerophilum grown in the presence of tungstate [WO4]2- and in the absence of
molybdate, [MoO4]2-.105 W-Nar is an active enzyme; however, it gradually lost activity. Also,
the nitrate reductase activity is decreased when the tungstate concentration in the environment
is increased.106
8) Molybdenum-substituted Tungsten enzyme, Acetylene Hydratase (AH)
Acetylene hydratase (AH) from Pelobacter acetylenicus has been characterized as an ironsulfur, tungsten enzyme. Fully active molybdenum containing AH has been purified from
cells grown in the absence of tungsten and exhibits different redox properties from those of
the tungsten enzyme.107 This Mo-AH is inactive under the strongly reducing conditions
necessary for activity of the tungsten enzyme. However, the molybdenum enzyme was shown
to have some activity upon careful reduction, which was lost upon further reduction.
18
Chapter 2-W-Nitrate Reductase
Tungsten - Nitrate Reductase
1. Introduction
Nitrate reductases (NRs) play key roles in the first step of biological nitrogen cycles108,109,110
i.e., assimilatory ammonification (to incorporate nitrogen into biomolecules), denitrification
(to generate energy for cellular function) and dissimilatory ammonification (to dissipate extra
energy by respiration). They always catalyze the reduction of nitrate to nitrite, and have been
classified into three groups, assimilatory nitrate reductases (Nas), respiratory nitrate
reductases (Nar) and periplasmic nitrate reductases (Nap). Nas belongs to the sulfite oxidase
family and is located in the cytoplasm. 111 It is the first enzyme of a reduction sequence for
nitrogen incorporation into the biomass that maintains the bioavailability of nitrate to plants,
algae, fungi, archaea and bacteria.112,113 Dissimilatory nitrate reductases, Nar and Nap belong
to the DMSO reductase family of mononuclear MPT containing molybdo-enzymes. They are
linked to respiratory electron transport systems and are located in the membrane and
periplasm, respectively. They catalyze the first step of the catabolic, anaerobic respiration
pathway in bacteria and archaea.115
Nitrate reduction, catalyzed by membrane bound respiratory nitrate reductase (Nar), is an
important step of the denitrification in the anaerobic respiratory pathways employed by a
diverse group of bacteria and archaea.114 Nar was found to contain a Mo cofactor in all
microbes from which it was isolated and belongs to the DMSO reductase family. 115 In
general, Nar becomes inactive by the addition of tungstate (WO42-) to the growth medium,116
although due to similar chemical properties W can replace Mo as the active site metal and
cannot only retain but increase its catalytic activity in E. coli TMAO reductase,102 the
Desulfovibrio alaskensis formate dehydrogenase117 and the Rhodobacter capsulatus DMSO
reductase.118 However, recently the nitrate reductase (Nar) from the hyperthermophilic
denitrifying archaeon Pyrobaculum aerophilum has been shown to retain its activity even at a
tungsten rich environment.105
P. aerophilum, a hyperthermophilic archaeon, is naturally exposed to high levels of tungsten,
a heavy metal that is abundant in high temperature environments. Tungsten was reported to
stimulate the growth of several mesophilc methanogens and some mesophilic and
thermophilic bacteria.115 The growth of P. aerophilum also depends on the presence of
tungstate in the growth medium which suggests the involvement of tungstoenzymes in
essential metabolic pathways. 119
19
Chapter 2-W-Nitrate Reductase
P. aerophilum is the only hyperthermophilic archaeon isolated that reduces nitrate via a
membrane bound respiratory nitrate reductase (Nar).119 Nar purified from P. aerophilum
grown in the absence of added molybdate (MoO42-) and with 4.5µM tungstate (WO42-) is a
tungsten containing enzyme, which is identical to Mo-Nar106 (previously isolated from P.
aerophilum), indicating that either metal can serve as the active site ion. The crystal structure
is similar to the previously reported Nar from E. coli,120 a heterodimeric enzyme termed as
NarGH where NarG hosts the metal (Mo or W) catalytic site. The metal is coordinated by two
metallopterin guanine dinucleotide (bis-MGD) ligands, a carboxyl group of Asp222 and a
water molecule. The NarH component possesses an iron-sulfur (FeS) redox active subunit.105
NarGH reduces nitrate to nitrite, changing the oxidation state of metal from +IV to +VI. Two
electrons and two protons are required for the reductive half reaction, resulting in the
formation of a water molecule and a nitrite ion (eq. 2.1).
NO3- + 2H+ + 2e-
NO2- + H2O (2.1)
An experimental study was carried out on small model complexes; an analogue of the active
site of dissimilatory nitrate reductase (Desulfovibrio desulfuricans) which is in the reduced
state contains a Mo atom bound by two metalopterindithiolene ligands and a cysteinate
residue. This study demonstrates that nitrate reduction by primary (direct) oxo transfer121 is a
feasible reaction pathway (Fig. 2.1).122
O
N
Y
S2
M
IV
NO3-
O
S3
M
S2
S4
S1
NO2-
Y
O
S3
S2
S4
S1
O
Y
M VI S3
S4
S1
O
HN
H2N
Y
H2O
S2
S1
H2O
M
N
H
M
S
O
O
O
P OR
O
Metalopterindithiolene cof actor,
R absent or a nucleotide
S3
S4
N
S
H
N
2H+
2e-
Fig. 2.1: Schematic description of the proposed mechansim122 for the nitrate reduction, where M=Mo
and Y= S-Cys. Also the metalopterindinucleotide cofactor is shown.
Here we have presented the density functional theory (DFT) study on chosen model
complexes derived from the protein X-ray crystal structure of P.aerophilum105 nitrate
reductase (Nar). The purpose of the study was to investigate (i) The effect on the reduction of
20
Chapter 2-W-Nitrate Reductase
nitrate when W replaces the Mo at the active site. (ii) The energy barriers on the potential
energy surface. (iii) The reason for the activity loss of Nars (respiratory nitrate reductase) in
the presence of W.
2. Computational Details
All the geometries were optimized using Gaussian 09 with the hybrid density functional
B3LYP123 and the LANL2DZ basis set124,125,126,127 augmented by polarization functions on
sulfur atoms (ζ = 0.421).128 The starting nitrate complex geometries for transition state
searches were generated by shortening and lengthening of forming and breaking bonds,
respectively. Frequency calculations proved transition states to have exactly one imaginary
frequency with the correct transition vector. Single point energies were computed with the
B3LYP functional and the Stuttgart-Dresden effective core potential basis set (SDD)129,130
augmented by polarization functions for all atoms except Mo, W and H (ζ = 0.600, 1.154,
0.864, and 0.421 for C, O, N, and S, respectively).128 Self-consistent reaction field (SCRF)
computations were performed on the optimized geometries to model the protein surrounding
the active site by a conductor like polarizable continuum method (CPCM)131 as implemented
in Gaussian 09.132,133 Default Gaussian 03 parameters were used for the evaluation of solutesolvent dispersion and repulsion interaction energies, 134,135 and solute cavitation energy
variations. 136 The molecular cavity was specified using a minimum radius (RMin) of 0.5Ǻ and
an overlap index (OFac) of 0.8.137
3. Active Site Models
Two types of active site models were designed on the basis of the protein X-ray crystal
structure of Pyrobaculum aerophilum (PDB ID 1R27)105 only differing in the metal center,
a containing Mo and b containing W at the active site. These active site models include the
metal center coordinated by two enedithiolene moieties of the pterin molecules, by Asp222 and
by H2O8538. His546, Asn52, Tyr220, Gly549 and Val578 residues were also included in the model
complexes as they may influence the catalytic reaction due to their proximity to the metal
center. Hydrogen atoms were added manually. His546 and Gly549 residues form hydrogen
bonds to the ionized Asp222 preventing it to rotate and become a bidentate ligand which then
would block the substrate binding site. Asn52 was included as its distance of 3.9Ǻ from the
metal center suggests that it is suitable for substrate coordination.105 During the optimizations,
alpha (α) carbon atoms and nitrogen atoms attached to the beta (β) carbon atoms of His546,
Asn52, Tyr220 and Asp222 were kept fixed to their crystal structure positions to mimic the steric
21
Chapter 2-W-Nitrate Reductase
constraints by the protein matrix. Carbon atom C7 and the nitrogen atom attached to carbon
atom C5 were kept fixed for residue Gly549. The MPT ligands were truncated at the pyran
rings and oxygen atoms of these pyran rings were also kept fixed (Fig.2.2).
First, hydrogen atoms were geometry optimized applying one negative overall charge
(assuming Mo/W at the +VI oxidation state), keeping all heavy-atoms fixed at their positions.
The resulting geometries served to generate the different starting geometries needed for
computing the mechanism for nitrate reduction.
The starting geometries for the substrate and product complexes are generated by slight
distortion of M-O and O-NO2 in the optimized transition state geometries, 5a and 5b.
Geometries with slightly elongated M-O distance and reduced O-NO2 distance are considered
as the starting geometries for the optimization of 4a and 4b educt-substrate complexes
whereas reduced M-O distance and elongated O-NO2 distance are considered as the starting
geometries for the optimization of 6a and 6b product complexes. The geometry optimizations
of these distorted geometries directly lead to complexes, 4a/4b and 6a/6b.
Fig. 2.2: Optimized oxidized active site model of Mo-Nar.
Atoms labeled (*) were kept fixed at their X-ray crystal structure positions.
22
Chapter 2-W-Nitrate Reductase
4. Results
Ø Optimized active site model complexes, 1a, 1b, 2a and 2b:
The protein X-ray crystal structure of P.aerophilum Nar from the PDB data base (PDB ID:
1R27)105 shows that at the active site the metal is coordinated by two metallopterin guanine
dinucleotide (bis-MGD) ligands, a carboxyl group of Asp222 and a water molecule.105
However, the distance of the oxygen atom (Owat) of this coordinated water molecule from the
metal center is 1.87 Ǻ which neither falls in the range expected for metal oxide (1.71 - 1.75
Ǻ),138,139 nor for water (2.0 - 2.3 Ǻ)140 ligands. Also, the distance between Owat and oxygen of
Asp222 (OAsp) is 1.59 Ǻ, which is only 0.1 Ǻ longer than the typical peroxo O-O- bond length
(1.49 Ǻ).
We have optimized two active site model complexes to clarify the nature of this oxo species;
1 (oxidation state of Mo/W is +IV, overall charge is -1) contains a water molecule and 2
(oxidation state of Mo/W is +VI, overall charge is -1) contains an oxide (O1) group attached
to the metal (Fig. 2.3). Geometry optimizations of active site model complexes 1 and 2 result
in distinctively different geometrical parameters of the metal coordination site relative to the
protein X-ray crystal structure geometry of NarGH.105 Optimized geometry data for the model
complexes 1a with M=Mo (1b, M=W) show that the dithiolenes are twisted less against each
other as the S1-S2-S3-S4 dihedral angle decreases from -18.3˚ to -6.4˚ for 1a (-2.5˚ for 1b) i.e.,
the coordination geometries are nearly trigonal prismatic. Bond distances between the metal
center, M and the dithiolene sulfur atoms, S decreases from ~2.455 Ǻ to ~2.393 Ǻ (~2.384 Ǻ)
when comparison is made with the protein X-ray crystal structure (Tables 2.2 and 2.3).
Elongated bond distances for M-Owat (from 1.874 Ǻ to 2.335 Ǻ (2.286 Ǻ)) and M-OAsp (from
1.97 Ǻ to 2.142 Ǻ (2.122 Ǻ)) are computed. But the main difference lies in the Mo-S2 bond
distance (from 2.537 Ǻ to 2.387 Ǻ (2.377 Ǻ)), in the bond angles between the OAsp, M and
Owat (from 49˚ to 66˚ (66˚)), and in the distance between the two oxygen atoms, OAsp-Owat
(from 1.596 Ǻ to 2.428 Ǻ (2.392Ǻ)).
Distorted octahedral coordination geometries result from geometry optimizations of oxidized
model complexes 2a (2b). Optimized data shows increase in the S1-S2-S3-S4 dihedral angles
(from -18.3˚ to -43.7˚ (-42.1˚)) and in the M-S bond distances (from ~ 2.455 Ǻ to ~2.474 Ǻ
(~2.461 Ǻ)). There is one longer M-S bond relative to other three M-S bonds in the optimized
oxidized model complexes and the protein X-ray crystal structure. However, it is M-S2 bond
(2.537 Ǻ) in the X-ray structure while M-S3 bond (2.591 Ǻ (2.549 Ǻ)) in the optimized
oxidized model complexes. These sulfur atoms (S2 in the X-ray crystal structure while S3 in
23
Chapter 2-W-Nitrate Reductase
the optimized oxidized model complexes) are at the trans position to the M=O ligand and oxo
is a trans- influencing ligand which causes the stretching of M-S bonds to which they are
trans.
Elongated bond angles between the OAsp, M and O1 (from 49˚ to 88˚ (88˚)), and distances
between the two oxygen atoms, OAsp-O1 (from 1.596 Ǻ to 2.684 Ǻ (2.647Ǻ)) are computed.
Slightly elongated M-OAsp distances (from 1.970 Ǻ to 2.083 Ǻ (2.040 Ǻ)) and shortened M-O1
distances (from 1.874 Ǻ to 1.755 Ǻ (1.764Ǻ)) are also observed (Tables 2.2 and 2.3).
Fig. 2.3: The chemical structure of the active site model complexes 1 and 2 derived from the protein
X-ray crystal structure of Nar (PDB ID 1R27)105
Fig. 2.4: Schematic description of the mechanism for nitrate reduction at the NR active site.
24
Chapter 2-W-Nitrate Reductase
Ø Optimized reduced complexes, 3a and 3b:
The reaction catalyzed by nitrate reductase is an oxo-transfer reaction, in which an oxygen
atom is transferred from nitrate to the reduced metal. As a consequence of the metal reduction
from MVI to MIV, the oxo group of the oxidized MVI is lost as hydroxo/water after proton
uptake. Optimizations of the reduced active site model complexes 3a (3b) without any
additional ligand, i.e. fivefold coordinate metal center give S1-S2-S3-S4 dihedral angles of
-0.2˚ (1.3˚), resulting in nearly tetragonal pyramidal geometries. The bond distances between
the metal center M and S of the dithiolenes are ~2.362 Ǻ (~2.353Ǻ) (Table 2.2 and 2.3). The
M-OAsp distance is 2.017 Ǻ (1.980Ǻ).
Ø Optimized substrate complexes, 4a and 4b:
First, nitrate gets loosely bound in the active site pocket by weak interactions with the active
site residues Asn52 and Gly549 resulting in the substrate complexes 4a (4b) (Fig. 2.4 and Fig.
2.5). The computed reaction energies for the substrate complex formation are exothermic, -9.6
kcal/mol (-7.6 kcal/mol) in the gas phase and -4.6 kcal/mol (0.2 kcal/mol) for the polarizable
continuum model relative to the separate substrate and educt complexes 3a (3b) (Table 2.1).
There is no significant change in geometrical parameters of the active site relative to the
reduced complexes 3a (3b) (Tables 2.2 and 2.3).
Ø Optimized transition state complexes, 5a and 5b:
Reduction of nitrate is a single step reaction in which the transfer of an oxygen atom proceeds
through transition state 5a (5b). The energy barrier computed for 5a, 34.4 kcal/mol in the gas
phase and 32.1 kcal/mol in the continuum is almost three times as large as compared to that of
5b, 12.0 kcal/mol in the gas phase and 11.0 kcal/mol in the continuum (Table 2.1). There is
also a remarkable difference in the geometries.
The Mo containing transition state (5a) has a distorted octahedral geometry. The optimized
data shows an increase in the S1-S2-S3-S4 dihedral angle from 2.0˚ to 30.5˚ and in the Mo-S
bond lengths from ~ 2.37 Ǻ to ~2.45Ǻ (Table 2.2) when comparison is made with the
optimized 4a geometry. The Mo-O and O-NO2 distances are 1.918 Ǻ and 1.723 Ǻ,
respectively. The Mo-OAsp bond distance is elongated from 2.029 Ǻ to 2.102 Ǻ.
The W containing transition state (5b) on the other hand has a distorted trigonal prismatic
geometry where the optimized data shows an increase in the S1-S2-S3-S4 dihedral angle from
1.2˚ to 7.6˚ and in the W-S bond lengths from ~ 2.37 Ǻ to ~2.45Ǻ (Table 2.3) as compared to
25
Chapter 2-W-Nitrate Reductase
the optimized 4b geometry. The W-O and O-NO2 bond distances are 1.942 Ǻ and 1.638 Ǻ,
respectively i.e., 5b can be considered to be an earlier transition state than 5a. The W-OAsp
distance is elongated from 1.986Ǻ to 2.079Ǻ.
In the optimized geometries 5a and 5b, NO3 - is coordinated to the metal at the active center
and also forms a hydrogen bond to the Asn55.
Ø Optimized product complexes, 6a and 6b:
The nitrate reduction results in metal oxo product complexes 6a (6b), having distorted
octahedral geometries. In the optimized geometries, 6a and 6b, NO2 - is loosely bound in the
active site pocket and makes hydrogen bonds with the active site residues Asn52 and Gly549.
Oxygen atom transfer is computed to be a slightly exothermic step for M=Mo where the
product complex (6a) has a relative energy of -7.6 kcal/mol in the gas phase and -1.9 kcal/mol
in the continuum (Table 2.1). The Mo-O bond distance is reduced from 1.918 Ǻ to 1.737 Ǻ
while the O-NO2- bond is broken (4.444 Ǻ) when comparison is made with the optimized 5a
geometry (Table 2.2). The S1-S2-S3-S4 dihedral angle is further increased from 30.5˚ to 54.5˚
while the Mo-S bond distances are increased from ~2.438 Ǻ to ~2.629 Ǻ (Table 2.2). The
Mo-OAsp bond distance is increased from 2.102 Ǻ to 2.133 Ǻ.
On the contrary, the W containing product complex (6b) is highly exothermic, with computed
relative energies of -43.3 kcal/mol in the gas phase and -34.7 kcal/mol in the continuum. The
W-O bond distance is reduced from 1.942 Ǻ to 1.757 Ǻ while the O-NO2- bond is broken
(5.133 Ǻ) relative to the optimized 5b geometry (Table 2.3). The S1-S2-S3-S4 dihedral angle of
the dithiolenes is changed from 7.6 ˚to -42.4˚, whereas the W-S bond distances are increased
from ~2.472 Ǻ to ~2.562Ǻ (Table 2.3). There is no significant change in the W-OAsp bond
distance (2.079 Ǻ instead of 2.076 Ǻ).
5. Discussion
To date, few archaeal Nars have been characterized from P. aerophilum,106 Haloarcula
marimortui 141,142 and Haloferax mediterranei.143 These archaeal Nars contain Mo cofactors at
their active sites. It is not clear how these microbes maintain their ability to respire with
nitrate using Mo-containing Nar in a high temperature environment that is naturally enriched
with W but depleted of molybdate (MoO42-).144 Early attempts to substitute tungsten for
molybdenum in molybdo-enzymes failed because the organism was incapable of growing on
the tungstate-containing medium. 145 However, the hyperthermophile P. aerophilum is a
26
Chapter 2-W-Nitrate Reductase
denitrifying archaeon requiring tungstate (WO42-) for growth although its Nar is a Mo cofactor
containing enzyme.119 Afshar et al. demonstrated that the external tungstate concentration
affects the denitrification pathway efficiency of this archaeon, resulting in the complete
denitrification only at high tungstate concentration.119
Recently, Nar purified from P. aerophilum grown in the absence of added molybdate and with
4.5µM tungstate has been reported114 which is a W containing enzyme. P. aerophilum Nar is
the first active nitrate reductase that contains a W cofactor. The presence of a W cofactor may
be reflective of high concentrations of this metal at high temperatures.3 As previously
described this enzyme can also accommodate Mo as the active site metal.106 To compare the
properties of Mo and W cofactors containing enzymes, DFT calculations were performed on
the active site model complexes derived from the protein X-ray crystal structure of P.
aerophilum.105
The computed energy barrier for the oxygen atom transfer from the nitrate to the metal center
is 34.4 kcal/mol for the Mo active site model complex, about triple the energy barrier of the
W active site model complex (12.0 kcal/mol) (Table 2.1). Thus, as compared to Mo-Nar, WNar should be more active, which is in contrast to experimental findings,146 but is similar to
the R. Capsulatus DMSO reductase;102,106,147 the W-substituted DMSO reductase was 17
times more active in the reduction of DMSO than the Mo-substituted enzyme. However, the
W-substituted DMSO reductase was inactive for the oxidation of dimethysulfide (DMS).147
Oxidation of the educt complex is close to thermoneutral for the Mo active site model
complex (-1.9 kcal/mol) but strongly exothermic for the W containing active site model
complex (-34.7 kcal/mol) (Table 2.1). The low relative energy for the oxidized W metal
complex makes the regeneration of the +IV oxidation state much more difficult as compared
to the Mo metal complex. The MVI to MIV reduction requires much more reductive power
with M=W then with M=Mo.
So, although the reduction of nitrate is stimulated when W replaces Mo in the active site of
Nar the catalytic cycle breaks after the reduction of nitrate to nitrite when the biochemical
reducer is not strong enough. It is similar/in agreement with the experimental findings which
suggest that Nar from cells grown at high WO42- concentrations appeared to be more labile
than the previously isolated Mo-Nar; i.e., the enzyme gradually lost activity.106 Also, the
nitrate reductase activity is decreased with the increase in tungstate concentration in the
27
Chapter 2-W-Nitrate Reductase
environment.106 In conclusion, the Nar isolated from P. aerophilum is a molybdenum
containing enzyme.
Table 2.1: Computed energies (kcal/mol) relative to the educt-substrate complex for the nitrate
reduction.
M=Mo
M=W
Educt
Substrate
Transition
Product
Complex
Complex
State Complex
Complex
0.0
-9.7
30.2
-11.6
//B3LYPa)
0.0
-9.6
34.4
-7.6
SDDb)
0.0
-4.6
32.1
-1.9
COSMOc)
0.0
-7.8
7.0
-52.6
//B3LYPa)
0.0
-7.6
12.0
-43.3
SDDb)
0.0
0.2
11.0
-34.7
COSMOc)
Where, a) B3LYP/Lanl2DZ(p), b) B3LYP/SDDp//B3LYP/Lanl2DZ (p),
c) COSMO-B3LYP/SDDp//B3LYP/Lanl2DZ(p) (see Computational details)
Scheme 2.1: Plot of computed reaction energies (kcal/mol) relative to separate substrate and educt
complex vs steps involved in the reaction mechanism.
Where, E = educt complex, ES = educt-substrate complex, TS = transition state, P = product complex.
28
Chapter 2-W-Nitrate Reductase
Table 2.2: Geometrical features of the optimized model complexes of the reaction mechanism for the
molybdenum containing nitrate reductase
Mo-S1 (Ǻ)
2.405
Reduced
Complex1
1a
2.409
Mo-S2 (Ǻ)
2.537
2.387
2.418
2.347
2.348
2.452
2.629
Mo-S3 (Ǻ)
2.395
2.380
2.591
2.345
2.349
2.422
2.421
Mo-S4 (Ǻ)
2.484
2.394
2.441
2.375
2.371
2.457
2.475
Mo-OAsp (Ǻ)
1.97
2.142
2.083
2.017
2.029
2.102
2.133
Mo-Owat (Ǻ)
1.874
2.335
-
-
-
-
-
Mo-O1 (Ǻ)
-
-
1.755
-
-
-
-
Mo-O (Ǻ)
-
-
-
-
-
1.918
1.737
O-NO2 (Ǻ)
-
-
-
-
1.310
1.723
-
OAsp-Owat/1 (Ǻ)
1.596
2.428
2.684
-
-
-
2.786
OAsp-Mo- Owat/1(˚)
49.0
65.5
88.3
-
-
-
91.5
S1-S2-S3-S4(˚)
-18.3
-6.4
-43.7
-0.2
2.0
30.5
54.5
Crystal
Structure
-
Oxidized
Complex2
2a
2.446
Reduced
Complex
3a
2.379
Educt
Complex
4a
2.370
Transition
State
5a
2.420
Product
Complex
6a
2.430
Where, = Water containing reduced complex, = Oxygen containing oxidized complex
1
2
Table 2.3: Geometrical features of the optimized model complexes of the reaction mechanism for the
tungsten containing nitrate reductase
W-S1 (Ǻ)
2.405
Reduced
Complex1
1b
2.397
W-S2 (Ǻ)
2.537
2.377
2.432
2.334
2.335
2.419
2.442
W-S3 (Ǻ)
2.395
2.373
2.549
2.337
2.337
2.424
2.562
W-S4 (Ǻ)
2.484
2.388
2.424
2.371
2.369
2.457
2.428
W-OAsp (Ǻ)
1.97
2.122
2.040
1.980
1.986
2.079
2.076
W-Owat (Ǻ)
1.874
2.286
-
-
-
-
-
W-O1 (Ǻ)
-
-
1.764
-
-
-
-
W-O (Ǻ)
-
-
-
-
-
1.942
1.757
O-NO2 (Ǻ)
-
-
-
-
1.310
1.638
-
OAsp-Owat/1 (Ǻ)
1.596
2.392
2.647
-
-
-
2.747
OAsp-W-Owat/1(˚)
49.0
65.6
87.9
-
-
-
91.2
S1-S2-S3-S4(˚)
-18.3
-6.3
-42.1
1.3
1.2
7.6
-42.4
Crystal
Structure
-
Oxidized
Complex2
2b
2.439
Reduced
Complex
3b
2.369
Educt
Complex
4b
2.363
Transition
State
5b
2.428
Product
Complex
6b
2.455
Where, 1 = Water containing reduced complex, 2 = Oxygen containing oxidized complex
29
Chapter 2-W-Nitrate Reductase
Fig.2.5: Optimized geometries for the Mo (a) and W (b) containing active site model complexes 1-6.
30
Chapter 2-W-Nitrate Reductase
Fig. 2.5: Continued…
31
Chapter 3-Ethylbenzene Dehydrogenase
Ethylbenzene Dehydrogenase
1. Introduction
Ethylbenzene dehydrogenase (EBDH) is a soluble periplasmic molybdoenzyme. It catalyzes
the oxygen independent stereospecific hydroxylation of ethylbenzene to (S)-1-phenylethanol
as the initial step of anaerobic ethylbenzene degradation. It is the first known example of
direct anaerobic oxidation of a non-activated hydrocarbon.148
Ethylbenzene → (S)-1-phenylethanol
(3.1)
EBDH belongs to the prokaryotic oxotransferase (DMSO) family of mononuclear
molybdenum enzymes and exhibits the highest sequence similarities to selenate reductase of
Thauera selenatis,149 dimethylsulfide dehydrogenase of Rhodovulum sulfidophilum,150
perchlorate and chlorate reductases of Dechloromonas sp., and membrane bound nitrate
reductases from archaebacteria and eubacteria. 151,152
In contrast to the oxidized state, this enzyme is inactivated by exposure to air in the reduced
state, possibly due to generation of reactive oxygen species at the reduced heme cofactor.153
Elucidation of the detailed role of EBDH in bio-mineralization of ethylbenzene, a major
component of crude oil, will help to improve the recovery of polluted ecosystems. EBDH
promises potential applications in chemical and pharmaceutical industry, as pure enantiomers
of alcohols are valuable as building blocks for physiologically active compounds; e.g.1phenylethanol itself is used as food and drinking flavoring agent and additive of cosmetics.
Also the enzyme seems to react with a relatively wide range of alkylaromatic and
alkylheterocyclic substrates.153,154 However, ethylbenzene is the native substrate of the
enzyme. The high affinity for ethylbenzene may result from the presence of hydrophobic
amino acid residues at the walls of a tunnel leading to the active center, which may facilitate
substrate transport into the enzyme interior and discrimination of other compounds. The
unsual catalytic versatility of EBDH may arise from a rather large active center cavity, which
seems not to pose significant steric constraints for para- and meta- substituents.155
Three bacterial species capable of anaerobic degradation of the aromatic hydrocarbon
ethylbenzene are known to date. All of these are denitrifying bacteria that belong to the genus
Azoarcus of the β-proteobacteria. For one of these strains, Azoarcus sp. EB-1, ethylbenzene is
the only known hydrocarbon utilized as growth substrate.156 The other two strains utilize
32
Chapter 3-Ethylbenzene Dehydrogenase
either ethylbenzene or an alternative hydrocarbon compound, namely toluene (strain EbN1) or
n-propylbenzene (strain PbN1).157
The X-ray protein structure of EBDH (PDB 2IVF),155 isolated from Aromatoleum
aromaticum strain EbN1 (formerly named Azoarcus sp.), is a heterotrimer. The first subunit
contains the molybdenum active site and one Fe-S cluster, while the second subunit carries
several Fe-S clusters and its third subunit binds a heme b (Fe-protoporphyrin IX).155 At the
active site, MoIV is coordinated with two metalopterin guanine dinucleotide (bisMGD)
ligands, one oxygen atom of Asp223 and one oxygen atom of an acetate bound at the sixth
coordination site. The two MGD molecules differ from each other: one has a closed pyran
ring, while the other assumes an open form (Fig. 3.1).
Fig. 3.1: The chemical structure of the active site of EBDH. Numbering follows the relevant
literature.158 MGD-P with an open pyran ring, MGD-Q with a closed pyran ring. R is (P2O7)-riboseguanine nucleotide155
The proposed reaction sequence starts with an oxidation of the MoIV ion to the MoVI state by
the removal of two electrons via the electron transport chain to the heme. MoIV binds a water
molecule releasing two protons via nearest His192 to the bulk solvent resulting in the
formation of a Mo-oxo group.155 This oxidized active site is the active species for the
hydroxylation of ethylbenzene (Fig. 3.2). Asp223, Lys450 and His192 seem to be in the best
position to take part in the hydrocarbon activation, together with the molybdenum oxo ligand
in the oxidized form of EBDH (Fig. 3.2).
33
Chapter 3-Ethylbenzene Dehydrogenase
Fig. 3.2: Sketch of the catalytic reaction starting from an oxo group attached to the Mo(VI).155
On the basis of the analysis of kinetic data,148 quantitative structure activity relationship
(QSAR) studies159 and X-ray crystal structure of the EBDH,155 a reaction mechanism was
proposed (Fig. 3.3) which explains the stereospecificity of EBDH in oxidation of
ethylbenzene. The reaction retains the stereochemistry, i.e., the pro-S hydrogen of C1 of
ethylbenzene is removed and replaced by the OH group to yield (S)-1-phenylethanol.148
34
Chapter 3-Ethylbenzene Dehydrogenase
Fig. 3.3: Hypothetical mechanism of ethylbenzene oxidation by EBDH.148
The mechanism of ethylbenzene hydroxylation at the molybdenum cofactor of EBDH
essentially involves the cleavage of a C-H bond (rate-limiting process) of the methylene group
in order to activate the hydrocarbon by a two electron transfer step or two one electron
transfer steps i.e. heterolytic cleavage and homolytic cleavage of the C-H bond, respectively.
These options result in the formation of a carbocation or a hydrocarbon radical intermediate,
35
Chapter 3-Ethylbenzene Dehydrogenase
respectively, together with a hydroxide coordinated to molybdenum (Fig.3.4). Subsequently,
this OH ligand must be shuttled back towards the activated (radical or carbocation)
hydrocarbon intermediate. This transition state would therefore constitute either a reaction of
two radicals (hydroxyl and alkyl) or a reaction of negatively charged OH- and a carbocation.
Finally, the oxidative half cycle of the reaction is followed by a series of electron transfers
toward the external electron acceptor reoxidizing the enzyme active center.
Fig. 3.4: Hypothetical variants of reaction mechanisms catalyzed by EBDH.
a) heterocyclic C-H cleavage, b) homolytic C-H cleavage. 159
Small model complexes were computed by Dmitriy Bykov. Results for small model
complexes derived from a protein crystal structure, where some atoms are fixed, are compared
with free complexes with and without imidazole to estimate geometrical and chemical
influences on the reactivity of the active site (Table. 3.1). The computed relative energies
indicate that the intermediate and product formation is more favourable in the presence of the
protein´s geometrical constraints. On the other hand there is no big difference (~3 kcal/mol)
among the model complexes with and without imidazole where all the atoms are free to move.
Table 3.1: Computed energy barriers [kcal/mol] relative to the educt substrate complex for
hydroxylation of ethyl benzene by small models of EBDH.
Small model complexes
SCRFa
SCRFb
SCRFc
Educt-substrate complex
0.0
0.0
0.0
Intermediate complex
34.0
47.9
44.4
Product complex
-7.7
13.4
16.4
Where, a = constraint containing complexes derived from the protein X-ray crystal structure, b = Free
complexes without imidazole, c = Free complexes containing imidazole.
36
Chapter 3-Ethylbenzene Dehydrogenase
Fig. 3.5: Optimized geometries of small model complexes for the EBDH mechanism.
In the following a detailed density functional theory (DFT) study of the reaction mechanism
for the hydroxylation of ethylbenzene is presented. Active site model complexes are derived
from the protein X-ray crystal structure (PDB 2IVF).155 Amino acid residues which may take
part in the mechanism are included. For comparison models with protonation of His192,
Lys450, Asp223 and a model without protonation are considered.
2. Computational Details
All the active site model geometries were optimized using Gaussian 09 with the hybrid
density functional B3LYP123 and the LANL2DZ basis set124,125,126,127 augmented by
polarization functions on sulfur atoms (ζ = 0.421).128 The self-consistent field (SCF)160
method was used with IntRep for the SCF procedure to account for integral symmetry and
NoVaracc for full integral accuracy. Whenever there was an SCF convergence problem the
QC161 option was used which involves linear searches when far from convergence. The
starting geometries for transition state (TS) searches were generated by shortening and
lengthening of forming and breaking bonds, respectively. These geometries were then preoptimized freezing the atoms that dominate the TS vector. Frequency calculations were
performed on these pre-optimized geometries to obtain accurate internal force constants. For
the frequency calculations #P was specified in the route section to produce some additional
output. IOP (7/33=1) was used to get the internal force constants which were then helpful for
the TS search. The optimized TS geometries were then slightly modified to generate the
starting geometries for the geometry optimization of educt substrate complexes and
intermediate complexes (from TS1) or intermediate complexes and product complexes (from
37
Chapter 3-Ethylbenzene Dehydrogenase
TS2). Single point energies were computed with the B3LYP functional and the StuttgartDresden effective core potential basis set (SDD)129,130 augmented by polarization functions for
all atoms except Mo, W and H (ζ =0.600, 1.154, 0.864, and 0.421 for C, O, N, and S,
respectively).128 Self-consistent reaction field (SCRF) computations were performed on the
optimized geometries to model the protein surrounding the active site by a conductor like
polarizable continuum method (CPCM)131 as implemented in Gaussian 09.132,133 Default
Gaussian 03 parameters were used for the evaluation of solute-solvent dispersion and
repulsion interaction energies,134,135 and solute cavitation energy variations.136 The molecular
cavity was specified using a minimum radius (RMin) of 0.5Ǻ and an overlap index (OFac) of
0.8.137
3. Active Site Models
Active site model complexes were derived from the protein X-ray crystal structure (PDB
2IVF).155 For the model complexes, the active site Mo coordinated with the two metallopterin
ligands (bisMGD), Asp223, Act1978 (acetate) group and nearby His192 were considered. Active
site residues, Trp87, Cys86, Pro88, Pro191, Val193, Gly222, Thr224, Phe446, Gly445, Ser447, Lys450,
Ala449, Ser151 were also considered as they may influence the catalytic reaction due to their
proximity to the metal center. MGD ligands were truncated at the pyrazine ring, Act1978 to
oxygen while Cys86, Pro191, Val193, Gly222, Thr224, Phe446, Gly445, Ser447, Ala449, Ser151 were
truncated at the alpha (α) carbon atoms. Pro88 was also truncated at the alpha (α) carbon atom
but its ring was left intact. Hydrogen atoms were added manually. During all the calculations,
alpha (α) carbon atoms of Cys86, Pro88, Pro191, Val193, Gly222, Thr224, Gly445, Ser447, Ala449,
and Ser151 were kept fixed at their crystal structure position to mimic the steric constraints by
the protein matrix. The carbon atoms bonded with the dithiolene moieties of MGD ligands
were also kept fixed.
First, hydrogen atoms were optimized applying one negative overall charge (assuming Mo at
the +VI oxidation state), keeping all heavy-atoms fixed at their positions. The resulting
geometries served as starting geometries for the generation of input geometries for relevant
structures for computing the mechanism for hydroxylation of ethylbenzene.
38
Chapter 3-Ethylbenzene Dehydrogenase
4. Results
The reaction for the hydroxylation of ethylbenzene by EBDH starts with the oxidized (MoVI)
enzyme. The EBDH active site center contains nearby residues, His192, Lys450 and Asp223,
which seems to be in the best position to involve in the hydroxylation of ethylbenzene. For
comparison, the active site centers are modeled in several forms; a non-protonated model and
models having protonation at His192, Lys450 or Asp223.
In all the optimized geometries for the non-protonated, protonated His192, protonated Lys450
and protonated Asp223 model complexes, hydrogen bonds are formed between Lys450, Asp223
and Gly222 as well as between His192 and the O1 atom of the oxo group ligated to the Mo atom
(Fig 3.6). These hydrogen bonds help in the stabilization of the active site center and the
hydroxylation of ethylbenzene. In the optimized geometries of non-protonated, protonated
His192 and protonated Lys450 model EBDH reaction mechanisms, hydrogen atoms of terminal
nitrogen atom of Lys450 form hydrogen bonds with the oxygen atom attached to the α carbon
atom of Gly222 and with the oxygen atom of Asp223 as well as between His192 and the O1 atom.
These hydrogen bonds are present in all the optimized complex geometries except the
optimized product complexes (P and L-P) where His192-O1 hydrogen bonds are missing (Fig.
3.7, 3.8 and 3.9).
In the optimized geometries of protonated Asp223 model EBDH reaction mechanism, one
carbonyl oxygen atom of Gly222 forms hydrogen bonds with a NH2 proton of Lys450 and the
proton of Asp223 while the His192-O1 hydrogen bond is present in all the optimized geometries
except A-E, A-ES, A-TS2’ and A-P (Fig. 3.10).
Fig. 3.6: The part of the active site of optimized model EBDH geometries chosen for the
graphical representation of optimized complex geometries.
39
Chapter 3-Ethylbenzene Dehydrogenase
1. Non-protonated Complexes:
Ø Optimized oxidized active site model complex E:
The oxidized active site model complex (E) of EBDH was geometry optimized where the
oxidation state of molybdenum is VI and the overall charge is -1. The optimized geometry is
distinctively different in geometrical parameters of the coordination site of the metal center in
comparison to the protein X-ray crystal structure of EBDH154. Optimized data show that the
dithiolenes are twisted more against each other as the S1-S2-S3-S4 dihedral angle changes from
-4.9˚ to 18.7˚ (Table 3.2), i.e. the coordination geometry is distorted trigonal prismatic. Bond
distances between Mo and dithiolene sulfurs, S are also increased from ~2.370 Ǻ to ~2.451 Ǻ
(Table 3.2).
Ø Optimized educt-substrate complex ES:
Geometry optimization of the educt-substrate model complex ES shows that the ethylbenzene
is loosely bound to the active site center where the methylene group of the substrate molecule
is facing the oxo ligand of the metal. The computed reaction energy for the ES complex
formation is exothermic for the gas phase model, -3.7 kcal/mol, while it is endothermic for the
continuum model, 12.0 kcal/mol with respect to separate substrate and educt complex E
(Table 3.6). Optimized data shows that there is no considerable change in geometrical
parameters of the active site relative to the oxidized active site model complex (E). Slight
decrease in the S1-S2-S3-S4 dihedral angle (from 18.7˚ to 18.0˚) and in the Mo-S distances
(from ~2.451 Ǻ to ~2.448 Ǻ) are observed.
In the optimized ES complex, pro-S-hydrogen (Hs) of C1 of ethylbenzene substrate is at a
distance of 2.525 Ǻ from the metal bound oxygen (O1) at the active site. The C1-Hs bond
distance is 1.099 Ǻ and the dihedral angle between the ethyl group and the benzene ring of
ethylbenzene is 68.9˚ (Table 3.2).
Hydroxylation of ethylbenzene is a two-step process; in the first step (TS1) the C1-Hs bond of
ethylbenzene breaks resulting in the formation of an intermediate having a metal bound
hydroxide and a carbocation or radical substrate. The second step (TS2) involves the rebound
of hydroxide from the metal center to the substrate intermediate resulting in the reduction of
Mo and the formation of (S)-1-phenylethanol (hydroxylized substrate).
40
Chapter 3-Ethylbenzene Dehydrogenase
Ø Optimized H-transfer transition state complexes TS1 and TS1’:
For the hydroxylation of ethylbenzene, the cleavage of the C1-Hs bond is essential which
could be by a two electron transfer step or two one electron transfer steps, i.e. heterolytic
(singlet) or homolytic (triplet) cleavage of C1-Hs bond, respectively. The activation of
ethylbenzene appears to be possible when the substrate molecule approaches the oxo ligand
with its methylene group. Two model complexes were optimized for the H-transfer transition
state considering these two types of cleavages.
Geometrical optimization of transition state model complex TS1 for heterolytic C1-Hs bond
cleavage shows a decrease in the S1-S2-S3-S4 dihedral angle (from 18.0˚ to 11.3˚) and in the
Mo-S bond distances (from ~2.448 Ǻ to ~2.430 Ǻ) relative to the optimized ES geometry
(Table 3.2). Elongated bond distances are observed for the Mo-O1 (from 1.767 Ǻ to 1.911 Ǻ)
and the C1-Hs bond of ethylbenzene (from 1.099 Ǻ to 1.492 Ǻ). The O1 and Hs distance is
reduced from 2.525 Ǻ to 1.120 Ǻ. The dihedral angle between the ethyl group and the
benzene ring of ethylbenzene is reduced from 68.9˚ to 9.5˚. The energy barrier for the
heterolytic C1-Hs bond cleavage and H-transfer from the ethylbenzene to the O1 atom is 28.5
kcal/mol in the gas phase and 46.9 kcal/mol in the continuum (Table 3.6). It is the rate
limiting step in the ethylbenzene hydroxylation.
In the optimized transition state model complex TS1’ for the homolytic C1-Hs bond cleavage
similar changes are observed in the S1-S2-S3-S4 dihedral angle (from 18.0˚ to 10.3˚) and in the
Mo-S bond distances (~2.448 Ǻ to ~2.426 Ǻ). Elongated Mo-O1 (from 1.767 Ǻ to 1.912 Ǻ)
and C1-Hs (from 1.099 Ǻ to 1.508 Ǻ) bonds are computed. The distance between O1 and Hs is
reduced from 2.525 Ǻ to 1.116 Ǻ when comparison is made to the optimized ES geometry.
The dihedral angle between the ethyl group and the benzene ring of ethylbenzene is reduced
from 68.9˚ to 8.8˚ (Table 3.2). The energy barrier for the homolytic C1-Hs bond cleavage and
H-transfer from the ethylbenzene to the O1 atom is 27.1 kcal/mol in the gas phase and 42.3
kcal/mol in the continuum (Table 3.6). This makes the homolytic C1-Hs bond cleavage
somewhat (~4 kcal/mol) more favorable as compared to heterolytic C1-Hs bond cleavage.
The educt substrate complex (ES) and the intermediate complex (I and I’) starting geometries
for geometry optimization were generated from these optimized transition state geometries.
Slight reduction of Mo-O1 and C1-Hs bonds together with an elongation of the O1-Hs distance
gave the starting geometry for the ES complex. On the other hand a slight elongation of the
41
Chapter 3-Ethylbenzene Dehydrogenase
Mo-O1 and C1-Hs bonds and reduction of the O1-Hs distance were performed to generate the
starting geometry for the intermediate complexes I and I’.
Ø Optimized intermediate complexes I and I’:
The heterolytic C1-Hs bond cleavage results in the formation of carbocation intermediate I
while the homolytic C1-Hs bond cleavage results in the formation of bi-radical type
intermediate I’.
The computed reaction energy for the formation of carbocation intermediate complex I is
endothermic with respect to separate substrate and educt (E) complex, 26.8 kcal/mol in gas
phase and 43.5 kcal/mol in the continuum (Table 3.6). Geometry optimization of I shows a
decrease in the S1-S2-S3-S4 dihedral angle (from 11.3˚ to 5.5˚) and in the Mo-S bond distances
(from ~2.43 Ǻ to ~2.407 Ǻ). The Mo-O1 and C1-Hs distances are elongated from 1.911 Ǻ to
2.029 Ǻ and 1.492 Ǻ to 3.60 Ǻ, respectively when comparison is made to the optimized TS1
geometry. The O1-Hs bond is reduced from 1.120 Ǻ to 0.983 Ǻ. The dihedral angle between
the ethyl group and the benzene ring of ethylbenzene is reduced from 9.5˚ to 4.5˚ (Table 3.2).
The computed reaction energy for the formation of radical intermediate, I’ as a result of
homolytic C1-Hs bond cleavage, is also endothermic (10.6 kcal/mol in the gas phase and 23.3
kcal/mol in the continuum (Table 3.6)) but less endothermic as compared to the energy for the
heterolytic C1-Hs bond cleavage. Optimized data shows the decrease in S1-S2-S3-S4 dihedral
angle (from 10.3˚ to 7.7˚) and in the Mo-S distances (from ~2.426 Ǻ to ~2.407 Ǻ), elongated
Mo-O1 (from 1.912 Ǻ to 2.014 Ǻ) and C1-Hs (1.508 Ǻ to 3.677 Ǻ) distances with respect to
optimized TS1’ geometry. The O1-Hs bond is reduced from 1.116 Ǻ to 0.983 Ǻ. The dihedral
angle between the ethyl group and the benzene ring of ethylbenzene is reduced from 8.8˚ to
0.0˚ (Table 3.2). The formation of the bi-radical type intermediate is (~20 kcal/mol) more
favorable as compared to the carbocation intermediate.
In order to generate starting geometries for the OH-transfer transition state complexes (TS2
and TS2’) O1-Hs was slightly reoriented, to facilitate the transfer of hydroxide from the metal
to the C1 of radical or carbocation type substrate intermediate, and the O1-C1 distance was
reduced in the optimized intermediate complex (I and I’).
Ø Optimized OH-transfer transition state complexes TS2 and TS2’:
The second step of the ethylbenzene hydroxylation involves the rebound of hydroxide from
the metal center to the carbocation or radical intermediate substrate resulting in the reduction
42
Chapter 3-Ethylbenzene Dehydrogenase
of Mo from oxidation state VI to IV and the formation of (S)-1-phenylethanol (hydroxylized
substrate).
The energy barrier for the transition state TS2 where an O1Hs anion is transferred from Mo to
the C1 of the carbocation intermediate is 32.0 kcal/mol in the gas phase and 45.1 kcal/mol in
the continuum with respect to separate substrate and educt (E) complex (Table 3.6). Geometry
optimization data shows a further decrease in the S1-S2-S3-S4 dihedral angle from 5.5˚ to -0.1˚
relative to optimized I geometry. Slight changes are found in the Mo-S and O1-Hs bond
distances. The Mo-O1Hs bond is elongated from 2.029 Ǻ to 2.102 Ǻ. The distance between
the C1 and O1Hs is decreased from 3.60 Ǻ to 2.415 Ǻ. The dihedral angle between the ethyl
group and the benzene ring of ethylbenzene is increased from 4.5˚ to 18.0˚ (Table 3.2).
The energy barrier for the transition state TS2’ where O1Hs is transferred from Mo to C1 of
the radical type intermediate is 35.1 kcal/mol in the gas phase and 54.6 kcal/mol in the
continuum (Table 3.6). Optimized data shows a change in the S1-S2-S3-S4 dihedral angle
(from 7.7˚ to -18.3˚) and in the Mo-S distances (from ~2.407 Ǻ to 2.453 Ǻ) when compared
with optimized I’ geometry. The Mo-O1Hs bond is elongated from 2.014 Ǻ to 2.181 Ǻ. A
slight change is found in the O1-Hs bond distance. The distance between the C1 and O1Hs is
decreased from 3.677 Ǻ to 2.290 Ǻ. The dihedral angle between the ethyl group and the
benzene ring of ethylbenzene is increased from 0.0˚ to 5.6˚ (Table 3.2).
Comparison between the energy barriers for the transition states TS2 and TS2’ shows that
TS2 is more favorable. So, the second step may involve the abstraction of a second electron
from the radical type intermediate (converting it to a carbocation intermediate) and then the
rebound of the hydroxide (O1Hs) group from the Mo to the carbocation intermediate resulting
in the formation of (S)-1-phenylethanol.
The starting geometry for the geometry optimization of product complex P was generated by
the slight modification in the optimized TS2 geometry, i.e. slight elongation of the Mo-O1Hs
and reduction of the C1-O1Hs distances.
Ø Optimized product complex P:
The final step in the hydroxylation of ethylbenzene is the transfer of O1Hs from the Mo to the
C1 of carbocation intermediate resulting in the formation of (S)-1-phenylethanol. The
computed reaction energy for the product bound complex P is slightly exothermic, -1.2
kcal/mol in the gas phase and endothermic in the continuum, 17.5 kcal/mol (Table 3.6)
43
Chapter 3-Ethylbenzene Dehydrogenase
relative to separate substrate and educt complex E. Geometry optimization shows a slight
change in the S1-S2-S3-S4 dihedral angle (from -0.1˚ to -1.7˚) and in the Mo-S distances (from
~2.403 Ǻ to ~2.395 Ǻ) when comparison is made with the optimized TS2 geometry. The
Mo-O1Hs distance is increased from 2.102 Ǻ to 2.242 Ǻ while the C1-O1Hs distance is
reduced from 2.415 Ǻ to 1.506 Ǻ. The dihedral angle between the ethyl group and the
benzene ring of hydroxyl-ethylbenzene is reduced from 18.0˚ to -10.8˚ (Table 3.2).
Table 3.2: Geometrical parameters of the optimized model complexes of the reaction mechanism for
the non-protonated EBDH model.
Mo-S1 (Ǻ)
X
2.369
E
2.432
ES
2.431
TS1
2.416
I
2.395
TS2
2.383
P
2.367
Homolytic
Cleavage/Triplet
TS1’
I’
TS2’
2.401
2.389 2.413
Mo-S2 (Ǻ)
2.427
2.439
2.437
2.446
2.430
2.408
2.423
2.443
2.434
2.467
Mo-S3 (Ǻ)
2.291
2.411
2.413
2.42
2.397
2.399
2.394
2.414
2.402
2.452
Mo-S4 (Ǻ)
2.391
2.520
2.511
2.438
2.407
2.423
2.397
2.444
2.404
2.479
-4.7
18.7
18.0
11.3
5.5
-0.1
-1.7
10.3
7.7
-18.3
Mo-O1 (Ǻ)
2.18
1.762
1.767
1.911
2.029
2.102
2.242
1.912
2.014
2.181
O1…Hs (Ǻ)
-
-
2.525
1.120
0.983
0.984
0.977
1.116
0.983
0.985
-
-
1.099
1.492
3.60
2.415
1.506
1.508
3.677
2.290
-
-
68.9
9.5
4.5
18.0
-10.8
8.8
0.0
5.6
Heterolytic Cleavage/Singlet
Non-prot
Dihedral,
S1-S2-S3-S4(˚)
C1…Hs/C1…O1
(Ǻ)
Ethyl-benzene
ring dihedral (˚)
Where, X = protein X-ray crystal structure data, E = educt complex, ES = educt-substrate complex,
TS1 = H-transfer transition state complex, I = intermediate complex,
TS2 = OH-transfer transition state, P = Product bound complex.
44
Chapter 3-Ethylbenzene Dehydrogenase
Fig. 3.7: Optimized geometries for the non-protonated EBDH reaction mechanism.
45
Chapter 3-Ethylbenzene Dehydrogenase
2. Protonated His Complexes:
Ø Optimized oxidized model complex H-E:
The oxidized active site model complex (H-E) of His192 protonated EBDH was geometry
optimized where the oxidation state of molybdenum is VI and the overall complex is neutral.
The optimized geometry is distinctively different in geometrical parameters of the metal
coordination site, in the same way as the optimized non-protonated E geometry, when
comparison is made to the protein X-ray crystal structure of EBDH.154 Optimized data show
that the dithiolenes are twisted more against each other as the S1-S2-S3-S4 dihedral angle
changes from -4.9˚ to 17.3˚, i.e. the coordination geometry is distorted trigonal prismatic.
Bond distances between Mo and dithiolene sulfurs, S are also increased from ~2.370 Ǻ to
~2.441 Ǻ (Table 3.3). The proton of protonated His192 (HHis) is making hydrogen bond
(distance is 1.360 Ǻ) with the O1 attached to the Mo while the HHis-NHis bond distance is
1.124 Ǻ (Fig 3.8).
Ø Optimized educt-substrate complex H-ES:
The computed reaction energy for the H-ES complex formation is exothermic for the gas
phase model, -4.8 kcal/mol, while it is endothermic for the continuum model, 16.0 kcal/mol
relative to separate substrate and educt (H-E) complex (Table 3.6). Optimized data shows that
there is no considerable change in geometrical parameters of the active site relative to the
oxidized active site model complex H-E. The S1-S2-S3-S4 dihedral angle is 16.8˚ and the MoS distance is ~2.439 Ǻ (Table 3.3). The O1-HHis (a hydrogen bond) and HHis-NHis bond
distances are 1.408 Ǻ and 1.103 Ǻ, respectively.
In the optimized H-ES complex, pro-S-hydrogen (Hs) of C1 of ethylbenzene substrate is at a
distance of 2.375 Ǻ from the metal bound oxygen (O1) at the active site. The C1-Hs bond
distance is 1.098 Ǻ and the dihedral angle between the ethyl group and the benzene ring of
ethylbenzene is -13.2˚ (Table 3.3).
Ø Optimized H-transfer transition state complexes H-TS1 and H-TS1’:
Geometrical optimization of transition state model complex H-TS1 for heterolytic C1-Hs bond
cleavage shows slight reductions in the S1-S2-S3-S4 dihedral angle (from 17.1˚ to 12.4˚) and in
the Mo-S bond distances (from ~2.440 Ǻ to ~2.420 Ǻ). Elongated bond distances are
observed for the Mo-O1 (from 1.803 Ǻ to 1.958 Ǻ) and the C1-Hs bond of ethylbenzene (from
1.098 Ǻ to 1.493 Ǻ). The O1 and Hs distance is reduced from 3.623 Ǻ to 1.134 Ǻ. The
46
Chapter 3-Ethylbenzene Dehydrogenase
dihedral angle between the ethyl group and the benzene ring of ethylbenzene is reduced from
85.2˚ to 6.5˚ with respect to the optimized H-ES complex geometry (Table 3.3). The O1-HHis
(a hydrogen bond) distance is reduced from 1.408 Ǻ to 1.308 Ǻ while the HHis-NHis bond
distance is increased from 1.103 Ǻ to 1.169 Ǻ. The energy barrier for the heterolytic C1-Hs
bond cleavage and H-transfer from the ethylbenzene to the O1 atom is 24.3 kcal/mol in the gas
phase and 46.1 kcal/mol in the continuum relative to separate substrate and educt (H-E)
complex (Table 3.6). It is the rate limiting step in the ethylbenzene hydroxylation.
Optimized data for the transition state model complex H-TS1’ for homolytic C1-Hs bond
cleavage shows that, in contrast to the optimized H-TS1 geometry, the proton (HHis) is
transferred from the protonated His192 to the O1 resulting in the formation of a water ligand
coordinated to the Mo active site metal. As a result, the O1-HHis distance is reduced from
1.408 Ǻ to 1.073 Ǻ while the HHis-NHis (a hydrogen bond) distance is increased from 1.103 Ǻ
to 1.467 Ǻ. A decrease in the S1-S2-S3-S4 dihedral angle (from 17.1˚ to 8.7˚) and in the Mo-S
bond distances (from ~2.440 Ǻ to ~2.422 Ǻ) is observed when comparison is made to the
optimized H-ES geometry. Elongated Mo-O1 (from 1.803 Ǻ to 2.007 Ǻ) and C1-Hs (from
1.093 Ǻ to 1.394 Ǻ) bonds are computed. The distance between O1 and Hs is reduced from
3.623 Ǻ to 1.205 Ǻ. The dihedral angle between the ethyl group and the benzene ring of
ethylbenzene is reduced from 85.2˚ to 8.5˚ (Table 3.3). The energy barrier for the homolytic
C1-Hs bond cleavage and H-transfer from the ethylbenzene to the O1 atom is 14.0 kcal/mol in
the gas phase and 32.7 kcal/mol in the continuum relative to separate substrate and educt
complex H-E (Table 3.6).
Ø Optimized intermediate complexes H-I and H-I’:
The computed reaction energy for the formation of carbocation intermediate complex H-I is
endothermic with respect to separate substrate and educt (H-E) complex, 21.0 kcal/mol in gas
phase and 41.7 kcal/mol in the continuum (Table 3.6). Geometry optimization of the H-I
complex shows that the proton (HHis) is transferred from the protonated His192 to the O1
attached to Mo resulting in the formation of a Mo coordinated water molecule. As a result, the
O1-HHis distance is reduced from 1.308 Ǻ to 1.092 Ǻ while the HHis-NHis (a hydrogen bond)
distance is increased from 1.169 Ǻ to 1.399 Ǻ. A decrease in the S1-S2-S3-S4 dihedral angle
(from 12.4˚ to 3.5˚) and in the Mo-S bond distances (from ~2.420 Ǻ to ~2.400 Ǻ) are
observed when compared to the optimized H-TS1 geometry. The Mo-O1 and C1-Hs distances
are elongated from 1.958 Ǻ to 2.097 Ǻ and 1.493 Ǻ to 2.675 Ǻ, respectively. The O1-Hs bond
47
Chapter 3-Ethylbenzene Dehydrogenase
is reduced from 1.134 Ǻ to 0.980 Ǻ. The dihedral angle between the ethyl group and the
benzene ring of ethylbenzene is reduced from 6.5˚ to 1.5˚ (Table 3.3).
The computed reaction energy for the formation of radical intermediate, H-I’ as a result of
homolytic C1-Hs bond cleavage, is also endothermic (2.8 kcal/mol in the gas phase and 20.0
kcal/mol for the polarizable continuum computation (Table 3.6) relative to the separate
substrate and educt (H-E) complex) but less endothermic as compared to the computed
energy for the heterolytic C1-Hs bond cleavage. Optimized data shows a decrease in the S1-S2S3-S4 dihedral angle (from 8.7˚ to 4.4˚) and in the Mo-S distances (from ~2.422 Ǻ to ~2.400
Ǻ), elongated Mo-O1 (from 2.007 Ǻ to 2.092 Ǻ) and C1-Hs (1.394 Ǻ to 2.589 Ǻ) distances
with respect to optimized H-TS1’ geometry. The O1-Hs bond is reduced from 1.205 Ǻ to
0.980 Ǻ. The dihedral angle between the ethyl group and the benzene ring of ethylbenzene is
reduced from 8.5˚ to 1.9˚ (Table 3.3). The O1-HHis distance is slightly increased from 1.073 Ǻ
to 1.090 Ǻ while the HHis-NHis (a hydrogen bond) distance is reduced from 1.467 Ǻ to 1.405
Ǻ.
Ø Optimized OH-transfer transition state complexes H-TS2 and H-TS2’:
The optimized H-I and H-I’ geometries were slightly modified to generate the starting
geometries for the optimization of OH-transfer transition state complexes, H-TS2 and
H-TS2’. The O1-HHis distance was slightly increased while the HHis-NHis distance was
reduced. The O1-Hs group was slightly reoriented and the O1-C1 distance was reduced to
facilitate the transfer of hydroxide from the metal to the C1 of the radical or carbocation type
substrate in the optimized (H-I and H-I’) intermediate geometries.
The energy barrier for the transition state (H-TS2) where an O1Hs anion is transferred from
Mo to the C1 of the carbocation intermediate is 29.0 kcal/mol in the gas phase and 48.7
kcal/mol in the continuum (Table 3.6). Geometry optimization data shows no considerable
change in the S1-S2-S3-S4 dihedral angle (from 3.5˚ to 3.2˚), in the Mo-S (from ~2.400 Ǻ to
~2.395 Ǻ) and in the O1-Hs (from 0.980 Ǻ to 0.981 Ǻ) bond distances relative to the
optimized H-I geometry. The Mo-O1Hs bond is elongated from 2.097 Ǻ to 2.230 Ǻ. The
distance between the C1 and O1Hs is decreased from 2.675 Ǻ to 2.233 Ǻ. The dihedral angle
between the ethyl group and the benzene ring of ethylbenzene is slightly decreased from 1.5˚
to 0.4˚ (Table 3.3). The O1-HHis (a hydrogen bond) distance is increased from 1.092 Ǻ to
1.421 Ǻ while the HHis-NHis bond distance is reduced from 1.399 Ǻ to 1.119 Ǻ.
48
Chapter 3-Ethylbenzene Dehydrogenase
The energy barrier for the transition state (H-TS2’) where O1Hs is transferred from Mo to C1
of the radical type intermediate is 36.4 kcal/mol in the gas phase and 52.4 kcal/mol in the
continuum (Table 3.6). Optimized data shows a change in the S1-S2-S3-S4 dihedral angle
(from 4.4˚ to -3.8˚) and in the Mo-S distances (from ~2.400 Ǻ to 2.430 Ǻ) as compared to the
optimized H-I’ geometry. The Mo-O1Hs bond is elongated from 2.092 Ǻ to 2.328 Ǻ. No
significant change is observed in the O1-Hs (from 0.980 Ǻ to 0.988 Ǻ) bond distance. The
distance between the C1 and O1Hs is decreased from 2.589 Ǻ to 2.073 Ǻ. The dihedral angle
between the ethyl group and the benzene ring of ethylbenzene is increased from 1.9˚ to 19.7˚
(Table 3.3). The O1-HHis (a hydrogen bond) distance is increased from 1.090 Ǻ to 1.390 Ǻ
while the HHis-NHis bond distance is reduced from 1.405 Ǻ to 1.132 Ǻ.
Ø Optimized product complex H-P:
The optimized H-TS2 geometry is modified (the O1-C1 distance is slightly reduced) for the
generation of starting geometry for geometry optimization of product bound complex H-P.
The computed reaction energy for the product bound complex H-P is endothermic relative to
separate substrate and educt (H-E) complex, 8.3 kcal/mol in the gas phase and 28.2 kcal/mol
in the continuum (Table 3.6). Geometry optimization shows a change in the S1-S2-S3-S4
dihedral angle (from 3.2˚ to -4.6˚) and in the Mo-S distances (from ~2.395 Ǻ to ~2.385 Ǻ).
The Mo-O1Hs distance is increased from 2.230 Ǻ to 2.669 Ǻ while the C1-O1Hs distance is
reduced from 2.233 Ǻ to 1.518 Ǻ when comparison is made with the optimized H-TS2
geometry. The dihedral angle between the ethyl group and the benzene ring of hydroxyethylbenzene is 49.6˚ (Table 3.3). The O1-HHis distance is increased from 1.421 Ǻ to 1.595 Ǻ
while the HHis-NHis bond distance is reduced from 1.119 Ǻ to 1.057 Ǻ.
49
Chapter 3-Ethylbenzene Dehydrogenase
Table 3.3: Geometrical parameters of the optimized model complexes of the reaction mechanism for
the His192 protonated EBDH.
Homolytic
Cleavage/Triplet
Heterolytic Cleavage/Singlet
Prot-His
X
H-E
H-ES
H-TS1
H-I
H-TS2
H-P
H-TS1’
H-I’
Mo-S1 (Ǻ)
2.369
2.429
2.431
2.407
2.383
2.384
2.329
2.418
2.390
HTS2’
2.386
Mo-S2 (Ǻ)
2.427
2.435
2.431
2.439
2.439
2.416
2.422
2.465
2.452
2.459
Mo-S3 (Ǻ)
2.291
2.409
2.408
2.412
2.390
2.392
2.421
2.399
2.382
2.441
Mo-S4 (Ǻ)
2.391
2.491
2.486
2.421
2.386
2.388
2.369
2.407
2.379
2.432
-4.9
17.3
16.8
12.4
3.5
3.2
-4.6
8.7
4.4
-3.8
Mo-O1 (Ǻ)
2.18
1.807
1.805
1.958
2.097
2.230
2.669
2.007
2.092
2.328
O1…Hs (Ǻ)
-
-
2.375
1.134
0.980
0.981
0.989
1.205
0.980
0.988
-
-
1.098
1.493
2.675
2.233
1.518
1.394
2.589
2.073
-
-
-13.2
6.5
1.5
0.4
49.6
8.5
1.9
19.7
O1-HHis (Ǻ)
-
1.360
1.408
1.308
1.092
1.421
1.595
1.073
1.090
1.390
HHis -NHis (Ǻ)
-
1.124
1.103
1.169
1.399
1.119
1.057
1.467
1.405
1.132
Dihedral,
S1-S2-S3-S4(˚)
C1…Hs/C1…O1
(Ǻ)
Ethyl-benzene
ring dihedral (˚)
Where, X = protein X-ray crystal structure data, H-E = educt complex, H-ES = educt-substrate
complex, H-TS1 = H-transfer transition state complex, H-I = intermediate complex,
H-TS2 = OH-transfer transition state complex, H-P = Product bound complex.
50
Chapter 3-Ethylbenzene Dehydrogenase
Fig. 3.8: Optimized geometries for the His192-protonated EBDH reaction mechanism.
51
Chapter 3-Ethylbenzene Dehydrogenase
3. Protonated Lys Complexes:
Ø Optimized oxidized model complex L-E:
The oxidized active site model complex (L-E) of Lys450 protonated EBDH derived from the
protein X-ray crystal structure was geometry optimized where the oxidation state of
molybdenum is VI and the overall complex is neutral. The optimized geometry is distinctively
different in geometrical parameters of the coordination site of the metal center in comparison
to the protein X-ray crystal structure of EBDH154 as well as to the optimized E and H-E
geometries. The optimized data shows that the dithiolenes are twisted more against each other
as the S1-S2-S3-S4 dihedral angle changes from -4.9˚ to 36.0˚. Bond distances between Mo and
dithiolene sulfurs, S also increased from ~2.370 Ǻ to ~2.461 Ǻ (Table 3.4).
Ø Optimized educt-substrate complex L-ES:
The computed reaction energy for the L-ES complex formation is exothermic for the gas
phase model, -8.1 kcal/mol, while it is endothermic for the continuum model, 12.4 kcal/mol
relative to the separate substrate and educt (L-E) complex (Table 3.6). Optimized data shows
that there is no considerable change in geometrical parameters of the active site relative to the
oxidized active site model complex L-E. The S1-S2-S3-S4 dihedral angle is 35.1˚ and Mo-S
distance is ~2.469 Ǻ (Table 3.4).
In the optimized L-ES complex, pro-S-hydrogen (Hs) of C1 of ethylbenzene substrate is at a
distance of 2.502 Ǻ from the metal bound oxygen (O1) at the active site. The C1-Hs bond
distance is 1.098 Ǻ and the dihedral angle between the ethyl group and the benzene ring of
ethylbenzene is -12.7˚ (Table 3.4).
Ø Optimized H-transfer transition state complexes L-TS1 and L-TS1’:
Geometrical optimization of transition state model complex L-TS1 for heterolytic C1-Hs bond
cleavage shows reductions in the S1-S2-S3-S4 dihedral angle (from 35.1˚ to 14.2˚) and in the
Mo-S bond distances (from ~2.469 Ǻ to ~2.430 Ǻ). Elongated bond distances are observed
for the Mo-O1 (from 1.753 Ǻ to 1.899 Ǻ) and the C1-Hs bond of ethylbenzene (from 1.098 Ǻ
to 1.493 Ǻ). The O1 and Hs distance is reduced from 2.502 Ǻ to 1.114 Ǻ. The dihedral angle
between the ethyl group and the benzene ring of ethylbenzene is increased from -12.7˚ to
11.8˚ relative to the optimized L-ES geometry (Table 3.4). The energy barrier for the
heterolytic C1-Hs bond cleavage and H-transfer from the ethylbenzene to the O1 atom is 22.5
52
Chapter 3-Ethylbenzene Dehydrogenase
kcal/mol in the gas phase and 47.5 kcal/mol in the continuum with respect to the separate
substrate and educt (L-E) complex (Table 3.6).
Optimized data for the transition state model complex L-TS1’ for homolytic C1-Hs bond
cleavage shows a decrease in the S1-S2-S3-S4 dihedral angle (from 35.1˚ to 7.4˚) and in the
Mo-S bond distances (from ~2.469 Ǻ to ~2.427 Ǻ). Elongated Mo-O1 (from 1.753 Ǻ to 1.912
Ǻ) and C1-Hs (from 1.098 Ǻ to 1.369 Ǻ) bonds are computed. The distance between O1 and
Hs is reduced from 2.502 Ǻ to 1.223 Ǻ. The dihedral angle between the ethyl group and the
benzene ring of ethylbenzene is increased from -12.7˚ to 16.0˚ as compared to the optimized
L-ES geometry (Table 3.4). The energy barrier for the homolytic C1-Hs bond cleavage and Htransfer from the ethylbenzene to the O1 atom is 30.7 kcal/mol in the gas phase and 36.5
kcal/mol in the continuum (Table 3.6).
Ø Optimized intermediate complexes L-I and L-I’:
The computed reaction energy for the formation of carbocation intermediate complex L-I is
endothermic relative to the separate substrate and educt (L-E) complex, 26.1 kcal/mol in gas
phase and 40.3 kcal/mol in the continuum (Table 3.6). Geometry optimization of L-I shows a
decrease in the S1-S2-S3-S4 dihedral angle (from 14.2˚ to 6.3˚) and in the Mo-S bond distances
(from ~2.430 Ǻ to ~2.403 Ǻ). The Mo-O1 and C1-Hs distances are elongated from 1.899 Ǻ to
2.010 Ǻ and 1.493 Ǻ to 3.794 Ǻ, respectively, as compared to the optimized L-TS1 geometry.
The O1-Hs bond is reduced from 1.114 Ǻ to 0.983 Ǻ. The dihedral angle between the ethyl
group and the benzene ring of ethylbenzene is reduced from 11.8˚ to 1.3˚ (Table 3.4).
The computed reaction energy for the formation of radical intermediate complex L-I’ as a
result of homolytic C1-Hs bond cleavage, is also endothermic (8.1 kcal/mol in the gas phase
and 25.5 kcal/mol when a polarizable continuum model is included (Table 3.6)) but less
endothermic as compared to the energy for the heterolytic C1-Hs bond cleavage. Optimized
data shows no significant change in the S1-S2-S3-S4 dihedral angle (from 7.4˚ to 8.0˚) and a
decrease in the Mo-S distances (from ~2.427 Ǻ to ~2.408 Ǻ), elongated Mo-O1 (from 1.912 Ǻ
to 1.981 Ǻ) and C1-Hs (1.369 Ǻ to 4.532 Ǻ) distances with respect to the optimized L-TS1’
geometry. The O1-Hs bond distance is reduced from 1.223 Ǻ to 0.982 Ǻ. The dihedral angle
between the ethyl group and the benzene ring of ethylbenzene is reduced from 16.0˚ to 3.5˚
(Table 3.4).
53
Chapter 3-Ethylbenzene Dehydrogenase
Ø Optimized OH-transfer transition state complexes L-TS2 and L-TS2’:
The energy barrier for the transition state L-TS2 where an O1Hs anion is transferred from Mo
to the C1 of the carbocation intermediate is 22.0 kcal/mol in the gas phase and 43.0 kcal/mol
in the polarizable continuum model (Table 3.6). Geometry optimization data shows a further
decrease in S1-S2-S3-S4 dihedral angle from 6.3˚ to 3.6˚ relative to the optimized L-I
geometry. Slight changes are found in the Mo-S and O1-Hs bond distances. The Mo-O1Hs
bond is elongated from 2.010 Ǻ to 2.057 Ǻ. The distance between the C1 and O1Hs is
decreased from 3.794 Ǻ to 2.701 Ǻ. The dihedral angle between the ethyl group and the
benzene ring of ethylbenzene is changed from 1.3˚ to -0.7˚ (Table 3.4).
The energy barrier for the transition state L-TS2’ where O1Hs is transferred from Mo to C1 of
the radical type intermediate is 29.9 kcal/mol in the gas phase and 53.4 kcal/mol in the
continuum (Table 3.6). Optimized data shows a change in the S1-S2-S3-S4 dihedral angle
(from 8.0˚ to -17.7˚) and in the Mo-S distances (from ~2.408 Ǻ to 2.446 Ǻ). The Mo-O1Hs
bond is elongated from 1.981 Ǻ to 2.155 Ǻ. No significant change is found in the O1-Hs (from
0.982 Ǻ to 0.985 Ǻ) bond distance. The distance between the C1 and O1Hs is decreased from
4.532 Ǻ to 2.370 Ǻ. No considerable change is observed in the dihedral angle between the
ethyl group and the benzene ring of ethylbenzene (from 3.5˚ to 3.9˚) relative to the optimized
L-I’ geometry (Table 3.4).
Ø Optimized product complex L-P:
The optimized L-TS2 geometry is modified (the O1-C1 distance is slightly reduced) for
generating the starting geometry for geometry optimization of product bound complex L-P.
The computed reaction energy for the product bound complex is exothermic in the gas phase,
-8.7 kcal/mol and endothermic for the continuum model, 12.5 kcal/mol (Table 3.6) relative to
separate substrate and educt (L-E) complex. Geometry optimization shows a slight change in
the S1-S2-S3-S4 dihedral angle (from 3.6˚ to 4.3˚) and in the Mo-S distances (from ~2.399 Ǻ to
~2.391 Ǻ) when comparison is made with the optimized L-TS2 geometry. The Mo-O1Hs
bond distance is increased from 2.057 Ǻ to 2.248 Ǻ while the C1-O1Hs distance is reduced
from 2.701Ǻ to 1.505 Ǻ. The dihedral angle between the ethyl group and the benzene ring of
hydroxy-ethylbenzene is increased from -0.7˚ to 60.4˚ (Table 3.4).
54
Chapter 3-Ethylbenzene Dehydrogenase
Table3.4: Geometrical parameters of the optimized model complexes of the reaction mechanism for
the Lys450 protonated EBDH.
Prot-Lys
Heterolytic Cleavage/Singlet
L-ES L-TS1
L-I
L-TS2
2.439
2.410
2.380
2.382
Mo-S1 (Ǻ)
L-E
2.436
Mo-S2 (Ǻ)
2.427
2.417
2.420
2.451
2.419
2.408
2.421
2.445
2.421
2.453
Mo-S3 (Ǻ)
2.291
2.429
2.427
2.415
2.401
2.393
2.391
2.400
2.408
2.434
Mo-S4 (Ǻ)
2.391
2.567
2.589
2.445
2.412
2.411
2.395
2.444
2.423
2.483
-4.9
36.0
35.1
14.2
6.3
3.6
4.3
7.4
8.0
-17.7
Mo-O1 (Ǻ)
2.18
1.751
1.753
1.899
2.010
2.057
2.248
1.912
1.981
2.155
O1…Hs (Ǻ)
-
-
2.502
1.114
0.983
0.980
0.979
1.223
0.982
0.985
-
-
1.098
1.493
3.794
2.701
1.505
1.369
4.532
2.370
-
-
-12.7
11.8
1.3
-0.7
60.4
16.0
3.5
3.9
Dihedral,
S1-S2-S3-S4(˚)
C1…Hs/C1…O1
(Ǻ)
Ethyl-benzene
ring dihedral (˚)
L-P
2.357
Homolytic Cleavage/Triplet
L-TS1’
L-I’
L-TS2’
2.417
2.381
2.414
X
2.369
Where, X = protein X-ray crystal structure data, L-E = educt complex, L-ES = educt-substrate
complex, L-TS1 = H-transfer transition state complex, L-I = intermediate complex,
L-TS2 = OH-transfer transition state, L-P = Product bound complex.
55
Chapter 3-Ethylbenzene Dehydrogenase
Fig. 3.9: Optimized geometries of Lys-protonated model structures relevant in the EBDH reaction
mechanism.
56
Chapter 3-Ethylbenzene Dehydrogenase
4. Protonated Asp223 Complexes:
Ø Optimized oxidized model complex A-E:
The oxidized active site model complex (A-E) of Asp223 protonated EBDH derived from
protein X-ray crystal structure is geometry optimized where the oxidation state of
molybdenum is VI and the overall complex is neutral. Geometrical optimization shows that
the Asp223 is detached from the Mo while the proton of Asp223 forms a hydrogen bond with
the oxygen atom of Gly222 when comparison is made with the protein X-ray crystal structure
of EBDH. The dithiolenes are not much twisted against each other as the S1-S2-S3-S4 dihedral
angle changes from -4.9˚ to 1.9˚. Bond distances between Mo and dithiolene sulfurs, S are
increased from ~2.370 Ǻ to ~2.416 Ǻ (Table 3.5). The Mo-O1 bond distance is decreased
from 2.18 Ǻ to 1.728 Ǻ.
Ø Optimized educt-substrate complex A-ES:
The computed reaction energy for the A-ES complex formation is exothermic for the gas
phase model, -3.3 kcal/mol, while it is endothermic for the continuum model, 16.5 kcal/mol
relative to the separate substrate and educt (A-E) complex (Table 3.6). Optimized data shows
that there is no considerable change in Mo-S distances relative to the oxidized active site
model complex A-E. The S1-S2-S3-S4 dihedral angle is slightly changed from 1.9˚ to 1.2˚
(Table 3.5).
In the optimized A-ES complex, pro-S-hydrogen (Hs) of C1 of ethylbenzene substrate is at a
distance of 3.478 Ǻ from the metal bound oxygen (O1) at the active site. The C1-Hs bond
distance is 1.100 Ǻ and the dihedral angle between the ethyl group and the benzene ring of
ethylbenzene is 84.0˚ (Table 3.5).
The oxidized model complex (A-E) starting geometry for geometry optimization was
generated from the optimized educt-substrate complex A-ES geometry.
Ø Optimized H-transfer transition state complexes A-TS1 and A-TS1’:
Geometrical optimization of transition state model complex A-TS1 for heterolytic C1-Hs bond
cleavage shows increase in the S1-S2-S3-S4 dihedral angle from 1.2˚ to 11.8˚. No considerable
change is observed for the Mo-S bond distances (from ~2.415 Ǻ to ~2.413 Ǻ). Elongated
bond distances are observed for the Mo-O1 (from 1.732 Ǻ to 1.910 Ǻ) and the C1-Hs of
ethylbenzene (from 1.100 Ǻ to 1.5146 Ǻ) as compared to the optimized A-ES geometry. The
O1 and Hs distance is reduced from 3.478 Ǻ to 1.119 Ǻ. The dihedral angle between the ethyl
57
Chapter 3-Ethylbenzene Dehydrogenase
group and the benzene ring of ethylbenzene is reduced from 84.0˚ to 10.9˚ (Table 3.5). The
energy barrier for the heterolytic C1-Hs bond cleavage and H-transfer from the ethylbenzene
to the O1 atom is 45.6 kcal/mol in the gas phase and 62.2 kcal/mol in the continuum relative
to the separate substrate and educt (A-E) complex (Table 3.6).
Optimized data for the transition state model complex A-TS1’ for homolytic C1-Hs bond
cleavage shows an increase in the S1-S2-S3-S4 dihedral angle (from 1.2˚ to 10.5˚) and in the
Mo-S bond distances (from ~2.415 Ǻ to ~2.420 Ǻ). Elongated Mo-O1 (from 1.732 Ǻ to 1.915
Ǻ) and C1-Hs (from 1.100 Ǻ to 1.419 Ǻ) bonds are computed. The distance between O1-Hs is
reduced from 3.478 Ǻ to 1.179 Ǻ as compared to the optimized A-ES geometry. The dihedral
angle between the ethyl group and the benzene ring of ethylbenzene is reduced from 84.0˚ to
18.9˚ (Table 3.5). The energy barrier for the homolytic C1-Hs bond cleavage and H-transfer
from the ethylbenzene to O1 atom is 36.7 kcal/mol in the gas phase and 47.6 kcal/mol in the
continuum (Table 3.6).
The optimized A-TS1 geometry was modified to generate the starting geometry for the
geometry optimization of educt-substrate complex A-ES while the intermediate complexes,
A-I and A-I’ starting geometries for geometry optimization were generated from the
optimized transition state geometries, A-TS1 and A-TS1’.
Ø Optimized intermediate complexes A-I and A-I’:
The computed reaction energy for the formation of carbocation intermediate complex A-I is
endothermic relative to the separate substrate and educt (A-E) complex, 43.8 kcal/mol in gas
phase and 53.6 kcal/mol in the continuum (Table 3.6). Geometry optimization of A-I shows a
decrease in the S1-S2-S3-S4 dihedral angle (from 11.8˚ to 4.6˚) and in the Mo-S bond distances
(from ~2.413 Ǻ to ~2.399 Ǻ). The Mo-O1 and C1-Hs distances are elongated from 1.910 Ǻ to
2.020 Ǻ and 1.504 Ǻ to 4.060 Ǻ, respectively as compared to the optimized A-TS1 geometry.
The O1-Hs bond is reduced from 1.119 Ǻ to 0.980 Ǻ. The dihedral angle between the ethyl
group and the benzene ring of ethylbenzene is reduced from 10.9˚ to 2.2˚ (Table 3.5).
The computed reaction energy for the formation of radical intermediate A-I’ as a result of
homolytic C1-Hs bond cleavage, is also endothermic (22.8 kcal/mol in the gas phase and 30.4
kcal/mol in the continuum (Table 3.6)). The optimized data shows a decrease in the S1-S2-S3S4 dihedral angle (from 10.5˚ to 5.8˚) and in the Mo-S distances (from ~2.420 Ǻ to ~2.399 Ǻ),
elongated Mo-O1 (from 1.915 Ǻ to 2.011 Ǻ) and C1-Hs (1.419 Ǻ to 4.211 Ǻ) distances
relative to the optimized A-TS1’ geometry. The O1-Hs bond is reduced from 1.179 Ǻ to 0.980
58
Chapter 3-Ethylbenzene Dehydrogenase
Ǻ. The dihedral angle between the ethyl group and the benzene ring of ethylbenzene is
reduced from 18.9˚ to 3.8˚ (Table 3.5).
Ø Optimized OH-transfer transition state complexes A-TS2 and A-TS2’:
The energy barrier for the transition state A-TS2 where an O1Hs anion is transferred from Mo
to the C1 of carbocation intermediate is 47.5 kcal/mol in the gas phase and 64.6 kcal/mol in
the continuum (Table 3.6). Geometry optimization data shows a decrease in S1-S2-S3-S4
dihedral angle (from 4.6˚ to 2.2˚) and in the Mo-S distances (from ~2.399 Ǻ to ~2.389 Ǻ).
The Mo-O1Hs bond is elongated from 2.020 Ǻ to 2.128 Ǻ as compared to the optimized A-I
geometry. The distance between the C1 and O1Hs is decreased from 4.060 Ǻ to 2.565 Ǻ. The
dihedral angle between the ethyl group and the benzene ring of ethylbenzene is slightly
changed from 2.2˚ to 3.7˚ (Table 3.5).
The energy barrier for the transition state A-TS2’ where O1Hs is transferred from Mo to C1 of
the radical type intermediate is 46.8 kcal/mol in the gas phase and 61.6 kcal/mol in the
continuum (Table 3.6). Optimized data shows increase in the S1-S2-S3-S4 dihedral angle (from
5.8˚ to 17.2˚) and in the Mo-S distances (from ~2.399 Ǻ to 2.462 Ǻ) relative to the optimized
A-I’ geometry. The Mo-O1Hs bond is reduced from 2.011 Ǻ to 1.998 Ǻ. The distance between
the C1-O1Hs is decreased from 4.211 Ǻ to 2.277 Ǻ. The dihedral angle between the ethyl
group and the benzene ring of ethylbenzene is increased from 3.8˚ to 5.6˚ (Table 3.5).
Ø Optimized product complex A-P:
The optimized A-TS2 geometry is modified (the O1-C1 distance is slightly reduced) for
generating the starting geometry for geometry optimization of product bound complex A-P.
The computed reaction energy for the A-P complex is endothermic, 12.9 kcal/mol for the gas
phase and 30.3 kcal/mol in the continuum relative to separate substrate and educt (A-E)
complex (Table 3.6). Geometry optimization shows a slight change in the S1-S2-S3-S4 dihedral
angle (from 2.2˚ to 1.9˚) and in the Mo-S distances (from ~2.389 Ǻ to ~2.384 Ǻ). The MoO1Hs bond distance is increased from 2.128 Ǻ to 2.261 Ǻ while the C1-O1Hs distance is
reduced from 2.565 Ǻ to 1.506 Ǻ. The dihedral angle between the ethyl group and the
benzene ring of hydroxyl-ethylbenzene is 60.5˚ (Table 3.5).
59
Chapter 3-Ethylbenzene Dehydrogenase
Table 3.5: Geometrical parameters of the optimized model complexes of the reaction mechanism for
the Asp223 protonated EBDH.
Homolytic
Cleavage/Triplet
Heterolytic Cleavage/Singlet
Prot-Asp
X
A-E
A-ES
A-TS1
A-I
A-TS2
A-P
A-TS1’
A-I’
Mo-S1 (Ǻ)
2.369
2.384
2.381
2.401
2.400
2.398
2.368
2.407
2.392
ATS2’
2.420
Mo-S2 (Ǻ)
2.427
2.394
2.396
2.410
2.408
2.393
2.392
2.432
2.421
2.430
Mo-S3 (Ǻ)
2.291
2.431
2.431
2.386
2.370
2.364
2.370
2.384
2.367
2.451
Mo-S4 (Ǻ)
2.391
2.456
2.451
2.453
2.418
2.401
2.405
2.456
2.417
2.547
-4.9
1.9
1.2
11.8
4.6
2.2
1.9
10.5
5.8
17.2
Mo-O1 (Ǻ)
2.18
1.728
1.732
1.910
2.020
2.128
2.261
1.915
2.011
1.998
O1…Hs (Ǻ)
-
-
3.478
1.119
0.980
0.979
0.979
1.179
0.980
0.980
-
-
1.100
1.504
4.060
2.565
1.506
1.419
4.211
2.277
-
-
84.0
10.9
2.2
3.7
60.5
18.9
3.8
5.6
Dihedral,
S1-S2-S3-S4(˚)
C1…Hs/C1…O1
(Ǻ)
Ethyl-benzene
ring dihedral (˚)
Where, X = protein X-ray crystal structure data, A-E = educt complex, A-ES = educt-substrate
complex, A-TS1 = H-transfer transition state complex, A-I = intermediate complex,
A-TS2 = OH-transfer transition state, A-P = Product bound complex.
60
Chapter 3-Ethylbenzene Dehydrogenase
Fig 3.9: Optimized geometries for model structures with protonated Asp relevant for the EBDH
reaction mechanism.
61
Chapter 3-Ethylbenzene Dehydrogenase
Table 3.6: Computed energy barriers [kcal/mol] relative to the separate substrate and edduct complex
for hydroxylation of ethyl benzene by EBDH models.
Multiplicity
Protonation
site
E
ES
TS1
I
TS2
P
Heterolytic Cleavage/Singlet
Homolytic Cleavage/Triplet
none
His
Lys
Asp
none
His
Lys
Asp
0.0
0.0
0.0
0.0
-
-
-
-
-4.6
-3.7
12.0
21.5
28.5
46.9
16.8
26.8
43.5
23.4
32.0
45.1
-8.3
-1.2
17.5
-5.2
-4.8
16.0
18.2
24.3
46.1
15.1
21.0
41.7
22.0
29.0
48.7
5.1
8.3
28.2
-8.8
-8.1
12.4
18.0
22.5
47.5
18.1
26.1
40.3
15.6
22.0
43.0
-11.8
-8.7
12.5
-4.0
-3.3
16.5
37.8
45.6
62.2
35.0
43.8
58.6
39.4
47.5
64.6
9.2
12.9
30.3
-
-
-
-
19.1
27.1
42.3
1.8
10.6
23.3
29.4
35.1
54.6
10.2
14.0
32.7
-3.1
2.8
20.0
30.4
36.4
52.4
20.7
30.7
36.5
3.2
8.1
25.5
25.0
29.9
53.4
26.6
36.7
47.6
13.0
22.8
30.4
39.4
46.8
61.6
-
-
-
-
//B3LYPa
SDDb
COSMOc
//B3LYPa
SDDb
COSMOc
//B3LYPa
SDDb
COSMOc
//B3LYPa
SDDb
COSMOc
//B3LYPa
SDDb
COSMOc
//B3LYPa
SDDb
COSMOc
Where, none = non-protonated EBDH, His = His192 protonated EBDH, Lys = Lys450 protonated
EBDH, Asp = Asp223 protonated EBDH. E = educt complex, ES = educt-substrate complex,
TS1 = H-transfer transition state complex, I = intermediate complex, TS2 = OH-transfer transition
state complex, P = product complex. a) B3LYP/Lanl2DZ(p), b) B3LYP/SDDp//B3LYP/Lanl2DZ (p),
c) COSMO-B3LYP/SDDp//B3LYP/Lanl2DZ(p) (see Computational details)
5. Discussion
The mechanism of ethylbenzene hydroxylation at the molybdenum cofactor of Aromatoleum
aromaticum EBDH has been investigated by using density functional theory (DFT). The
mechanism involves two transition states; TS1 for the cleavage of the C1-Hs bond of
ethylbenzene and the transfer of Hs from the substrate to metal bound oxygen atom (O1), and
TS2 for the rebound of O1Hs from the metal to the C1 of the substrate intermediate resulting in
the formation of (S)-1-phenylethanol (a hydroxylized substrate) and reduced metal.
Heterolytic and homolytic cleavages of C1-Hs bond of ethylbenzene were considered. The
heterolytic cleavage of the C1-Hs bond (TS1) leads to the formation of a carbocation
intermediate. The rebound of the O1Hs (TS2) anion to the carbocation intermediate leads to
the formation of the hydroxylized substrate. The homolytic cleavage of the C1-Hs bond (TS1’)
leads to the formation of a radical type intermediate. The reaction is followed by the rebound
of a O1Hs (TS2’) radical from the Mo to the radical substrate resulting in the formation of the
hydroxylized substrate.
62
Chapter 3-Ethylbenzene Dehydrogenase
Scheme 3.1: Plot of computed reaction energies (kcal/mol) relative to educt complex vs steps involved
in the different level of computations for the non-protonated EBDH reaction mechanism.
Where, OPT = geometry optimization, SDD = single point energy calculation in the gas phase,
COSMO = single point energy calculation in the continuum.
Three different levels of computation (OPT, SDD and COSMO) were considered for all the
geometries involved in the mechanism. The graphical representation (Scheme 3.1) of different
level of computations for the non-protonated EBDH pathway shows that the computed
relative energies are increased with the level of computations. There is small energy change
from the level of geometry optimization (OPT) to the single point energy calculations in the
gas phase (SDD) but a major difference is evident between the SDD and the polarizable
continuum (COSMO) results. For the OPT and SDD, the computed relative energies for the
educt-substrate ES complex formation are exothermic relative to the separate substrate and
educt (E) complex while it is endothermic for the COSMO (see Scheme 3.1 and Table 3.6).
Most likely the interaction of the two separate entities, substrate and model complex, with a
polarizable continuum is overestimated in comparison with the educt, where the active site
complex is ‘wrapped’ around the substrate. In the COSMO, a solvent accessible surface is
formed around the molecule, so both substrate and the educt (E) complex have their own
solvent surfaces. Dispersion interactions between the substrate and the surrounding molecules
63
Chapter 3-Ethylbenzene Dehydrogenase
are underestimated with the DFT method used and in the substrate educt complex ES only the
outer surface interacts with the polarizable continuum in the COSMO approach. Due to this
reason the energy increases relative to the separate substrate and educt (E) complex from
SDD to COSMO models and more reliable energy profile results from energies relative to the
substrate educt complexes ES, as shown in Scheme 3.3 and Table 3.7.
Table 3.7: Computed energy barriers [kcal/mol] relative to the edduct substrate complex for
hydroxylation of ethyl benzene by EBDH models.
Multiplicity
Protonation
site
ES
TS1
I
TS2
P
Heterolytic Cleavage/Singlet
Homolytic Cleavage/Triplet
none
His
Lys
Asp
none
His
Lys
Asp
0.0
0.0
0.0
0.0
-
-
-
-
26.1
32.2
35.0
21.4
30.5
31.5
28.0
35.7
33.1
-3.7
2.6
5.5
23.5
29.1
30.1
20.5
25.9
25.7
27.3
33.8
32.7
10.5
13.2
12.2
26.7
30.6
35.1
26.9
34.2
27.9
24.4
30.1
30.6
-3.0
-0.6
0.1
41.7
48.9
45.7
39.0
47.1
37.1
43.3
50.8
48.1
13.2
16.2
13.8
23.7
30.9
30.3
6.3
14.3
11.3
34.0
38.8
42.6
15.5
18.9
16.7
2.3
7.6
3.9
35.7
41.2
36.4
29.5
38.8
24.1
12.0
16.2
13.1
33.7
38.0
41.0
30.6
40.0
31.1
16.9
26.1
13.9
43.3
50.2
45.1
-
-
-
-
//B3LYPa
SDDb
COSMOc
//B3LYPa
SDDb
COSMOc
//B3LYPa
SDDb
COSMOc
//B3LYPa
SDDb
COSMOc
//B3LYPa
SDDb
COSMOc
Where, none = non-protonated EBDH, His = His192 protonated EBDH, Lys = Lys450 protonated
EBDH, Asp = Asp223 protonated EBDH. ES = educt-substrate complex, TS1 = H-transfer transition
state complex, I = intermediate complex, TS2 = OH-transfer transition state complex,
P = product complex. a) B3LYP/Lanl2DZ(p), b) B3LYP/SDDp//B3LYP/Lanl2DZ (p),
c) COSMO-B3LYP/SDDp//B3LYP/Lanl2DZ(p) (see Computational details)
In the computed unprotonated EBDH pathway, the energy barrier for the heterolytic cleavage
of C1-Hs bond (TS1) is 35.0 kcal/mol in the polarizable continuum (the rate limiting step).
This heterolytic cleavage leads to the formation of a carbocation intermediate (I) which is 3.4
kcal/mol lower in energy than the TS1 complex. The energy barrier for the rebound of O1Hs
anion to the carbocation intermediate (TS2) is 33.1 kcal/mol in the polarizable continuum
(Table 3.7). The energy barrier for the homolytic cleavage of C1-Hs bond (TS1’) is 30.3
kcal/mol in the polarizable continuum. This homolytic cleavage leads to the formation of a
radical type intermediate (I’) which is 19 kcal/mol lower in energy than TS1’. The energy
barrier for the rebound of O1Hs radical from the Mo to the radical type intermediate substrate
(TS2’) is 42.6 kcal/mol in the polarizable continuum. This is the rate limiting step in the
unprotonated EBDH pathway followed by homolytic cleavage of C1-Hs bond. The energy
required for the formation of hydroxylized substrate containing product complex P is 5.5
64
Chapter 3-Ethylbenzene Dehydrogenase
kcal/mol in the polarizable continuum relative to the educt substrate (ES) complex (Table
3.7).
In the computed protonated His192 EBDH pathway, the energy barrier for the heterolytic
cleavage of the C1-Hs bond (H-TS1) is 30.1 kcal/mol in the polarizable continuum relative to
the educt substrate (H-ES) complex. The carbocation intermediate complex H-I is 4.4
kcal/mol lower in energy than the transition state H-TS1 (Table 3.7). The energy barrier for
the transition state H-TS2 is 32.7 kcal/mol in the polarizable continuum, which corresponds
to the rate limiting step. In the homolytic cleavage pathway, the energy barrier for the H-TS1’
is 16.7 kcal/mol in the polarizable continuum relative to the educt substrate (H-ES) complex.
The radical type intermediate (H-I’) is 12.8 kcal/mol lower in energy than the transition state
H-TS1’. The energy barrier associated with H-TS2’ for the rebound of O1Hs radical from the
Mo to the radical type intermediate substrate is 36.4 kcal/mol in the polarizable continuum,
which is the rate limiting step. The formation of the hydroxylized substrate containing product
complex H-P is endothermic, 12.2 kcal/mol in the polarizable continuum, relative to the educt
substrate (H-ES) complex (Table 3.7).
To evaluate the effect of protonation and non-protonation of His192, energy barriers are
compared in Table 3.8, also with the data reported by Szaleniec et al.162 According to our
results the protonation of His192 is helpful for the hydrogen abstraction as it reduces the
energy barrier for the radical type C1-Hs bond cleavage from 30.3 kcal/mol (TS1’) to 16.7
kcal/mol (H-TS1’) in the presence of a polarizable continuum relative to the educt-substrate
(ES) complexes. However, a small effect of His 192 protonation is observed on the energy
barriers associated with the heterolytic type C1-Hs bond cleavage TS1= 35.0 kcal/mol and
H-TS1=30.1 kcal/mol. Also, the energy barrier for the second step (H-TS2), which is rate
determining, in the prot-His192 pathway is equal (prot-His192 is lower by 0.3 kcal/mol) to that
of the non-prot (TS2) EBDH pathway (Table 3.8, Scheme 3.3).
According to the reported data,162 the radical type C1-Hs bond cleavage is energetically more
favorable than the heterolytic cleavage in both non-prot and prot His192 pathways. The same
effect is observed in our computational results, where TS1’ is ~5 kcal/mol lower in energy
than TS1 and H-TS1’ is ~13 kcal/mol lower in energy than H-TS1. However, comparing the
computed energy barriers with the reported data, the computed energy barrier for H-TS1’ is
~14 kcal/mol lower than the energy barrier reported for the homolytic cleavage of C1-Hs bond,
while the energy barrier for H-TS1 is ~6 kcal/mol lower than reported for the heterolytic C1Hs bond cleavage (see Table 3.8). According to Szaleniec et al. 162 for the non-protonated
65
Chapter 3-Ethylbenzene Dehydrogenase
EBDH, it was not possible to identify a transition state (TS2) associated with the O1Hs
rebound as the optimization of an intermediate species led directly to the product complex.
However, we were able to localize this transition state (TS2) for the O1Hs rebound to the
carbocation intermediate. The energy barrier associated with it (TS2) is very similar to that of
H-TS2, for the protonated His192 (Table 3.8).
Table 3.8: Computed and reported energy barriers [kcal/mol] relative to the educt substrate complex
for hydroxylation of ethyl benzene by the non-protonated, protonated His192 and protonated Lys450
EBDH models.
Protonation
site
TS1
TS2
Heterolytic Cleavage/Singlet
None
His
Lys
a
b
a
b
a
26.1
23.5
26.7
32.2 33.2 29.1
31.1
30.6
35.0 31.0 30.1
36.3
35.1
28.0
27.3
24.4
35.7
33.8
31.7
30.1
33.0
32.7
28.2
30.6
Homolytic Cleavage/Triplet
None
His
a
b
a
b
23.7
15.5
30.9
29.7
18.9
26.1
30.3
29.7
16.7
30.8
34.0
35.7
38.8
27.8
41.2
31.5
42.6
24.9
36.4
27.5
Lys
a
29.5
38.8
24.1
33.7
38.0
41.0
//B3LYP
SDD
COSMO
//B3LYP
SDD
COSMO
Where, a = this work, b = data reported in Ref.162.
The protonated Lys450 EBDH pathway was also considered, as the Lys450 residue seems to be
in a perfect position to take part in the hydroxylation of ethylbenzene. The protonation of
Lys450 is 9.8 kcal/mol lower in energy (including a polarizable continuum model) as
compared to the protonation of His192 (Scheme 3.2). The energy barrier for the heterolytic
C1-Hs bond cleavage (L-TS1) is 35.1 kcal/mol in the polarizable continuum (energy similar to
the energy barriers associated with non-protonated transition state TS1), which is the rate
limiting step. The heterolytic cleavage leads to the formation of a carbocation intermediate
(L-I) which is 7.2 kcal/mol lower in energy than the L-TS1 complex. The energy barrier
associated with the L-TS2 for the rebound of the O1Hs anion to the carbocation intermediate
is 30.6 kcal/mol in the polarizable continuum (Table 3.7). The energy barrier for the
homolytic cleavage of C1-Hs bond (L-TS1’) is 24.1 kcal/mol in the polarizable continuum
relative to the educt substrate (L-ES) complex. The radical type intermediate (L-I’), formed
as a result of homolytic C1-Hs bond cleavage, is 11.0 kcal/mol lower in energy than the
transition state L-TS1’. The energy barrier associated with L-TS2’ for the rebound of the
O1Hs radical from the Mo to the radical type intermediate substrate is 41.0 kcal/mol in the
polarizable continuum, which is the rate limiting step. The formation of the hydroxylized
substrate containing product complex L-P is thermoneutral, 0.1 kcal/mol in the polarizable
continuum, relative to the educt substrate (L-ES) complex (Table 3.7).
66
Chapter 3-Ethylbenzene Dehydrogenase
Scheme 3.2: Plot of computed reaction energies (kcal/mol) relative to His192 protonated educt complex
vs steps involved in the protonated EBDH reaction mechanism.
Although the energy barrier associated with L-TS1’ (i.e. the energy of L-TS1’ relative to
L-ES) for the Lys450 protonated EBDH model is higher in energy than for the His192
protonated EBDH model (i.e. H-TS1’ vs H-ES), the Lys450 protonated EBDH pathway is
more favorable. As shown in Scheme 3.2, L-TS1’ of protonated Lys450 is lower in energy
than the H-TS1’ for protonated His192 and the L-TS2 of prot-Lys450 is about the same energy
as the H-TS1’ of prot-His192. Also, among all the computed product complexes P, H-P, L-P
and A-P, L-P is the only product complex which is thermoneutral (~0.1 kcal/mol) relative to
the educt-substrate complex L-ES (Table 3.7, Scheme 3.2 and 3.3). In all other cases the
product complex formation is considerably endothermic relative to the educt substrate
complex (unprotonated: 5.5 kcal/mol, His192 protonated: 12.2 kcal/mol and Asp223 protonated:
13.8 kcal/mol).
67
Chapter 3-Ethylbenzene Dehydrogenase
Scheme 3.3: Plot of computed reaction energies (kcal/mol) relative to educt-substrate complex vs steps
involved in the non-protonated, prot His and prot Lys EBDH reaction mechanism.
Like His192 and Lys450 residues, Asp223 residue also seems to be in a perfect position to take
part in the hydroxylation of ethylbenzene. So, the protonated Asp223 EBDH pathway was also
considered. The relative energy for the protonation of Asp223 is very close (~0.6 kcal/mol
lower) to that for the His192 (Scheme 3.2). The energy barrier for the heterolytic C1-Hs bond
cleavage (A-TS1) is 45.7 kcal/mol in the polarizable continuum. This heterolytic cleavage
leads to the formation of a carbocation intermediate (A-I) which is only 8.6 kcal/mol lower in
energy than the A-TS1 complex. The energy barrier associated with A-TS2 for the rebound of
O1Hs anion to the carbocation intermediate is 48.1 kcal/mol in the polarizable continuum
(Table 3.7). The energy barrier for the homolytic cleavage of C1-Hs bond (A-TS1’) is 31.1
kcal/mol in the polarizable continuum relative to the educt substrate (A-ES) complex. As a
result of homolytic C1-Hs bond cleavage, a radical type intermediate (A-I’) is formed which is
17.2 kcal/mol lower in energy than the transition state A-TS1’. The energy barrier associated
with A-TS2’ for the rebound of O1Hs radical from the Mo to the radical type intermediate
substrate is 45.1 kcal/mol in the polarizable continuum, which makes it the rate limiting step
68
Chapter 3-Ethylbenzene Dehydrogenase
for this pathway. The formation of hydroxylized substrate containing product complex A-P is
also endothermic, 13.8 kcal/mol in the polarizable continuum, relative to the educt substrate
(A-ES) complex (Table 3.7).
Comparing the computed results of protonated Asp223 EBDH pathway with the other model
EBDH pathways, the prot-Asp223 is the worst as the energy barrier for the C1-Hs bond
cleavage (A-TS1’) is ~7-15 kcal/mol higher than the prot-His192 and prot-Lys450. Also, the
energy barrier for the second step (A-TS2) in the pathway of prot-Asp223 is higher (~3
kcal/mol) in energy than the A-TS2’ which is higher in case of non-prot, prot-His192 and protLys450 EBDH pathways.
Comparing heterolytic and homolytic pathways, both TS1’ and I’ are energetically more
favorable than TS1 and I, respectively, for each protonation state (Scheme 3.1, 3.2 and 3.3).
However, transition state TS2 is lower in energy than transition state TS2’ in all except the
Asp223 protonated model EBDH pathway. The TS2 is also lower in energy than the TS1 in the
non-protonated and Lys450 protonated model pathways. For the His192 and Asp223 protonated
model pathways the energy differences between transition state TS1 and transition state TS2
are also very small (~2.5 kcal/mol) but TS2 is higher in energy than TS1 (Table 3.7, Scheme
3.2).
6. Conclusion
Among the model EBDH pathways Lys450 protonated EBDH offers energetically the best
pathway. The protonation of Lys450 is computed to be 9.8 kcal/mol lower in energy than the
protonation of His192 (protonation of Asp223 is similar to His192). L-TS1 is the rate limiting
step in the heterolytic pathway and it is lower in energy than the transition state H-TS1 of
prot-His192. For the homolytic pathway, the rate limiting step is associated with transiton state
L-TS2’ which is lower in energy than the rate limiting steps (H-TS2 and H-TS2’) of protHis192 (Scheme 3.2). However, the overall lowest barrier pathway results when ionic and
radical pathways are mixed. So, based on the computational results, the mechanism of
ethylbenzene hydroxylation starts with a homolytic C1-Hs bond cleavage (L-TS1’) resulting
in the formation of a radical type intermediate (L-I’) and then in order to continue the reaction
by the O1Hs anion transfer, an electron needs to be transferred to transform the di-radical to
the ionic intermediate. Then the transfer of O1Hs anion from the Mo to the cationic substrate
(L-TS2) results in the formation of product bound complex (L-P).
69
Chapter 3-Ethylbenzene Dehydrogenase
In the meantime a similar mechanism has been presented by Szaleniec et al.162,163 where the
hydroxylation of ethylbenzene, as well as different experimentally active substrates, by the
EBDH enzyme has been investigated by quantum chemical methods. However, only the
unprotonated and protonated His192 model complexes of EBDH were considered in these
studies. The reported computational results point towards the radical type C-H cleavage as the
initial reaction and the rate limiting step. It was also suggested that His192 residue of the active
site is apparently involved in the reaction mechanism but it is most probably in the
unprotonated state. So, non-protonated EBDH enzyme was favoured by Szaleniec et al.162
However, we have found a better possibility where Lys450 is in the protonated state.
Comparing the Lys450 protonated EBDH and reported non-protonated EBDH pathways data,
the transition state L-TS1’ is 5.6 kcal/mol lower in energy than the reported non-protonated
transition state for the cleavage of C1-Hs bond (Table 3.8). According to Szaleniec et al.162 it
was not possible to identify a transition state (TS2) associated with the O1Hs rebound.
However, we were able to localize this transition state (L-TS2) which is 30.6 kcal/mol in the
polarizable continuum (0.4 kcal/mol lower in energy than L-TS1 and 6.5 kcal/mol higher in
energy than the L-TS1’) (Table 3.8). So, according to our results, Lys450 protonated EBDH
enzyme shows the energetically best pathway for the hydroxylation of ethylbenzene.
70
Chapter 4-Acetylene Hydratase
Acetylene Hydratase
1. Introduction
Acetylene hydratase (AH) of Pelobacter acetylenicus is a tungsten (W) containing iron-sulfur
enzyme. It is the only member of the third class of W enzymes, the acetylene hydratase family
besides the aldehyde oxidoreductase family and the formyl-methanofuran dehydrogenase
family. It can also be grouped into the DMSO reductase family of mononuclear Mo/W
enzymes on the basis of protein sequence homology and metal coordination. AH catalyzes a
non-redox reaction, the hydration of acetylene to acetaldehyde as part of an anaerobic
degradation pathway of unsaturated hydrocarbons.98
C2H2 + H2O → CH3CHO (23)
The oxidation state of W is unchanged throughout the catalytic cycle.98 AH is the only
enzyme capable of carrying out this reaction, although a nitrogenase can reduce acetylene to
ethylene.164,165
Based on the studies of biomimetic complexes of AH166 and redox titration,167 it has been
demonstrated that WIV participates in the catalysis of acetylene hydration, whereas WVI is
inactive.167 AH is extremely oxygen-sensitive and its activity is lost irreversibly upon
exposure to air as its [4Fe-4S] cluster is converted to [3Fe-4S].99 Interestingly, the enzyme
needs to be activated by reduction of the W center from WVI to WIV and it requires a strong
reductant for this activation. The [4Fe-4S] cluster is thought to facilitate this activation step.168
Once the W is reduced, no further changes in the oxidation state of metal occur during
catalytic activity. For the stability of the reduced tungstoprotein, bis(dithiolene) coordination
through metallopterin cofactors is proposed to be an obligate requirement.169 The role of the
pterin part of AH may involve provision of structural stability through hydrogen bonding with
the apoprotein and/or participation in the electron flow path between the tungsten and Fe-S
centers during the necessary stages of activation or deactivation.167
The protein X-ray crystal structure of AH100 from Pelobacter acetylenicus reveals a
mononuclear W center in the active site and a nearby iron-sulfur [4Fe-4S] cluster.166 In the
active site W is coordinated by two metallopterin guanine dinucleotide cofactors (MGD), a
sulfur atom of cysteine and an oxygen species, which was assigned to be a water molecule
because of its distance (2.04 Ǻ) from the W center. The location of the [4Fe-4S] cluster is not
far from the W center.100 (Fig. 4.1)
71
Chapter 4-Acetylene Hydratase
Fig. 4.1: Cofactors and active site of AH. (A) The tungsten atom (blue) is coordinated by the dithiolene
groups of both MGD cofactors and the side chain of Cys-141. A water molecule completes the slightly
distorted octahedral geometry. This water is also hydrogen-bonded to Asp-13, a residue adjacent to the
[4Fe-4S] cluster ligand Cys-12. (B) The binding pocket positions an acetylene molecule directly above
the water molecule and Asp-13. (C) Bond distances of 2.04 Å to W and 2.41 Å to the OH atom of
Asp-13 indicate a highly activated water molecule positioned right below a binding pocket for
acetylene. (Pictures taken from reference 100)
The proposed catalytic mechanism of AH depends on the nature of the oxygen ligand bound
to the W. The distance of oxygen from the metal is 2.04 Ǻ which falls in the range expected
for both hydroxo ligands (1.9-2.1 Ǻ) and coordinated water (2.0-2.3 Ǻ). According to Seiffert
et al.100 the two possibilities lead to two different mechanisms that do not take place through
organometallic intermediates. A hydroxo ligand would constitute a strong nucleophile and a
water molecule would constitute an electrophile. In both these mechanisms, Asp13 is assumed
to be protonated and to donate a hydrogen bond to the water/hydroxide molecule attached to
the W.166
Seiffert et al.100 suggest an electrophilic addition mechanism (Scheme 4.1) for the hydration
of acetylene. As the bound water molecule gains a partial positive charge through protonated
Asp13, it directly attacks the triple bond of acetylene as an electrophile.
72
Chapter 4-Acetylene Hydratase
Scheme 4.1: Reaction mechanism of acetylene hydratase suggested by Seiffert et al.100
Density functional theory (DFT) calculations on small models of AH by Antony and Bayse170
(Scheme 4.2) show that the displacement of a water molecule by an acetylene molecule is
exothermic. Based on this result, they suggest the nucleophilic attack of a water molecule at
the acetylene η2 –bound to W to form vinyl alcohol assisted by protonated Asp13.
Scheme 4.2: Reaction mechanism of acetylene hydratase suggested by Antony and Bayse.170
Vincent et al. 171 computed high energy barriers (higher than 40 kcal/mol) for both the
mechanisms with DFT methods and therefore ruled out both. Instead they speculated that the
reaction starts with the displacement of the water molecule, by acetylene bound to the metal
center in an η2 fashion. The reaction proceeds through intermediate vinylidene (W=C=CH2)
and carbene (W=C(OH)CH3) complexes (see Scheme 4.3). However, the energy barriers for
the formation of these intermediates are also quite high (28 and 34 kcal/mol, respectively).171
73
Chapter 4-Acetylene Hydratase
Scheme 4.3: Reaction mechanism of the acetylene hydratase suggested by Vincent et al.171
74
Chapter 4-Acetylene Hydratase
Recently, Himo et al172 performed quantum chemical calculations on considerably larger
models of the active site of AH derived from the protein X-ray crystal structure and proposed
a five step mechanism. This mechanism starts with the displacement of the WIV-bound water
molecule with η2-acetylene in an exothermic step (REACT → INT1). The water molecule,
activated by the ionized Asp13, performs nucleophilic attack on the acetylene resulting in the
formation of a vinyl anion intermediate (INT2) which is stabilized by metal coordination. The
protonated Asp13 then acts as an acid and donates a proton to the vinyl anion generating a
vinyl alcohol intermediate (INT3). This is the rate limiting step (energy barrier of 23
kcal/mol). The two final steps involve tautomerization of vinyl alcohol to acetaldehyde
(PROD) with the help of Asp13 and the W metal center (Scheme 4.4).172
Scheme 4.4: Reaction mechanism of the acetylene hydratase suggested by Himo et al.172
75
Chapter 4-Acetylene Hydratase
Here, we have investigated the two mechanisms for catalysis, proposed by Seiffert et al.,100
which do not take place through organometallic intermediates. The first proposed mechanism,
favoured by Seiffert et al.,100 involves electrophilic attack on acetylene by the water molecule
activated by the nearby Asp13. The OH group of Asp13 forms a hydrogen bond with the
oxygen atom of the water molecule, leading to a partial positive charge at the oxygen atom
that can then act as an electrophile. This water molecule may become sufficiently acidic to
protonate the triple bond of acetylene generating a vinyl cation intermediate. The oxygen
atom of nearby water molecule activated by some base then attacks the vinyl cation
intermediate resulting in the formation of vinyl alcohol.
This vinyl alcohol then tautomerizes to acetaldehyde (Scheme 4.5). Suenobou et al.173
computed the energy barrier for the tautomerization of vinyl alcohol to acetaldehyde with and
without the assistance of a water molecule. He suggests that when the reaction is catalysed by
a water molecule, the energy barrier decreases from 55.8 kcal/mol to 29.6 kcal/mol in the gas
phase. Lledós et al.174 suggests that the intervention of a chain of two water molecules further
reduces the potential energy barrier to 21.8 kcal/mol.
Scheme 4.5: Proposed electrophilic reaction mechanism suggested by Seiffert et al.100
In the second proposed mechanism, a metal bound water molecule is activated by the W
center, which acts as a Lewis acid, generating a W-bound hydroxide ligand and protonated
Asp13. The nucleophilic attack of coordinated hydroxide at the acetylene results in the
formation of vinyl alcohol which then tautomerized to acetaldehyde (Scheme 4.6).
Scheme 4.6: Proposed nucleophilic reaction mechanism suggested by Seiffert et al.100
76
Chapter 4-Acetylene Hydratase
Considering the two proposed mechanisms, DFT calculations were performed on small and
large model complexes of the AH active site designed on the basis of the protein X-ray crystal
structure of AH.100
2. Computational Details
2.1. Small model complexes:
All the small model geometries were optimized using Gaussian 03175 with the density
functional BP86176,177,178 and the LANL2DZ basis set.124,125,126,127 The self-consistent field
(SCF)160 method was used with the IntRep option for the SCF procedure to account for
integral symmetry and NoVaracc for full integral accuracy. Whenever there was an SCF
convergence problem, the QC161 option was used which involves linear searches when far
from convergence. The starting geometries for transition state searches were generated by
shortening and lengthening of forming and breaking bonds, respectively. Single point energies
were computed with the B3LYP123 functional and the Stuttgart-Dresden effective core
potential basis set (SDD)129,130 augmented by polarization functions for all atoms except W
and H (ζ = 0.600, 1.154, 0.864, and 0.421 for C, O, N, and S, respectively).128 In addition selfconsistent reaction field (SCRF) computations were performed on the optimized geometries to
model the protein surrounding the active site by a conductor like polarizable continuum
method (CPCM)131 with a dielectric constant of 4 and a solvent radius of 1.4 Ǻ. The
molecular cavity was specified using a minimum radius (RMin) of 0.5 Ǻ and an overlap index
(OFac) of 0.8.137
2.2. Large model complexes:
All the large model geometries were optimized using Gaussian 03175 with the hybrid density
functional B3LYP123 and the LANL2DZ basis set124,125,126,127 augmented by polarization
function on sulfur atoms (ζ = 0.421).128 The self-consistent field (SCF)160 method was used
with the same parameters as for the small model complexes. The starting geometries for
transition state searches were generated by shortening and lengthening of forming and
breaking bonds, respectively. Single point energies were computed with the same parameters
as for the small model complexes.
2.3. Large model complexes with water molecules:
All the water containing large model geometries were optimized using Gaussian 09 with the
hybrid density functional B3LYP123 and the LANL2DZ basis set124,125,126,127 augmented by
77
Chapter 4-Acetylene Hydratase
polarization function on sulfur atoms (ζ = 0.421).128 The Self consistent field (SCF)160 method
was used with the same parameters as for the small and large model complexes. The starting
geometries for transition state (TS) searches were generated by shortening and lengthening of
forming and breaking bonds, respectively. These geometries were then pre-optimized freezing
the crucial atoms dominating the transition vector. Frequency calculations were performed on
these pre-optimized geometries to get the internal force constants. For the frequency
calculations #P was specified in the route section to produce some additional output. IOP
(7/33=1) was used to get the internal force constants which were then supplied as starting
values in the TS search. Single point energies were computed with the B3LYP functional and
the Stuttgart-Dresden effective core potential basis set (SDD)129,130 augmented by polarization
functions for all atoms except Mo, W and H (ζ =0.600, 1.154, 0.864, and 0.421 for C, O, N,
and S, respectively).128 Self-consistent reaction field (SCRF) computations were performed on
the optimized geometries to model the effect of protein surrounding the active site by the
conductor like polarizable continuum method (CPCM)131 as implemented in Gaussian
09.132,133 In order to make it consistent with the results of small and large model complexes,
the default Gaussian 03 procedure and parameters were used with solute-solvent dispersion
and repulsion interaction energies,134,135 and solute cavitation energy variations.136 The
molecular cavity was specified using a minimum radius (RMin) of 0.5Ǻ and overlap index
(OFac) of 0.8.137
3. Active site Models
Active site model complexes were designed on the basis of the protein X-ray crystal structure
of Pelobacter acetylenicus (PDB-ID: 2E7Z).100
3.1. Small model complexes:
Small active site models, designed on the basis of the protein crystal structure,100 were
considered to identify the most probable reaction mechanism. These models include the W
metal center coordinated with two molybdopterin ligands (MGD), a metal bound water
(Wat1862) molecule, a cysteinate (Cys141) ligand and an additional aspartate (Asp13) residue.
The water molecule (Wat1424), nearby Asp13 and Wat1862, were also considered in case of the
electrophilic reaction mechanism (Scheme 4.7 A). Cys141 was truncated to a H3CS- group,
Asp13 to acetate (CH3COO-) and MGD to 2, 3-dithiolato but-2-ene (enedithiolate). Hydrogen
atoms were added manually. Beta (β) carbon atom of acetate and methyl carbon atoms of the
78
Chapter 4-Acetylene Hydratase
ene-dithiolato ligands were kept fixed during the calculations to their crystal structure
positions to mimic the steric constraints of the protein matrix (Fig. 4.2).
Fig. 4.2: Optimized reduced active site small model of AH.
Atoms labeled (*) were kept fixed at their X-ray crystal structure positions.
3.2. Large model complexes:
From the protein crystal structure,100 it was deduced that Asp13 forms hydrogen bonds to the
oxygen species attached to W as well as to the peptide bond of Cys12 and to the side chain of
Trp179. So, Trp179, Trp293 and Trp472 were included for the large models to account for the
effect of second shell ligands on the energy profile and reaction mechanism. Hydrogen atoms
were added manually. During the optimizations, alpha (α) carbon atoms and nitrogen atoms
attached to the beta (β) carbon atoms of Asp13, Trp179, Trp293 and Trp472 were kept fixed to
their crystal structure positions to mimic the steric constraints by the protein matrix. Nitrogen
attached to the beta (β) carbon atom of Cys141 was also kept fixed. The MGD ligands were
truncated to pyran rings and the oxygen atoms of these pyran rings were kept fixed (Fig. 4.3).
79
Chapter 4-Acetylene Hydratase
Fig. 4.3: Optimized reduced active site large model of AH.
Atoms labeled (*) were kept fixed at their X-ray crystal structure positions.
3.3. Large model complexes with water molecules:
The protein X-ray crystal structure100 shows that there are at least 16 well defined water
molecules in a vestibule directly adjacent to the active site and these molecules may help in
the catalytic activity, the hydration of acetylene to acetaldehyde. So, Wat1209, Wat1212, and
Wat1432 water molecules were considered in the large active site models. Ala137, Met138, Ile113,
Ile142, and Phe611 were also considered to keep the water molecules at their locations as they
form hydrogen bonds to these water molecules. Hydrogen atoms were added manually. Alpha
(α) carbon atoms and nitrogen atoms attached to the β carbon atoms of Asp13, Ile113, Trp179,
Trp472, Phe611; β carbon atom of Ala137, nitrogen atom attached to the β carbon atom of Cys141,
α carbon of Ile142, α and β carbon atoms of Trp293, C4 of Met138, the oxygen atoms of the
pyran rings of dithiolenes were kept fixed to their crystal structure positions during
optimizations to mimic the steric constraints by the protein matrix (Fig. 4.4).
For all the active site model complexes, first, the hydrogen atoms were optimized, applying
two negative charges for the nucleophilic pathway (assuming W at the +IV oxidation state
and Asp13 in the deprotonated form) while one negative charge for the electrophilic pathway
(assuming W at the +IV oxidation state and Asp13 in the protonated form), keeping all the
heavy-atom fixed at their positions. The resulting geometries served as starting geometries
when generating input geometries for the study of the mechanism for acetylene hydratase
(Fig. 4.4). The educt-substrate complexes (ES) and the alcohol product complexes (EP1)
80
Chapter 4-Acetylene Hydratase
starting geometries for geometry optimizations were generated from the optimized transition
state (TS) geometries.
Fig. 4.4: Optimized reduced active site large model (water containing) complex of AH.
Atoms labeled (*) were kept fixed at their X-ray crystal structure positions. Hydrogen atoms were
excluded to get the clear view of the selected active site model.
Scheme 4.7: Schematic description of the mechanism for the acetylene hydration at the Acetylene
hydratase, where A = Electrophilic pathway, B = Nucleophilic pathway, ES =educt-substrate complex,
TS = transition state, EP1 = alcoholic product, EP2 = tautomerized (to aldehyde) product.
81
Chapter 4-Acetylene Hydratase
4. Results
4.1. Small model complexes:
·
Electrophilic pathway (SE):
We started our search for the hydration of acetylene on small model complexes derived from
the protein X-ray crystal structure,100 where Asp13 was considered to be protonated. The Wbound water molecule (Wat1862) is activated by the nearby Asp13 residue and second water
molecule (Wat1424). This activated water molecule (Wat1862) then attacks the acetylene
substrate (Scheme 4.7A). Thus, starting with this activated Wat1862 molecule, transition state
involves the electrophilic attack on the triple bond of acetylene with the simultaneous transfer
of protons among Asp13, Wat1862, Wat1424 and acetylene. The proton from Asp13 residue
(-COOH) is transferred to the Wat1862 molecule and one proton of Wat1862 molecule is
transferred to the alpha carbon atom (Cα or C1) of acetylene. From the Wat1424, one proton is
transferred to Asp13 while its electron donating part (-OH) is transferred to the second carbon
atom (Cβ or C2) of acetylene. This proton shuttle results in the formation of vinyl alcohol
which subsequently may tautomerize to aldehyde (Scheme 4.7A).
Fig 4.5: The chemical structure of the active site model complexes for the electrophilic pathway
derived from the protein X-ray crystal structure of AH.100
Ø Optimized active site model complex SE-E:
The reduced active site model complex SE-E derived from the protein X-ray crystal structure
of AH100 was geometry optimized where oxidation state of tungsten is IV and the overall
charge is -1. The optimized geometry of active site model complex SE-E is distinctively
different in geometrical parameters of the coordination site of the metal center in comparison
to the protein X-ray crystal structure of AH.100 The optimized data shows a reduction in the
82
Chapter 4-Acetylene Hydratase
S1-S2-S3-S4 dihedral angle (from -31.4˚ to -20.8˚) and in the bond distances between tungsten
(W) and ene-dithiolate sulfur atoms (S) (from ~2.442 Ǻ to ~2.415 Ǻ). The W-O1 bond
distance is increased from 2.041 Ǻ to 2.279 Ǻ (Table 4.1).
Ø Optimized educt substrate complex SE-ES:
In the optimizated educt-substrate complex SE-ES, the acetylene is loosely bound to the
active site where a hydrogen bond is formed between the H5 of acetylene and the oxygen
atom (O2) of Wat1424 (see Fig. 4.2). The optimization data shows no considerable change in
the S1-S2-S3-S4 dihedral angle (from -20.8˚ to -20.4˚) and in the W-S bond distances (from
~2.415 Ǻ to ~2.413 Ǻ) when comparison is made with the optimized SE-E geometry. The WO1 distance is increased from 2.279 Ǻ to 2.300Ǻ while the O1-H1 bond is reduced from 1.142
Ǻ to 1.087 Ǻ. The H1-O2 and H7-O1 distances are increased from 1.318 Ǻ to 1.430 Ǻ and
from 1.392 Ǻ to 1.435 Ǻ, respectively (Table 4.1).
In the optimized SE-ES complex, H5 of acetylene is at a distance of 1.994 Ǻ from the O2 of
Wat1424, forming a hydrogen bond. The H1-C1 and the C1-C2 distances are 4.994 Ǻ and 1.236
Ǻ, respectively. The H5-C1-C2 and C1-C2-H6 bond angles of acetylene are 179.8˚ and 179.7˚,
respectively (Table 4.1).
Ø Optimized transition state complex SE-TS:
Geometry optimization of the transition state complex SE-TS shows no considerable change
in the S1-S2-S3-S4 dihedral angle (from -20.4˚ to -19.8˚), in the W-S bond distances (from
~2.413 Ǻ to ~2.412 Ǻ) as well as in the W-O1 (from 2.300 Ǻ to 2.317 Ǻ) and O1-H1 (from
1.087 Ǻ to 1.083 Ǻ) bond distances relative to the optimized SE-ES geometry (Table 4.1).
The H1-C1 and C2-O2 distances are reduced from 4.994 Ǻ to 1.651 Ǻ and 4.274 Ǻ to 2.110 Ǻ,
respectively. The H5-C1-C2 (from 179.8˚ to 139.4˚) and C1-C2-H6 (179.7˚ to 163.8˚) bond
angles are reduced. The energy barrier for the transition state is 28.5 kcal/mol in the
continuum (30.4 kcal /mol in the gas phase) relative to the educt-substrate (SE-ES) complex
(Table 4.3).
The educt substrate (SE-ES) complex and the alcohol product (SE-EP1) complex starting
geometries for geometry optimization were generated from the optimized transition state
(SE-TS) geometry. Slight reduction of W-O1 bond together with the elongation of the H1-C1
and the C2-O2 distances generate the starting geometry for the SE-ES complex. On the other
83
Chapter 4-Acetylene Hydratase
hand a slight elongation of the W-O1 bond and reduction of the H1-C1 and C2-O2 distances
were performed to generate the starting geometry for the SE-EP1 complex.
Ø Optimized alcoholic product complex SE-EP1:
The formation of a vinyl alcohol complex, an initial product of acetylene hydration, is an
exothermic reaction with respect to the SE-ES complex, -36.2 kcal/mol in the continuum
(-33.6 kcal/mol in the gas phase) (Table 4.3). In the optimized alcoholic product (SE-EP1)
complex, the vinyl alcohol is bound to the active site W metal. Geometry optimization of the
SE-EP1 complex shows a slight change in the S1-S2-S3-S4 dihedral angle (from -19.8˚ to 22.5˚) and in the W-S bond distances (from ~2.412 Ǻ to ~2.421 Ǻ) (Table 4.1). The W-O1
bond is reduced from 2.317 Ǻ to 2.118 Ǻ while the O1-H1 distance is increased from 1.083 Ǻ
to 4.892 Ǻ. The C1-C2 distance is increased from 1.269 Ǻ to 1.357 Ǻ indicating the formation
of C=C double bond (~1.33 Ǻ). The C2-O2 bond is reduced from 2.110 Ǻ to 1.401 Ǻ whereas
the O2-H4 bond is broken (distance is increased from 1.022 Ǻ to 1.401 Ǻ) resulting in the
formation of vinyl alcohol. On the other hand the reduction of the H4-O3 distance from 1.597
Ǻ to 1.057 Ǻ and the elongation of O4-H7 distance (from 1.070 Ǻ to 1.613 Ǻ) results in the
restoration of protonated Asp13. Finally, the reduction of O1-H7 (from 1.470 Ǻ to 1.010 Ǻ)
indicates the formation of a water ligand attached to the active site metal. The H5-C1-C2 and
C1-C2-H6 bond distances are reduced from 139.4˚ to 119.4˚ and from 163.8˚ to 123.2˚,
respectively (Table 4.1).
The tautomerized acetaldehyde product complex SE-EP2 starting geometry for geometry
optimization was generated by a slight modification in the vinyl alcohol part of the optimized
SE-EP1 complex geometry. The O2-H3 bond was broken together with the reduction of C1-H3
and C2-O2 distances to generate the starting geometry for the SE-EP2 complex.
Ø Optimized tautomerized product complex SE-EP2:
The computed energy barrier for the tautomerization of vinyl alcohol (C2H3OH) to
acetaldehyde (CH3CHO, without the educt complex) with the assistance of two water
molecules is 20.7 kcal/mol in the continuum (20.3 kcal/mol in the gas) (see Table 4.12). The
computed reaction energy for the tautomerized product complex SE-EP2 is also exothermic,
-10.4 kcal/mol in the continuum (-9.9 kcal/mol in the gas) relative to the SE-EP1 and -46.6
kcal/mol in continuum (-43.5 kcal/mol) relative to the SE-ES complex (Table 4.3). Geometry
optimization shows a slight change in the S1-S2-S3-S4 dihedral angle (from -22.5˚ to -19.3˚).
No considerable change is observed in the W-S bond distances (from ~2.421 Ǻ to ~2.423Ǻ).
84
Chapter 4-Acetylene Hydratase
The elongation of the C1-C2 bond (from 1.357 Ǻ to 1.512 Ǻ which is typical for a C-C single
bond (~1.53 Ǻ)) and the reduction of the C2-O2 bond (1.401 Ǻ to 1.286 Ǻ which shows the
C=O bond) indicates the formation of the aldehyde product (Table 4.1).
Fig. 4.6: Optimized geometries for the small model complexes involved in the electrophilic pathway
of acetylene hydration by AH.
Table 4.1: Geometrical parameters of the optimized model complexes of the electrophilic reaction
mechanism for acetylene hydration by the small model complexes of AH.
X
SE-E
SE-ES
SE-TS
SE-EP1
SE-EP2
W-S1 (Ǻ)
2.432
2.435
2.434
2.433
2.446
2.443
W-S2 (Ǻ)
2.489
2.405
2.404
2.414
2.398
2.395
W-S3 (Ǻ)
2.511
2.401
2.418
2.395
2.419
2.410
W-S4 (Ǻ)
2.336
2.420
2.397
2.404
2.422
2.444
S1-S2-S3-S4 (˚)
-31.4
-20.8
-20.4
-19.8
-22.5
-19.3
W-O1 (Ǻ)
2.041
2.279
2.300
2.317
2.118
2.141
O1-H1 (Ǻ)
-
1.142
1.087
1.083
4.892
-
H1-O2 (Ǻ)
-
1.318
1.430
3.027
-
-
H1-C1 (Ǻ)
-
-
4.994
1.651
1.095
-
C1-C2 (Ǻ)
-
-
1.236
1.269
1.357
1.512
C2-O2 (Ǻ)
-
-
4.274
2.110
1.401
1.286
O2-H4 (Ǻ)
-
1.046
1.047
1.022
1.401
-
H4-O3 (Ǻ)
-
1.522
1.518
1.597
1.057
1.085
O4-H7 (Ǻ)
-
1.109
1.089
1.070
1.613
1.535
H7-O1 (Ǻ)
-
1.392
1.435
1.470
1.010
1.028
H5-C1-C2 (˚)
-
-
179.8
139.4
119.4
-
C1-C2-H6 (˚)
-
-
179.7
163.8
123.2
-
Where, X = protein X-ray crystal structure data,100 SE-E = educt complex, SE-ES = educt-substrate
complex, SE-TS = transition state complex, SE-EP1 = alcoholic product complex,
SE-EP2 = aldehyde product complex
85
Chapter 4-Acetylene Hydratase
·
Nucleophilic Pathway SN:
The second mechanism proposed by Seiffert et al. 100 is the nucleophilic attack (Scheme 4.7B)
of a W-bound water molecule (Wat1862) on the acetylene substrate. DFT calculations for this
reaction mechanism were carried out on small model complexes derived from the protein
X-ray crystal structure, where Asp13 was considered to be in anionic form. The W center,
which acts as a Lewis acid activates water molecule (Wat1862), generates a W-bound
hydroxide and protonated Asp13. The transition state for the nucleophilic attack of this Wbound hydroxide at the alpha carbon atom (Cα or C1) of acetylene also involves the
simultaneous transfer of a proton from the Asp13 to the second acetylene carbon atom (Cβ or
C2) resulting in the formation of vinyl alcohol which subsequently may tautomerize to
aldehyde.
Fig. 4.7: The chemical structure of the active site model complexes for the nucleophilic pathway
derived from the protein X-ray crystal structure of AH.100
Ø Optimized active site model complex SN-E:
The reduced active site model complex SN-E derived from the protein X-ray crystal structure
of AH100 was geometry optimized where the oxidation state of tungsten is IV and the overall
charge is -2. The optimized geometry of active site model complex SN-E is similar in
geometrical parameters of the metal coordination site in comparison to the protein X-ray
crystal structure of AH.100 No considerable change is observed in the S1-S2-S3-S4 dihedral
angle (from -31.4˚ to -27.4˚) and in the bond distances between tungsten (W) and
86
Chapter 4-Acetylene Hydratase
enedithiolene sulfurs (S) (from ~2.442 Ǻ to ~2.443 Ǻ) (Table 4.2). The W-O1 bond distance is
increased from 2.041 Ǻ to 2.257 Ǻ.
Ø Optimized educt substrate complex SN-ES:
In the educt substrate complex SN-ES, a W-bound water molecule is activated by the Wcenter generating a W-bound hydroxide (OH) and protonated Asp13. Geometrical optimization
of the SN-ES complex shows no considerable change in the S1-S2-S3-S4 dihedral angle (from
-27.4˚ to -26.7˚). On the other hand, the W-S bond distances are increased from ~2.443 Ǻ to
~2.467 Ǻ when comparison is made with the optimized SN-E geometry. The W-O1 and H1-O3
distances are reduced from 2.257 Ǻ to 2.075 Ǻ and from 1.821 Ǻ to 1.021 Ǻ, respectively
while the O1-H1 distance is increased from 1.036 Ǻ to 3.736 Ǻ (Table 4.2).
In the optimized SN-ES complex geometry, the acetylene is loosely bound to the active site
where a hydrogen bond is formed between the H3 of acetylene and the oxygen atom (O1) of
W-bound hydroxide. The distance between the O1 -C1 is 2.687 Ǻ, C1-C2 is 1.246 Ǻ and C2-H1
is 2.078 Ǻ. The H3-C1-C2 and C1-C2-H4 bond angles of acetylene are 172.2˚ and 175.4˚,
respectively (Table 4.2).
Ø Optimized transition state complex SN-TS:
Geometry optimization of the transition state complex SN-TS shows no considerable change
in the S1-S2-S3-S4 dihedral angle (from -26.7˚ to -26.3˚) and in the W-S bond distances (from
~2.467 Ǻ to ~2.469 Ǻ) when compared with the optimized SN-ES geometry. The W-O1 (from
2.075 Ǻ to 2.043 Ǻ), O1-C1 (from 2.687 Ǻ to 2.305 Ǻ) and C2-H1 (from 2.078 Ǻ to 1.705 Ǻ)
distances are reduced while the C1-C2 distance is slightly elongated from 1.246 Ǻ to 1.269Ǻ
(Table 4.2). The H3-C1-C2 (from 172.2˚ to 161.1˚) and C1-C2-H4 (from 175.4˚ to 143.9˚) bond
angles of acetylene are reduced. The energy barrier for the formation of the transition state
(SN-TS) complex is 17.0 kcal/mol in the continuum (15.7 kcal/mol in the gas phase) relative
to the SN-ES complex (Table 4.3).
The educt substrate complex SN-ES and the alcohol product complex SN-EP1 starting
geometries for geometry optimization were generated from the optimized transition state
(SN-TS) geometry. Slight reduction of the W-O1 bond together with the elongation of the O1C1 and C2-H1 distances gives the starting geometry for the SN-ES complex. On the other hand
a slight elongation of the W-O1 and O3-H1 bonds and reduction of the O1-C1 and C2-H1
distances were performed to generate the starting geometry for the SN-EP1 complex.
87
Chapter 4-Acetylene Hydratase
Ø Optimized alcoholic product complex SN-EP1:
The formation of the initial product of acetylene hydration, a vinyl alcohol, is an exothermic
step, -29.3 kcal/mol (-31.3 kcal/mol) relative to the SN-ES complex (Table 4.3). In this
alcoholic product complex SN-EP1, the O1H2 part of the vinyl alcohol is bound to the active
site W metal. Geometry optimization of the SN-EP1 complex shows a slight change in the S1S2-S3-S4 dihedral angle (from -26.3˚ to -22.0˚) and in the W-S bond distances (from ~2.466 Ǻ
to ~2.431 Ǻ) with respect to the optimized SN-TS geometry. The W-O1 and H1-O3 distances
are increased from 2.043 Ǻ to 2.314 Ǻ and from 1.065 Ǻ to 2.038 Ǻ, respectively. The O1-C1
(from 2.305 Ǻ to 1.440 Ǻ) and C2-H1 (from 1.705 Ǻ to 1.110 Ǻ) distances are reduced while
the C1-C2 bond is elongated from 1.269 Ǻ to 1.351 Ǻ. The H3-C1-C2 and C1-C2-H4 bond
angles of acetylene are reduced from 161.1˚ to 127.5˚ (a typical H-C=C bond angle is
~121.3˚) and 143.9˚ to 123.8˚, respectively (Table 4.2).
The tautomerized acetaldehyde product (SN-EP2) complex starting geometry for geometry
optimization was generated by a slight modification in the vinyl alcohol part of the optimized
SN-EP1 complex geometry. The O2-H3 bond was broken together with the reduction of C1-H3
and C2-O2 distances to generate the starting geometry for the SN-EP2 complex.
Ø Optimized tautomerized product complex SN-EP2:
The computed reaction energy for the tautomerized product complex is again strongly
exothermic, -50.2 kcal/mol (-54.2 kcal/mol) relative to the SN-ES complex (Table 4.3).
Geometry optimization of the SN-EP2 complex shows a change in the S1-S2-S3-S4 dihedral
angle (from -22.0˚ to -28.8˚) back to roughly the value of the optimized SN-E complex and in
the W-S bond distances (from ~2.431 Ǻ to ~2.467 Ǻ) when comparison is made with the
optimized SN-EP1 geometry. The W-O1 (from 2.314 Ǻ to 2.002 Ǻ) and O1-C1 (from 1.440 Ǻ
to 1.299 Ǻ) bonds are reduced while C1-C2 bond is elongated from 1.351 Ǻ to 1.496 Ǻ.
Fig. 4.8: Optimized geometries for the small model complexes involved in the nucleophilic pathway of
acetylene hydration by AH.
88
Chapter 4-Acetylene Hydratase
Table 4.2: Geometrical parameters of the optimized model complexes of the nucleophilic reaction
mechanism for acetylene hydration by the small model complexes of AH.
X
SN-E
SN-ES
SN-TS
SN-EP1
SN-EP2
W-S1 (Ǻ)
2.432
2.413
2.490
2.495
2.439
2.486
W-S2 (Ǻ)
2.489
2.438
2.437
2.452
2.422
2.446
W-S3 (Ǻ)
2.511
2.505
2.477
2.469
2.427
2.481
W-S4 (Ǻ)
2.336
2.417
2.463
2.459
2.434
2.455
S1-S2-S3-S4 (˚)
-31.4
-27.4
-26.7
-26.3
-22.0
-28.8
W-O1 (Ǻ)
2.041
2.257
2.075
2.043
2.314
2.002
O1-H1 (Ǻ)
-
1.036
3.736
-
-
-
H1-O3 (Ǻ)
-
1.821
1.021
1.065
2.038
-
O1-C1 (Ǻ)
-
-
2.687
2.305
1.440
1.299
C1-C2 (Ǻ)
-
-
1.246
1.269
1.351
1.496
C2-H1 (Ǻ)
-
-
2.078
1.705
1.110
-
H3-C1-C2 (˚)
-
-
174.2
161.1
127.5
-
C1-C2-H4 (˚)
-
-
175.4
143.9
123.8
Where, X = protein X-ray crystal structure data,100 SN-E = educt complex, SN-ES = educt-substrate
complex, SN-TS = transition state complex, SN-EP1 = alcoholic product complex,
SN-EP2 = tautomerized product complex
89
Chapter 4-Acetylene Hydratase
Table 4.3: Computed energies [kcal/mol] relative to the edduct-substrate complex for stationary
points relevant in the hydration of acetylene by small model complexes of AH.
Electrophilic Pathway, SE
Nucleophilic Pathway, SN
//BP86a
ES
0.0
0.0
SDDb
COSMOc
TS
EP1
EP2
30.1
12.9
//BP86a
30.4
15.7
SDDb
28.5
17.0
COSMOc
-27.9
-27.1
//BP86a
-33.6
-31.3
SDDb
-36.2
-29.3
COSMOc
-34.0
-53.8
//BP86a
-43.5
-54.2
SDDb
-46.6
-50.2
COSMOc
Where, ES = educt-substrate complex, TS = transition state complex,
EP1 = alcohol product complex, EP2 = tautomerized product complex.
a) BP86/Lanl2DZ(p), b) B3LYP/SDDp// BP86/Lanl2DZ (p),
c) COSMO-B3LYP/SDDp// BP86/Lanl2DZ(p) (see Computational details)
90
Chapter 4-Acetylene Hydratase
4.2. Large model complexes:
Computed relative energies for the small model complexes indicate that the nucleophilic
reaction pathway is more favorable relative to the electrophilic pathway. Now in order to
identify the most probable reaction mechanism for the hydration of acetylene large AH model
complexes were analyzed considering the surrounding amino acid residues, Trp179, Trp293 and
Trp472, that may take part in the reaction.
·
Electrophilic pathway LE:
Ø Optimized active site model complex LE-E:
The reduced active site large model complex LE-E of AH was geometry optimized where the
oxidation state of tungsten is IV and the overall charge is -1. The optimized geometry of the
active site model complex LE-E is distinctively different in geometrical parameters of the
metal coordination site in comparison to the protein X-ray crystal structure of AH100 as well as
to the optimized SE-E geometry (see Table 4.1 and 4.4). The optimized data shows a
reduction in the S1-S2-S3-S4 dihedral angle (from -31.4˚ to -17.4˚) and in the bond distance
between tungsten (W) and dithiolate sulfur atoms (S) (from ~2.442 Ǻ to ~2.388 Ǻ) (Table
4.4) when comparison is made to the protein X-ray crystal structure of AH. The W-O1 bond
distance is increased from 2.041 Ǻ to 2.182 Ǻ.
Ø Optimized educt substrate complex LE-ES:
Geometry optimization of educt-substrate complex LE-ES shows that Wat1862 is detached
from the W center. The W-O1 bond distance is increased from 2.182 Ǻ to 4.229 Ǻ while no
considerable change is observed in the O1-H1 bond of Wat1862 (from 0.976 Ǻ to 0.979 Ǻ). A
reduction in the S1-S2-S3-S4 dihedral angle (from -17.4˚ to -2.3˚) and in the W-S bond
distances (from ~2.388 Ǻ to ~2.364 Ǻ) is observed when comparison is made with the
optimized LE-E geometry (Table 4.4).
In the optimized LE-ES complex, H5 of acetylene is at a distance of 2.041 Ǻ from the O2 of
Wat1424, forming a hydrogen bond. The H1-C1 and the C1-C2 distances are 2.385 Ǻ and 1.225
Ǻ, respectively. The H5-C1-C2 and C1-C2-H6 bond angles of acetylene are 178.9˚ and 178.7˚,
respectively (Table 4.4).
91
Chapter 4-Acetylene Hydratase
Ø Optimized transition state complex LE-TS:
In the optimized transition state complex LE-TS geometry, a hydronium ion (H3O+) is formed
with the transfer of a proton (H7) from the Asp13 to the O1 of Wat1862 leaving Asp13 an anion
(see Fig 4.9). The optimized data shows a change in the S1-S2-S3-S4 dihedral angle (from -2.3˚
to -12.6˚), in the W-S bond distances (from ~2.364 Ǻ to ~2.350 Ǻ) and in the W-O1 (from
4.229 Ǻ to 4.134 Ǻ) relative to the optimized LE-ES geometry (Table 4.4). The O1-H1 bond
of Wat1862 is elongated from 0.797 Ǻ to 1.097 Ǻ while the H1-C1 and the C2-O2 distances are
reduced from 2.385 Ǻ to 1.540 Ǻ and 3.104 Ǻ to 1.930 Ǻ, respectively. The H5-C1-C2 and the
C1-C2-H6 bond angles of acetylene are reduced from 178.9˚ to 125.6˚ and from 178.7˚ to
155.0˚, respectively. The energy barrier for the formation of this transition state is 54.6
kcal/mol in the polarizable continuum (50.8 kcal /mol in the gas phase) relative to the eductsubstrate (LE-ES) complex (Table 4.6).
The optimized transition state complex LE-TS geometry is slightly modified to generate the
starting geometries for the geometry optimization of the educt substrate complex LE-ES and
the alcohol product complex LE-EP1. A slight reduction of W-O1, O1-H1, O4-H7 and O2-H4
bonds together with the elongation of the H1-C1, C2-O2, O3-H4 and O1-H7 distances generate
the starting geometry for the LE-ES complex. On the other hand a slight elongation of the WO1, O1-H1, O4-H7 and O2-H4 bonds and reduction of the H1-C1, C2-O2, O3-H4 and O1-H7
distances were performed to generate the starting geometry for the LE-EP1 complex.
Ø Optimized alcoholic product complex LE-EP1:
Geometry optimization of vinyl alcohol complex LE-EP shows a change in the S1-S2-S3-S4
dihedral angle (from -12.6˚ to 6.5˚) and in the W-S bond distances (from ~2.350 Ǻ to ~2.365
Ǻ) (Table 4.4). The W-O1 and the O1-H1 distances are increased from 4.134 Ǻ to 4.688 Ǻ and
from 1.097 Ǻ to 3.303 Ǻ, respectively. The C1-C2 distance is increased from 1.278 Ǻ to 1.346
Ǻ and the C2-O2 bond is reduced from 1.930 Ǻ to 1.406 Ǻ whereas the O2-H4 bond is broken
(distance is increased from 1.045 Ǻ to 5.021 Ǻ) resulting in the formation of vinyl alcohol. On
the other hand the reduction of the H4-O3 distance from 1.477 Ǻ to 1.014 Ǻ and the elongation
of the O4-H7 distance from 1.314 Ǻ to 1.859 Ǻ results in the restoration of protonated Asp13.
The H5-C1-C2 and C1-C2-H6 bond angles of acetylene are further reduced from 125.6˚ to
120.3˚ and from 155.0˚ to 123.6˚ (Table 4.4).
The formation of vinyl alcohol complex is an exothermic reaction with respect to the LE-ES
complex, -32.5 kcal/mol in the polarizable continuum (-28.9 kcal/mol in the gas phase) (Table
92
Chapter 4-Acetylene Hydratase
4.6). Although the formation of vinyl alcohol is an exothermic step, this mechanism was not
pursued further due to the high energy barrier of the transition state LE-TS.
Fig. 4.9: Optimized geometries for the large model complexes involved in the electrophilic pathway of
acetylene hydration by AH.
Table 4.4: Geometrical parameters of the optimized model complexes of the electrophilic reaction
mechanism for acetylene hydration by the large model complexes of AH.
X
LE-E
LE-ES
LE-TS
LE-EP1
W-S1 (Ǻ)
2.432
2.429
2.374
2.359
2.380
W-S2 (Ǻ)
2.489
2.366
2.362
2.358
2.367
W-S3 (Ǻ)
2.511
2.356
2.357
2.333
2.355
W-S4 (Ǻ)
2.336
2.400
2.362
2.348
2.359
S1-S2-S3-S4 (˚)
-31.4
-17.4
-2.3
-12.6
6.5
W-O1 (Ǻ)
2.041
2.182
4.229
4.134
4.688
O1-H1 (Ǻ)
-
0.976
0.979
1.097
3.303
H1-C1 (Ǻ)
-
-
2.385
1.540
1.086
C1-C2 (Ǻ)
-
-
1.225
1.278
1.346
C2-O2 (Ǻ)
-
-
3.104
1.930
1.406
O2-H4 (Ǻ)
-
0.987
0.999
1.045
5.021
H4-O3 (Ǻ)
-
1.788
1.683
1.477
1.014
O4-H7 (Ǻ)
-
1.068
1.097
1.314
1.859
H7-O1 (Ǻ)
-
2.729
1.341
1.113
0.982
H5-C1-C2 (˚)
-
-
178.9
125.6
120.3
C1-C2-H6 (˚)
-
-
178.7
155.0
123.6
Where, X = protein X-ray crystal structure data,100 LE-E = educt complex, LE-ES = educt-substrate
complex, LE-TS = transition state complex, LE-EP1 = alcoholic product complex.
93
Chapter 4-Acetylene Hydratase
·
Nucleophilic pathway LN:
Ø Optimized active site model complex LN-E:
The reduced active site model complex LN-E of AH was geometry optimized where the
oxidation state of tungsten is IV and the overall charge is -2. Unlike SN-E optimized
geometry, the optimized active site model complex LN-E geometry is distinctively different
in geometrical parameters of the metal coordination site in comparison to the protein X-ray
crystal structure of AH.100 The optimized data shows a slight reduction in the S1-S2-S3-S4
dihedral angle (from -31.4˚ to -22.1˚) and in the bond distances between tungsten (W) and
dithiolate sulfur atoms (S) (from ~2.442 Ǻ to ~2.386 Ǻ) (Table 4.5). The W-O1 bond distance
is increased from 2.041 Ǻ to 2.203 Ǻ.
Ø Optimized educt substrate complex LN-ES:
Geometry optimization of the LN-ES complex shows a slight change in the S1-S2-S3-S4
dihedral angle (from -22.1˚ to -26.6˚) and in the W-S bond distances (from ~2.386 Ǻ to
~2.365 Ǻ) when comparison is made with the optimized LN-E geometry. The W-O1 and
H1-O3 distances are slightly reduced from 2.203 Ǻ to 2.196 Ǻ and from 1.559 Ǻ to 1.544 Ǻ,
respectively (Table 4.5).
In the optimized LN-ES complex geometry, the distance between the O1-C1 is 3.779 Ǻ, C1-C2
is 1.223 Ǻ and C2-H1 is 4.483 Ǻ. The H3-C1-C2 and C1-C2-H4 bond angles of acetylene are
179.6˚ and 177.8˚, respectively (Table 4.5).
Ø Optimized transition state complex LN-TS:
Geometry optimization of the transition state complex LN-TS shows a slight change in the
S1-S2-S3-S4 dihedral angle (from -26.6˚ to -24.0˚) and in the W-S bond distances (from ~2.365
Ǻ to ~2.384 Ǻ) when compared with the optimized LN-ES geometry. The W-O1 (from 2.196
Ǻ to 2.067 Ǻ), O1-C1 (from 3.779 Ǻ to 2.212 Ǻ) and C2-H1 (from 4.483 Ǻ to 1.433 Ǻ)
distances are reduced while the C1-C2 distance is slightly elongated from 1.223 Ǻ to 1.266Ǻ
(Table 4.5). The H3-C1-C2 and C1-C2-H4 bond angles of acetylene are reduced from 179.6˚ to
117.8˚ and from 177.8˚ to 138.1˚, respectively. The energy barrier for the formation of the
transition state complex LN-TS is 35.2 kcal/mol in the continuum (44.6 kcal/mol in the gas
phase) relative to the LN-ES complex (Table 4.6).
94
Chapter 4-Acetylene Hydratase
Ø Optimized alcoholic product complex LN-EP1:
The formation of the vinyl alcohol product complex LN-EP1 is an exothermic step, -27.8
kcal/mol (-26.8 kcal/mol) relative to the LN-ES complex (Table 4.6). In this alcoholic
product (LN-EP1) complex, the O1H2 part of the vinyl alcohol is bound to the active site W
metal. Geometry optimization of the LN-EP1 complex shows no considerable change in the
S1-S2-S3-S4 dihedral angle (from -24.0˚ to -24.4˚). The W-S bond distances are slightly
decreased from ~2.384 Ǻ to ~2.375 Ǻ with respect to the optimized LN-TS geometry. The
W-O1 and H1-O4 distances are increased from 2.067 Ǻ to 2.306 Ǻ and from 1.158 Ǻ to 2.676
Ǻ, respectively. The O1-C1 (from 2.212 Ǻ to 1.381 Ǻ) and C2-H1 (from 1.433 Ǻ to 1.086 Ǻ)
distances are reduced while C1-C2 bond is elongated from 1.266 Ǻ to 1.353 Ǻ. The H3-C1-C2
bond angle of acetylene is slightly increased from 117.8˚ to 122.5˚ while the C1-C2-H4 bond
angle is reduced from 138.1˚ to 119.1˚, respectively (Table 4.5).
Fig. 4.10: Optimized geometries for the large model complexes involved in the nucleophilic pathway
of acetylene hydration by AH.
95
Chapter 4-Acetylene Hydratase
Table 4.5: Geometrical parameters of the optimized model complexes of the nucleophilic reaction
mechanism for acetylene hydration by the large model complexes of AH.
X
LN-E
LN-ES
LN-TS
LN-EP1
W-S1 (Ǻ)
2.432
2.405
2.379
2.411
2.403
W-S2 (Ǻ)
2.489
2.367
2.340
2.346
2.353
W-S3 (Ǻ)
2.511
2.383
2.369
2.391
2.346
W-S4 (Ǻ)
2.336
2.388
2.373
2.387
2.397
S1-S2-S3-S4 (˚)
-31.4
-22.1
-26.6
-24.0
-24.4
W-O1 (Ǻ)
2.041
2.203
2.196
2.067
2.306
O1-H1 (Ǻ)
-
1.022
1.022
-
-
H1-O3 (Ǻ)
-
1.559
1.544
1.158
2.676
O1-C1 (Ǻ)
-
-
3.779
2.212
1.381
C1-C2 (Ǻ)
-
-
1.223
1.266
1.353
C2-H1 (Ǻ)
-
-
4.483
1.433
1.086
H3-C1-C2 (˚)
-
-
179.6
117.8
122.5
C1-C2-H4 (˚)
-
-
177.8
138.1
119.1
Where, X = protein X-ray crystal structure data,100 LN-E = educt complex, LN-ES = educt-substrate
complex, LN-TS = transition state complex, LN-EP1 = alcoholic product complex.
Table 4.3: Computed energies [kcal/mol] relative to the edduct-substrate complex for stationary
points relevant in the hydration of acetylene by small model complexes of AH.
Electrophilic Pathway, LE
Nucleophilic Pathway, LN
//B3LYPa
ES
0.0
0.0
SDDb
COSMOc
TS
EP1
47.1
47.2
//B3LYPa
50.8
44.6
SDDb
54.6
35.2
COSMOc
-18.3
-21.5
//B3LYPa
-28.9
-26.8
SDDb
-32.5
-27.8
COSMOc
Where, ES = educt-substrate complex, TS = transition state complex, EP1 = alcohol product
complex. a) B3LYP/Lanl2DZ(p), b) B3LYP/SDDp//B3LYP/Lanl2DZ (p),
c) COSMO-B3LYP/SDDp//B3LYP/Lanl2DZ(p) (see Computational details)
96
Chapter 4-Acetylene Hydratase
4.3. Large model complexes including water molecules:
Computational results from the large model complexes also favour the nucleophilic pathway.
Although the relative energies for the formation of vinyl alcohol products are comparable
with the results from the small model complexes, the energy barriers are considerably higher
for both mechanistic options. The optimized geometry for the transition state following the
large model nucleophilic reaction pathway features a cisoid arrangement of hydrogen atoms
in acetylene while it is transoid in the small model nucleophilic pathway.
According to the protein X-ray crystal structure,100 there are at least 16 well defined water
molecules in a vestibule directly adjacent to the active site. Considering these water molecules
may help for the hydration of acetylene; we have included four water molecules (Wat1209,
Wat1212, Wat1424, and Wat1432) which are in the proximity of the W metal center and connected
together through hydrogen bonding (Fig 4.7). Ala137, Met138, Ile113, Ile142, and Phe611 residues
were also included to keep these water molecules at their location, as they may be involved in
hydrogen bonds to these water molecules. DFT studies were carried out on these water
containing large model complexes to check the possible assistance of additional water
molecules in the active site.
·
Electrophilic reaction pathway (XE):
Fig. 4.11: The chemical structure of the large active site model (water containing) complexes for the
electrophilic pathway derived from the protein X-ray crystal structure of AH.100
97
Chapter 4-Acetylene Hydratase
Ø Optimized active site model complex XE-E:
The reduced active site large model (water containing) complex XE-E of AH was geometry
optimized where oxidation state of tungsten is IV and the overall charge is -1. The optimized
geometry of active site model complex XE-E is distinctively different in geometrical
parameters of the metal coordination site in comparison to the protein X-ray crystal structure
of AH100 as well as to the optimized SE-E and LE-E geometries (see Table 4.1, 4.4 and 4.7).
The optimized data shows a reduction in the S1-S2-S3-S4 dihedral angle (from -31.4˚ to -24.9˚)
and in the bond distance between tungsten (W) and dihtiolate sulfur atoms (S) (from ~2.442 Ǻ
to ~2.377 Ǻ) when comparison is made with the protein X-ray crystal structure of AH
(Table 4.7). The W-O1 bond distance is increased from 2.041 Ǻ to 2.326 Ǻ as H1 is forming
hydrogen bond with the O2 (see Fig 4.11). The O1-H1 bond distance is 1.107 Ǻ. The O2-H3
and H3-O5 (hydrogen bond) distances are 1.001 Ǻ and 1.727Ǻ, respectively.
Ø Optimized educt substrate complex XE-ES:
In the optimizated educt-substrate complex XE-ES geometry, the acetylene is loosely bound
to the active site pocket. The hydrogen bonding is present between all the water molecules as
well as between the H4 of Wat1424 and the O3 of Asp13 (see Fig 4.11). Geometry optimization
of XE-ES complex shows a slight change in the S1-S2-S3-S4 dihedral angle (from -24.9˚ to
-26.6˚) and in the W-S bond distances (from ~2.377 Ǻ to ~2.379 Ǻ) when comparison is made
with the optimized XE-E geometry. The W-O1 and O1-H1 bond distances are reduced from
2.326 Ǻ to 2.192Ǻ and from 1.107 Ǻ to 0.982 Ǻ, respectively. The H4-O3 and H7-O1 distances
are increased from 1.714 Ǻ to 2.033 Ǻ and from 1.545 Ǻ to 2.477 Ǻ, respectively. The O3-H7
bond is reduced from 1.042 Ǻ to 1.016 Ǻ (Table 4.1). No considerable change is observed in
the O2-H3 bond (from 1.001 Ǻ to 0.995 Ǻ) while H3-O5 (H-bond) distance is increased from
1.727 Ǻ to 1.809 Ǻ.
In the optimized XE-ES complex geometry, the H1-C1, C1-C2 and C2-O2 distances are 2.357
Ǻ, 1.226 Ǻ and 2.979 Ǻ, respectively. The H5-C1-C2 and C1-C2-H6 bond angles of acetylene
are 176.2˚ and 175.3˚, respectively (Table 4.7).
Ø Optimized transition state complex XE-TS:
In the optimized transition state complex XE-TS geometry, the water molecules, Wat1432, and
Wat1209 are also involved in the process of vinyl alcohol formation. The O2-H3 bond of
Wat1424 (distance increased from 0.995 Ǻ to 1.438 Ǻ) is broken and H3 is transferred from
98
Chapter 4-Acetylene Hydratase
Wat1424 to the O5 of Wat1432 making it a hydronium ion while the electron donating part (-OH)
of Wat1424 is transferred to the substrate (see Fig. 4.12). The energy barrier for the formation
of this transition state is 23.1 kcal/mol in the continuum (29.2 kcal /mol in the gas phase)
relative to the educt-substrate (XE-ES) complex (Table 4.9).
Geometry optimization of the transition state XE-TS complex shows no considerable change
in the S1-S2-S3-S4 dihedral angle (from -26.6˚ to -26.2˚) and in the W-S bond distances (from
~2.379 Ǻ to ~2.384 Ǻ). The W-O1 bond is slightly reduced from 2.192 Ǻ to 2.122 Ǻ and the
O1-H1 bond is elongated from 0.982 Ǻ to 1.162 Ǻ) relative to the optimized XE-ES geometry
(Table 4.7). The H1-C1 and C2-O2 distances are reduced from 2.357 Ǻ to 1.471 Ǻ and 2.979 Ǻ
to 1.477 Ǻ, respectively. The C1-C2 bond is elongated from 1.226 Ǻ to 1.349 Ǻ. The H4-O3
distance is reduced from 2.033 Ǻ to 1.695 Ǻ. The O3-H7 and H7-O1 distances are elongated
from 1.016 Ǻ to 1.023 Ǻ and from 2.477 Ǻ to 2.730 Ǻ, respectively. The O2-H3 distance is
increased from 0.995 Ǻ to 1.438 Ǻ while the H3-O5 distance is reduced from 1.809 Ǻ to 1.063
Ǻ. The H5-C1-C2 and C1-C2-H6 bond angles of acetylene are reduced from 176.2˚ to 110.8˚
and from 175.3˚ to 128.7˚, respectively.
The educt substrate complex XE-ES and the alcohol product complex XE-EP1 starting
geometries for the geometry optimization were generated from the optimized transition state
complex XE-TS geometry. Slight reduction of W-O1 and O2-H3 bonds together with the
elongation of the H1-C1 and C2-O2 distances gave the starting geometry for the XE-ES
complex. On the other hand a slight elongation of the W-O1 bond and reduction of the O2-H3,
H1-C1 and C2-O2 distances were performed to generate the starting geometry for the XE-EP1
complex.
Ø Optimized alcoholic product complex XE-EP1:
The formation of vinyl alcohol complex XE-EP1 is an exothermic reaction with respect to the
XE-ES complex, -29.8 kcal/mol in continuum (-28.7 kcal/mol in the gas phase) (Table 4.9).
Geometry optimization result shows a slight change in the S1-S2-S3-S4 dihedral angle (from 26.2˚ to -23.5˚) and in the W-S bond distances (from ~2.384 Ǻ to ~2.373 Ǻ) (Table 4.7). The
W-O1 and O1-H1 distances are increased from 2.122 Ǻ to 2.252 Ǻ and from 1.162 Ǻ to 3.156
Ǻ, respectively. The C1-C2 distance is increased from 1.349 Ǻ to 1.358 Ǻ. The H1-C1 and C2O2 bond distances are reduced from 1.471 Ǻ to 1.087 Ǻ and from 1.477 Ǻ to 1.378 Ǻ,
respectively, whereas the O2-H4 distance is increased from 0.998 Ǻ to 1.048 Ǻ. The reduction
of H4-O3 distance from 1.695 Ǻ to 0.994 Ǻ and the elongation of O3-H7 distance (from 1.023
99
Chapter 4-Acetylene Hydratase
Ǻ to 2.649 Ǻ) results in the restoration of protonated Asp13. Finally, the reduction of O1-H7
(from 2.730 Ǻ to 0.983 Ǻ) results in the formation of a water ligand attached to the active site
metal. The O2-H3 distance is reduced from 1.438 Ǻ to 1.048 Ǻ while the H3-O5 distance is
increased from 1.063 Ǻ to 1.463 Ǻ. The H5-C1-C2 bond angle of acetylene is slightly
increased from 110.8˚ to 118.5˚ while the C1-C2-H6 is slightly reduced from 128.7˚ to 120.9˚
(Table 4.7).
The tautomerized acetaldehyde product (XE-EP2) complex starting geometry for geometry
optimization was generated by a slight modification in the vinyl alcohol part of the optimized
XE-EP1 complex geometry. The O2-H3 bond was broken together with the reduction of
C1-H3 and C2-O2 distances to generate the starting geometry for the XE-EP2 complex.
Ø Optimized tautomerized product complex XE-EP2:
The computed reaction energy for the tautomerized product complex XE-EP2 is exothermic
relative to the XE-ES complex, -35.2 kcal/mol in continuum (-32.9 kcal/mol) (Table 4.9).
Geometry optimization shows no considerable change in the S1-S2-S3-S4 dihedral angle (from
-23.5˚ to -23.1˚) and in the W-S bond distances (from ~2.373 Ǻ to ~2.379Ǻ). The W-O1 bond
is reduced from 2.252 Ǻ to 2.181 Ǻ. The elongation of C1-C2 bond (from 1.358 Ǻ to 1.494 Ǻ)
and the reduction of C2-O2 bond (from 1.048 Ǻ to 1.256 Ǻ) indicate the formation of the
acetaldehyde product complex (Table 4.7).
Fig. 4.12: Optimized geometries for the large model (water containing) complexes involved in the
electrophilic pathway of acetylene hydration by AH.
100
Chapter 4-Acetylene Hydratase
Table 4.7: Geometrical parameters of the optimized model complexes of the electrophilic reaction
mechanism for acetylene hydration by the large model complexes (containing water molecules) of
AH.
X
XE-E
XE-ES
XE-TS
XE-EP1
XE-EP2
W-S1 (Ǻ)
2.432
2.412
2.421
2.417
2.400
2.399
W-S2 (Ǻ)
2.489
2.396
2.394
2.381
2.390
2.395
W-S3 (Ǻ)
2.511
2.348
2.339
2.369
2.351
2.359
W-S4 (Ǻ)
2.336
2.350
2.362
2.367
2.348
2.363
S1-S2-S3-S4 (˚)
-31.4
-24.9
-26.6
-26.2
-23.5
-23.1
W-O1 (Ǻ)
2.041
2.326
2.192
2.122
2.252
2.181
O1-H1 (Ǻ)
-
1.107
0.980
1.162
3.156
-
H1-C1 (Ǻ)
-
-
2.357
1.471
1.087
-
C1-C2 (Ǻ)
-
-
1.226
1.349
1.358
1.494
C2-O2 (Ǻ)
-
-
2.979
1.477
1.378
1.256
O2-H4 (Ǻ)
-
0.997
0.983
0.998
-
-
O2-H3 (Ǻ)
1.001
0.995
1.438
1.048
-
H3-O5 (Ǻ)
1.727
1.809
1.063
1.463
-
H4-O3 (Ǻ)
-
1.714
2.033
1.695
0.994
-
O3-H7 (Ǻ)
-
1.042
1.016
1.023
-
-
H7-O1 (Ǻ)
-
1.545
2.477
2.730
0.983
-
H5-C1-C2 (˚)
-
-
176.2
110.8
118.5
-
C1-C2-H6 (˚)
-
-
175.3
128.7
120.9
-
Where, X = protein X-ray crystal structure data,100 XE-E = educt complex, XE-ES = educt-substrate
complex, XE-TS = transition state complex, XE-EP1 = alcoholic product complex,
XE-EP2 = aldehyde product complex
101
Chapter 4-Acetylene Hydratase
·
Nucleophilic reaction pathway XN:
Ø Optimized active site model complex XN-E:
The reduced active site large (water containing) model complex XN-E of AH is geometry
optimized where the oxidation state of tungsten is IV and the overall charge is -2. The
optimized geometry of XN-E is distinctively different in geometrical parameters of the metal
coordination site in comparison to the protein X-ray crystal structure of AH100 as well as to
the optimized SN-E and LN-E geometries (see Table 4.2, 4.5 and 4.8). The optimized data
shows a change in the S1-S2-S3-S4 dihedral angle (from -31.4˚ to -26.3˚) and in the bond
distances between tungsten (W) and dithiolate sulfur atoms (S) (from ~2.442 Ǻ to ~2.383 Ǻ)
relative to the protein X-ray crystal structure of AH. The W-O1 bond distance is increased
from 2.041 Ǻ to 2.167 Ǻ. The O1-H1 and H1 -O3 distances are 1.013 Ǻ and 3.633 Ǻ,
respectively (Table 4.8).
Ø Optimized educt substrate complex XN-ES:
Geometry optimization of the XN-ES complex shows a change in the S1-S2-S3-S4 dihedral
angle (from -26.3˚ to -29.4˚) and in the W-S bond distances (from ~2.383 Ǻ to ~2.414 Ǻ)
when comparison is made with the optimized XN-E geometry. The W-O1 and H1-O3
distances are reduced from 2.167 Ǻ to 2.024 Ǻ and from 3.633 Ǻ to 1.014 Ǻ, respectively
(Table 4.8).
In the optimized XN-ES complex geometry, the distance between the O1-C1 is 2.727 Ǻ, C1-C2
is 1.232 Ǻ and C2-H1 is 2.057 Ǻ. The H3-C1-C2 and C1-C2-H4 bond angles of acetylene are
173.2˚ and 172.2˚, respectively (Table 4.8).
Ø Optimized transition state complex XN-TS:
Geometry optimization of the transition state complex XN-TS shows no considerable change
in the the S1-S2-S3-S4 dihedral angle (from -29.4˚ to -27.6˚) and in the W-S bond distances
(from ~2.414 Ǻ to ~2.416 Ǻ) when compared with the optimized XN-ES geometry. The WO1 (from 2.024 Ǻ to 2.009 Ǻ), O1-C1 (from 1.014 Ǻ to 1.051 Ǻ) and C2-H1 (from 2.057 Ǻ to
1.733 Ǻ) distances are reduced while the C1-C2 distance is slightly elongated from 1.232 Ǻ to
1.246Ǻ (Table 4.8). The H3-C1-C2 and C1-C2-H4 bond angles of acetylene are reduced from
173.2˚ to 163.0˚ and from 172.2˚ to 154.7˚, respectively. The energy barrier for the formation
of this transition state complex XN-TS is 14.4 kcal/mol in the continuum (13.4 kcal/mol in
the gas phase) relative to the XN-ES complex (Table 4.10).
102
Chapter 4-Acetylene Hydratase
The educt substrate complex XN-ES and the alcohol product complex XN-EP1 starting
geometries for geometry optimization were generated from the optimized transition state
(XN-TS) geometry. Slight reduction of W-O1 bond together with the elongation of the O1-C1
and C2-H1 distances gave the starting geometry for the XN-ES complex. On the other hand a
slight elongation of the W-O1 and O3-H1 bonds and reduction of the O1-C1 and C2-H1
distances were performed to generate the starting geometry for the XN-EP1 complex.
Ø Optimized alcoholic product complex XN-EP1:
The formation of initial product of acetylene hydration, a vinyl alcohol, is an exothermic step,
-38.3 kcal/mol in the polarizable continuum (-47.3 kcal/mol) relative to the XN-ES complex
(Table 4.10). In this alcoholic product complex XN-EP1, the O1H2 part of the vinyl alcohol is
bound to the active site W metal. Geometry optimization of the XN-EP1 complex shows a
change in the S1-S2-S3-S4 dihedral angle (from -27.6˚ to -25.1˚) and in the W-S bond distances
(from ~2.416 Ǻ to ~2.380 Ǻ) with respect to the optimized XN-TS geometry. The W-O1
(from 2.009 Ǻ to 2.202 Ǻ) and the H1-O3 (from 1.051 Ǻ to 3.327 Ǻ) distances are increased.
The O1-C1 (from 2.262 Ǻ to 1.414 Ǻ) and C2-H1 (from 1.733 Ǻ to 1.086 Ǻ) distances are
reduced while the C1-C2 bond is elongated from 1.246 Ǻ to 1.342 Ǻ. The H3-C1-C2 and C1-C2H4 bond angles of acetylene are reduced from 163.0˚ to 124.5˚ and from 154.7˚ to 122.7˚,
respectively (Table 4.8).
The tautomerized acetaldehyde product (XN-EP2) complex starting geometry for geometry
optimization was generated by a slight modification in the vinyl alcohol part of the optimized
XN-EP1 complex geometry. The O2-H3 bond was broken together with the reduction of
C1-H3 and C2-O2 distances to generate the starting geometry for the XN-EP2 complex.
Ø Optimized tautomerized product complex XN-EP2:
The computed reaction energy for the formation of tautomerized product complex XN-EP2 is
again strongly exothermic, -51.4 kcal/mol in the continuum (-61.4 kcal/mol) relative to the
XN-ES complex (Table 4.10). Geometry optimization result shows only minor changes in the
S1-S2-S3-S4 dihedral angle (from -25.1˚ to -27.7˚) and in the W-S bond distances (from ~2.380
Ǻ to ~2.406 Ǻ) when comparison is made with the optimized XN-EP1 geometry. No change
is observed in the W-O1 bond distance. The O1-C1 bond is reduced from 1.414 Ǻ to 1.272 Ǻ
while the C1-C2 bond is elongated from 1.342 Ǻ to 1.498 Ǻ.
103
Chapter 4-Acetylene Hydratase
Fig. 4.13: Optimized geometries for the large model (water containing) complexes involved in the
nucleophilic pathway of acetylene hydration by W containing AH.
Table 4.8: Geometrical parameters of the optimized model complexes of the nucleophilic reaction
mechanism for acetylene hydration by the large model (water containing) complexes of AH.
X
XN-E
XN-ES
XN-TS
XN-EP1
XN-EP2
W-S1 (Ǻ)
2.432
2.402
2.413
2.425
2.387
2.399
W-S2 (Ǻ)
2.489
2.388
2.362
2.373
2.361
2.383
W-S3 (Ǻ)
2.511
2.367
2.410
2.403
2.369
2.412
W-S4 (Ǻ)
2.336
2.376
2.469
2.464
2.404
2.428
S1-S2-S3-S4 (˚)
-31.4
-26.3
-29.4
-27.6
-25.1
-27.7
W-O1 (Ǻ)
2.041
2.167
2.024
2.009
2.202
2.202
O1-H1 (Ǻ)
-
1.013
-
-
-
-
H1-O3 (Ǻ)
-
3.633
1.014
1.051
3.327
-
O1-C1 (Ǻ)
-
-
2.727
2.262
1.414
1.272
C1-C2 (Ǻ)
-
-
1.232
1.246
1.342
1.498
C2-H1 (Ǻ)
-
-
2.057
1.733
1.086
-
H3-C1-C2 (˚)
-
-
173.2
163.0
124.5
-
C1-C2-H4 (˚)
-
-
172.2
154.7
122.7
-
Where, X = protein X-ray crystal structure data,100 XN-E = educt complex, XN-ES = educt-substrate
complex, XN-TS = transition state complex, XN-EP1 = alcoholic product complex,
XN-EP2 = aldehyde product complex
104
Chapter 4-Acetylene Hydratase
·
Nucleophilic reaction pathway for acetylene hydration by Mo-containing AH (Mo):
Molybdenum dependent active acetylene hydratase enzyme was purified from P.acetylenicus
cells which had been grown in the absence of tungsten.179 The activity of Mo-dependent AH
for acetylene hydration was believed to be similar to the W-dependent AH. In order to check
the influence of the nature of metal on the acetylene hydration, we have computed the same,
water molecules containing, large active site model complexes (XN) derived from the protein
X-ray crystal structure of AH100 for acetylene hydration (via nucleophilic pathway) where
only the W is replaced by Mo atom at the active site.
Ø Optimized active site model complex Mo-E:
The reduced active site large (water containing) model complex Mo-E of AH is geometry
optimized where the oxidation state of tungsten is IV and the overall charge is -2. The
optimized geometry of Mo-E is distinctively different in geometrical parameters of the metal
coordination site in comparison to the optimized XN-E geometry. Although, the optimized
data shows no considerable change in the S1-S2-S3-S4 dihedral angle (from -26.3˚ to -29.5˚),
the bond distances between molybdenum (Mo) and dithiolate sulfur atoms (S) increases from
~2.383 Ǻ to ~2.394 Ǻ (Table 4.9). The M-O1 bond distance is increased from 2.167 Ǻ (W-O1)
to 2.242 Ǻ (Mo-O1). A slight change is observed in the O1-H1 bond distance from 1.013 Ǻ to
1.000 Ǻ while the H1-O3 distance is decreased from 3.633 Ǻ to 1.923 Ǻ.
Ø Optimized educt substrate complex Mo-ES:
Geometry optimization of the Mo-ES complex shows no change in the S1-S2-S3-S4 dihedral
angle (from -29.5˚ to -29.0˚) but the Mo-S bond distances are increased from ~2.394 Ǻ to
~2.425 Ǻ when comparison is made with the optimized Mo-E geometry. The Mo-O1 and
H1-O3 distances are reduced from 2.242 Ǻ to 2.047 Ǻ and from 1.923 Ǻ to 1.015 Ǻ,
respectively (Table 4.9).
In the optimized Mo-ES complex geometry, the distance between the O1-C1 is 2.700 Ǻ, C1-C2
is 1.233 Ǻ and C2-H1 is 2.052 Ǻ. The H3-C1-C2 and C1-C2-H4 bond angles of acetylene are
172.6˚ and 172.0˚, respectively (Table 4.9).
Ø Optimized transition state complex Mo-TS:
Geometry optimization of the transition state complex Mo-TS shows no considerable change
in the the S1-S2-S3-S4 dihedral angle (from -29.0˚ to -27.7˚) and in the Mo-S bond distances
105
Chapter 4-Acetylene Hydratase
(from ~2.425 Ǻ to ~2.428 Ǻ) when compared with the optimized Mo-ES geometry. The
Mo-O1 (from 2.047 Ǻ to 2.029 Ǻ), O1-C1 (from 2.700 Ǻ to 2.262 Ǻ) and C2-H1 (from 2.052 Ǻ
to 1.725 Ǻ) distances are reduced while C1-C2 distance is slightly elongated from 1.233 Ǻ to
1.247Ǻ (Table 4.9). The H3-C1-C2 and C1-C2-H4 bond angles of acetylene are reduced from
172.6˚ to 162.7˚ and from 172.0˚ to 155.3˚, respectively. The energy barrier for the formation
of this transition state complex Mo-TS is 13.3 kcal/mol in the continuum (13.1 kcal/mol in
the gas phase) relative to the Mo-ES complex (Table 4.10). The energy barrier for Mo-TS is
similar to the energy barrier for the tungsten containing complex XN-TS (14.4 kcal/mol in the
polarizable continuum).
The educt substrate complex Mo-ES and the alcohol product complex Mo-EP1 starting
geometries for geometry optimization were generated from the optimized transition state
complex Mo-TS geometry. Slight reduction of the Mo-O1 bond together with the elongation
of the O1-C1 and C2-H1 distances gave the starting geometry for the Mo-ES complex. On the
other hand a slight elongation of the Mo-O1 and O3-H1 bonds and reduction of the O1-C1 and
C2-H1 distances were performed to generate the starting geometry for the Mo-EP1 complex.
Ø Optimized alcoholic product complex Mo-EP1:
The formation of initial product of acetylene hydration, a vinyl alcohol, is an exothermic step,
-42.7 kcal/mol in the polarizable continuum (-51.3 kcal/mol) relative to the Mo-ES complex
(Table 4.10). The formation of Mo-EP1 is ~4 kcal/mol lower in energy than the formation of
XN-EP1. In this alcoholic product (Mo-EP1) complex, the O1H2 part of the vinyl alcohol is
bound to the active site Mo metal. Geometry optimization of the Mo-EP1 complex shows no
considerable change in the S1-S2-S3-S4 dihedral angle (from -27.7˚ to -25.7˚) but the Mo-S
bond distances are decreased from ~2.428 Ǻ to ~2.389 Ǻ relative to the optimized Mo-TS
geometry. The Mo-O1 (from 2.029 Ǻ to 2.235 Ǻ) and the H1-O3 (from 1.052 Ǻ to 2.606 Ǻ)
distances are increased. The O1-C1 (from 2.262 Ǻ to 1.409 Ǻ) and C2-H1 (from 1.725 Ǻ to
1.086 Ǻ) distances are reduced while the C1-C2 bond is elongated from 1.247 Ǻ to 1.343 Ǻ (a
typical C=C bond length). The H3-C1-C2 and C1-C2-H4 bond angles of acetylene are further
reduced from 162.7˚ to 124.1˚ and from 155.3˚ to 122.7˚, respectively (Table 4.9).
The tautomerized acetaldehyde product complex Mo-EP2 starting geometry for geometry
optimization was generated by a slight modification in the vinyl alcohol part of the optimized
Mo-EP1 complex geometry. The O2-H3 bond was broken together with the reduction of
C1-H3 and C2-O2 distances to generate the starting geometry for the Mo-EP2 complex.
106
Chapter 4-Acetylene Hydratase
Ø Optimized tautomerized product complex Mo-EP2:
The computed reaction energy for the formation of tautomerized product complex Mo-EP2 is
again strongly exothermic, -55.5 kcal/mol in the continuum and -63.7 kcal/mol in the gas
phase (again ~4 kcal/mol lower in energy than XN-EP2) relative to the Mo-ES complex
(Table 4.10). Geometry optimization shows no change in the S1-S2-S3-S4 dihedral angle (from
-25.7˚ to -25.6˚) while the Mo-S bond distances are increased from ~2.389 Ǻ to ~2.404 Ǻ
when comparison is made with the optimized Mo-EP1 geometry. The Mo-O1 bond distance is
reduced from 2.235 Ǻ to 2.129 Ǻ. The O1-C1 bond is reduced from 1.409 Ǻ to 1.260 Ǻ while
the C1-C2 bond is elongated from 1.343 Ǻ to 1.497 Ǻ (a typical C-C bond).
Table 4.9: Geometrical parameters of the optimized model complexes of the nucleophilic reaction
mechanism for acetylene hydration by the large model (water containing) complexes of AH.
XN-E
Mo-E
Mo-ES
Mo-TS
Mo-EP1
Mo-EP2
Mo-S1 (Ǻ)
2.402
2.405
2.427
2.439
2.394
2.402
Mo-S2 (Ǻ)
2.388
2.376
2.368
2.383
2.366
2.380
Mo-S3 (Ǻ)
2.367
2.391
2.416
2.408
2.382
2.410
Mo-S4 (Ǻ)
2.376
2.402
2.488
2.482
2.415
2.425
S1-S2-S3-S4 (˚)
-26.3
-29.5
-29.0
-27.7
-25.7
-25.6
Mo-O1 (Ǻ)
2.167
2.242
2.047
2.029
2.235
2.129
O1-H1 (Ǻ)
1.013
1.0
4.176
-
-
-
H1-O3 (Ǻ)
3.633
1.923
1.015
1.052
2.606
-
O1-C1 (Ǻ)
-
-
2.700
2.262
1.409
1.260
C1-C2 (Ǻ)
-
-
1.233
1.247
1.343
1.497
C2-H1 (Ǻ)
-
-
2.052
1.725
1.086
-
H3-C1-C2 (˚)
-
-
172.6
162.7
124.1
-
C1-C2-H4 (˚)
-
-
172.0
155.3
122.7
-
Where, XN-E = educt complex of the nucleophilic reaction mechanism for acetylene hydration by the
large model (water contining) complex of W-AH. Mo-E = educt complex, Mo-ES = educt-substrate
complex, Mo-TS = transition state complex, Mo-EP1 = alcoholic product complex,
Mo-EP2 = aldehyde product complex
107
Chapter 4-Acetylene Hydratase
Fig 4.14: Optimized geometries for the large model (water containing) complexes involved in
nucleophilic pathway of acetylene hydration by Mo containing AH.
Table 4.10: Computed energy barriers [kcal/mol] relative to the edduct-substrate complex for
hydration of acetylene by the large model (water containing) complexes of AH.
Electrophilic
Nucleophilic
Nucleophilic
Pathway, XE
Pathway, XN
Pathway, Mo
//B3LYPa
0.0
ES
0.0
0.0
SDDb
COSMOc
TS
EP1
EP2
EP3
26.5
14.3
14.3
//B3LYPa
29.2
13.4
13.1
SDDb
23.1
14.4
13.3
COSMOc
-36.1
-47.1
-50.7
//B3LYPa
-28.7
-47.3
-51.3
SDDb
-29.8
-38.3
-42.7
COSMOc
-34.0
-56.7
-57.3
//B3LYPa
-32.9
-61.4
-63.7
SDDb
-35.2
-51.4
-55.5
COSMOc
-56.5
-58.9
//B3LYPa
-62.5
-65.4
SDDb
-58.6
-61.6
COSMOc
-
Where, ES = educt-substrate complex, TS = transition state complex, EP1 = alcohol product
complex, EP2 = tautomerized product complex, EP3 = product complex, where acetaldehyde
is replaced by surrounding water molecule but acetaldehyde is also present in the structure.
a) B3LYP/Lanl2DZ(p), b) B3LYP/SDDp//B3LYP/Lanl2DZ (p),
c) COSMO-B3LYP/SDDp//B3LYP/Lanl2DZ(p) (see Computational details).
108
Chapter 4-Acetylene Hydratase
·
Optimized energy barriers for the tautomerization of vinyl alcohol to acetaldehyde:
In the absence of the educt complex, the energy barriers for the tautomerization of vinyl
alcohol (VA) to the acetaldehyde (AA) were computed with (intermolecular) and without
(intramolecular) the assistance of water molecules using the density functional theory (DFT)
calculations.
Ø Intramolecular 1A:
The computed energy barrier for the intramolecular conversion of 1A-VA to 1A-AA is 58.8
kcal/mol in the continuum (57.7 kcal/mol in the gas phase) relative to the vinyl alcohol
molecule 1A-VA (Table 4.12). In the optimized transition state 1A-TS, the C3-O5 (form 1.395
Ǻ to 1.324 Ǻ) and C1-H6 (from 2.578 Ǻ to 1.542 Ǻ) distances are reduced while the O5-H6
distance is increased from 0.980 Ǻ to 1.324 Ǻ when comparison is made with the optimized
1A-VA geometry (Table 4.11). An imaginary frequency of 2168ί cm-1 corresponds to the
stretching vibration modes of O5-H6 and C1-H6 in 1A-TS and confirmed the saddle point
character between the 1A-VA and 1A-AA.
The computed relative energy for the acetaldehyde formation 1A-AA is exothermic, -13.8
kcal/mol in the continuum (-14.4) relative to the 1A-VA (Table 4.12). In the optimized
1A-VA, the C3-O5 (form 1.324 Ǻ to 1.244 Ǻ) and C1-H6 (from 1.542 Ǻ to 1.093 Ǻ) distances
are further reduced while the O5-H6 distance is increased from 1.324 Ǻ to 2.615 Ǻ when
comparison is made with the optimized 1A-TS geometry (Table 4.11).
Fig. 4.15: Optimized geometries for the intramolecular tautomerization of vinyl alcohol to
acetaldehyde.
109
Chapter 4-Acetylene Hydratase
Ø Intermolecular - Single water molecule catalyzed reaction 2A :
Suenobu et al.173 suggest that the energy barrier for the conversion of VA to AA decreases
with the assistance of a water molecule (intermolecular). The computed energy barrier for the
water catalyzed conversion of 2A-VA to 2A-AA is 30.7 kcal/mol in the continuum (27.8
kcal/mol) relative to the 2A-VA (Table 4.12). An imaginary frequency of 1711ί cm-1
corresponds to the stretching vibration modes of O5-H6, O7-H9 and C1-H9 in 2A-TS and
confirmed the saddle point character between the 2A-VA and 2A-AA. In the optimized
transition state 2A-TS, the C3-O5 (form 1.381 Ǻ to 1.313 Ǻ), H6-O7 (form 1.679 Ǻ to 1.187
Ǻ), and C1-H9 (from 4.228 Ǻ to 1.463 Ǻ) distances are reduced while the O5-H6 (from 0.999
Ǻ to 1.263 Ǻ) and O7-H9 (from 0.975 Ǻ to 1.223 Ǻ) distances are increased when comparison
is made with the optimized 2A-VA geometry (Table 4.11).
The computed relative energy for the acetaldehyde 2A-AA formation in the water catalyzed
reaction is also exothermic, -11.5 kcal/mol in the continuum (-12.9 kcal/mol in the gas)
relative to the 2A-VA. The computed relative energy for the formation of 2A-AA is similar to
the formation of 1A-AA (see Table 4.12). In the optimized 2A-AA, the H6 is transferred from
the O5 of 2A-VA to the O7 of water molecule while the H9 from the water molecule is
transferred to the C1 of 2A-VA (see Fig. 4.16). The C3-O5 (form 1.313 Ǻ to 1.251 Ǻ), H6-O7
(form 1.187 Ǻ to 0.988 Ǻ), and C1-H9 (from 1.463 Ǻ to 1.093 Ǻ) distances are further reduced
while the O5-H6 (from 1.263 Ǻ to 1.841 Ǻ) and O7-H9 (from 1.223 Ǻ to 2.326 Ǻ) distances
are increased when comparison is made with the optimized 2A-TS geometry (Table 4.11).
Fig. 4.16: Optimized geometries for the single water molecule catalyzed, intermolecular
tautomerization of vinyl alcohol to acetaldehyde.
110
Chapter 4-Acetylene Hydratase
Ø Intermolecular – reaction catalyzed by two water molecules 3A:
Lledós et al.174 suggest that the intervention of a chain of two water molecules reduces
dramatically the potential energy barrier. The computed energy barrier for the conversion of
3A-VA to 3A-AA catalyzed by two water molecules is 20.7 kcal/mol in the continuum (20.3
kcal/mol) relative to the 3A-VA (Table 4.12). An imaginary frequency of 1318ί cm-1
corresponds to the stretching vibration modes of O5-H6, O7-H6, O7-H9, O11-H9, O11-H13 and
C1-H13 in 3A-TS and confirmed the saddle point character between the 3A-VA and 3A-AA.
In the optimized transition state 3A-TS, the C3-O5 (from 1.364 Ǻ to 1.304 Ǻ), H6-O7 (from
1.562 Ǻ to 1.125 Ǻ), H9-O11 (from 1.625 Ǻ to 1.186 Ǻ), and C1-H13 (from 2.172 Ǻ to 1.467
Ǻ) distances are reduced while the O5-H6 (from 1.020 Ǻ to 1.306 Ǻ), O7-H9 (from 1.005 Ǻ to
1.216 Ǻ) and O11-H13 (from 0.991 Ǻ to 1.188 Ǻ) distances are increased when comparison is
made with the optimized 3A-VA geometry (Table 4.11).
The computed relative energy for the acetaldehyde 3A-AA formation is also exothermic,
-12.0 kcal/mol in the continuum (-11.0 kcal/mol) relative to the 3A-VA. The computed
relative energy for the formation of 3A-AA is similar to the formation of 1A-AA and 2A-AA
(see Table 4.12). In the optimized 3A-AA, the H6 is transferred from the O5 of 3A-VA to the
O7 of one water molecule. The H9 from O7 of water molecule is transferred to the O11 of the
second water molecule while the H13 is transferred from the O11 to the C1 of 3A-VA (see Fig.
4.17). The C3-O5 (form 1.304 Ǻ to 1.252 Ǻ), H6-O7 (form 1.125 Ǻ to 0.995 Ǻ), H9-O11 (form
1.186 Ǻ to 1.002 Ǻ), and C1-H13 (from 1.467 Ǻ to 1.098 Ǻ) distances are further reduced
while the O5-H6 (from 1.306 Ǻ to 1.712 Ǻ), O7-H9 (from 1.216 Ǻ to 1.643 Ǻ) and O11-H13
(from 1.188 Ǻ to 2.129 Ǻ) distances are increased when comparison is made to the optimized
3A-TS geometry (Table 4.11).
Fig. 4.17: Optimized geometries for the intermolecular tautomerization of vinyl alcohol to
acetaldehyde catalyzed by two water molecules.
111
Chapter 4-Acetylene Hydratase
Ø Tautomerization of vinyl alcohol to acetaldehyde in the educt complex WN:
According to the computational results for the tautomerization of vinyl alcohol (VA) to
acetaldehyde (AA) (without educt complex), the energy barrier decreases when the reaction is
catalyzed by the chain of two water molecules. Therefore, we have computed the vinyl
alcohol to acetaldehyde tautomerization in the educt complex WN. WN-TS complex
geometry for the transition state search was generated by the addition of two water molecules
in the optimized SN-EP1 geometry. The WN-VA and WN-AA geometries for the geometry
optimization were generated by the slight modifications in the optimized WN-TS geometry.
In the optimized WN-VA geometry, a water molecule forms the hydrogen bond with the
Asp13 as well as with the other water molecule (Fig. 4.18). The computed energy barrier for
the WN-TS is 18.9 kcal/mol in the continuum (28.5 kcal/mol) relative to the WN-VA
complex (Table 4.12). In the optimized transition state WN-TS geometry, the C3-O5 bond is
reduced from 1.410 Ǻ to 1.351 Ǻ. The O5-H6, O7-H9 and O11-H13 bond distances are 1.5324
Ǻ, 1.214 Ǻ and 1.171 Ǻ, respectively. The H6-O7 , H9-O11 and C1-H13 distances are 1.062 Ǻ,
1.226 Ǻ and 1.505 Ǻ, respectively (Table 4.11).
The computed relative energy for the formation of acetaldehyde containing product complex
WN-AA is also exothermic, -13.0 kcal/mol in the continuum (-9.7 kcal/mol) relative to the
WN-VA complex (Table 4.12). The C3-O5 (from 1.351 Ǻ to 1.299 Ǻ), H6-O7 (form 1.062 Ǻ
to 1.003 Ǻ), H9-O11 (form 1.226 Ǻ to 1.023 Ǻ), and C1-H13 (from 1.505 Ǻ to 1.103 Ǻ)
distances are reduced while the O5-H6 (from 1.534 Ǻ to 3.572 Ǻ), O7-H9 (from 1.214 Ǻ to
1.613 Ǻ) and O11-H13 (from 1.171 Ǻ to 6.303 Ǻ) distances are increased when comparison is
made with the optimized WN-VA geometry (Table 4.11).
Fig. 4.18: Optimized geometries for the intermolecular tautomerization of vinyl alcohol to
acetaldehyde catalyzed by two water molecules in the educt complex, WN. (Labelling of the atoms are
according to the reaction 3A (Fig 4.17)).
112
Chapter 4-Acetylene Hydratase
Table 4.11: Geometrical parameters of the optimized model complexes for the tautomerization of
vinyl alcohol (VA) to acetaldehyde (AA).
C3-O5
O5-H6
C1-H6
H6-O7
O7-H9
C1-H9
H9-O11
O11-H13
C1-H13
(Ǻ)
(Ǻ)
(Ǻ)
(Ǻ)
(Ǻ)
(Ǻ)
(Ǻ)
(Ǻ)
(Ǻ)
1A-VA
1.395
0.980
2.578
-
-
-
-
-
-
1A-TS
1.324
1.324
1.542
-
-
-
-
-
-
1A-AA
1.244
2.615
1.093
-
-
-
-
-
-
2A-VA
1.381
0.999
-
1.679
0.975
4.228
-
-
-
2A-TS
1.313
1.263
-
1.187
1.223
1.463
-
-
-
2A-AA
1.251
1.841
-
0.988
2.326
1.093
-
-
-
3A-VA
1.364
1.020
-
1.562
1.005
-
1.625
0.991
2.172
3A-TS
1.304
1.306
-
1.125
1.216
-
1.186
1.188
1.467
3A-AA
1.252
1.712
-
0.995
1.643
-
1.002
2.129
1.098
WN-VA
1.410
1.00
-
-
-
-
-
-
-
WN-TS
1.351
1.534
-
1.062
1.214
-
1.226
1.171
1.505
WN-AA
1.299
3.572
-
1.003
1.613
-
1.023
6.303
1.103
Where, 1A = intramolecular reaction, 2A = single water molecule catalyzed reaction, 3A = reaction
catalyzed by two water molecules, WN = small model complex (SN-EP1) with two additional water
molecules, VA = vinyl alcohol, TS = transition state, AA = acetaldehyde.
(see Fig 4.15, 4.16, 4.17 and 4.18).
113
Chapter 4-Acetylene Hydratase
Table 4.12: Computed energy barriers (kcal/mol) for the tautomerization of vinyl alcohol to the
acetaldehyde.
VA
1A
2A
3A
WN
0.0
0.0
0.0
0.0
TS
AA
62.6
-9.6
//B3LYPa
57.7
-14.4
SDDb
58.8
-13.8
COSMOc
26.4
-7.3
//B3LYPa
27.8
-12.9
SDDb
30.7
-11.5
COSMOc
13.3
-5.3
//B3LYPa
20.3
-11.0
SDDb
20.7
-12.0
COSMOc
18.3
-8.6
//BP86d
28.5
-9.7
SDDe
18.9
-13.0
COSMOf
Where, 1A = intramolecular reaction, 2A = single water molecule catalyzed reaction, 3A = reaction
catalyzed by two water molecules, WN = small model complex (SN-EP1) with two additional water
molecules, VA = vinyl alcohol, TS = transition state, AA = acetaldehyde.
a) B3LYP/Lanl2DZ(p), b) B3LYP/SDDp//B3LYP/Lanl2DZ (p),
c) COSMO-B3LYP/SDDp//B3LYP/Lanl2DZ(p), d) BP86/Lanl2DZ(p),
e) B3LYP/SDDp//BP86/Lanl2DZ(p), f) COSMO-B3LYP/SDDp//BP86/Lanl2DZ(p)
(see Computational details).
114
Chapter 4-Acetylene Hydratase
Scheme 4.8: Graphical representation of the computed energy barriers (kcal/mol) for the
tautomerization of vinyl alcohol to acetaldehyde.
Where, VA = vinyl alcohol, TS = transition state, AA = acetaldehyde.
5. Discussion
Acetylene hydratase (AH) is a unique tungsten-containing enzyme as it does not appear to
catalyze a redox reaction. The protein X-ray crystal structure of AH100 provides important
clues on the catalytic mechanism and the role of the W center. For catalysis, the enzyme is
activated by the reduction of W center from WVI to WIV, as only WIV participates in the
catalysis of acetylene hydration, and the [4Fe-4S] cluster facilitates this step. The reactive
species in AH is either a hydroxo or a water molecule coordinated to the W center.
Considering the nature of the oxo-ligand attached to the W, we have computed two pathways
(as suggested by Seiffert et al.100): an electrophilic pathway, where the coordinated water
molecule would act as an electrophile and a nucleophilic pathway, where the hydroxide would
act as a nucleophile.
In the electrophilic pathway, Asp13 was considered to be protonated. The W-bound water
molecule (Wat1862) is activated by the nearby protonated Asp13 residue and a second water
molecule (Wat1424). The transition state involves the electrophilic attack of activated Wat1862
molecule on the triple bond of acetylene with the simultaneous transfer of protons among
115
Chapter 4-Acetylene Hydratase
Asp13, Wat1862, Wat1424 and acetylene. The proton from the Asp13 residue (-COOH) is
transferred to the Wat1862 molecule and one proton of the Wat1862 molecule is transferred to
the alpha carbon atom (Cα or C1) of acetylene. From the Wat1424, one proton is transferred to
Asp13 while its electron donating part (-OH) is transferred to the second carbon atom (Cβ or
C2) of acetylene. This proton shuttle results in the formation of vinyl alcohol which
subsequently may tautomerize to aldehyde (Scheme 4.7A).
In the nucleophilic pathway, Asp13 was considered to be in the anionic form. The W center,
which acts as a Lewis acid activates water molecule Wat1862, generates a W-bound hydroxide
and protonated Asp13. The transition state involves the nucleophilic attack of this W-bound
hydroxide at the alpha carbon atom (Cα or C1) of acetylene together with the simultaneous
transfer of a proton from the Asp13 to the second acetylene carbon atom (Cβ or C2) resulting
in the formation of vinyl alcohol which subsequently may tautomerize to aldehyde (Scheme
4.7B).
The density functional theory (DFT) computations were, first, performed on the small model
complexes derived from the protein X-ray crystal structure of AH.100 For the small model
complexes W metal center coordinated with two molybdopterin ligands (MGD), a metal
bound water (Wat1862) molecule, a cysteinate (Cys141) ligand and an additional aspartate
(Asp13) residue was considered. The water molecule (Wat1424), present near Asp13 and water
molecule Wat1862, were also considered in case of the electrophilic reaction mechanism
(Scheme 4.7 A). The computational results for the small model complexes show that the
nucleophilic pathway is energetically more favorable than the electrophilc pathway. The
energy barrier for the electrophilic pathway transition state SE-TS (28.5 kcal/mol in the
continuum) is higher than the energy barrier for the nucleophilic pathway transition state
SN-TS (17.0 kcal/mol). The vinyl alcohol product SE-EP1 formation is ~7 kcal/mol lower in
energy than the SN-EP1. The final step may involve the tautomerization of vinyl alcohol
intermediate to the acetaldehyde. The formation of acetaldehyde product SN-EP2 is ~4
kcal/mol lower in energy than the formation of SE-EP2 (Table 4.3).
The energy barrier for the tautomerization of vinyl alcohol (VA) to acetaldehyde (AA) was
computed (without educt complex) without the assistance of a water molecule and with the
assistance of one or two water molecules (Table 4.12 and Scheme 4.8). The computed energy
barrier for the intramolecular conversion of VA to AA was 58.8 kcal/mol in the polarizable
continuum. This energy barrier decreases to 30.7 kcal/mol when the reaction is catalysed by a
water molecule. The same effect was explained by the Suenobou et al.173 where the energy
116
Chapter 4-Acetylene Hydratase
barrier (in the gas phase) decreases from 55.8 kcal/mol to 29.6 kcal/mol when the reaction is
catalysed by a water molecule. As the energy barrier drops dramatically with the assistance of
two water molecules as suggested by Lledós et al.,174 we then computed the energy barrier of
20.7 kcal/mol in the continuum for the tautomerization of 3A-VA to 3A-AA in the presence
of two water molecules (Table 4. 12 and Scheme 4.8).
The tautomerization of VA to AA is then computed in the educt complex, WN, which was
generated by the introduction of two water molecules in the optimized small model SN-EP1
complex. The computed energy barrier for this conversion of VA to AA in WN-TS is 18.9
kcal/mol in the continuum (Table 4.12). Now, when we compare this energy barrier of
WN-TS with the energy barrier for alcohol formation in the electrophilic (SE-TS = 28.5
kcal/mol) and nucleophilic (SN-TS = 17.0 kcal/mol) pathways, the tautomerization of VA to
AA seems to be the rate limiting step in the nucleophilic pathway.
The large model complexes were then analyzed to identify the most probable reaction
mechanism. In the large model complexes some surrounding amino acid residues, Trp179,
Trp293 and Trp472, which may take part in the reaction were also considered. The
computational results for the large model complexes also favors the nucleophilic pathway for
the formation of vinyl alcohol as the energy barrier for LN-TS is ~20 kcal/mol lower than the
energy barrier for LE-TS (Table 4.6). The formation of vinyl alcohol complexes LE-EP1 and
LN-EP1 are exothermic for both pathways; however, LN-EP1 is ~4 kcal/mol lower in energy
than LN-EP1. Although the relative energies for the vinyl alcohol formation were comparable
with the results from the small model complexes, the energy barriers are considerably higher
for LN-TS and LE-TS to continue the mechanism (see Table 4.3 and 4.6).
Now, according to the protein X-ray crystal structure of AH, there are at least 16 well defined
water molecules adjacent to the active site. These water molecules may help in the hydration
of acetylene. So, in the large model complex we included four water molecules (Wat1209,
Wat1212, Wat1424, and Wat1432) which are in the proximity of W metal center and connected
together through the hydrogen bonding (Fig 4.7). To keep these water molecules at their
location, some surrounding residues, Ala137, Met138, Ile113, Ile142, and Phe611, were also
included. The computational results for the water molecules containing large model
complexes also favor the nucleophilic pathway for vinyl alcohol formation. The energy
barrier for the nucleophilic pathway transition state complex XN-TS (14.4 kcal/mol) is ~9
kcal/mol lower than the energy barrier for the electrophilic pathway transition state complex
XE-TS (23.1 kcal/mol). In the small model and the large model computational results, the
117
Chapter 4-Acetylene Hydratase
vinyl alcohol containing electrophilic pathway EP1 complexes (SE-EP1 and LE-EP1) were
lower in energy than the nucleophilic pathway EP1 complexes (SN-EP1 and LN-EP1).
However, in the water molecules containing large model complexes, the energy for the
formation of vinyl alcohol product XN-EP1 is ~8 kcal/mol lower than the XE-EP1. The
relative energies for the acetaldehyde product complexes XN-EP2 and XE-EP2 are also
exothermic but the XN-EP2 is ~16 kcal/mol lower in energy than the XE-EP2. In the
nucleophilic pathway, the W-center is regenerated by the replacement of acetaldehyde with
one of the surrounding water molecule. The reaction energy for this step (XN-EP3) is also
exothermic, -7.2 kcal/mol in the polarizable continuum and -1.1 kcal/mol in the gas phase,
relative to the XN-EP2 complex (Table 4.10).
In 1997, Rosner et al.179 purified a molybdenum dependent active acetylene hydratase enzyme
P.acetylenicus cells which had been grown in the absence of tungsten. The molecular mass
and the first 10 amino acids of the N-terminus of Mo-dependent AH was identical to the Wdependent AH enzyme. Now, to find the influence of metal on the acetylene hydration we
have computed the same, water molecules containing, large active site model complexes
(nucleophilic reaction pathway) only differing in the metal (W is replaced by Mo) at the
active site center. The computational results shows that the energy barrier for Mo-TS (13.3
kcal/mol) is similar to the W containing transition state model complex XN-TS (14.4
kcal/mol) (Table 4.10). The relative energies for the formation of vinyl alcohol (Mo-EP1=
-55.5 kcal/mol, XN-EP1 = -51.4 kcal/mol in the continuum) and acetaldehyde (Mo-EP2=
-61.6 kcal/mol, XN-EP2 = -58.6 kcal/mol) are also similar (Table 4.10). So, no decisive
influence of the metal on the hydration of acetylene was observed when W is replaced with
Mo.
Based on the computational results we here present the most likely mechanism for the
hydration of acetylene by the acetylene hydratase (AH) enzyme which is essentially the
nucleophilic reaction pathway proposed by Seiffert et al.100 DFT studies were carried out on
the active site model complexes derived from the protein X-ray crystal structure of AH (PDBID: 2E7Z)100 relevant along this mechanistic pathway. In this mechanism, the water (Wat1424)
molecule is coordinated to the W center and Asp13 is assumed to be in anionic form. The role
of W is to activate the Wat1424 molecule making it a Lewis acid. This activated Wat1424 then
donates one of its proton to the anionic Asp13 forming the W-bound hydroxide and protonated
Asp13. The W-bound hydroxide then attacks the Cα or C1 atom of acetylene together with the
transfer of proton from the Asp13 to its Cβ or C2 atom, resulting in the formation of a vinyl
118
Chapter 4-Acetylene Hydratase
alcohol intermediate complex. The energy barrier for this step is 14.4 kcal/mol in the
polarizable continuum. The final step corresponds to the tautomerization of vinyl alcohol
intermediate to the acetaldehyde via intermolecular assistance of two water molecules. The
energy barrier for this step is 18.9 kcal/mol in the polarizable continuum (calculated only for
the small model complexes, SN). This step should be the rate-limiting step in the nucleophilic
pathway (see Table 4.10 and 4.12). However, the energy associated with the rate limiting step
is higher in the acetylene hydration reaction mechanisms proposed in the literature: the
mechanism proposed by Vincent et al.171 has a maximum energy barrier as high as 34.0
kcal/mol while the mechanism proposed by the Himo et al.172 has an energy barrier of 23.9
kcal/mol. The mechanism presented here is more favorable by 5 kcal/mol.
119
Chapter 5-Selenate Reductase
Selenate Reductase
1. Introduction
The chalcogen elements (i.e., elements of group 16 of the periodic table) oxygen (O), sulfur
(S) and selenium (Se) fulfill a wide range of essential biological functions. All three elements
are constituents of functional groups in biomolecules that participate in redox reactions.180,181
There are close similarities but also striking differences between sulfur and selenium in terms
of their chemistry and biochemistry. Both S and Se are present in proteins as constituents of
the natural amino acids cysteine, methionine, selenocysteine and selenomethionine182 and also
occur as substrates e.g., for the sulfite oxidase and selenate reductase enzymes, respectively.
These are mononuclear molybdenum enzymes. All mononuclear molybdoenzymes contain a
molybdenum cofactor, Moco, which consists of either one or two organic moieties of
metalopterin (MPT) or some of its nucleotide variants, coordinated to Mo through an
enedithiolate motif. Based on the active site composition, i.e., the number of MPT and type of
additional ligands, these enzymes are generally grouped into three families (Fig. 1.2); the
xanthine oxidase (XO) family,9,40-42,47 the sulfite oxidase (SO) family9,53-55 and the
dimethylsulfoxide reductase (DMSOR) family.9,59-61
Selenate reductase (SeR) from Thauera selenatis149,183 is a soluble periplasmic
molybdoenzyme that catalyzes the two electron reduction of selenate (SeO42-) to selenite
(SeO32-).
SeO42- + 2H+ + 2e-
SeO32- + H2O (5.1)
This reduction is associated with the respiratory electron transfer chains that generates an
electro-chemical gradient across the cytoplasmic membrane of bacteria.184,149 Selenate is the
oxidized form of selenium which is highly soluble and can present significant hazards to
health and the environment. Selenate detoxification can be done by the reduction of selenate
to selenite catalyzed by the microbial reductase. The microbes that can reduce selenate are not
restricted to any particular group/subgroup of prokaryotes and examples are found throughout
the bacterial domain. 185
Physiologically, SeR is the terminal reductase supporting anaerobic growth on acetate in the
presence of selenate. It shows substrate specificity and does not reduce nitrate, arsenate or
sulfate, but does reduce chlorate.186 It is the only dissimilatory selenate reductase known and
is of considerable interest as a novel member of the molybdenum enzymes. SeR contains an
120
Chapter 5-Selenate Reductase
active site that is characteristic of the prokayrotic oxotransferase (DMSO reductase) family of
molybdenum enzymes.187 Periplasmic SeR from Thauera selenatis is a heterotrimeric
enzyme. It comprises three subunits, SeR-A, SeR-B and SeR-C. (Fig.5.1) SeR-A is the
catalytic subunit containing the molybdenum active site coordinated with the two
molybdopterin guanine dinucleotide (bis-MGD) ligands and a hydroxide group in the reduced
form. SeR-B contains a number of cysteine rich motifs that coordinate a [3Fe-4S] and three
[4Fe-4S] iron sulfur clusters. SeR-C contains b-type cytochrome that is rarely found in
periplasmic proteins.187,187,188,189,190
Fig. 5.1: (A) Schematic representations of selenate reduction SerABC from Thauera selenatis.186
SerABC receives electrons from cytochrome c4, which is reduced by quinol-cytochrome c (QCR)
oxidoreductase coupled with quinol oxidation. The dashed arrows represent electron flow; Q,
quinones; QH2, quinols; cytc4, cytochrome c4; [4Fe-4S], iron-sulfur cluster; [3Fe-4S], iron-sulfur
cluster; MoCo, molybdenum cofactor; SeO42-, selenate; SeO32-, selenite.191 (B) Active site of selenate
reductase depicted from the X-ray absorption spectroscopy.187
The SeR enzyme is stable and active upon incubation at temperatures upto 60˚C with an
optimum activity recorded at 65˚C. The SeR-C component appears to be least stable once
above 60˚C and perhaps the loss of this contributes to the overall instability of the SeR
complex.186,192 The location as a soluble protein in the periplasm and thermostability of SeR is
consistent with the other molybdoenzymes from mesophilic bacteria e.g. TMAO reductase
from E.coli.102
121
Chapter 5-Selenate Reductase
Analogue reaction systems for the reduction of selenate to selenite using bis(dithiolene)
complexes of MoIV and WIV demonstrate that chemically both metals can function as a
catalyst for the selenate reduction. However, to date no evidence has been presented regarding
the effect of tungsten on selenate reduction in an enzyme system.184 Recently it has been
observed that T. selenatis can grow readily on a tungsten rich medium and SeR isolated under
such conditions show 20-fold reduction in the selenate reductase activity, and 23-fold
increased affinity for selenate. The tungsten substitution may have enhanced the bond
strengths of the W-substrate complexes leading to the observed higher substrate affinity
which in turns makes the complex kinetically slower that the equivalent Mo-substrate
complexes, when measured at the mesophilic range of temperatures.191
Sulfite oxidase (SO) is the name giving member of the sulfite oxidase family of mononuclear
molybdenum enzymes. It catalyzes the oxidation of sulfite to sulfate through an oxygen atom
transfer (OAT) reaction.31
SO32- + H2O
SO42- + 2H+ + 2e- (5.2)
In the active site of reduced SO the central Mo atom is five-fold coordinated in a square
pyramidal fashion by one metalopterin (MPT) ligand, one terminal oxygen atom, one sulfur
atom of cysteine and one water or hydroxide. In the oxidized form of the enzyme the
water/hydroxyl is probably replaced by a second oxo group. The MPT forms a tricylic ring
system with a pyran ring fused to the pyrazine ring of the pterin and is not conjugated by an
additional nucleotide (Fig. 5.2).31 The cysteine and dithiolene moieties have been implicated
in the tuning of the flexibility of the equatorial oxo group towards the OAT reaction.
Fig. 5.2: Schematic representation of the active site of oxidized sulfite oxidase (SO)193
The mode of sulfite attack to the Mo center could be oxoanionic as suggested or sulfur lone
pair attack.194 Detailed theoretical studies were carried out by Sarkar et al, 195 using DFT
calculations. The author investigated the reaction of the substrate (HSO3-) with the
computational model complex [MoVIO2(mpt)(Cys)]- (1) of Moco, derived from the X-ray
122
Chapter 5-Selenate Reductase
crystal structure of native SO,31 and with the experimental model complex196
[MoVIO2(mnt)2]2- (2) (Scheme 5.1). The calculations show that the initial step in the oxygen
atom transfer reaction of 1 and 2 with HSO31- is the oxoanionic binding of the substrate to the
MoVI resulting in the formation of an intermediate complex (Scheme 5.1). This intermediate
complex participates in product formation through a six-membered {MoOeqSOHOax}
transition state involving breaking of the Mo-Oeq bond and formation of the Ssulfite-Oeq bond
(Scheme 5.1).
Scheme 5.1: Reaction scheme of sulfite oxidation to sulfate suggested by Sarkar et al.195
123
Chapter 5-Selenate Reductase
Nitrate reductases (NRs) play key roles in the first step of biological nitrogen
cycles197,198,199 i.e., assimilatory ammonification (to incorporate nitrogen into biomolecules),
denitrification (to generate energy for cellular function) and dissimilatory ammonification (to
dissipate extra energy by respiration). NRs have been classified into three groups,
assimilatory nitrate reductases (Nas), respiratory nitrate reductases (Nar) and periplasmic
nitrate reductases (Nap) (Fig. 5.3). Nas belongs to the sulfite oxidase family and is located in
the cytoplasm. 200 Dissimilatory nitrate reductases, Nar and Nap belong to the DMSO
reductase family of mononuclear MPT containing molybdo-enzymes. They are linked to
respiratory electron transport systems and are located in the membrane and periplasm,
respectively. Although Nas, Nap and Nar are different at their active site composition (Fig.
5.3) and belong to two different families of mononuclear molybdoenzymes, they all catalyze
the same reaction; reduction of nitrate to nitrite.
NO3- + 2H+ + 2e-
NO2- + H2O (5.3)
Fig. 5.3: Active site composition of nitrate reductases.
2. Project: I
When we compare the active site composition of SO, SeR and NRs we find that SO has one
MPT ligand and it oxidizes sulfite to sulfate, SeR has two MPT ligands and it reduces selenate
to selenite, while NRs reduce nitrate to nitrite and is able to do this with either one or with
two MPT’s coordinated to Mo. Now, the question arises whether the active site itself is
special in some way for the oxidation/reduction process or it is the substrates or the different
active sites behave the same way and it is the role of the protein to make it specific.
124
Chapter 5-Selenate Reductase
To answer these questions, we have performed the DFT calculations for the reaction of HSO3 and HSeO3 - with the computational model complex, [MoVIO2(S2C2Me2)SMe]- (A) of Moco,
derived from the X-ray crystal structure of native SO,31 and with the experimental model
complex196 [MoVIO2(mnt)2]2- (B) through oxoanionic binding of substrate at the Mo center.
2.1. Computational Details
All calculations were performed with the Gaussian 03175 program package using B3LYP123
hybrid functional of density functional theory (DFT). For the geometry optimizations, the
LANL2DZ basis set124,125,126,127 augmented by polarization functions on all atoms except Mo
and H (ζ =0.600, 1.154, 0.864, 0.421, 0.338 for C, O, N, S and Se, respectively)128 was
employed. The optimized minima and transition-state structures were characterized by
frequency calculations with the same method and basis set to verify that all minima have no
imaginary frequency and each transition state has exactly one. Single point energies were
computed with the B3LYP functional and the Stuttgart-Dresden effective core potential basis
set (SDD)129,130 augmented by polarization functions for all atoms except Mo, and H
(ζ =0.600, 1.154, 0.864, 0.421, 0.338 for C, O, N, S and Se, respectively).128 The selfconsistent reaction field (SCRF) computations were performed on the optimized geometries
by a conductor like polarizable continuum method (CPCM)131 with a dielectric constant of 4
and solvent radius of 1.4Ǻ.
2.2. Active Site Models
Model complexes studied here were the same used by Sarkar et al.195 where the initial active
site geometry for [MoVIO2(S2C2Me2)SMe]- (A) was derived from the protein X-ray crystal
structure of the native SO enzyme31 while the initial geometry of [MoVIO2(mnt)2]2- (B) was
obtained from the crystal structure (Fig. 5.4) reported for this compound.196 The complex B is
similar to the active site composition of SeR190,191 as both contain two MPT ligands.
Fig. 5.4: Active site composition of initial active site model geometries A and B.
125
Chapter 5-Selenate Reductase
2.3. Results
·
Molybdenum (VI) dithiolene complex [MoVIO2(S2C2Me2)SMe]- with HSeO3- A1:
Ø Optimized educt [MoVIO2(S2C2Me2)SMe]- complex A1-E:
The molybdenum dithiolene complex A1-E, derived from the protein X-ray crystal structure
of the native SO enzyme,31 was geometry optimized where the oxidation state of the
molybdenum is VI and overall charge on the complex is -1. The optimized data shows that the
Mo-Oax and Mo-Oeq bond lengths are 1.724 Ǻ and 1.736 Ǻ, respectively. The dithiolene Mo-S
bond distances are ~2.510 Ǻ and the Mo-SCH3 bond distance is 2.452 Ǻ (Table 5.1).
Ø Optimized transition state complex for educt-substrate complex formation A1-TS1:
Oxygen atom transfer (OAT), from square pyramidal complex [MoVIO2(S2C2Me2)SMe]1(A1-E) to HSeO3- is initiated by a
transition state [MoVIO2(S2C2Me2)SMe(HSeO3)]2-
(A1-TS1). The energy barrier for A1-TS1 is 10.0 kcal/mol in the polarizable continuum (47.3
kcal/mol in the gas phase) relative to the separate substrate, HSeO3-, and educt (A1-E)
complex (Table 5.3). One imaginary frequency of 49ί cm-1 corresponds predominantly to the
stretching vibration mode of Mo-OSe.
The optimized A1-TS1 has a distorted square pyramidal geometry. The optimized data shows
a slight decrease in the Mo-Oax bond distance from 1.724 Ǻ to 1.719 Ǻ when comparison is
made with the optimized A1-E geometry. The dithiolene Mo-S bond distances are increased
from ~2.510 Ǻ to ~2.543 Ǻ. Because of the oxoanionic approach of selenite, the -SCH3 group
moves from an equatorial to an axial position and the Mo-SCH3 bond distance is increased
from 2.452 Ǻ to 2.474 Ǻ. At this stage, the HSeO3- is loosely bound to the Mo center. The
distance between Mo-OSe is 3.159Å (Table 5.1).
Ø Optimized educt-substrate intermediate complex A1-ES:
The
computed
reaction
energy
for
the
formation
of
intermediate
complex
[MoVIO2(S2C2Me2)SMe(HSeO3)]2- (A1-ES) is endothermic, 9.7 kcal/mol in the polarizable
continuum (40.0 kcal/mol) relative to the separate substrate, HSeO3-, and educt (A1-E)
complex (Table 5.3). Geometry optimization of A1-ES shows a decrease in the Mo-OSe
distance from 3.159 Ǻ to 2.231 Å while the dithiolene Mo-S bond distances are further
increased from ~2.543 Å to 2.600 Å. The Mo-Oax bond distance is also increased from 1.719
Ǻ to 1.754 Ǻ whereas the Mo-Oeq bond length remains unchanged. The Mo-SCH3 bond
126
Chapter 5-Selenate Reductase
distance is increased from 2.474 Å to 2.570 Å as compared to the optimized A1-TS1
geometry. The Oeq-Se distance is 3.372 Ǻ (Table 5.1).
Ø Optimized transition state complex for oxygen atom transfer (OAT) A1-TS2:
Oxygen atom transfer occurs through a transition state A1-TS2 of distorted square pyramidal
geometry with a relative energy of 37.4 kcal/mol in the polarizable continuum (68.0 kcal/mol
in the gas phase) (Table 5.3). Geometry optimization of A1-TS2 shows that the Mo-Oeq bond
length is elongated from 1.738 Å to 1.945 Å and the Oeq-Se distance is decreased from 3.372
Å to 1.969 Å while the Mo-OSe distance is increased from 2.231 Ǻ to 4.675 Ǻ when
comparison is made with the optimized A1-ES geometry. The Mo-Oax bond distance is
decreased from 1.754 Ǻ to 1.721 Ǻ. At this stage, the dithiolene Mo-S bond distances,
elongated in the previous step, are decreased from ~2.600 to ~2.476 Å. The -SCH3 group is
moved back to an equatorial position (Mo-SCH3 = 2.439 Å). An imaginary frequency of 339ί
cm-1 corresponds to the stretching vibration modes of Mo-Oeq and Oeq-Se (HSeO3-) and
confirms the saddle point character between the selenite and selenate complexes (Table 5.1).
Ø Optimized product bound complex A1-EP:
To complete OAT reaction, a product bound complex [MoIVO(S2C2Me2)SMe(HSeO4)]2(A1-EP) is formed as a result of breaking of Mo-Oeq bond and formation of Oeq-Se bond in
the optimized A1-TS2 geometry. The computed reaction energy for the formation of A1-EP
is endothermic, 35.9 kcal/mol and 62.8 kcal/mol are computed with and without a continuum
model, respectively, relative to the separate substrate, HSeO3- and educt (A1-E) complex
(Table 5.3). The A1-EP is higher in energy than A1-ES by 27.7 kcal/mol in the continuum
(22.8 kcal/mol in the gas phase). At this stage, the OAT reaction is complete and HSeO4- is
loosely bound to the MoIV centre via one oxo group in A1-EP.
The optimized data shows that the Mo-Oeq bond and Mo-Oax bond distances are elongated
from 1.945 Å to 2.296 Å and from 1.721 Ǻ to 1.731 Ǻ, respectively. The Oeq-Se (HSeO3-)
bond length is decreased from 1.969 Å to 1.669 Å as compared to the optimized A1-TS2
geometry. The dithiolene Mo-S bond distances are further decreased from ~2.476 Ǻ to ~2.410
Å (Table 5.1).
Ø Optimized reduced complexes A1-P and A1-P’:
Loss of the oxidized substrate HSeO4- from A1-EP gives the reduced complex
[MoIVO(S2C2Me2)SMe]1- (A1-P) with tetrahedral geometry in an endothermic reaction, 34.7
127
Chapter 5-Selenate Reductase
kcal/mol in the continuum (33.8 kcal/mol in gas phase) relative to the separate substrate,
HSeO3- and educt (A1-E) complex (Table 5.3). Complex A1-P is slightly exothermic (-1.2
kcal/mol in the continuum) relative to the A1-EP complex. The optimized data shows the
reduction in the Mo-Oax bond length (from 1.731 Ǻ to 1.713 Ǻ) and in the Mo-S bond
distances (from ~2.410 Ǻ to ~2.350 Ǻ) when comparison is made with the optimized A1-EP
geometry. The Mo-SCH3 bond is also reduced from 2.434 Ǻ to 2.421 Ǻ when comparison is
made with the optimized A1-EP geometry (Table 5.1).
Direct
replacement
of
IV
HSeO4-
in
A1-EP
by
a
water
molecule
leads
to
1-
[Mo O(S2C2Me2)SMe(H2O)] , A1-P’ which has also square pyramidal geometry as A1-EP.
This ligand exchange is computed to be endothermic by 28.7 kcal/mol (26.9 kcal/mol when
no continuum model is applied) relative to the separate substrate, HSeO3- and educt (A1-E)
complex (Table 5.3). The optimized data shows a reduction in the Mo-Oax bond length (from
1.731 Ǻ to 1.713 Ǻ) and in the Mo-S bond distances (from ~2.410 Ǻ to ~2.391 Ǻ) as
compared to the optimized A1-EP geometry. The Mo-SCH3 bond is slightly elongated from
2.434 Ǻ to 2.442 Ǻ when comparison is made with the optimized A1-EP geometry (Table
5.1). The Mo-OH2O bond length is 2.383 Ǻ.
Table 5.1: Selected bond lengths [Å] of optimized stationary points along the reaction path for oxygen
atom transfer from [MoVIO2(S2C2Me2)SMe]1- to HSeO3-.
A1-E
A1-TS1
A1-ES
A1-TS2
A1-EP
A1-P
A1-P’
Mo-Oax (Ǻ)
1.724
1.719
1.754
1.721
1.731
1.713
1.713
Mo-Oeq (Ǻ)
1.736
1.736
1.738
1.945
2.296
-
-
Mo-OH2O (Ǻ)
-
-
-
-
-
-
2.383
Mo-OSe (Ǻ)
-
3.159
2.231
4.675
-
-
-
Oeq-Se (Ǻ)
-
-
3.372
1.969
1.669
-
-
Mo-S1 (Ǻ)
2.562
2.605
2.700
2.484
2.410
2.350
2.375
Mo-S2 (Ǻ)
2.457
2.481
2.499
2.468
2.409
2.350
2.406
Mo-SCH3 (Ǻ)
2.452
2.474
2.570
2.439
2.434
2.421
2.442
Where, A1-E = Mo-dithiolene educt complex, A1-TS1 =transition state complex for educt-substrate
complex formation, A1-ES = educt-substrate complex, A1-TS2 = transition state complex for oxygen
atom transfer, A1-EP = product bound complex, A1-P = reduced product complex without water
molecule, A1-P’ = reduced product complex with water molecule.
128
Chapter 5-Selenate Reductase
Fig. 5.5: Optimized geometries along the reaction pathway for the oxygen atom transfer from
[MoVIO2(S2C2Me2)SMe]1- to HSeO3-.
129
Chapter 5-Selenate Reductase
·
Molybdenum (VI) dithiolene complex [MoVIO2(S2C2Me2)SMe]- with HSO3- A2:
Ø Optimized educt [MoVIO2(S2C2Me2)SMe]- complex A2-E:
The molybdenum dithiolene complex A2-E is identical to the optimized educt complex A1-E,
derived from the protein X-ray crystal structure of the native SO enzyme31 (Table 5.2).
Ø Optimized transition state complex for educt-substrate complex formation A2-TS1:
Oxygen atom transfer (OAT), from square pyramidal [MoVIO2(S2C2Me2)SMe]1- (A2-E) to
HSO3- is initiated by a
transition state [MoVIO2(S2C2Me2)SMe(HSO3)]2- (A2-TS1). The
energy barrier for A2-TS1 is 25.4 kcal/mol in the polarizable continuum (58.0 kcal/mol in the
gas phase) relative to the separate substrate, HSO3-, and educt (A2-E) complex (Table 5.3).
This energy barrier is 15.4 kcal/mol (in the continuum) higher than the energy barrier for the
A1-TS2. One imaginary frequency of 55ί cm-1 corresponds predominantly to the stretching
vibration mode of Mo-OSulfite.
The optimized A2-TS1 has a distorted square pyramidal geometry. The optimized data shows
no considerable change in the Mo-Oax bond distance (from 1.724 Ǻ to 1.721 Ǻ) while the
dithiolene Mo-S bond distances are increased from ~2.510 Ǻ to ~5.548 Ǻ when comparison is
made with the optimized A2-E geometry. Like in A1-TS1, the -SCH3 group moves from an
equatorial to an axial position because of the oxoanionic approach of sulfite and the Mo-SCH3
bond distance is increased from 2.452 Ǻ to 2.482 Ǻ. At this stage, the HSO3- is loosely bound
to the Mo center. The distance between Mo-OSulfite is 3.012Å (Table 5.2).
Ø Optimized educt-substrate intermediate complex A2-ES:
The
computed
VI
relative
energy
for
the
formation
of
intermediate
complex
2-
[Mo O2(S2C2Me2)SMe(HSO3)] (A2-ES) is endothermic, 25.9 kcal/mol in the continuum
(55.7 kcal/mol) (Table 5.3). The relative energy for the formation of A2-ES is similar to the
relative energy for the A2-TS1 (A2-TS1 is 0.5 kcal/mol higher in energy). Geometry
optimization of A2-ES shows a decrease in the Mo-OSulfite distance from 3.012 Ǻ to 2.244 Å
while the dithiolene Mo-S bond distances are increased from ~2.548 Å to ~2.597 Å relative to
the optimized A2-TS1 geometry. The Mo-Oax bond distance is increased from 1.721 Ǻ to
1.757 Ǻ whereas the Mo-Oeq bond length remains unchanged. The Mo-SCH3 bond distance is
increased from 2.482 Å to 2.577Å (Table 5.2). The Oeq-SSulfite distance is 3.522 Ǻ.
130
Chapter 5-Selenate Reductase
Ø Optimized transition state complex for oxygen atom transfer (OAT) A2-TS2:
The energy barrier for the formation of distorted square pyramidal transition state A2-TS2 is
48.8 kcal/mol in the polarizable continuum (79.4 kcal/mol in the gas) relative to the separate
substrate, HSO3- and educt (A2-E) complex (Table 5.3). Geometry optimization of A2-TS2
shows that the Mo-Oeq bond length is elongated from 1.735 Å to 1.908 Å and the Oeq-SSulfite
distance is decreased from 3.552 Å to 1.979 Å whle the Mo-OSulfite distance is increased from
2.244 Ǻ to 4.494 Ǻ as compared to the optimized A2-ES geometry. The Mo-Oax bond
distance is decreased from 1.757 Ǻ to 1.722 Ǻ. At this stage, the dithiolene Mo-S bond
distances, elongated in the previous step, are decreased from ~2.597 to ~2.490 Å. The -SCH3
group is moved back to an equatorial position (Mo-SCH3 = 2.448 Å) (Table 5.2). An imaginary
frequency of 380ί cm-1 corresponds to the stretching vibration modes of Mo-Oeq and OeqSSulfite (HSO3 -) and confirms the saddle point character between the sulfite and sulfate
complexes.
Ø Optimized product bound complex A2-EP:
The
computed
relative
energy
for
the
formation
of
product
bound
complex
[MoIVO(S2C2Me2)SMe(HSO4)]2- (A2-EP) is endothermic, 33.2 kcal/mol in the continuum
(61.1 kcal/mol in the gas) (Table 5.3). The A2-EP is higher in energy than A2-ES by only 7.3
kcal/mol in the continuum (5.4 kcal/mol in the gas phase). At this stage, the OAT reaction is
complete and HSO4- is loosely bound to a MoIV centre in A2-EP.
The optimized data shows that the Mo-Oeq bond and Mo-Oax bond distances are elongated
from 1.908 Å to 2.340 Å and from 1.722 Ǻ to 1.728 Ǻ, respectively when comparison is made
with the optimized A2-TS2 geometry. The Oeq-SSulfite (HSO3-) bond distance is decreased
from 1.979 Å to 1.542 Å. The dithiolene Mo-S bond distances are further decreased from
~2.491 Ǻ to ~2.411 Å (Table 5.2).
Ø Optimized reduced complexes A2-P and A2-P’:
Loss of the oxidized substrate HSeO4- from A2-EP gives the reduced complex
[MoIVO(S2C2Me2)SMe]1- (A2-P) with tetrahedral geometry in an endothermic reaction, 30.5
kcal/mol in the continuum (31.0 kcal/mol in gas phase) relative to the separate substrate,
HSO3-, and educt (A2-E) complex (Table 5.3). Complex A2-P is slightly exothermic (-2.7
kcal/mol in the continuum) relative to the A2-EP complex. The optimized data shows the
reduction in the Mo-Oax bond length (from 1.728 Ǻ to 1.713 Ǻ) and in the Mo-S bond
131
Chapter 5-Selenate Reductase
distances (from ~2.411 Ǻ to ~2.350 Ǻ). The Mo-SCH3 bond is also reduced from 2.436 Ǻ to
2.421 Ǻ when comparison is made with the optimized A2-EP geometry (Table 5.2).
Direct replacement of HSO4- in A2-EP by a water molecule leads to the formation of
[MoIVO(S2C2Me2)SMe(H2O)]1-, A2-P’ which has also square pyramidal geometry as A2-EP.
This ligand exchange is computed to be endothermic by 24.4 kcal/mol in the polarizable
continuum (24.2 kcal/mol in the gas phase) relative to the separate substrate, HSO3- and educt
(A2-E) complex (Table 5.3). The optimized data shows the reduction in the Mo-Oax bond
distance (from 1.728 Ǻ to 1.713 Ǻ) and in the Mo-S bond distances (from ~2.411 Ǻ to ~2.391
Ǻ). The Mo-SCH3 bond is slightly elongated from 2.436 Ǻ to 2.442 Ǻ when comparison is
made with the optimized A2-EP geometry (Table 5.2). The Mo-OH2O bond length is 2.383 Ǻ.
Table 5.2: Selected bond lengths [Å] of optimized geometries along the reaction pathway for oxygen
atom transfer from [MoVIO2(S2C2Me2)SMe]1- to HSO3-.
A2-E
A2-TS1
A2-ES
A2-TS2
A2-EP
A2-P
A2-P’
Mo-Oax (Ǻ)
1.724
1.721
1.757
1.722
1.728
1.713
1.713
Mo-Oeq (Ǻ)
1.736
1.736
1.735
1.908
2.340
-
-
Mo-OH2O (Ǻ)
-
-
-
-
-
-
2.383
Mo-OSulfite (Ǻ)
-
3.012
2.244
4.494
-
-
-
Oeq-SSulfite (Ǻ)
-
-
3.522
1.979
1.542
-
-
Mo-S1 (Ǻ)
2.562
2.615
2.702
2.503
2.411
2.350
2.375
Mo-S2 (Ǻ)
2.457
2.480
2.491
2.478
2.410
2.350
2.406
Mo-SCH3 (Ǻ)
2.452
2.482
2.577
2.448
2.436
2.421
2.442
Where, A2-E = Mo-dithiolene educt complex, A2-TS1 =transition state complex for educt-substrate
complex formation, A2-ES = educt-substrate complex, A2-TS2 = transition state complex for oxygen
atom transfer, A2-EP = product bound complex, A2-P = reduced product complex without water
molecule, A2-P’ = reduced product complex with water molecule.
132
Chapter 5-Selenate Reductase
Fig. 5.6: Optimized geometries along the reaction pathway for oxygen atom transfer from
[MoVIO2(S2C2Me2)SMe]1- to HSO3-.
133
Chapter 5-Selenate Reductase
Table 5.3: Relative energies (kcal/mol) computed for stationary points along the OAT from
[MoVIO2(S2C2Me2)SMe]1- to HSeO3- (A1) and HSO3- (A2)
A1
A2
//B3LYPa
E
0.0
0.0
SDDb
COSMOc
TS1
ES
TS2
EP
P
P’
43.3
58.5
//B3LYPa
47.3
58.0
SDDb
10.0
25.4
COSMOc
39.4
56.3
//B3LYPa
40.0
55.7
SDDb
9.7
25.9
COSMOc
59.3
74.3
//B3LYPa
68.0
79.4
SDDb
37.4
48.8
COSMOc
47.7
51.0
//B3LYPa
62.8
61.1
SDDb
35.9
33.2
COSMOc
17.1
18.6
//B3LYPa
33.8
31.0
SDDb
34.7
30.5
COSMOc
11.8
13.3
//B3LYPa
26.9
24.2
SDDb
28.7
24.4
COSMOc
Where, E = educt complex, TS1 =transition state complex for educt-substrate complex formation,
ES = educt-substrate complex, TS2 = transition state complex for oxygen atom transfer,
EP = product bound complex, P = reduced product complex without water molecule,
P’ = reduced product complex with water molecule.
A1 = Mo(VI)dithiolene complex [MoVIO2(S2C2Me2)SMe]- with HSeO3-,
A2 = Mo(VI)dithiolene complex [MoVIO2(S2C2Me2)SMe]- with HSO3-, a) B3LYP/Lanl2DZ(p),
b) B3LYP/SDDp//B3LYP/Lanl2DZ(p), c) COSMO-B3LYP/SDDp//B3LYP/Lanl2DZ(p)
(see Computational details).
134
Chapter 5-Selenate Reductase
·
Molybdenum (VI) maleonitriledithiolate (mnt) complex with HSeO3- B1:
Ø Optimized educt [MoVIO2(mnt)2]2- complex B1-E:
The [MoVIO2(mnt)2]2- complex, B1-E (distorted octahedral geometry) obtained from the
reported crystal structure,196 was geometry optimized where the oxidation state of the
molybdenum is VI and the overall charge on the complex is -2. The optimized data shows that
both the Mo-Oax and Mo-Oeq bonds are of equal length, i.e., 1.732 Ǻ. The dithiolate Mo-S
bond distances are ~2.622 Ǻ where the dithiolate Mo-S2 and Mo-S4 bonds are trans to Oeq and
trans to Oax, respectively, and are of equal length (2.742 Ǻ). The Mo-S1 and Mo-S3 bonds are
also equal in length (2.502 Ǻ) but shorter than the Mo-S2 and Mo-S4 bonds (Table 5.4).
Ø Optimized transition state complex for educt-substrate complex formation B1-TS1:
Oxygen atom transfer (OAT), from distorted octahedral [MoVIO2(mnt)2]2- (B1-E) to HSeO3- is
initiated by a transition state [MoVIO2(mnt)2(HSeO3)]3- (B1-TS1). The energy barrier for
B1-TS1 is 27.2 kcal/mol in the polarizable continuum (89.2 kcal/mol in the gas phase)
relative to the separate substrate, HSeO3-, and educt (B1-E) complex (Table 5.6). An
imaginary frequency of 59ί cm-1 corresponds to the Mo-OSe stretching mode.
The optimized data shows no considerable change in the Mo-Oax and Mo-Oeq bond distances
while the dithiolate Mo-S bond distances are increased from ~2.622 Ǻ to ~2.687 Ǻ when
comparison is made with the optimized B1-E geometry. The two dithiolate Mo-S bonds (trans
to oxo), Mo-S2 and Mo-S4 are elongated from 2.742 Ǻ to 2.781 Ǻ and from 2.742 Ǻ to 2.898
Ǻ, respectively. At this stage, the HSeO3- is loosely bound to the MoVI center and the Mo-OSe
distance is 2.794 Ǻ (Table 5.4).
Ø Optimized educt-substrate intermediate complex B1-ES:
The
computed
relative
energy
for
the
formation
of
intermediate
complex
[MoVIO2(mnt)2(HSeO3)]3- (B1-ES) is endothermic, 26.0 kcal/mol in the continuum (89.2
kcal/mol) (Table 5.6). Geometry optimization of B1-ES shows a decrease in the Mo-OSe
distance from 2.794 Ǻ to 2.359 Å while the dithiolate Mo-S bond distances are increased from
~2.687 Å to 2.718 Å. The dithiolate Mo-S2 bond is decreased from 2.781 Ǻ to 2.771 Ǻ while
no considerable change is observed in the Mo-S4 bond (from 2.898 Ǻ to 2.901 Ǻ). The
Mo-Oax and the Mo-Oeq bond distances are increased from 1.728 Ǻ to 1.741 Ǻ and from 1.728
Ǻ to 1.734 Ǻ, respectively, when comparison is made with the optimized B1-TS1 geometry.
The Oeq-Se distance is 3.147 Ǻ (Table 5.4).
135
Chapter 5-Selenate Reductase
Ø Optimized transition state complex for oxygen atom transfer (OAT) B1-TS2:
The energy barrier for the transition state B1-TS2 is endothermic, 53.8 kcal/mol in the
polarizable continuum (114.0 kcal/mol in the gas phase) relative to the separate substrate,
HSeO3- and the educt (A1-E) complex (Table 5.6). An imaginary frequency of 386ί cm-1
corresponds to the stretching vibration modes of Mo-Oeq and Oeq-Se (HSeO3 -) and confirms
the saddle point character between the selenite and selenate complexes.
Geometry optimization of B1-TS2 shows an increase in the Mo-Oeq bond length from 1.734 Å
to 1.947 Å and in the Mo-OSe distance from 2.359 Ǻ to 4.689 Ǻ while the Oeq-Se distance is
decreased from 3.147 Å to 1.965 Å. The Mo-Oax (from 1.741 Ǻ to 1.728 Ǻ) and the dithiolate
Mo-S (from ~2.718 Ǻ to ~2.603 Å) bond distances are decreased as compared to the
optimized B1-ES geometry (Table 5.4). The dithiolate Mo-S bonds (trans to oxo), Mo-S2 and
Mo-S4 are decreased from 2.771 Ǻ to 2.585 Ǻ and from 2.901 Ǻ to 2.779 Ǻ, respectively
Ø Optimized product bound complex B1-EP:
To complete OAT reaction, a product bound intermediate complex [MoIVO(mnt)2(HSeO4)]3(B1-EP) is formed as a result of breaking of the Mo-Oeq bond and formation of the Oeq-Se
bond in the optimized B1-TS2 geometry. B1-EP has a distorted octahedral geometry. The
computed relative energy for the formation of B1-EP (distorted square pyramidal geometry)
is endothermic, 35.9 kcal/mol in the polarizable continuum (62.8 kcal/mol in the gas) (Table
5.6). At this stage, the OAT reaction is complete and HSeO4- is loosely bound to a MoIV
centre.
The optimized data shows a decrease in the Mo-Oax bond distance from 1.728 Ǻ to 1.702 Ǻ
while the Mo-Oeq bond is elongated from 1.947 Å to 2.349 Å relative to the optimized
B1-TS2 geometry. The Oeq-Se (HSeO3-) bond distance is decreased from 1.965 Å to 1.663 Å.
The dithiolate Mo-S bond distances are further decreased from ~2.603 Ǻ to ~2.568 Å (Table
5.4). The dithiolate Mo-S2 bond is decreased from 2.585 Ǻ to 2.472 Ǻ while the Mo-S4 bond
is increased from 2.779 Ǻ to 2.799 Ǻ.
Ø Optimized reduced complex B1-P:
The loss of the oxidized substrate, HSeO4- gives the reduced complex [MoIVO(mnt)2]2- (B1-P)
with square pyramidal geometry in an endothermic reaction, 7.1 kcal/mol in the polarizable
continuum (7.7 kcal/mol in gas phase) relative to the separate substrate, HSeO3-, and educt
(B1-E) complex (Table 5.6). The optimized data shows no considerable change in the Mo-Oax
136
Chapter 5-Selenate Reductase
bond distance (from 1.702 Ǻ to 1.701 Ǻ) while the dithiolate Mo-S bond distances are
decreased from ~2.568 Ǻ to ~2.458 Ǻ where all Mo-S bonds are of equal length when
comparison is made with the optimized B1-EP geometry (Table 5.4).
Table 5.4: Selected bond lengths [Å] of optimized geometries along the reaction pathway for oxygen
VI
2-
atom transfer from [Mo O2(mnt)2] to HSeO3-.
B1-E
B1-TS1
B1-ES
B1-TS2
B1-EP
B1-P
Mo-Oax (Ǻ)
1.732
1.728
1.741
1.728
1.702
1.701
Mo-Oeq (Ǻ)
1.732
1.728
1.734
1.947
2.349
-
Mo-OSe (Ǻ)
-
2.794
2.359
4.689
-
-
Oeq-Se (Ǻ)
-
-
3.147
1.965
1.663
-
Mo-S1 (Ǻ)
2.502
2.561
2.638
2.482
2.447
2.458
Mo-S2 (Ǻ)
2.742
2.781
2.771
2.585
2.472
2.458
Mo-S3 (Ǻ)
2.502
2.508
2.561
2.565
2.554
2.458
Mo-Sa (Ǻ)
2.742
2.898
2.901
2.779
2.799
2.458
Where, B1-E = Mo-mnt educt complex, B1-TS1 =transition state complex for educt-substrate
complex formation, B1-ES = educt-substrate complex, B1-TS2 = transition state complex for oxygen
atom transfer, B1-EP = product bound complex, B1-P = reduced product complex.
137
Chapter 5-Selenate Reductase
Fig. 5.7: Optimized geometries along the reaction pathway for oxygen atom transfer from
[MoVIO2(mnt)2]2- to HSeO3-.
138
Chapter 5-Selenate Reductase
·
Molybdenum (VI) maleonitriledithiolate (mnt) complex with HSO3- B2:
Ø Optimized educt [MoVIO2(mnt)2]2- complex B2-E:
The [MoVIO2(mnt)2]2- complex, B2-E is identical to the optimized educt complex B1-E
obtained from the reported crystal structure196 (Table 5.5).
Ø Optimized transition state complex for educt-substrate complex formation B2-TS1:
The energy barrier for the transition state [MoVIO2(mnt)2(HSO3)]3- (B2-TS1) associated with
the formation of educt substrate complex is 37.0 kcal/mol in the polarizable continuum (93.6
kcal/mol in the gas phase) (Table 5.6). An imaginary frequency of 27ί cm-1 corresponds to the
Mo-OSulfite stretching mode.
The optimized data shows no considerable change in the Mo-Oax (from 1.732 Ǻ to 1.734 Ǻ)
and Mo-Oeq (from 1.732 Ǻ to 1.728 Ǻ) bond distances when comparison is made with the
optimized B2-E geometry (Table 5.5). The ditholate Mo-S2 and Mo-S4 bonds are elongated
from 2.742 Ǻ to 2.784 Ǻ and from 2.742 Ǻ to 2.903 Ǻ, respectively. At this stage, the HSO3 is loosely bound to the MoVI center and the Mo-OSe distance is 2.575 Ǻ (Table 5.5).
Ø Optimized educt-substrate intermediate complex B2-ES:
The
computed
reaction
energy
for
the
formation
of
intermediate
complex
[MoVIO2(mnt)2(HSO3)]3- (B2-ES) is endothermic, 20.5 kcal/mol in the polarizable continuum
(93.0 kcal/mol) relative to the separate substrate, HSO3-, and educt (B2-E) complex (Table
5.6). B2-ES complex is ~6 kcal/mol lower in energy than B1-ES. Geometry optimization of
B2-ES shows a decrease in the Mo-OSulfite distance from 2.575 Ǻ to 2.500 Å. No considerable
change is observed in the dithiolate Mo-S2 and the Mo-S4 bond distances as well as in the
Mo-Oax and the Mo-Oeq bond distances when comparison is made to the optimized B2-TS1
geometry. The Oeq-SSulfite distance is 3.256 Ǻ (Table 5.5).
Ø Optimized transition state complex for oxygen atom transfer (OAT) B2-TS2:
The energy barrier for the transition state B2-TS2, associated with the oxygen atom transfer,
is endothermic, 53.7 kcal/mol in the polarizable continuum (114.1 kcal/mol in the gas phase)
relative to the substrate, HSO3-, and the educt (B2-E) complex (Table 5.6). This energy
barrier of B2-TS2 is similar to the energy barrier for B1-TS2 (see Table 5.6). An imaginary
frequency of 434ί cm-1 corresponds to the stretching vibration modes of Mo-Oeq and
139
Chapter 5-Selenate Reductase
Oeq-SSulfite (HSO3 -) and confirms the saddle point character between the sulfite and sulfate
complexes.
Geometry optimization of B2-TS2 shows an increase in the Mo-Oeq bond length from 1.729 Å
to 1.905 Å and Mo-OSulfite distance from 2.509 Ǻ to 4.470 Ǻ while the Oeq-SSulfite distance is
reduced from 3.256 Å to 1.974 Å. No considerable change is observed in the Mo-Oax (from
1.737 Ǻ to 1.734 Ǻ) bond distance. The dithiolate Mo-S2 and Mo-S4 bond distances are
decreased from 2.782 Ǻ to 2.622 Å and from 2.903 Ǻ to 2.779 Å as compared to the
optimized B2-ES geometry (Table 5.5).
Ø Optimized product bound complex B2-EP:
The computed relative energy for the formation of product bound intermediate complex
[MoIVO(mnt)2(HSO4)]3- (B2-EP) is also endothermic, 31.0 kcal/mol in the polarizable
continuum (93.8 kcal/mol in the gas) (Table 5.6).
The optimized data shows a decrease in the Mo-Oax bond distance from 1.734 Ǻ to 1.702 Ǻ
while the Mo-Oeq bond is elongated from 1.905 Å to 2.345 Å relative to the optimized
B2-TS2 geometry. The Oeq-SSulfite (HSO3 -) bond distance is decreased from 1.974 Å to 1.537
Å and the dithiolate Mo-S2 bond is decreased from 2.622 Ǻ to 2.472 Ǻ while the Mo-S4 bond
is increased from 2.779 Ǻ to 2.806 Ǻ (Table 5.5).
Ø Optimized reduced complex B2-P:
The loss of oxidized substrate, HSO4- gives the reduced complex [MoIVO(mnt)2]2- (B2-P)
with square pyramidal geometry in an exothermic reaction, -10.6 kcal/mol in the polarizable
continuum (-7.6 kcal/mol in gas phase) relative to the separate substrate, HSO3-, and educt
(B2-E) complex (Table 5.6). The optimized data shows no considerable change in the Mo-Oax
bond distance (from 1.702 Ǻ to 1.701 Ǻ) while the Mo-S bond distances are decreased from
~2.571 Ǻ to ~2.458 Ǻ where all Mo-S bonds are of equal length when comparison is made
with the optimized B2-EP geometry (Table 5.5).
140
Chapter 5-Selenate Reductase
Table 5.5: Selected bond lengths [Å] of optimized geometries along the reaction pathway for oxygen
atom transfer from [MoVIO2(mnt)2]2- to HSO3-.
B2-E
B2-TS1
B2-ES
B2-TS2
B2-EP
B2-P
Mo-Oax (Ǻ)
1.732
1.734
1.737
1.734
1.702
1.701
Mo-Oeq (Ǻ)
1.732
1.728
1.729
1.905
2.345
-
Mo-OSulfite (Ǻ)
-
2.575
2.509
4.470
-
-
Oeq-SSulfite (Ǻ)
-
-
3.256
1.974
1.537
-
Mo-S1 (Ǻ)
2.502
2.588
2.600
2.488
2.446
2.458
Mo-S2 (Ǻ)
2.742
2.784
2.782
2.622
2.472
2.458
Mo-S3 (Ǻ)
2.502
2.534
2.543
2.572
2.561
2.458
Mo-Sa (Ǻ)
2.742
2.903
2.903
2.779
2.806
2.458
Where, B2-E = Mo-mnt educt complex, B2-TS1 =transition state complex for educt-substrate
complex formation, B2-ES = educt-substrate complex, B2-TS2 = transition state complex for oxygen
atom transfer, B2-EP = product bound complex, B2-P = reduced product complex.
141
Chapter 5-Selenate Reductase
Fig. 5.8: Optimized geometries along the reaction pathway for oxygen atom transfer from
[MoVIO2(mnt)2]2- to HSO3-.
142
Chapter 5-Selenate Reductase
Table 5.6: Relative energies (kcal/mol) computed for stationary points along the OAT from
[MoVIO2(mnt)2]2- to HSeO3- (B1) and HSO3- (B2)
B1
B2
//B3LYPa
E
0.0
0.0
SDDb
COSMOc
TS1
ES
TS2
EP
P
89.8
92.7
//B3LYPa
89.2
93.6
SDDb
27.2
37.0
COSMOc
89.4
92.8
//B3LYPa
89.2
93.0
SDDb
26.0
20.5
COSMOc
106.6
109.1
//B3LYPa
114.0
114.1
SDDb
53.8
53.7
COSMOc
94.2
84.0
//B3LYPa
106.7
93.8
SDDb
46.4
31.0
COSMOc
-5.8
-17.6
//B3LYPa
7.7
-7.6
SDDb
7.1
-10.6
COSMOc
Where, E = educt complex, TS1 =transition state complex for educt-substrate complex formation,
ES = educt-substrate complex, TS2 = transition state complex for oxygen atom transfer,
EP = product bound complex, P = product complex. B1 = Mo (VI) maleonitriledithiolate complex
with HSeO3-, B2 = Mo (VI) maleonitriledithiolate complex with HSO3-. a) B3LYP/Lanl2DZ(p),
b) B3LYP/SDDp//B3LYP/Lanl2DZ(p), c) COSMO-B3LYP/SDDp//B3LYP/Lanl2DZ(p)
(see Computational details).
143
Chapter 5-Selenate Reductase
2.4. Discussion
Sulfite oxidase (SO), selenate reductase (SeR) and nitrate reductases (NRs) are among the
mononuclear molybdenum enzymes involved in the catalysis of redox reactions in biological
activities. The active site composition of SO has one MPT ligand and it oxidizes the sulfite to
sulfate, SeR has two MPT ligands and it reduces the selenate to selenite, while NRs reduces
nitrate to nitrite by either one or with two MPT’s. In this context the question arises whether
the active site itself is special in some way for the oxidation/reduction process of the one or
the other substrate or the different active sites behave the same way and it is the role of the
protein to make it specific. In order to address these questions, density functional theory
(DFT)
studies
have
been
performed
on
the
computational
model
complex,
[MoVIO2(S2C2Me2)SMe]- (A) derived from the X-ray crystal structure of native SO,31 and on
the experimental model complex196 [MoVIO2(mnt)2]2- (B, coordination geometry similar to the
active site of SeR) for the oxidation of selenite and sulfite. For both the computational and
experimental model complexes, two transition states (TS1 and TS2) are involved in the
oxygen atom transfer (OAT) reaction from the MoVI to the substrate.
Three different levels of computation (OPT, SDD and COSMO) were considered for all the
geometries involved in the OAT reaction mechanisms. The relative energies show a slight
change between the level of optimization (OPT) and the single point energy calculations in
the gas phase (SDD) but a major difference is evident between the SDD and the polarizable
continuum (COSMO) results where the relative energy decreases in the presence of
continuum models (see Table 5.3 and 5.6). This may be because both the reactant and the
substrate molecules are negatively charged and the association of two negatively charged ions
increases the overall energy of the reaction pathway in the gas phase due to the repulsive
interaction between two similar charged species. This repulsive interaction is stabilized in the
continuum model resulting in the decrease of energies (see Table 5.3 and 5.6).
In the computed reaction pathway for the OAT from model complex A to HSeO3- (A1), the
energy barrier associated with A1-TS1, required for the formation of educt substrate complex,
is 10.0 kcal/mol in the polarizable continuum. The educt-substrate A1-ES complex is only 0.3
kcal/mol lower in energy than A1-TS1. The energy barrier associated with the OAT from Mo
to HSeO3- (A1-TS2) is 37.4 kcal/mol in the polarizable continuum, the rate limiting step. The
relative energy for the formation of HSeO4- bound product complex A1-EP is 35.9 kcal/mol
in the polarizable continuum. A1-EP is 1.5 kcal/mol lower in energy than the A1-TS2. The
relative energy for the formation of reduced complex A1-P (starting geometry is generated by
144
Chapter 5-Selenate Reductase
the removal of oxidized substrate HSeO4- from A1-EP) is 34.7 kcal/mol while relative energy
for the direct replacement of HSeO4- with water molecule (A1-P’) in A1-EP is 28.7 kcal/mol
in the continuum. A1-P’ is ~6 kcal/mol lower in energy than A1-P (Table 5.3).
In the computed reaction pathway for the OAT from model complex A to HSO3- (A2), the
energy barrier associated with A2-TS1 is 25.4 kcal/mol in the polarizable continuum i.e.
much higher then for HSeO3 - (10.0 kcal/mol). The computed relative energy for the formation
of educt-substrate complex A2-ES (25.9 kcal/mol) is similar to the relative energy for A2TS1. The energy barrier associated with OAT reaction (A2-TS2, the rate limiting step) is 48.8
kcal/mol in the polarizable continuum relative to separate substrate and educt complex A2-E.
The barrier for OAT in the substrate complex, however (i.e. A2-TS2 vs A2-ES) is 22.9
kcal/mol. The activation energy for OAT in the HSeO3- complex is 27.7 kcal/mol (A1-TS2 vs
A1-ES) and hence larger by 4.8 kcal/mol. The formation of HSO4- bound product complex
A2-EP is endothermic, 33.2 kcal/mol in the polarizable continuum relative to A2-E and
HSO3-. The energy relative to the substrate complex is only 7.3 kcal/mol (26.2 kcal/mol for
HSeO3- as substrate i.e. A1-EP vs A1-E). With the involvement of one water molecule to
replace the oxidized substrate the OAT reaction starting from the substrate complex becomes
slightly exothermic (-1.5 kcal/mol) for HSO3- but is computed to be considerably endothermic
for HSeO3- as substrate (+19.0 kcal/mol) (compare Table 5.3).
In summary, when the active site with one enedithiolato ligand (A) should oxidize a substrate
it is a much better catalyst for HSO3- than for HSeO3- (lower barrier, reaction energy close to
thermoneutral). In a reduction reaction HSeO4- requires a smaller activation energy (8.7
kcal/mol, A1-P’ vs A1-TS2) than HSO4- (24.4 kcal/mol, A2-P’ vs A2-TS2) but shows
considerable exothermicity (-19.0 kcal/mol) (Scheme 5.1).
145
Chapter 5-Selenate Reductase
Scheme 5.1: Plot of computed reaction energies (kcal/mol) relative to separate substrate and educt
complex vs steps involved in the OAT from educt complex A to HSeO3- and HSO3-.
In the computed reaction pathway for the OAT from model complex B to HSeO3- (B1), the
energy barrier associated with B1-TS1 is 27.2 kcal/mol in the polarizable continuum. The
relative energy for the formation of educt substrate B1-ES complex is 26.0 kcal/mol in the
polarizable continuum. B1-ES is 1.2 kcal/mol lower in energy than the B1-TS1. The energy
barrier associated with B1-TS2, the rate limiting step, is 53.8 kcal/mol in the polarizable
continuum. This results in an energy barrier of 27.8 kcal/mol for OAT in the substrate
complex (B1-TS2 vs B1-ES), which is almost identical to the value computed for model A1
(27.7 kcal/mol). The relative energy for the formation of the HSeO4- bound product complex
B1-EP is 46.4 kcal/mol in the polarizable continuum. The loss of HSeO4- from the B1-EP
complex gives a B1-P complex in an exothermic reaction, -39.3 kcal/mol in the polarizable
continuum (Table 5.6).
In the computed reaction pathway for the OAT from model complex B to HSO3- (B2), the
energy barrier for the formation of educt-substrate complex (B2-TS1) is 37.0 kcal/mol in the
146
Chapter 5-Selenate Reductase
polarizable continuum. The relative energy for the formation of B2-ES is 20.5 kcal/mol in the
polarizable continuum. B2-ES is 16.5 kcal/mol lower in energy than B2-TS1. The B2-ES
intermediate is therefore in a much deeper dip on the potential energy surface than both the
HSeO3- intermediate complex and the A2-ES HSO3- intermediate complex. The energy
barrier associated with the B2-TS2 is 53.7 kcal/mol in the polarizable continuum. The barrier
for OAT in the substrate complex B2-ES is therefore 33.2 kcal/mol. This value is
significantly higher than for the HSeO3 - oxidation (27.8 kcal/mol) and also higher than for the
HSO3- oxidation with the A2 model (22.9 kcal/mol). The relative energy for the HSO4- bound
product complex B2-EP is 31.0 kcal/mol in the polarizable continuum. The loss of HSO4from the B2-EP gives a reduced complex B2-P in an exothermic reaction, -41.6 kcal/mol in
the polarizable continuum (Table 5.6). The overall reaction from separate educt complex E
and substrate to separate product complex P and oxidized substrate is computed to be
exothermic for HSO3- (B2, -10.6 kcal/mol) but endothermic for HSeO3- as substrate (B1, 7.1
kcal/mol).
In both cases (B1 and B2) the release of the product, HSeO4- and HSO4-, respectively, is
computed to be strongly exothermic. However, this will probably not be the case for the
enzyme. Fivefold coordinate product complex P adopts a square pyramidal geometry, which
is not possible in the presence of the protein framework which restricts the flexibility of the
large molybdopterin ligands. More meaningful are therefore the reaction energies leading to
the product complexes P or the water complex P’ and released product. However, distorted
octahedral starting geometries for the geometry optimization of [MoO(OH2)(mnt)2]2- lead to
square pyramidal [MoO(mnt)2]2- with the water molecule only bound through hydrogen bonds
without coordination to Mo (Scheme 5.2, Table 5.6).
Considering all computational results one can state that model A with one enedithiolate ligand
is well suited for the oxidation of sulfite. It has a lower barrier than for the selenite oxidation
and also than model complex B with two chelating ligands.
The reduction is always easier for selenate than for sulfate (lower activation, stronger
exothermicity). The hydrogen sulfate complex needs less activation for OAT with model A
than with model B. However, the reduction is much more exothermic with A than with B.
Of course the energy profile obtained here for small model complexes may be significantly
modeifoed in the presence of the protein part of the enzyme.
147
Chapter 5-Selenate Reductase
Scheme 5.2: Plot of computed reaction energies (kcal/mol) relative to separate substrate and educt
complex vs steps involved in the OAT from educt complex B to HSeO3- and HSO3-.
3. Project: II
Wang et al.184 depicted a reaction pathway for the reduction of selenate to selenite, (Fig. 5.9)
where the oxidation state of metal changes from +IV to +VI. Two electrons and two protons
are required for the reductive half reaction, resulting in the formation of a water molecule and
a selenite ion (Eq 5.1).
SeO42- + 2H+ + 2e-
SeO32- + H2O (5.1)
148
Chapter 5-Selenate Reductase
Fig. 5.9: Selenate reduction reaction based on the active site structures postulated from X-ray
absorption spectroscopy depicted by Wang et al.184
Here, we have investigated additional ways of binding the substrate with the active site
besides the reported analogous way (Fig. 5.9) by using the density functional theory (DFT).
Experimental model complex [MoIV(OH)(mnt)2]1- (1) obtained from the reported crystal
structure196 was used as an initial geometry.
3.1. Computational Details
All calculations were performed with the Gaussian 03175 program package using B3LYP123
hybrid functional of density functional theory (DFT). For the geometry optimizations, the
LANL2DZ basis set124,125,126,127 augmented by polarization functions on all atoms except Mo
and H (ζ =0.600, 1.154, 0.864, 0.421, 0.338 for C, O, N, S and Se, respectively)128, was
employed. The optimized minima and transition-state structures were characterized by
frequency calculations with the same method and basis set to verify that all minima have no
imaginary frequency and each transition state has exactly one. Single point energies were
computed with the B3LYP functional and the Stuttgart-Dresden effective core potential basis
set (SDD)129,130 augmented by polarization functions for all atoms except Mo, and H.128 The
self-consistent reaction field (SCRF) computations were performed on the optimized
geometries by a conductor like polarizable continuum method (CPCM)131 with a dielectric
constant of 4 and solvent radius of 1.4Ǻ.
149
Chapter 5-Selenate Reductase
3.2. Active Site Models
The initial active site geometry for the model complex [MoIV(OH)(mnt)2]1- (1) studied here
were obtained from the reported crystal structure.196
Fig. 5.10: Active site composition of initial active site geometry for selenate reductase obtained from
the reported crystal structure.196
3.3. Results
For the oxygen atom transfer (OAT) reaction, in which HSeO4- reacts with the bis-dithiolene
Mo-OH complex [MoIV(OH)(mnt)2]- (1) having square pyramidal geometry, different
coordination modes between the two molecules were considered:
Fig. 5.11: Different ways of binding the substrate to the active site,
where, Ln = bis-maleonitriledithiolate (mnt) ligand.
Ø Optimized educt [MoIV(OH)(mnt)2]- complex 1:
The [MoIVO (mnt)2]- complex, 1 obtained from the reported crystal structure,196 was geometry
optimized where the oxidation state of the molybdenum is IV and the overall charge on the
complex is -1. The optimized data shows that the dithiolenes are not twisted against each
other as the S1-S2-S3-S4 dihedral angle is 0.0˚ and the Mo-S bond distances are ~2.381 Ǻ. The
Mo-O1 and O1-H1 bond distances are 1.882 Ǻ and 0.971 Ǻ, respectively (Table 5.11).
150
Chapter 5-Selenate Reductase
·
1A:
In this case, the interaction between the two molecules is complemented by the attachment of
selenate by its oxygen atom Oa to the Mo center as well as by a bond between the hydrogen
atom of selenate (HSe) and the hydroxo oxygen (O1) attached to the Mo centre (Fig. 5.11).
Ø Optimized educt-substrate complex, 1A-ES:
The computed reaction energy for the formation of educt-substrate complex, 1A-ES is
endothermic, 7.2 kcal/mol in the polarizable continuum (20.1 kcal/mol in the gas phase)
relative to the separate substrate and educt (1) complex (Table 5.11). The optimized data
shows an increase in the dithiolene S1-S2-S3-S4 dihedral angle (from 0.0˚ to 20.3˚) and in the
Mo-S bond distances (from ~2.381 Ǻ to ~2.411 Ǻ) when comparison is made with the
optimized 1 geometry. The Mo-O1 bond distance is increased from 1.882 Ǻ to 2.034 Ǻ. The
O1-HSe distance (hydrogen bond distance) is 1.552 Ǻ and the HSe-Ob bond distance is 1.015 Ǻ.
The Mo-Oa and Oa-Se bond distances are 2.116 Ǻ and 1.708 Ǻ, respectively (Table 5.7).
Ø Optimized transition state complex, 1A-TS:
The energy barrier for the oxygen atom transfer from substrate to the Mo center (1A-TS) is
13.1 kcal/mol in the polarizable continuum (27.5 kcal/mol in the gas phase) relative to the
separate substrate and educt (1) complex (Table 5.11). An imaginary frequency of 157ί cm-1
corresponds to the stretching vibration modes of Mo-Oa and Oa-Se (HSeO4-) and confirms the
saddle point character between the selenate and selenite complexes.
Geometry optimization of 1A-TS shows an increase in the dithiolene S1-S2-S3-S4 dihedral
angle (from 20.3˚ to 37.7˚) and in the Mo-S bond distances (from ~2.411 Ǻ to ~2.489 Ǻ)
when comparison is made with the optimized 1A-ES geometry. The Mo-O1 (from 2.034 Ǻ to
2.015 Ǻ), HSe-Ob (from 1.015 Ǻ to 0.992 Ǻ) and Mo-Oa (from 2.116 Ǻ to 1.826 Ǻ) bond
distances are decreased while O1-HSe and Oa-Se bond distances are increased from 1.552 Ǻ to
1.715 Ǻ and from 1.708 Ǻ to 2.217 Ǻ, respectively (Table 5.7).
Ø Optimized product bound complex, 1A-EP:
During the geometry optimization of 1A-EP, the hydrogen H1 from the hydroxo group of Mo
complex is shifted to the selenite complex resulting in the formation of di-oxo MoVI complex
and the selenous acid H2SeO3 which are not the desired products. The relative energy for the
151
Chapter 5-Selenate Reductase
formation of 1A-EP complex is exothermic, -26.0 kcal/mol in the polarizable continuum
(-16.7 kcal/mol in the gas phase) (Table 5.11).
The optimized data shows an increase in the dithiolene S1-S2-S3-S4 dihedral angle (from 37.7˚
to 54.0˚) and in the Mo-S bond distances (from ~2.489 Ǻ to ~2.589 Ǻ) when comparison is
made with the optimized 1A-TS geometry. A decrease in the Mo-O1 (from 2.015 Ǻ to 1.723
Ǻ) and Mo-Oa (from 1.826 Ǻ to 1.783 Ǻ) bond distances are observed. The Oa-Se distance is
increased from 2.217 Ǻ to 3.391 Ǻ (Table 5.7).
Ø Optimized oxidized product complex, 1A-P:
The removal of selenous acid H2SeO3 leads to the formation of dioxo Mo complex
[MoVIO2(mnt)2]2- (1A-P). The computed reaction energy for the formation 1A-P is
exothermic (-17.0 kcal/mol) in the polarizable continuum while it is endothermic (9.0
kcal/mol) in the gas phase relative to the separate substrate and educt (1) complex (Table
5.11).
The optimized data shows no change in the dithiolene S1-S2-S3-S4 dihedral angle while the
Mo-S bond distances are increased from ~2.589 Ǻ to ~2.622 Ǻ when comparison is made
with the optimized 1A-EP geometry. The Mo-Oa bond distance is decreased from 1.783 Ǻ to
1.732 Ǻ (Table 5.7).
152
Chapter 5-Selenate Reductase
Table 5.7: Selected bond lengths [Å] of optimized geometries along the reaction pathway for oxygen
atom transfer from HSeO4- to [MoIVOH(mnt)2]- in the 1A coordination mode of substrate binding.
A
1
1A-ES
1A-TS
1A-EP
1A-P
Mo-O1 (Ǻ)
1.882
2.034
2.015
1.723
1.732
O1-H1 (Ǻ)
0.971
0.966
0.968
-
-
O1-HSe (Ǻ)
-
1.552
1.715
-
-
HSe-Ob(Ǻ)
-
1.015
0.992
0.995
-
Mo-Oa (Ǻ)
-
2.116
1.826
1.783
1.732
Oa-Se (Ǻ)
-
1.708
2.217
3.391
-
Mo-S1 (Ǻ)
2.385
2.402
2.454
2.491
2.502
Mo-S2 (Ǻ)
2.386
2.432
2.517
2.662
2.742
Mo-S3 (Ǻ)
2.377
2.411
2.475
2.475
2.502
Mo-S4 (Ǻ)
2.377
2.400
2.509
2.726
2.742
S1-S2-S3-S4 (˚)
0.0
20.3
37.7
54.0
54.1
Where, 1 = Mo(IV) bis-maleonitriledithiolate complex, 1A-ES = educt-substrate complex,
1A-TS = transition state complex, 1A-EP = product bound complex,
1A-P = oxidized product complex.
153
Chapter 5-Selenate Reductase
Fig. 5.12: Optimized geometries along the reaction pathway for oxygen atom transfer from HSeO4- to
[MoIVOH(mnt)2]- in the 1A coordination mode of substrate binding.
154
Chapter 5-Selenate Reductase
·
1B:
In this case, the interaction between the educt (1) and substrate molecule is complemented by
the attachment of selenate by its oxygen atom O3 to the Mo center as well as by a bond
between the hydroxo hydrogen atom of Mo (H1) and the hydroxo oxygen atom (Ob) of
selenate (Fig. 5.11).
Ø Optimized educt-substrate complex, 1B-ES:
The computed reaction energy for the formation of educt-substrate complex 1B-ES is
endothermic, 15.3 kcal/mol in the polarizable continuum (26.2 kcal/mol in the gas phase)
relative to the separate substrate and educt (1) complex (Table 5.11). The optimized data
shows a slight increase in the Mo-O1 and O1-H1 bond distances from 1.882 Ǻ to 1.891 Ǻ and
from 0.971 Ǻ to 0.980 Ǻ, respectively, when comparison is made with the optimized 1
geometry. The H1-Ob, Mo-Oa and Oa-Se bond distances are 1.968 Ǻ, 2.365 Ǻ and 1.664 Ǻ,
respectively. The dithiolene S1-S2-S3-S4 dihedral angle (from 0.0˚ to 33.0˚) and the Mo-S
bond distances (from ~2.381 Ǻ to ~2.418 Ǻ) are increased (Table 5.8).
Ø Optimized transition state complex, 1B-TS:
The energy barrier for the oxygen atom transfer from substrate to the Mo center (1B-TS) is
20.1 kcal/mol in the polarizable continuum (39.1 kcal/mol in the gas phase) relative to the
separate substrate and educt (1) complex (Table 5.11). This energy barrier associated with
1B-TS is 7 kcal/mol higher than for 1A-TS. An imaginary frequency of 376ί cm-1
corresponds to the stretching vibration modes of Mo-Oa and Oa-Se (HSeO4-) and confirms the
saddle point character between the selenate and selenite complexes.
Geometry optimization of 1B-TS shows an increase in the Mo-O1 (from 1.891 Ǻ to 1.954 Ǻ),
H1-Ob (from 1.968 Ǻ to 1.994 Ǻ) and Oa-Se (from 1.664 Ǻ to 2.028 Ǻ) bond distances when
comparison is made with the optimized 1B-ES geometry. No change is observed in the O1-H1
bond distance while the Mo-Oa distance is decreased from 2.365 Ǻ to 1.902 Ǻ. An increase is
observed in the dithiolene S1-S2-S3-S4 dihedral angle (from 33.0˚ to 43.7˚) and in the Mo-S
bond distances (from ~2.418 Ǻ to ~2.485 Ǻ) (Table 5.8).
Ø Optimized product bound complex, 1B-EP:
During the geometry optimization of 1B-EP, the hydrogen H1 from the hydroxo group of Mo
complex is shifted to the hydroxo oxygen atom Ob of selenite complex resulting in the
155
Chapter 5-Selenate Reductase
formation of SeO2, H2O and di-oxo MoVI complex which are not the desired products. The
computed reaction energy for the formation of 1B-EP complex is exothermic, -19.5 kcal/mol
in the polarizable continuum (-11.5 kcal/mol in the gas phase) relative to the separate
substrate and educt (1) complex (Table 5.11).
The optimized data shows a decrease in the Mo-O1 (from 1.954 Ǻ to 1.763 Ǻ), H1-Ob (from
1.994 Ǻ to 0.999 Ǻ) and Mo-Oa (from 1.902 Ǻ to 1.731 Ǻ) bond distances when comparison
is made with the optimized 1B-TS geometry. The O1-H1 and the Oa-Se bond distances are
increased from 0.980 Ǻ to 1.679 Ǻ and from 2.028 Ǻ to 3.511 Ǻ, respectively. An increase is
observed in the dithiolene S1-S2-S3-S4 dihedral angle (from 43.7˚ to 52.3˚) and in the Mo-S
bond distances (from ~2.485 Ǻ to ~2.593 Ǻ) (Table 5.8).
Ø Optimized oxidized product complex, 1B-P:
The removal of SeO2 and H2O molecules from 1B-EP leads to the formation of dioxo Mo
complex [MoVIO2(mnt)2]2- (1B-P) which has the computed relative energy of -3.5 kcal/mol in
the polarizable continuum while it is endothermic (26.4 kcal/mol) in the gas phase (Table
5.11).
The optimized data shows a decrease in the Mo-O1 bond distance from 1.763 Ǻ to 1.732 Ǻ
while no change is observed in the Mo-Oa bond distance when comparison is made with the
optimized 1B-EP geometry. No considerable change is observed in the dithiolene S1-S2-S3-S4
dihedral angle (from 52.3˚ to 54.1˚) while the Mo-S bond distances are increased from ~2.593
Ǻ to ~2.622 Ǻ (Table 5.8).
156
Chapter 5-Selenate Reductase
Table 5.8: Selected bond lengths [Å] of optimized geometries along the reaction pathway for oxygen
atom transfer from HSeO4- to [MoIVOH(mnt)2]- in the 1B coordination mode of substrate binding.
B
1
1B-ES
1B-TS
1B-EP
1B-P
Mo-O1 (Ǻ)
1.882
1.891
1.954
1.763
1.732
O1-H1 (Ǻ)
0.971
0.980
0.980
1.679
-
H1- Ob (Ǻ)
-
1.968
1.994
0.999
-
Mo-Oa (Ǻ)
-
2.365
1.902
1.731
1.732
Oa-Se (Ǻ)
-
1.664
2.028
3.511
-
Mo-S1 (Ǻ)
2.385
2.339
2.435
2.480
2.502
Mo-S2 (Ǻ)
2.386
2.368
2.480
2.713
2.742
Mo-S3 (Ǻ)
2.377
2.485
2.520
2.505
2.502
Mo-S4 (Ǻ)
2.377
2.481
2.506
2.675
2.742
S1-S2-S3-S4 (˚)
0.0
33.0
43.7
52.3
54.1
Where, 1 = Mo(IV) bis-maleonitriledithiolate complex, 1B-ES = educt-substrate complex,
1B-TS = transition state complex, 1B-EP = product bound complex,
1B-P = oxidized product complex.
157
Chapter 5-Selenate Reductase
Fig. 5.13: Optimized geometries along the reaction pathway for oxygen atom transfer from HSeO4- to
[MoIVOH(mnt)2]- in the 1B coordination mode of substrate binding.
158
Chapter 5-Selenate Reductase
·
1C:
In this case, the interaction among the two reactants is complemented by the attachment of
selenate by its oxygen atom Oa to the Mo center as well as by the hydrogen bond between
hydroxo hydrogen atom of Mo (H1) and the oxygen atom (Oc) of selenate (Fig. 5.11).
Ø Optimized educt-substrate complex, 1C-ES:
The computed reaction energy for the formation of educt-substrate complex 1C-ES is
endothermic, 9.7 kcal/mol in the polarizable continuum (20.1 kcal/mol in the gas phase)
relative to the separate substrate and educt (1) complex (Table 5.11). The optimized data
shows a decrease in the Mo-O1 bond distance from 1.882 Ǻ to 1.848 Ǻ while the O1-H1 bond
distance is increased from 0.971 Ǻ to 1.016 Ǻ when comparison is made with the optimized 1
geometry. The H1-Oc, Mo-Oa and Oa-Se bond distances are 1.593 Ǻ, 2.468 Ǻ and 1.657 Ǻ,
respectively. The dithiolene S1-S2-S3-S4 dihedral angle (from 0.0˚ to 32.4˚) and the Mo-S
bond distances (from ~2.381 Ǻ to ~2.442 Ǻ) are increased (Table 5.9).
Ø Optimized transition state complex, 1C-TS:
The energy barrier for the oxygen atom transfer from substrate to the Mo (1C-TS) is 17.7
kcal/mol in the polarizable continuum (28.9 kcal/mol in the gas phase) relative to the separate
substrate and educt (1) complex (Table 5.11). This energy barrier associated with 1C-TS is
4.6 kcal/mol higher in energy than the 1A-TS while 2.4 kcal/mol lower in energy than the
1B-TS. An imaginary frequency of 398ί cm-1 corresponds to the stretching vibration modes of
Mo-Oa and Oa-Se (HSeO4-) and confirms the saddle point character between the selenate and
selenite complexes.
Geometry optimization of 1C-TS shows an increase in the Mo-O1 (from 1.848 Ǻ to 1.922 Ǻ),
H1-Oc (from 1.593 Ǻ to 1.758 Ǻ) and Oa-Se (from 1.657 Ǻ to 2.023 Ǻ) bond distances when
comparison is made with the optimized 1C-ES geometry. The O1-H1 and Mo-Oa bond
distances are decreased from 1.016 Ǻ to 0.995 Ǻ and from 2.468 Ǻ to 1.910 Ǻ, respectively.
An increase is observed in the dithiolene S1-S2-S3-S4 dihedral angle (from 32.4˚ to 45.3˚) and
in the Mo-S bond distances (from ~2.442 Ǻ to ~2.496 Ǻ) (Table 5.9).
Ø Optimized product bound complex, 1C-EP:
During the geometry optimization of 1C-EP, the hydrogen H1 from the hydroxo group of Mo
complex is shifted to the oxygen atom Oc of selenite complex resulting in the formation of
159
Chapter 5-Selenate Reductase
H2SeO3 and di-oxo MoVI complex which are also not the desired products. The computed
relative energy for the formation of 1C-EP complex is exothermic, -27.3 kcal/mol in the
polarizable continuum (-18.7 kcal/mol in the gas phase) (Table 5.11).
The optimized data shows a decrease in the Mo-O1 (from 1.922 Ǻ to 1.758 Ǻ), H1-Oc (from
1.758 Ǻ to 0.996 Ǻ) and Mo-Oa (from 1.910 Ǻ to 1.731 Ǻ) bond distances as compared to the
optimized 1C-TS geometry. The O1-H1 and the Oa-Se distances are increased from 0.995 Ǻ to
1.727 Ǻ and from 2.023 Ǻ to 3.341 Ǻ, respectively. An increase is observed in the dithiolene
S1-S2-S3-S4 dihedral angle (from 45.3˚ to 52.6˚) and in the Mo-S bond distances (from ~2.496
Ǻ to ~2.596 Ǻ) (Table 5.9).
Ø Optimized oxidized product complex, 1C-P:
The removal of H2SeO3 molecule leads to the formation of dioxo Mo complex
[MoVIO2(mnt)2]2- (1C-P). The computed reaction energy for 1C-P is exothermic (-17.0
kcal/mol ) in the polarizable continuum while it is endothermic (9.0 kcal/mol) in the gas phase
relative to the separate substrate and educt (1) complex (Table 5.11).
The optimized data shows a decrease in the Mo-O1 bond distance from 1.758 Ǻ to 1.732 Ǻ
while no change is observed in the Mo-Oa bond distance when comparison is made to the
optimized 1C-EP geometry. No considerable change is observed in the dithiolene S1-S2-S3-S4
dihedral angle (from 52.6˚ to 54.1˚) while the Mo-S bond distances are increased from ~2.596
Ǻ to ~2.622 Ǻ (Table 5.9).
160
Chapter 5-Selenate Reductase
Table 5.9: Selected bond lengths [Å] of optimized geometries along the reaction pathway for oxygen
atom transfer from HSeO4- to [MoIVOH(mnt)2]- in the 1C coordination mode of substrate binding.
C
1
1C-ES
1C-TS
1C-EP
1C-P
Mo-O1 (Ǻ)
1.882
1.848
1.922
1.758
1.732
O1-H1 (Ǻ)
0.971
1.016
0.995
1.727
-
H1- Oc (Ǻ)
-
1.593
1.758
0.996
-
Mo-Oa (Ǻ)
-
2.468
1.910
1.731
1.732
Oa-Se (Ǻ)
-
1.657
2.023
3.341
-
Mo-S1 (Ǻ)
2.385
2.412
2.427
2.480
2.502
Mo-S2 (Ǻ)
2.386
2.362
2.494
2.711
2.742
Mo-S3 (Ǻ)
2.377
2.490
2.533
2.507
2.502
Mo-S4 (Ǻ)
2.377
2.503
2.529
2.684
2.742
S1-S2-S3-S4 (˚)
0.0
32.4
45.3
52.6
54.1
Where, 1 = Mo(IV) bis-maleonitriledithiolate complex, 1C-ES = educt-substrate complex,
1C-TS = transition state complex, 1C-EP = product bound complex,
1C-P = oxidized product complex.
161
Chapter 5-Selenate Reductase
Fig. 5.14: Optimized geometries along the reaction pathway for oxygen atom transfer from HSeO4- to
[MoIVOH(mnt)2]- in the 1C coordination mode of substrate binding.
162
Chapter 5-Selenate Reductase
·
1D:
In this case, the interaction between the two molecules is complemented by the attachment of
selenate by its hydroxo oxygen atom Ob to the Mo center as well as by a bond between the
hydroxo hydrogen atom of Mo (H1) and the oxygen atom (Oc) of selenate (Fig. 5.11).
Ø Optimized educt-substrate complex, 1D-ES:
The geometry optimization of 1D-ES leads to the transfer of hydroxo hydrogen H1 from the
Mo to the selenate resulting in the formation of H2SeO4 and reduced mono-oxo complex
[MoIVO(mnt)2]2-. The computed reaction energy for the formation of 1D-ES is exothermic
(-3.2 kcal/mol) in the polarizable continuum while it is endothermic (4.5 kcal/mol) in the gas
phase relative to the separate substrate and educt (1) complex (Table 5.11). The optimized
data shows a decrease in the Mo-O1 bond distance from 1.882 Ǻ to 1.729 Ǻ while the O1-H1
bond distance is increased from 0.971 Ǻ to 1.561 Ǻ when comparison is made with the
optimized 1 geometry. The H1-Oc distance is 1.018Ǻ. No considerable change is observed in
the dithiolene S1-S2-S3-S4 dihedral angle (from 0.0˚ to -3.3˚) while the Mo-S bond distances
are increased from ~2.381 Ǻ to ~2.440 Ǻ (Table 5.10).
Table 5.10: Selected bond lengths [Å] of optimized geometries along the reaction pathway for oxygen
atom transfer from HSeO4- to [MoIVOH(mnt)2]- in the 1D coordination mode of substrate binding.
D
1
1D-ES
Mo-O1 (Ǻ)
1.882
1.729
O1-H1 (Ǻ)
0.971
1.561
H1- Oc (Ǻ)
-
1.018
Mo-Ob (Ǻ)
-
-
Ob-Se (Ǻ)
-
-
Mo-S1 (Ǻ)
2.385
2.430
Mo-S2 (Ǻ)
2.386
2.435
Mo-S3 (Ǻ)
2.377
2.440
Mo-S4 (Ǻ)
2.377
2.454
S1-S2-S3-S4 (˚)
0.0
-3.3
Where, 1 = Mo(IV) bis-maleonitriledithiolate complex, 1D-ES = educt-substrate complex.
163
Chapter 5-Selenate Reductase
Fig. 5.15: Optimized geometries along the reaction pathway for oxygen atom transfer from HSeO4- to
[MoIVOH(mnt)2]- in the 1D coordination mode of substrate binding.
Table 5.11: Relative energies (kcal/mol) computed for stationary points along the OAT from HSeO4to [MoIVOH(mnt)2]- (1) in the four different coordination modes of substrate binding.
1A
1B
1C
1D
//B3LYPa
E
0.0
0.0
0.0
SDDb
0.0
COSMOc
ES
TS
EP
P
19.6
25.3
18.5
4.7
//B3LYPa
20.1
26.2
20.1
4.5
SDDb
7.2
15.3
9.7
-3.2
COSMOc
35.9
38.9
35.2
27.5
39.1
28.9
13.1
20.1
17.7
COSMOc
-6.4
3.3
-6.1
//B3LYPa
-16.7
-11.5
-18.7
-26.0
-19.5
-27.3
COSMOc
56.3
38.2
56.3
//B3LYPa
9.0
26.4
9.0
-17.0
-3.5
-17.0
//B3LYPa
SDDb
-
SDDb
-
SDDb
-
COSMOc
Where, E = Mo(IV) bis-maleonitriledithiolate complex (1), ES = educt-substrate complex,
TS =transition state complex, EP = product bound complex, P = oxidized product complex.
1A, 1B, 1C, 1D are the different modes of substrate coordination to the educt complex (1).
a) B3LYP/Lanl2DZ(p), b) B3LYP/SDDp//B3LYP/Lanl2DZ (p),
c) COSMO-B3LYP/SDDp//B3LYP/Lanl2DZ(p) (see Computational details).
164
Chapter 5-Selenate Reductase
Discussion:
Density functional theory (DFT) studies have been performed to investigate various ways of
binding the substrate with the active site including but not limited to the reported way (Fig.
5.9). For the oxygen atom transfer, (OAT) in which HSeO4- reacts with the bis-dithiolene MoOH complex [MoIV(OH)(mnt)2]- (1), different coordination modes between the two molecules
have been considered (Fig. 5.11):
In structures 1A-1C HSeO4- coordinates to Mo through an oxygen atom (Oa), while in 1D it
coordinates through the hydroxo oxygen (Ob) atom of the substrate.
In case of 1A, the interaction is complemented by a bond between the hydrogen atom (HSe) of
selenate and the hydroxo oxygen atom (O1) attached to the Mo centre. The computed reaction
energy for this complex 1A-ES is 7.2 kcal/mol in the continuum (20.1 kcal/mol in the gas
phase) relative to the separate substrate, HSeO4- and educt (1) complex. Transition state
1A-TS, has a distorted octahedral geometry and energy barrier of 13.1 kcal/mol in the
continuum (27.5 kcal/mol in the gas phase). It is characterized by one imaginary frequency of
157ί cm-1 and a transition vector that shows mainly stretching mode of Mo-Oa. Its computed
barrier is the lowest we could find. Geometry optimization of [MoVIO(OH)(mnt)2HSeO3]2leads to the formation of complex 1A-EP containing the dioxo Mo complex [MoVIO2(mnt)2]2and selenous acid (H2SeO3), where both hydrogen atoms of selenous acid form hydrogen
bonds with one oxo group (Oa) of the Mo complex. The complex 1A-EP was not the expected
result from OAT reaction from selenate to the reduced MoIV complex. In the gas phase this
complex, however, is probably more stable than the expected hydroxo-oxo-complex with
HSeO3-. The relative energy for the formation of this product bound complex (1A-EP) is 26.0 kcal/mol in the polarizable continuum (-16.7 kcal/mol in the gas phase). The loss of
H2SeO3 from the
1A-EP gives the dioxo Mo complex [MoVIO2(mnt)2]2- (1A-P) in an
exothermic step (-17.0 kcal/mol) in the polarizable continuum while in an endothermic step
(9.0 kcal/mol) without the continuum relative to the separate substrate, HSeO4- and educt (1)
complex .
In case of 1B, there is an additional hydrogen bond between the hydroxo hydrogen (H1) of the
Mo complex and the hydroxo oxygen atom (Ob) of the HSeO4-. The computed relative energy
for such a MoIV starting complex 1B-ES is 15.3 kcal/mol in the continuum (26.2 kcal/mol in
the gas phase). An OAT transition state 1B-TS with a distorted octahedral geometry and an
energy barrier of 20.1 kcal/mol in the polarizable continuum (39.2 kcal/mol in the gas phase)
165
Chapter 5-Selenate Reductase
could be found. It is characterized by an imaginary frequency of 376ί cm-1 and a transition
vector that shows the stretching mode of Mo-Oa. However, geometry optimization of
[MoVIO(OH)(mnt)2HSeO3]2- did not lead to the expected HSeO3 - complex, but the complex
1B-EP, in which SeO2 and H2O are loosely bound to [MoVIO2(mnt)2]2-. 1B-EP has a
computed relative energy of -19.5 kcal/mol in the polarizable continuum (-11.5 kcal/mol in
the gas phase). This structure may be rationalized as a result of a proton being transferred
from the hydroxo group at Mo to the hydroxo group of the reduced substrate rearranging to
water, selenium dioxide and a dioxo Mo complex. Loss of water and selenium dioxide from
1B-EP gives the Mo complex, [MoVIO2(mnt)2]2- (1B-P), which has a relative energy of -3.5
kcal/mol in the polarizable continuum (26.4 kcal/mol in the gas phase).
In case of 1C, the hydrogen bond is formed between the Mo bound hydroxo hydrogen (H1)
and an oxo group (Oc) of selenate. The relative energy for such a reduced complex 1C-ES is
9.7 kcal/mol in the polarizable continuum (20.1 kcal/mol in the gas phase). The distorted
octahedral transition state 1C-TS has an energy barrier of 17.7 kcal/mol (28.9 kcal/mol in the
gas phase). 1C-TS is characterized by an imaginary frequency of 398ί cm-1 and a transition
vector that shows the stretching mode of Mo-Oa. The H2SeO3 bound complex
[MoVIO2(mnt)2(H2SeO3)]2- (1C-EP) formed as a result of geometry optimization, has a
relative energy of -27.3 kcal/mol in the polarizable continuum (-18.7 kcal/mol in the gas
phase). 1C-EP is not the desired complex as hydrogen (H1) from the hydoxo group of Mo
complex is shifted to the selenite complex forming selenous acid and dioxo Mo complex.
Loss of selenous acid gives the Mo complex, [MoVIO2(mnt)2]2- (1C-P), which has the relative
energy of -17.0 kcal/mol in the polarizable continuum (9.0 kcal/mol in the gas phase).
In case of 1D, we tried to explore the possibility of the hydoxo (ObHSe) transfer from the
selenate to the Mo complex 1. However, during the optimization of a starting geometry
(1D-ES) the hydrogen atom (H1) is shifted from the Mo complex to the selenate forming
selenic acid (H2SeO4) and the mono oxo MoIV complex. It has a relative energy of -3.2
kcal/mol in the polarizable continuum (4.5 kcal/mol in the gas phase).
166
Chapter 5-Selenate Reductase
Scheme 5.3: Plot of computed reaction energies (kcal/mol) relative to separate substrate and educt
complex vs steps involved in the OAT from HSeO4- to educt complex 1 in the four different
coordination modes of substrate binding.
The best way for selenate reduction by the chosen model complex corresponds to 1A. While
an oxygen atom is transferred from the selenate to Mo the substrate OH group forms a
hydrogen bond with the hydroxyl oxygen atom coordinated to Mo. Barriers computed for
alternative arrangements are higher by only a few kcal/mol. The difficulties in locationg the
expected product complexes where the HSeO3- product is attached to a Mo complex
demonstrate the necessity to investigate this reaction by a larger model system representing
the enzyme active site more realistically, i.e. including part of the neighbouring amino acids
of the protein part.
167
Conclusion
Conclusion
Different mononuclear Mo/W containing active site model complexes derived from the
protein X-ray crystal structure of their native enzymes were computed to elucidate
mechanistic details.
Nitrate reductase is an enzyme that catalyzes the reduction of nitrate to nitrite. The first Wcontaining nitrate reductase (Nar) isolated from Pyrobaculum aerophilum was computed for
nitrate reduction where both Mo and W model complexes were considered. The energy
barrier for OAT transfer in W-Nar is lower than the Mo-Nar; however, Mo-Nar is the best
choice for the reduction of nitrate (oxidation of educt complex is less exothermic than
W-Nar). The computed results indicate that although the reduction of nitrate is stimulated
when W replaces Mo in the active site of Nar the catalytic cycle breaks after the reduction of
nitrate to nitrite when the biochemical reducer is not strong enough to reduce the metal center.
Ethylbenzene dehydrogenase (EBDH) is an enzyme that catalyzes the oxygen-independent,
stereospecific hydroxylation of ethylbenzene to (S)-1-phenylethanol. EBDH active site
models were computed to find the most probable mechanism, ionic or radical pathway.
Models with protonation of His192, Lys450, Asp223 and a model without protonation were
investigated for comparison. Computed relative energies indicate that the overall lowest
energy barrier pathway results when ionic and radical pathways are mixed and the Lys450
protonated EBDH model shows the energetically best pathway for the hydroxylation of
ethylbenzene.
Acetylene hydratase (AH) of Pelobacter acetylenicus is a W containing iron-sulfur enzyme
that catalyzes the transformation of acetylene to acetaldehyde. Based on the computational
results for AH active site models the most likely nucleophilic mechanism for the hydration of
acetylene by the acetylene hydratase (AH) enzyme is the one where the water (Wat1424)
molecule is coordinated to the W center and Asp13 is assumed to be in anionic form. For the
AH, first small model complexes were computed but we have more reliable results when the
surrounding amino acid residues were included.
Sulfite oxidase (SO), selenate reductase (SeR) and nitrate reductases (NRs) are among the
mononuclear molybdenum enzymes involved in the catalysis of metabolic redox reactions.
SeR is an enzyme that catalyzes the reduction of selenate to selenite and has two
molybdopterin (MPT) ligands, SO oxidizes the sulfite to sulfate and has one MPT ligand,
while NRs reduces nitrate to nitrite by either one or with two MPT ligands at the active site.
168
Conclusion
The sulfite model A (the computational model complex, [MoVIO2(S2C2Me2)SMe]-), which
resembles the SO active site is clearly the best choice (lowest energy barrier, minor
exothermicity) for sulfite oxidation. For the reduction of selenate, however, a smaller
activation is computed for model A, but the reaction is less exothermic with model B
([MoVIO2(mnt)2]2-, resembling SeR), which resembles the SeR active site.
The simple active site model complexes of SeR were computed to investigate different ways
of binding the substrate and the OAT reaction. Unfortunately, the results are little conclusive.
Larger models might be needed to obtain more meaningful computational results as the active
site environment may have the influence on the reactions catalyzed by the enzyme.
169
References
References
1
The Catalytic Site Atlas at The European Bioinformatics Institute. Retrieved 4 April 2007
2
Holm, R. H., Kennepohl, P., and Solomon, E. I., Chem. Rev., 1996, 96, 2239-2314.
3
Kletzin, A., and Adams, M. W., FEMS Microbiol Rev., 1996, 18, 5-63.
4
Greenwood, N. N., and Earnshaw, A., The Chemistry of the Elements, Pergamon Press,
Oxford, UK, 1984, p.1170.
5
Lawson, D. M., Williams, C. E., White, D. J., Choay, A. P., Mitchenhall, L. A., and Pau, R.
N., J. Chem. Soc., Dalton Trans., 1997, 3981-3984.
6
Lawson, D. M., Williams, C. E., Mitchenhall, L. A., and Pau, R. N., Structure., 1998, 6,
1529-1539.
7
Sigel, A., Sigel, H., Metal Ions in Biological Systems 39; Marcel Dekker: New York, 2002.
8
Hille, R., Trends Biochem. Sci., 2002, 27, 360.
9
Hille, R., Chem. Rev., 1996, 96, 2575-2816.
10
Holm, R. H. Chem. Rev., 1987, 87, 1401.
11
Young, C. G., and Wedd, A. G., in Molybdenum. Molybdopterin Containing Enzymes,
Encyclopedia of Inorganic Chemistry, Vol. 4 (R. B. King, ed.), John Wiley & Sons, New
York, 1994, pp. 2330-2346.
12
Garner, C. D., Banham, R., Cooper, S. J., Davies, E. S., and Stewart, L. J., in Enzymes and
Proteins Containing Molybdenum or Tungsten (I. Bertini, A. Sigel, and H. Sigel, eds.),
Handbook on Metalloproteins, Marcel Dekker, New York, 2001, pp. 1023-1090.
13
Howard, J. B., Rees, D. C., Chem. Rev., 1996, 96, 2965.
14
Burgess, B. K., Lowe, D. J., Chem. Rev., 1996, 96, 2983.
15
Campbell, W. H., Annu. Rev. Plant Physiol., 1999, 50, 277-303.
16
Sekimoto, H., Seo, M., Dohmae, N., Takio, K., Kamiya, Y., and Koshiba, T., J. Biol.
Chem., 1997, 272, 15280-15285.
17
Gladyshev, V. N., Khangulov, S. V., and Stadtman, T. C., Biochemistry., 1996, 35, 212-
223.
18
Lehmann, M., Tshisuaka, B., Fetzner, S., and Lingens, F., J. Biol. Chem., 1995, 270,
14420-14429.
170
References
19
Bläse, M., Bruntner, C., Tshisuaka, B., Fetzner, S., and Lingens, F., J. Biol. Chem., 1996,
271, 23068-23079.
20
Canne, C., Stephan, I., Finsterbusch, J., Lingens, F., Kappl, R., Fetzner, S., and
Hüttermann, J., Biochemistry., 1997, 36, 9780-9790.
21
Weiner, J. H., Rothery, R. A., Sambasivarao, D., and Trieber, C. A., Biochim. Biophys.
Acta., 1992, 1102, 1-18.
22
Massey, V., Komai, H., Palmer, G., and Elion, G. B., J. Bio. Chem., 1970, 245, 2837-2844.
23
Huang, D.Y., Furukawa, A., and Ichikawa, Y., Arc. Biochem. Biophys., 1999, 364, 264-
272.
24
Rajagopalan, K. V., in Molybdenum and Molybdenum Containing Enzymes (M. P.
Coughlan, ed.), Pergamon Press, New York, 1980, pp. 241-272.
25
Ljungdahl, L. G., Trends Biochem. Sci., 1976, 1, 63-65.
26
Ljungdahl, L. G., and Andreesen, J. R., Meth. Enzymol., 1978, 53, 360-372.
27
Andreesen, J. R., and Ljungdahl, L. G., J. Bacteriol., 1973, 116, 867-873.
28
Andreesen, J. R., and El Ghazzawi, E., Arch. Microbiol., 1974, 96, 103-110.
29
Yamamoto, I., Saiki, T., Liu, S.-M., and Ljungdahl, L. G., J. Biol. Chem., 1983,258, 1826-
1832.
30
Johnson, M. K., Rees, D. C., and Adams, M. W. W., Chem. Rev., 1996, 96, 2817-2840.
31
Kisker, C., Schindelin, H., Pacheco, A., Wehbi, W. A., Garrett, R. M., Rajagopalan, K. V.,
Enemark, J. H., and Rees, D. C., Cell., 1997, 91, 973-983.
32
Kisker, C., Schindelin, H., and Rees, D. C., Annu. Rev. Biochem., 1997, 66, 233-267.
33
Fischer, K., Barbier, G. G., Hecht, H. J., Mendel, R. R., Campbell, W. H., Schwarz, G.,
Plant. Cell., 2005, 17, 1167-1179.
34
Kisker, C., Schindelin, H., Baas, D., Retey, J., Meckenstock, R. U., and Kroneck, P. M.,
FEMS Microbiol Rev., 1998, 22, 503-521.
35
Stiefel, E. I., J Chem Soc Dalton Trans., 1997, 3915- 3923.
36
Ellis, P. J., Conrads, T., Hille, R., and Kuhn, P., Structure., 2001, 9, 125-132.
37
Hille R., European Journal of Inorganic Chemistry., 2006, 10, 1905–2095.
38
Hille, R., and Sprecher, H., J. Biol. Chem., 1987, 262, 10914–10917.
39
Wootton, J. C., Nicolson, R. E., Cock, J. M., Walters, D. E., Burke, J.F., Doyle, W.A., et al.
171
References
Biochim Biophys Acta., 1991, 1057, 157–185.
40
Beedham, C., Prog. Med. Chem., 1987, 24, 85–121.
41
Turner, N. A., Doyle, W. A., Ventom, A. M. and Bray, R. C., Eur. J. Biochem., 1995, 232,
646–657.
42
Hille, R., Arch. Biochem. Biophys., 2005, 433, 107–116.
43
Krenitsky, T. A., Neil, S. M., Elion, G. B. and Hitchings, G. H., Arch. Biochem. Biophys.,
1972, 150, 585–599.
44
Bray, R. C. In The Enzymes, Boyer, P. D., Ed., 1975, 12, 299.
45
Romao, M. J., Archer, M., Moura, I., Moura, J. J. G., LeGall, J., Engh, R., Schneider, M.,
Hof, P., and Huber, R., Science., 1995, 270, 1170-1176.
46
Huber, R., Hof, P., Duarte, R. O., Moura, J. J. G., Moura, I., LeGall, J., Hille, R., Archer,
M., and Romao, M. J., Proc. Natl. Acad. Sci. USA., 1996, 93, 8846-8851.
47
Harrison, R., Free Radic. Biol. Med., 2002, 33, 774–797.
48
Murray, K. N., Watson, J. G., Chaykin, S., J. Biol. Chem., 1966, 241, 4798.
49
Komai, H., Massey, V., Palmer, G., J. Biol. Chem., 1969, 244, 1692.
50
Olson, J. S., Ballou, D. P., Palmer, G., Massey, V., J. Biol. Chem., 1974, 249, 4350-4362.
51
Hille, R., Massey, V., J. Biol. Chem., 1981, 256, 9090.
52
Porras, A. G., Olson, J. S., Palmer, G., J. Biol. Chem., 1981, 256, 9096.
53
Rajagopalan, K.V., Bio Chem Elem., 1984, 3, 149.
54
Kessler, D. L., Rajagopalan, K. V., J. Biol. Chem., 1972, 247, 6566.
55
MacLeod, R. M., Farkas, W., Fridovich, I., Handler, P., J. Biol. Chem., 1961, 236, 1841.
56
Campbell, W.H., Structure and synthesis of higher plant nitrate reductase, 1989, See ref.
102a, pp. 123-154.
57
Campbell, W.H., Plant Physiol., 1992, 99, 693-699.
58
Campbell, W.H., Annu. Rev. Plant Physiol., 1999, 50, 277-303.
59
Barrett, E. L., and Kwan, H. S., Annu. Rev. Microbiol., 1985, 39, 131-149.
60
Sigel, A., Sigel, H., Metal Ions in Biological Systems 39; Marcel Dekker: New York, 2002,
p. 202-204.
61
McAlpine, A. S., McEwan, A. G., and Bailey, S., J. Mol. Biol., 1998, 275, 613-623.
62
Santos, J. P. D., Lobbi-Nivol, C., Giordano, C. C. G., and MeÂjean, V., J. Mol. Biol., 1998,
284, 421-433.
63
Richardson, D. J., Berks, B. C., Russell, D. A., Spiro, S., Taylor, C. J., Cell. Mol. Life
Sci., 2001, 58, 165-178.
172
References
64
Campbell, W. H., Cell. Mol. Life Sci., 2001, 58, 194-204.
65
Richardson, D. J., and Watmough, N. J., Curr. Opin. Chem. Biol., 1999, 3, 207-219.
66
Hofmann. M., J. Biol. Inorg. Chem., 2009, 14, 1023-1035.
67
Blasco, F., Iobbi, C., Giordano, G., Chippaux, M., and Bonnefoy, V., Mol. Gen. Genet.,
1989, 249-256.
68
Lin, J. T., Stewart, V., Adv. Microb. Physiol., 1998, 39, 1-30.
69
Majumdar, A., Sarkar, S., Coord. Chem. Rev., 2011, 255, 1039-1054.
70
Boyington J. C., Gladyshev, V. N., Khangulov, S. V., Stadman, T. C., Sun, P. D., Science.,
1997, 275, 1305-1308.
71
Jormakka, M., Tornroth, S., Byrne, B., Iwata, S., Science., 2002, 295, 1863-1868.
72
Costa, C., Teixeira, M., LeGall, J., Moura, J.J.G., Moura, I., J. Biol. Inorg. Chem., 1997, 2,
198–208.
73
George, G.N., Costa, C., Moura, J.J.G., Moura, I., J. Am. Chem. Soc., 1999, 121, 2625–
2626.
74
Brondino, C. D., Passeggi, M.C.G., Caldeira, J., Almendra, M.J., Feio, M.J., Moura, J.J.G.,
Moura, I., J. Biol. Inorg. Chem., 2004, 9, 145-151.
75
Messerschmidt, A., Niessen, H., Abt, D., Einsle, O., Schink, B., and Kroneck, P. M. H.,
Proc. Natl. Acad. Sci. U. S. A., 2004, 101 (2), 11571-11576.
76
Ellis, P. J., Conrads, T., Hille, R., and Kuhn, P., Structure, 2001, 9, 125-132.
77
Almendra, M. J., Brondino, C. D., Gavel, O., Pereira, A. S., Tavares, P., Bursakov, S.,
Duarte, R., Caldeira, J., Moura, J. J. G., Moura, I., Biochemistry., 1999, 38, 16366–16372.
78
de Bok, F. A. M., Hagedoorn, P. L., Silva, P. J., Hagen, W. R., Schiltz, E., Fritsche, K.,
Stams, A. J. M., Eur. J..Biochem., 2003, 270, 2476–2485.
79
Bevers, L. E., Hagedoorn, P. –L., Hagen, W. R., Coord. Chem. Rev., 2009, 253, 269-290.
80
Sigel, A., Sigel, H., Metal Ions in Biological Systems 39; Marcel Dekker: New York, 2002,
p. 677-680.
81
Mukund, S., and Adams, M. W. W., J. Biol. Chem., 1991, 266, 14208-14216.
82
Ma, K., Hutchins, A., Sung, S. –H. S., and Adams, M. W. W., PNAS, 1997, 94, 9608-9613.
83
Heider, J., Ma, K., and Adams, M. W. W., J. Bacteriol., 1995, 177, 4757-4764.
84
Sigel, A., Sigel, H., Metal Ions in Biological Systems 39; Marcel Dekker: New York, 2002,
p. 680-684.
85
Chan, M. K., Mukund, S., Kletzin, A., Adams, M. W. W., Rees, D. C., Science., 1995, 267
(5203), 1463-1469.
173
References
86
Mukund, S., and Adams, M. W. W., J. Biol. Chem., 1993, 268, 13592-13600.
87
Roy. R., Mukund, S., Schut, G. J., Dunn, D. M., Weiss, R., and Adams, M. W. W., J.
Bacteriol., 1999, 181, 1171-1180.
88
Sigel, A., Sigel, H., Metal Ions in Biological Systems 39; Marcel Dekker: New York, 2002,
p. 687-688.
89
Mukund, S., and Adams, M. W. W., J. Biol. Chem., 1995, 270, 8389-8392.
90
Sigel, A., Sigel, H., Metal Ions in Biological Systems 39; Marcel Dekker: New York, 2002,
p. 690-691.
91
92
White, H., Strobl, G., Feicht, R., and Simon, H., Eur. J. Biochem., 1989, 184, 89-96.
Sigel, A., Sigel, H., Metal Ions in Biological Systems 39; Marcel Dekker: New York, 2002,
p. 685-686.
93
Hensgens, C. M. H., Hagen, W. R., and Hansen, T. H., J. Bacteriol., 1995, 177, 6195-6200.
94
Zellner, G., Jargon, A., Arch. Microbiol., 1997, 168, 480-485.
95
Yamamoto, I., Saiki, T., Liu, S. –M., and Ljungdahl, L. G., J. Biol. Chem., 1983, 258,
1826-1832.
96
Almendra, M. J., Brondino, C. D., Gavel, O., Pereira, S., Tavares, P., Bursakov, S., Duarte,
R., Caldeira, J., Moura, J. J. G., and Moura, I., Biochemistry., 1999, 38, 16366-16372.
97
Schmitz, R. A., Linder, D., Stetter, K. O., and Thauer, R. K., Arch. Microbiol., 1991, 156,
427-434.
98
Rosner, B., and Schink, B., J. Bacteriol., 1995, 177, 5767-5772.
99
Meckenstock, R. U., Kreiger, R., Ensign, S., Kroneck, P. M. H., and Schink, B., Eur. J.
Biochem., 1999, 264, 176-182.
100
Seiffert, G. B., Ullmann, G. M., Messerschmidt, A., Schink, B., Kroneck, P. M. H., and
Einsle, O., Proc. Natl. Acad. Sci. USA., 2007, 104, 3073-3077.
101
Johnson, J. L., Rajagopalan, K. V., and Cohan. H. J., J. Biol. Chem., 1974, 249, 5046-
5055.
102
Buc, J., Santini, C. –L., Giordano, R., Czjek, M., Wu, L. –F., and Giordano, G., Mol.
Microbiol., 1999, 32, 159-168.
103
Stewat, L. J., Bailey, S., Bennett, B., Charnock, J. M., Garner, C. D., and McAlpine, A. S.,
J. Mol. Biol., 2000, 299, 593-600.
104
Sung, K. –M., and Holm, R. H., J. Am. Chem. Soc., 2001, 123, 1931-1943.
105
Jormakka, M., Richardson, D., Byrne, B., and Iwata, S., Structure., 2004, 12, 95-104.
106
Afshar, S., Johnson, E., de Vries, D., and Schröder, I., J. Bacteriol., 2001, 183, 5491-5495.
174
References
107
Kisker, C., Schindelin, H., Baas, D., Rétey, J., Meckenstock, R. U., Kriger, R., and
Kroneck, P. M. H., FEMS Microbiol. Rev., 1999, 178, 163-167.
108
Stolz, J. F., and Basu, P., Chem. Bio. Chem., 2002, 3, 198-206.
109
Kroneck, P. M. H., Abt, D. J., Met. Ions. Biol. Syst., 2002, 39, 369-403.
110
Cabello, P., Roldan, M. D., Moreno-vivian, C., Microbiology., 2004, 150, 3527-3546.
111
Richardson, D. J., and Watmough, N. J., Curr. Opin. Chem. Biol., 1999, 3, 207-219.
112
Campbell, W. H., Annu. Rev. Plant Physiol. Plant Mol.Biol., 1999, 50, 277-303.
113
Lin, J. T., Stewart, V., Adv. Microb. Physiol., 1998, 39, 1-30.
114
De Vries, S., Momcilovic, M., Strampraad, M. J. F., Whitelegge, J. P., Baghai, A., and
Schröder, I., Biochemistry., 2010, 49, 9911-9921.
115
Zumft, W. G., Microbiol. Mol. Biol. Rev., 1997, 61, 533-616.
116
Enoch, H. G., and Lester, R. L., J. Bacteriol., 1972, 110, 1032-1040.
117
Brondino, C. D., Passeggi, M. C., Caldeira, J., Almendra, M. J., Feio, M. J., Moura, J. J.,
and Moura, I., J. Biol. Inorg. Chem., 2004, 9, 145–151.
118
Stewart, L. J., Bailey, S., Bennett, B., Charnock, J. M., Garner, C. D., and McAlpine, A.
S., J. Mol. Biol., 2000, 299, 593–600.
119
Afshar, S., Kim, C., Monbouquette, H. G., and Schröder, I., Applied And Environmental
Microbiology., 1998, 64 (8), 3004-3008.
120
Chaudhry, G. R., and MacGregor, C. H., J. Biol. Chem., 1983, 258, 5819–5827.
121
Holm, R. H., Chem. Rev., 1987, 87, 1401-1449.
122
Jiang, J., and Holm, R. H., Inorg. Chem., 2005, 44, 1068-1072.
123
Lee, C., Yang, W., Parr, R. G., Phys Rev B., 1988, 37, 785-789.
124
Hay, P. J., Wadt, W. R., J. Chem. Phys., 1985, 82, 270-283.
125
Wadt, W. R., Hay, P. J., J. Chem. Phys., 1985, 82, 284-293.
126
Hay, P. J., Wadt, W. R., J. Chem. Phys., 1985, 82, 299-310.
127
Dunning, T. H., Hay, P. J., In Schaefer HF III (ed) Modern theoretical chemistry, 1976,
vol 3. Plenum, New York.
128
Huzinaga, S., Andzelm, J., Klobukowski, M., Radzio-Andzelm, E., Sakai, Y., Tatewaki,
H., Gaussian basis sets for molecular orbital calculations, Elesevier, 1984, Amsterdam
129
Andrae, D., Haeussermann, U., Dolg, M., Preuss, H., Theor Chim Acta., 1990, 77, 123-
141.
130
Andrae, D., Haeussermann, U., Dolg, M., Preuss, H., Theor Chim Acta., 1991, 78, 247-
266.
175
References
131
Klamt, A., Jonas, V., Buerger, T., Lohrenz, J.C.W., J. Phys. Chem., 1998, 102, 5074-5085.
132
Barone, V., Cossi, M., J. Phys Chem A., 1998, 102, 1995.
133
Cossi, M., Rega, N., Scalmani, G., Barone, V., J. Comp. Chem., 2003, 24, 669-681.
134
Florsi, F., and Tomasi, J., J. Comp. Chem., 1989, 10, 616.
135
Florsi, F., and Tomasi, J., and Pascual-Ahuir, J. L., J. Comp. Chem., 1991, 12, 784.
136
Cammi, R., Cappelli, C., Corni, S., and Tomasi, J., J. Phys. Chem. A., 2000, 104, 9874-
9879.
137
Pascual-Ahuir, J. L., Silla, E., and Tuñón, I., J. Comp. Chem., 1994, 15, 1127-1138.
138
George, G. N., Turner, N. A., Bray, R. C., Morpeth, F. F., Boxer, D. H., and Cramer, S. P.,
Biochem. J., 1989, 259, 693-700.
139
McAlpine, A. S., McEwan, A. G., and Bailey, S., J. Mol. Biol., 1998, 275, 613-623.
140
Cotton, F. A., Wing, R. M., Inorg. Chem., 1965, 4, 867-873.
141
Yoshimatsu, K., Iwasaki, T., and Fujiwara, T., FEBS Lett., 2002, 516, 145-150.
142
Yoshimatsu, K., Sakurai, T., and Fujiwara, T., FEBS Lett., 2000, 470, 216-220.
143
Martinez-Espinosa, R. M., Richardson, D. J., Butt, J. N., and Bonete, M. J., Biochem. Soc.
Trans., 2006, 34, 115-117.
144
Klenitz, A., and Adams, M. W. W., FEMS Microbiol. Rev., 1996, 18, 5-63.
145
Sigel, A., Sigel, H., Metal Ions in Biological Systems 39; Marcel Dekker: New York, 2002.
146
De Vries, S., Momcilovic, M., Strampraad, M. J. F., Whitelegge, J. P., Baghai, A., and
Schröder, I., Biochemistry., 2010, 49, 9911-9921.
147
Stewart, L. J., Bailey, S., Bennett, B., Charnock, J. M., Garner, C. D., and McAlpine, A.
S., J. Mol. Biol., 2000, 299, 593–600.
148
Szaleniec, M., Hagel, C., Menke, M., Nowak, P., Witko, M., and Heider, J., Biochemistry.,
2007, 46, 7637-7646.
149
Schröder, I., Rech, S., Macy, J. M., J. Biol. Chem., 1997, 272, 23765-23768.
150
McDevitt, C. A., Hugenholtz, P., Hanson, G. R., and MsEwan, A. G., Mol. Microbiol.,
2002, 44, 1575-1587.
151
Liedo, B., Martinez-Espinosa, R. M., Marhuenda-Egea, F. C., and Bonete, M. J., Biochim
Biophys. Acta., 2004, 1674, 50-59.
152
Yoshimatsu, K., Iwasaki, T., and Fujiwara, T., FEBS Lett., 2002, 516, 145-150.
153
Kniemeyer, O., and Heider, J., J. Biol. Chem., 2001, 276, 21381-21386.
154
Johnson, H. A., Pelletier, D. A., and Spormann, A. M., J. Bacteriol., 2001, 183, 4536-
4542.
176
References
155
Kloer, D. P., Hagel, C., Heider, J., and Schulz, G. E., Structure., 2006, 14, 1377-1388.
156
Ball, A. A., Johnson, H. A., Reinhard, M., and Spormann, A. M., J. Bacteriol., 1996, 178,
5755-5761.
157
Rabus, R., and Widdel, F., Arch. Microbiol., 1995, 163, 96-103.
158
Fischer, B., Enemark, J. H., and Basu, P., J. Inorg. Bio. Chem., 1998, 72, 13-21.
159
Szaleniec, M., Witko, M., and Heider, J., J. Mol. Catal. A: Chem., 2008, 286, 128-136.
160
Schlegel, H. B., and McDouall, J. J., in Computational Advances in Organic Chemistry,
Ed. Ögretir, C., and Csizmadia, I. G., (Kluwer Academic, The Netherlands, 1991) 167-185.
161
Bacskay, G. B., Chem. Phys., 1981, 61, 385-404.
162
Szaleniec, M., Borowski, T., Schühle, K., Witko, M., and Heider, J., J. Am. Chem. Soc.,
2010, 132, 6014-6024.
163
Szaleniec, M., Salwiński, A., Borowski, T., Heider, J., and Witko, M., J. Quantum. Chem.,
2011, 000, 000.
164
Schink, B., Arch. Microbiol., 1985, 142, 295-301.
165
Burgess, B. K., and Lowe, D. L. Chem. Rev., 1996, 96, 2983-3012.
166
Yadav, J., Das, S. K., and Sakar, S., J. Am. Chem. Soc., 1997, 119, 4315-4316.
167
Einsle, O., Niessen, H., Abt, D. J., Seiffert, G., Schink, B., Huber, R., Messeschmidt, A.,
and Kroneck, P. M. H., Acta Crystallogr. Sect. F: Struct. Biol. Cryst. Commun., 2005, 61,
299-301.
168
Bevers, L. E., Hagedoorn, P. L., Hagen, W. R., Coord. Chem. Rev., 2009, 253, 269-290.
169
Johnson, M. K., Rees, D. C., and Adams, M. W. W., Chem. Rev., 1996, 96, 2817-2859.
170
Antony, A., and Bayse, C. A., Organometallics., 2009, 28, 4938-4944.
171
Vincent, M. A., Hillier, I. H., Periyasamy, G., and Burton, N. A., Dalton. Trans., 2010, 39,
3816-3822.
172
Liao, R. Z., Yu, J. G., and Himo. F., PNAS., 2010, 107, 22523-22527.
173
Suenobu, K., Nagaoka, M., and Yamabe, T., Journal of Molecular Structure (Theochem).,
1999, 461-462, 581-588.
174
LLedós, A., Bertrán, J., and ventura, O. N., International Journal of Quantum Chemistry.,
1986, xxx, 467-477.
175
Richardson, D. J., and Watmough, N. J., Curr. Opin. Chem. Biol., 1999, 3, 207-219.
176
Slater, J. C. Quantum Theory of Molecules and Solids, Vol. 4: The Self-Consistent Field
for Molecules and Solids. McGraw-Hill: New York, 1974.
177
Perdew, J. P., Phys. Rev. B., 1986, 34, 7406.
177
References
178
Perdew, J. P., Phys. Rev. B., 1986, 33, 8822-8824.
179
Rosner, B. M., Rainey, F. A., Kroppenstedt, R. M., and Schink, B., FEMS Microbiology
Letters., 1997, 148, 175-180.
180
Protein Sensors and Reactive Oxygen Species, Part A: Selenoproteins and Thioredoxin
(Eds.: Sies, H., Packer, L), Academic Press, San Diego, 2002, pp. 1-461.
181
Protein Sensors and Reactive Oxygen Species, Part B: Thiol Enzymes and Proteins (Eds.:
Sies, H., Packer, L), Academic Press, San Diego, 2002, pp. 1-364.
182
Jacob, C., Giles, G. I., Giles, N. M., and Sies, H., Angew. Chem. Int. Ed., 2003, 42, 4742-
4758.
183
Kraft, T., Bowen, A., Theis, F., Macy, J. M., DNA Sequence, 2000, 10, 365-377.
184
Wang, J. –J., Tessier, C., and Holm, R. H., Inorg. Chem., 2006, 45, 2979-2988.
185
Watts, C. A., Ridley, H., Dridge, E. J., Leaver, J. T., Reilly, A. J., Richardson, D. J., and
Butler, C. S., 10th Nitrogen Cycle meeting 2004, 173-174.
186
Dridge, E. J., Watts, C. A., Jepson, B. J. N., Line, K., Santini, J. M., Richardson, D. J., and
Butler, C. S., Biochem. J., 2007, 408, 19-28.
187
Maher, M. J., Santini, J., Pickering, I. J., Prince, R. C., Macy, J. M., and George, G. N.,
Inorg. Chem., 2004, 43, 402-404.
188
Krafft, T., Bowen, A., Theis, F., and Macy, J. M., DNA Seq., 2000, 10, 365-377.
189
Maher, M. J., and Macy, J. M., Acta Cryst., 2002, D58, 706-708.
190
Schröder, I., Rech, S., Kraft, T., and Macy, J. M., The Journal Of Biological Chemistry.,
1997, 272, 23765-23768.
191
Kuroda, M., Yamashita, M., Miwa, E., Imao, K., Fujimoto, N., Ono, H., Nagano, K., Sei,
K., ans Ike, M., Journal of Bacteriology., 2011, 193, 2141-2148.
192
Dridge, E. J., and Butler, C. S., Biochimie., 2010, 92, 1268-1273.
193
Kappler, U., Biochimica et Biophysica Acta., 2011, 1807, 1-10.
194
George, G. N., Garrett, R. M., Graf, T., Prince, R. C., Rajagopalan, K. V., J. Am. Chem.
Soc., 1998, 120, 4522-4523.
195
Pal, K., Chaudhury, P. K., and Sarkar, S., Chem. Asian. J., 2007, 2, 956-964.
196
Das, S. K., Chaudhury, P. K., Biswas, D., and Sarkar, S., J. Am. Chem. Soc., 1994, 116,
9061-9070.
197
Stolz, J. F., and Basu, P., Chem. Bio. Chem., 2002, 3, 198-206.
198
Kroneck, P. M. H., Abt, D. J., Met. Ions. Biol. Syst., 2002, 39, 369-403.
199
Cabello, P., Roldan, M. D., Moreno-vivian, C., Microbiology., 2004, 150, 3527-3546.
178
References
200
Richardson, D. J., and Watmough, N. J., Curr. Opin. Chem. Biol., 1999, 3, 207-219.
179