Practical approaches to biological inorganic chemistry

Practical Approaches
to Biological Inorganic
Chemistry

Edited by

Robert R. Crichton
Batiment Lavoisier
Universite´ Catholique de Louvain
Louvain-la-Neuve, Belgium

Ricardo O. Louro
ITQB, Universidade Nova de Lisboa
Oeiras, Portugal

AMSTERDAM • WALTHAM • HEIDELBERG • LONDON • NEW YORK • OXFORD
PARIS • SAN DIEGO • SAN FRANCISCO • SYDNEY • TOKYO


Elsevier
Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK
Copyright Ó 2013 Elsevier B.V. All rights reserved.
No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means
electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher
Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0)
1865 843830; fax (+44) (0) 1865 853333; email: Alternatively you can submit your request online
by visiting the Elsevier web site at and selecting Obtaining permission to use Elsevier
material
Notice
No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products

liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the
material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and
drug dosages should be made
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
A catalog record for this book is available from the Library of Congress
ISBN: 978-0-444-56351-4
For information on all Elsevier publications visit
our website at www.store.elsevier.com
Printed and bound in China
12 13 11 10 9 8 7 6 5 4 3 2 1


Preface

Shrouded in the mists of scientific antiquity (things move so quickly that even a decade or two seems a long time),
in reality a little less than 30 years ago e the Federation of European Biochemical Societies, better known by its
acronym FEBS, invited the Belgian Biochemical society to organise their annual Congress in Belgium. For the
first time in the history of these meetings (since the inaugural Congress, in London in 1964), two half day
symposia were organised on the subject of metalloproteins. At the end of the second of these, a group of what in
those days were called inorganic biochemists met to enjoy a drink together in the bar of the Sheraton Hotel. The
outcome was that two of those present, one of whom is co-editor of the present volume, together with Cees Veeger
were entrusted with the task of organising a FEBS Workshop Course on Inorganic Biochemistry. The first of these
was held at the Hotel Etap in Louvain-la Neuve at the end of April, 1985. The origins of this book can be traced
back to the long series of Advanced Courses which have followed that pioneering start.
At that very first Course, the pattern was established of organising lectures to introduce the subject and to
present a theoretical background to the methods which could be used to study metals in biological systems,
together with practical sessions in smaller groups. The final lectures were then devoted to specific examples. It is
interesting, and perhaps not too surprising, that after an introduction to ligand field theory by Bob Williams, and

metal coordination in biology by Jan Reedijk, X-ray, EPR, NMR, Mo¨ssbauer and EXAFS spectroscopy of
metalloproteins were on the programme. The practicals included NMR, EPR and Mo¨ssbauer as well as Cees
Veeger’s favourite, biochemical analysis of Fe and S in FeeS proteins. There was an evening lecture by Helmut
Beinert (then on sabbatical in Konstanz) entitled ‘Limitations of Spectroscopic Studies on Metalloproteins and
Chemical Analysis of Metals in Proteins’. While the lecturers were shuffled around from year to year, Fred Hagen,
Antonio Xavier, Alfred Trautwein, and Dave Garner represented the cornerstone of the spectroscopic part of the
course over the early years.
Since then, over the period from 1985 until now we have organised some 20 courses, and trained over 800
students, most of whom were doctoral or post-doctoral students when they came on the course. It is a source of
great pride and satisfaction that many of the former students still enjoy active and distinguished careers in the area
of Biological Inorganic Chemistry, as we now call the subject. Even more rewarding are the number of former
participants who now form the staff of the course, notably the other co-editor, who has also taken on the mantle of
co-organiser of the most recent courses. Indeed, with the exception of Rob Robson, who taught the Molecular
Biology lectures and practical for many years, the other authors contributing to this book, Frank Neese, Fred
Hagen, Eckhard Bill, Martin Feiters, Christophe Leger and Margarida Archer are all alumni of the ‘Louvain-laNeuve’ course.
Our intention in editing this volume is that it can serve as a starting point for any student who wants to
study metals in biological systems. The presentations by the authors represent a distillation of what they have
taught over a number of years in the advanced course. We begin with an overview of the roles of metal ions in
biological systems, which we hope will serve as taster for the reader, who will find a much more detailed
account in the companion work to this volume (Crichton, 2012). Thereafter, after an introduction to that most
erudite of discipline (at least for non-inorganic chemists) ligand field theory, augmented by a good dose of
how molecular orbital theory can predict the properties of catalytic metal sites. This leads naturally into
a sequence which describes the physicochemical methods which can be used to study metals in biology,
concluding with an overview of the application of the powerful methods of modern genetics to
metalloproteins.
ix


x

Preface

The considerations expressed by that pioneer of analytical precision Helmut Beinert in his 1985 evening
lecture in Louvain-la-Neuve are as relevant today as they were then. Use as many techniques as possible to analyse
your sample e the more information from different approaches you have, the better we will understand your
protein. Do not waste expensive and sensitive methods on shoddy impure samples, and conversely do not employ
primitive technical means to analyse highly purified samples, which have required enormous investment to obtain
them. And above all recognise that the key to metalloprotein characterisation is collaboration. Do not think you
can simply phagocytise a technique from the laboratory of a colleague who knows the method inside out e it is
much richer to collaborate, incorporating his or her know-how into your research. And you will be the richer for it.
Bonne chance, good luck, boa sorte e and we look forward to greet you on one of the courses which will, we
hope, continue into the future. Hopefully, this little introductory text will not only whet your appetite, but help you
to find your way about the myriad practical methods which can be used to study metals in biological systems.
Robert R. Crichton and Ricardo O. Louro
Louvain-la-Neuve, July, 2012


Chapter 1

An Overview of the Roles of Metals
in Biological Systems
Robert R. Crichton
Batiment Lavoisier, Universite´ Catholique de Louvain, Louvain-la-Neuve, Belgium

Chapter Outline
Introduction: Which Metals Ions and Why?
Some Physicochemical Considerations on Alkali Metals
NaD and KD e Functional Ionic Gradients
Mg2D e Phosphate Metabolism
Ca2D and Cell Signalling

Zinc e Lewis Acid and Gene Regulator
Iron and Copper e Dealing with Oxygen
Ni and Co e Evolutionary Relics
Mn e Water Splitting and Oxygen Generation
Mo and V e Nitrogen Fixation

1
3
3
5
6
10
12
13
16
18

INTRODUCTION: WHICH METALS IONS AND WHY?
In the companion book to this one, ‘Biological Inorganic Chemistry 2nd edition’ (Crichton, 2011), we explain in
greater detail why life as we know it would not be possible with just the elements found in organic chemistry e
namely carbon, oxygen, hydrogen, nitrogen, phosphorus and sulfur. We also need components of inorganic
chemistry as well, and in the course of evolution nature has selected a number of metal ions to construct living
organisms. Some of them, like sodium and potassium, calcium and magnesium, are present at quite large
concentrations, constituting the so-called ‘bulk elements’, whereas others, like cobalt, copper, iron and zinc, are
known as ‘trace elements’, with dietary requirements that are much lower than the bulk elements.
Just six elements e oxygen, carbon, hydrogen, nitrogen, calcium and phosphorus e make up almost 98.5% of
the elemental composition of the human body by weight. And just 11 elements account for 99.9% of the human
body (the five others are potassium, sulfur, sodium, magnesium and chlorine). However, between 22 and 30
elements are required by some, if not all, living organisms, and of these are quite a number are metals. In addition
to the four metal ions mentioned above, we know that cobalt, copper, iron, manganese, molybdenum, nickel,

vanadium and zinc are essential for humans, while tungsten replaces molybdenum in some bacteria. The essential
nature of chromium for humans remains enigmatic.
Just why these elements out of the entire periodic table (Figure 1.1) have been selected will be discussed here.
However, their selection was presumably based not only on suitability for the functions that they are called upon to
Practical Approaches to Biological Inorganic Chemistry, 1st Edition. />Copyright Ó 2013 Elsevier B.V. All rights reserved.

1


2

Practical Approaches to Biological Inorganic Chemistry

FIGURE 1.1 An abbreviated periodic table of the elements showing the metal ions discussed in this chapter.

play in what is predominantly an aqueous environment, but also on their abundance and their availability in the
earth’s crust and its oceans (which constitute the major proportion of the earth’s surface).
The 13 metal ions that we will discuss here fall naturally into four groups based on their chemical properties. In
the first, we have the alkali metal ions Naþ and Kþ. Together with Hþ and ClÀ, they bind weakly to organic
ligands, have high mobility, and are therefore ideally suited for generating ionic gradients across membranes and
for maintaining osmotic balance. In most mammalian cells, most Kþ is intracellular, and Naþ extracellular, with
this concentration differential ensuring cellular osmotic balance, signal transduction and neurotransmission. Naþ
and Kþ fluxes play a crucial role in the transmission of nervous impulses both within the brain and from the brain
to other parts of the body.
The second group is made up by the alkaline earths, Mg2þ and Ca2þ. With intermediate binding strengths to
organic ligands, they are, at best semi-mobile, and play important structural roles. The role of Mg2þ is intimately
associated with phosphate, and it is involved in many phosphoryl transfer reactions. Mg-ATP is important in
muscle contraction, and also functions in the stabilisation of nucleic acid structures, as well as in the catalytic
activity of ribozymes (catalytic RNA molecules). Mg2þ is also found in photosynthetic organisms as the metal
centre in the light-absorbing chlorophylls. Caþ is a crucial second messenger, signalling key changes in cellular

metabolism, but is also important in muscle activation, in the activation of many proteases, both intra- and
extracellular, and as a major component of a range of bio-minerals, including bone.
Zn2þ, which is arguably not a transition element,1 constitutes the third group on its own. It is moderate to
strong binding, is of intermediate mobility and is often found playing a structural role, although it can also fulfil
a very important function as a Lewis acid. Structural elements, called zinc fingers, play an important role in the
regulation of gene expression.
The other eight transition metal ions, Co, Cu, Fe, Mn, Mo, Ni, V and W form the final group. They bind tightly
to organic ligands and therefore have very low mobility. Since they can exist in various oxidation states, they
participate in innumerable redox reactions, and many of them are involved in oxygen chemistry. Fe and Cu are
constituents of a large number of proteins involved in electron transfer chains. They also play an important role in
oxygen-binding proteins involved in oxygen activation as well as in oxygen transport and storage. Co, together
with another essential transition metal, Ni, is particularly important in the metabolism of small molecules like
carbon monoxide, hydrogen and methane. Co is also involved in isomerisation and methyl transfer reactions.
A major role of Mn is in the catalytic cluster involved in the photosynthetic oxidation of water to dioxygen in
plants, and, from a much earlier period in geological time, in cyanobacteria. Mo and W enzymes contain a pyranopterindithiolate cofactor, while nitrogenase, the key enzyme of N2 fixation contains a molybdenumeironesulfur cofactor, in which V can replace Mo when Mo is deficient. Other V enzymes include
1. IUPAC defines a transition metal as “an element whose atom has an incomplete d sub-shell, or which can give rise to cations with an
incomplete d sub-shell.”


Chapter j 1

3

An Overview of the Roles of Metals in Biological Systems

haloperoxidases. To date no Cr-binding proteins have been found, adding to the lack of biochemical evidence for
a biological role of the enigmatic Cr.

SOME PHYSICOCHEMICAL CONSIDERATIONS ON ALKALI METALS
Before considering, in more detail, the roles of the alkali metals, Naþ and Kþ, and the alkaline earth metals, Mg2þ

and Ca2þ, it may be useful to examine some of their physicochemical properties (Table 1.1). We can observe, for
example that Naþ and Kþ have quite significantly different unhydrated ionic radii, whereas, the hydrated radii are
much more similar. It therefore comes as no surprise that the pumps and channels which carry them across
membranes, and which can easily distinguish between them, as we will see shortly, transport the unhydrated ions.
Although not indicated in the table, it is clear that Naþ is invariably hexa-coordinate, whereas Kþ and Ca2þ can
adjust to accommodate 6, 7 or 8 ligands. As we indicated above, both Naþ and Kþ are characterised by very high
solvent exchange rates (around 109/s), consistent with their high mobility and their role in generating ionic
gradients across membranes. In contrast, the mobility of Mg2þ is some four orders of magnitude slower, consistent
with its essentially structural and catalytic. Perhaps surprisingly, Ca2þ has a much higher mobility (3 Â 108/s),
which explains why it is involved in cell signalling via rapid changes on Ca2þ fluxes.
The selective binding of Ca2þ by biological ligands compared to Mg2þ can be explained by the difference in
their ionic radius, as we pointed out above. Also, for the smaller Mg2þ ion, the central field of the cation dominates
its coordination sphere, whereas for Ca2þ, the second and possibly even the third, coordination spheres have an
important influence resulting in irregular coordination geometry. This allows Ca2þ, unlike Mg2þ to bind to a large
number of centres at once.
The high charge density on Mg2þ as a consequence of its small ionic radius ensures that it is an excellent Lewis
acid in reactions notably involving phosphoryl transfers and hydrolysis of phosphoesters. Typically, Mg2þ
functions as a Lewis acid, either by activating a bound nucleophile to a more reactive anionic form (e.g. water to
hydroxide anion), or by stabilising an intermediate. The invariably hexacoordinate Mg2þ often participates in
structures where the metal is bound to four or five ligands from the protein and a phosphorylated substrate. This
leaves one or two coordination positions vacant for occupation by water molecules, which can be positioned in
a particular geometry by the Mg2þ to participate in the catalytic mechanism of the enzyme.

NAD AND KD e FUNCTIONAL IONIC GRADIENTS
How, we might ask, do the pumps and channels responsible for transport across membranes distinguish between
Naþ and Kþ ions? Studies over the last 50 years or so of synthetic and naturally occurring small molecules which
bind ions have established the basic rules of ion selectivity. Two major factors appear to be of capital importance,

TABLE 1.1

Properties of Common Biological Cations
Ionic
radius (A˚)

Hydrated
radius (A˚)

Ionic
volume (A˚3)

Hydrated
volume (A˚3)

Naþ

0.95

2.75

3.6

88.3

8 Â 108

7e13

Kþ

1.38

2.32

11.0

52.5

109

4e6

Cation

2þ

Mg

2þ

Ca

0.65
0.99

(From Maguire and Cowan, 2002).

4.76
2.95

1.2

4.1

Exchange
rate (secÀ1)

5

453

10

108

3 Â 10

Transport
number

12e14
8

8e12


4

Practical Approaches to Biological Inorganic Chemistry

namely the molecular composition and the stereochemistry (essentially the size) of the binding site. Synthetic
˚ ), Naþ (0.95 A

˚ ), Kþ (1.35 A
˚ ) and Rbþ (radius
molecules have been created which selectivity bind Liþ (radius 0.60 A
˚
1.48 A) by simply adjusting the cavity size to match the ion (Dietrich, 1985). Now that we have the crystal structures
of membrane transport proteins, we can begin to understand how ion selectivity is accomplished (MacKinnon,
2004; Gouax and MacKinnon, 2005). The Naþ-selective binding sites in the Naþ-dependant leucine transporter
LeuT and the Kþ-selective binding sites in the Kþ channel have been determined, providing a direct comparison of
selectivity for Naþ and Kþ. The Naþ and Kþ ions are completely dehydrated, both the Naþ and the Kþ sites contain
oxygen ligands, but by far the most important factor distinguishing Naþ and Kþ sites is the size of the cavity formed
by the binding site, which agrees well with the rules already learned from host/guest chemistry. What determines
alkali metal cation selectivity, similar to that observed in ion binding by small molecules, is that the protein selects
for a particular ion, Naþ or Kþ, by providing an oxygen-lined binding site of the appropriate cavity size.
Mammalian cells maintain a high intracellular Kþ (around 140 mM) and low intracellular Naþ (around 12 mM)
through the action of the Naþ, Kþ-ATPase present in the plasma membrane. The overall reaction catalysed is:
3Naþ(in) þ 2Kþ (out) þ ATP þ H2O 5 3Naþ (out) þ 2Kþ (in) þ ADP þ Pi
The extrusion of three positive charges for every two which enter the cell, results in a transmembrane potential of
50e70 mV, which has enormous physiological significance, controlling cell volume, allowing neurons and muscle
cells to be electrically excitable, and driving the active transport of important metabolites such as sugars and
amino acids. More than one-third of ATP consumption by resting mammalian cells is used to maintain this intracellular Naþ À Kþ gradient (in nerve cells this can rise to up to 70%).
This thermodynamically unfavourable exchange is achieved by ATP-mediated phosphorylation of the
Naþ,Kþ-ATPase followed by dephosphorylation of the resulting aspartyl phosphate residue, which drives
conformational changes that allow ion access to the binding sites of the pump from only one side of the membrane
at a time. The ATPase exists in two distinct conformations, E1 and E2, which differ in their catalytic activity and
their ligand specificity (Figure 1.2). The E1 form, which has a high affinity for Naþ, binds Naþ, and the E1.3Naþ
form then reacts with ATP to form the “high-energy” aspartyl phosphate ternary complex E1 ~ P.3Naþ. In relaxing
to its “low-energy” conformation E2-P, the bound Naþ is released outside the cell. The E2-P, which has a high
affinity for Kþ, binds 2Kþ, and the aspartyl phosphate group is hydrolysed to give E2.2Kþ, which then changes
conformation to the E1 form, releasing its 2Kþ inside the cell. The structures of a number of P-type ATPases,
including the Naþ - Kþ-ATPase and the Ca2þÀATPase of the Sarcoplasmic reticulum have been determined and

are shown in Figure 1.3.

FIGURE 1.2 A model for the active transport of Naþ and Kþ by the Naþ-Kþ-ATPase.


Chapter j 1

An Overview of the Roles of Metals in Biological Systems

5

FIGURE 1.3 Overall structures and ion-binding site architectures of two P-type ATPases, rabbit sarcoplasmic reticulum Ca2þ-ATPase
(SERCA) and pig Naþ,Kþ-ATPase. The upper panel depicts rabbit SERCA (E1 Protein Data Base [PDB] entry 1T5S) and pig Naþ-Kþ-ATPase
(E2:Pi, PDB entry 3KDP). N-, P-, and A-domains are coloured red, blue and yellow, respectively; the b-subunit and g-subunit of Naþ,KþATPase wheat and cyan. The lower panel depicts the ion-binding sites, viewed approximately perpendicular to the membrane plane from the
extracytoplasmic side, in the E1 state. Ion liganding residues are shown as sticks, transmembrane helices and calcium ions in SERCA are
indicated by numbers and grey spheres, respectively, and the sites superposed as transparent spheres onto the Naþ,Kþ-ATPase model. Putative
binding sites for the third sodium ion in the Naþ,Kþ-ATPase are indicated as grey ellipses. (From Bublitz et al., 2010. Reproduced Copyright
2010 with permission from Elsevier).

MG2D e PHOSPHATE METABOLISM
The intracellular concentration of free Mg2þ is about 5 Â 10À3 M, so that although Mg2þ-binding to enzymes is
relatively weak (Ka not more than 105MÀ1) and most Mg2þ-dependent enzymes have adequate local concentrations of Mg2þ for their activity. Mg2þ is the most abundant divalent cation in the cytosol of mammalian cells,
binds strongly to ATP and ADP, and is therefore extensively involved in intermediary metabolism and in nucleic
acid metabolism. However, like Zn2þ, it is a difficult metal ion to study, since it is spectroscopically silent, with the
consequence that many spectroscopic studies on Mg2þ enzymes utilise Mn2þ as a replacement metal ion.


6

Practical Approaches to Biological Inorganic Chemistry

Of the five enzymes selected in the Enzyme Function Initiative, recently established to address the challenge
of assigning reliable functions to enzymes discovered in bacterial genome projects, but for which functions
have not yet been attributed (Gerlt et al., 2011), three of them are Mg2þ-dependent. We discuss two of them
briefly here.
The haloalkanoic acid dehalogenase superfamily (HADSF) (>32,000 nonredundant members) catalyse a diverse
range of reactions that involve the Mg2þ-dependent formation of a covalent intermediate with an active site Asp.
Despite being named after a dehalogenase, the vast majority are involved in phosphoryl transfer reactions (Allen and
Dunaway-Mariano, 2004, 2009). While ATPases and phosphatases are the most prevalent, the haloacid dehalogenase
(HAD) family can carry out many different metabolic functions, including membrane transport, signal transduction
and nucleic-acid repair. Their physiological substrates cover an extensive range of both size and shape, ranging from
phosphoglycolate, the smallest organophosphate substrate, to phosphoproteins, nucleic acids, phospholipids,
phosphorylated disaccharides, sialic acids and terpenes.
In HAD enzymes, Asp mediates carbon-group transfer to water (in the dehalogenases) and phosphoryl-group
transfer to a variety of acceptors. Thus, the HAD superfamily is unique in catalysing both phosphoryl-group
transfer (top) and carbon-group transfer (bottom) (Figure 1.4a). The roles of the four loops that comprise the
catalytic scaffold are shown in Figure 1.4b. The activity ‘switch’ is located on loop 4 of the catalytic scaffold
(yellow) which positions one carboxylate residue to function as a general base for the dehalogenases and either
two or three carboxylates to bind the Mg2þ cofactor essential for the phosphotransferases. CO represents the
backbone carbonyl oxygen of the moiety that is two residues downstream from the loop 1 nucleophile (red). The
side-chain at this position is also used as an acid-base catalyst by phosphatase and phosphomutase HAD members.
Loop 2 (green) and loop 3 (cyan) serve to position the nucleophile and substrate phosphoryl moiety. Figure 1.4c
presents a ribbon diagram of the fold supporting the catalytic scaffold of phosphonatase.
The members of another large superfamily of Mg2þ enzymes, the enolase superfamily (with more than 6000
nonredundant members) catalyse diverse reactions, including b-eliminations (cycloisomerisation, dehydration
and deamination) and 1,1-proton transfers (epimerisation and racemisation). The three founder members of the
family are illustrated by mandelate racemase, muconate lactonising enzyme and enolase (Figure 1.5). They all
catalyse reactions in which the a-proton of the carboxylate substrate is abstracted by the enzyme, generating an
enolate anion intermediate. This intermediate, which is stabilised by coordination to the essential Mg2þ ion of the
enzyme, is then directed to different products in the enzyme active sites.

CA2D AND CELL SIGNALLING
Calcium ions play a major role as structural components of bone and teeth, but are also crucially important in cell
signalling. To prevent the precipitation of phosphorylated or carboxylated calcium complexes, many of which are
insoluble, the cytosolic levels of Ca2þ in unexcited cells must be kept extremely low, much lower than that in the
extracellular fluid and in intracellular Ca2þ stores. This concentration gradient gives cells the opportunity to use
Ca2þ as a metabolic trigger e the cytosolic Ca2þ concentration can be abruptly increased for signalling purposes
by transiently opening Ca2þ channels in the plasma membrane or in an intracellular membrane. These increases in
intracellular free Ca2þ concentration can regulate a wide range of cellular processes, including fertilisation,
muscle contraction, secretion, learning and memory and ultimately cell death, both apoptotic and necrotic.
Extracellular signals often act by causing a transient rise in cytosolic Ca2þ levels, which, in turn, activates
a great variety of enzymes through the action of Ca2þ-binding proteins like calmodulin, as we will discuss in detail
below: this triggers such diverse processes as glycogen breakdown, glycolysis and muscle contraction. In the
phosphoinositide cascade (Figure 1.6), binding of the external signal (often referred to as the agonist2 when it
provokes a positive response) to the surface receptor R (step 1) activates phospholipase C, either through a G
2. Many drugs have been developed either as agonist or antagonists to receptor-mediated signalling pathways, e.g. b-blockers block the action
of the endogenous catecholamines adrenaline (epinephrine) and noradrenaline (norepinephrine) on b-adrenergic receptors.


Chapter j 1

An Overview of the Roles of Metals in Biological Systems

7

FIGURE 1.4 The catalytic scaffold in the haloacid dehalogenase (HAD) family of phosphotransferases. (a) In HAD enzymes, Asp mediates
carbon-group transfer to water (in the dehalogenases) and phosphoryl-group transfer to a variety of acceptors. Thus, the HAD superfamily is
unique in catalyzing both phosphoryl-group transfer (top) and carbon-group transfer (bottom). (b) Schematic of the roles of the four loops that
comprise the catalytic scaffold. The activity ‘switch’ is located on loop 4 of the catalytic scaffold (yellow) which positions one carboxylate
residue to function as a general base for the dehalogenases and either two or three carboxylates to bind the Mg2þ cofactor essential for the

phosphotransferases. CO represents the backbone carbonyl oxygen of the moiety that is two residues downstream from the loop 1 nucleophile
(red). The side-chain at this position is also used as an acid-base catalyst by phosphatase and phosphomutase HAD members. Loop 2 (green)
and loop 3 (cyan) serve to position the nucleophile and substrate phosphoryl moiety. (c) Ribbon diagram (core domain: loop 1, red; loop 2,
cyan; loop 3, green; loop 4, yellow; cap domain: specificity loop, blue) of the fold supporting the catalytic scaffold of phosphonatase (1FES).
(From Allen and Dunaway-Mariano, 2004. Copyright 2004, with permission from Elsevier).

protein which uses the energy of guanosine triphosphate hydrolysis to liberate a subunit capable of activating the
next partner in the cascade (2) or alternatively (not shown) by activating a tyrosine kinase. The activated phospholipase C, then hydrolyses phosphatidylinositol-4,5-bisphosphate (PIP2) in the plasma membrane to InsP3 (IP3
in the figure) and diacylglycerol (DG) (3). InsP3 stimulates the release of Ca2þ, sequestered in the endoplasmic
reticulum (4), and this in turn activates numerous cellular processes through Ca2þ-binding proteins, such as


8

Practical Approaches to Biological Inorganic Chemistry

FIGURE 1.5 The substrates, enolate anion intermediates, and products of the MR; MLE and enolase reactions (adapted from Gerlt et al.,
2005).

FIGURE 1.6 The phosphoinositide cascade.


Chapter j 1

An Overview of the Roles of Metals in Biological Systems

9

calmodulin (CaM) (5). The membrane-associated DG activates protein kinase C (6) to phosphorylate and activate
other enzymes, like glycogen phosphorylase. This step also requires Ca2þ.

Calmodulin is a small, dumbbell-shaped protein, abundant in the cytoplasm of most cells of higher organisms,
which has been highly conserved throughout evolution It is made up of two globular domains each of which can
bind two Ca2þ ions, connected by a flexible linker. When all four Ca2þ sites are filled, the linker forms a flexible
seven turn long a-helix, and the protein undergoes a change in conformation, which does not alter its overall
dimensions, but opens up its two Ca2þ-binding lobes, exposing previously hidden hydrophobic residues,
particularly Met. In Figure 1.7a, the structure on the left shows calmodulin without calcium, while that on the right

FIGURE 1.7 (a) Calmodulin without calcium (left, PDB 1cfd) and after calcium binds (right, PDB 1cll). The linker region between the two
Ca2þ-binding domains is in pink and the Ca2þ ions are shown in light blue. (b) Left calmodulin bound to two different target enzymes: calmodulindependent protein kinase II-alpha (top, PDB 1cm1) and myosin light chain kinase (bottom, PDB 2bbm). Only a small piece of the target protein
chain (in red) is included. Right calmodulin bound to anthrax bacteria oedema factor toxin (PDB 1k93). The entire toxin protein is shown in red.


10

Practical Approaches to Biological Inorganic Chemistry

shows calmodulin after calcium binds. The two hydrophobic regions are represented in green and yellow (C and
Met S atoms, respectively), and we can see that with calcium bound, the hydrophobic residues form two grooves
(red stars), waiting to grip around the target protein, while the linker (pink) has formed a long alpha-helix
separating the two calcium-binding domains. A second, and much more dramatic conformation change then
occurs, collapsing the elongated structure of calmodulin to a hairpin conformation, which enables it to wrap
around the binding domain of the target enzyme, gripping the target protein between its two globular domains.
This is illustrated in Figure 1.7b (left panel) for calmodulin bound to calmodulin-dependent protein kinase
II-alpha (top) and myosin light chain kinase (bottom). Only a small piece of the target protein chain (red) is
included, with the flexible linker of calmodulin, (purple), allowing it to adopt to the slightly different shapes of the
two targets. In the case of the oedema factor toxin from the anthrax bacteria shown in Figure 1.7b (left panel), we
see a quite different binding geometry. This time, the whole toxin protein is shown; once calmodulin binds, it
induces a conformational change in the toxin which activates its adenylyl cyclase activity, thereby depleting the
host cell’s energy stores.
An unusual feature of calmodulin is that, unlike other Ca2þ-binding proteins which usually only interact with

a specific target protein, calmodulin interacts with a wide range of targets. A comparison of amino acid sequences
of calmodulin-binding domains of target proteins suggests that calmodulin principally recognises positively
charged amphipathic helices. Upon binding to the target peptide (compare Figure 1.7a and b), the long central
helix of uncomplexed calmodulin unwinds and bends to form a globular structure that encloses the target polypeptide within a hydrophobic tunnel.

ZINC e LEWIS ACID AND GENE REGULATOR
After iron, zinc is the second most abundant trace element in the human body: an average adult has about 3 g of
Zn. Some 95% of Zn is intracellular. It is essential for growth and development in all forms of life, has been
proposed to have beneficial therapeutic and preventative effects on infectious diseases, including a shortening of
the length of the common cold in man. Zn is found in more than 300 enzymes, where it plays both a catalytic
and a structural role. It is the only metal to have representatives in each of the six fundamental classes of
enzymes recognised by the International Union of Biochemistrydoxidoreductases (e.g. alcohol dehydrogenase): transferases (RNA polymerase): hydrolases (carboxypeptidase A): lyases (carbonic anhydrase):
isomerases (phosphomannose isomerase): and ligases (pyruvate carboxylase, aminoacyl-tRNA synthases). Zinc
is involved in enzymes in both a catalytic and a structural role. Many nucleic acid-binding proteins have
essential Zn atoms in characteristic structures called ‘zinc fingers’ which are widely involved in the regulation of
the transcription and translation of the genetic message. Figure 1.8 is a representation of the Cys2His2 zinc finger
motif, consisting of an a helix and an antiparallel b sheet. The zinc ion (green) is coordinated by two histidine
residues and two cysteine residues.
The first zinc enzyme to be discovered was carbonic anhydrase in 1940; it represents the archetype of monozinc enzymes, with a central catalytically active Zn2þ atom bound to three protein ligands, and the fourth distorted
tetrahedral site occupied by a water molecule. The mechanism of action of mononuclear zinc enzymes depends on
the Zn2þ-OH2 centre, which can participate in the catalytic cycle in three distinct ways (Figure 1.9) e either by
ionisation, to give zinc-bound hydroxyl ion (in carbonic anhydrase), polarisation by a general base (as in
carboxypeptidases) or displacement of the ÀOH2 ligand by the substrate (in alkaline phosphatase). In the case of
carbonic anhydrase, the zinc ion functions as a powerful electrophilic catalyst by providing some or all of the
following properties: (i) an activated water molecule for nucleophilic attack, (ii) polarisation of the carbonyl of the
bond to be cleaved, (iii) stabilisation of the negative charge which develops in the transition state.
The coordination chemistry and the main features of the mechanism of carbonic anhydrase are illustrated in
Figure 1.10, and involve the following steps: (i) deprotonation of the coordinated water molecule with a pKa ~ 7,
in a process facilitated by general base catalysis involving His 64. This residue is too far away from the
Zn2þ-bound water to directly remove its proton, but it is linked to it by two intervening water molecules, forming

Chapter j 1

An Overview of the Roles of Metals in Biological Systems

11

FIGURE 1.8 Cartoon representation of the Cys2His2 zinc finger motif, consisting of an a helix and an antiparallel b sheet. The zinc ion
(green) is coordinated by two histidine residues and two cysteine residues.

FIGURE 1.9 The zinc-bound water can either be ionized to zinc-bound hydroxide, polarised by a general base to generate a nucleophile for
catalysis, or displaced by the substrate.


12

Practical Approaches to Biological Inorganic Chemistry

FIGURE 1.10 (a) The active site of human carbonic anhydrase and (b) the main features of the mechanism of action of carbonic anhydrase.

a hydrogen-bonded network which acts as a proton shuttle, (ii) the zinc-bound hydroxide then carries out
a nucleophilic attack on the carbon dioxide substrate to generate a hydrogen carbonate intermediate [(His)3ZnOCO2H]þ which (iii) is displaced by H2O to release bicarbonate and complete the catalytic cycle. The key to
understanding the role of the Zn2þ ion is that its charge makes the bound water molecule more acidic than free
H2O, and enables it to act as a source of OHÀ even at neutral pH values.

IRON AND COPPER e DEALING WITH OXYGEN
Both iron and copper play a very important role in the living world, and both seem to be essential for life, although
iron may not be essential for lactic acid bacteria. On the basis of their chemistry and biochemistry, it seems
probable that the early chemistry of life in an oxygen-free environment used water soluble Fe(II), whereas copper

was present essentially as highly insoluble sulfides of Cu(I). The advent of oxygen was a catastrophic event for
most living organisms, and can be considered to be the first general irreversible pollution of the earth. The
oxidation of iron resulted in the loss of its bioavailability as Fe(II) was replaced by insoluble Fe(III), whereas
the oxidation of insoluble Cu(I) led to soluble Cu(II). Further, the advent of an oxidising atmosphere exposed the
potential toxicity of both elements through their capacity to generate oxygen-free radicals. A new iron
biochemistry became possible after the advent of oxygen, with the development of chelators of Fe(III), which
rendered iron once again accessible, and with the control of the potential toxicity of iron by its storage in a water
soluble, non-toxic, bio-available storage protein (ferritin). Biology also discovered that whereas enzymes involved
in anaerobic metabolism were designed to operate in the lower portion of the redox spectrum (attaining values of
close to þ0.6 V for iron itself), the arrival of dioxygen created the need for a new redox active metal which could
attain higher redox potentials. Copper, now bioavailable, was ideally suited to exploit the oxidizing power of
dioxygen. The arrival of copper also coincided with the development of multicellular organisms which had
extracellular cross-linked matrices capable of resisting attack by oxygen-free radicals. After the initial ‘iron age’,
subsequent evolution moved, not towards a ‘copper age’, but rather to an ‘iron-copper’ age.
The extensive range of biological functions carried out by both of these metal ions, range through oxygen
transport by haemoglobins, haemerythrins and haemocyanins, through electron transfer by cytochromes, Fe-S
proteins and plastocyanins, to multiple reactions involving oxygen activation and detoxification e the list is
seemingly endless. However, here we focus briefly on one enzyme which combines the powerful redox chemistry
of iron and copper acting in concert, cytochrome oxidase, CcO, the terminal component of the respiratory chain in


Chapter j 1

An Overview of the Roles of Metals in Biological Systems

13

aerobic organisms, CcO catalyses the one electron reduction of four reduced cytochrome c (c2þ) molecules and
the four electron reduction of dioxygen to water:
O2 þ 4Hþ þ 4c2þ / 2H2O þ 4c3þ

In mammals, CcO spans the mitochondrial inner membrane and catalyses the reduction of molecular dioxygen to
water at the rate of up to 250 molecules of O2 per second. The energy released in this process is coupled to the
translocation of protons, which in turn contributes to the chemiosmotic gradient required for ATP synthesis. Since
the electrons and protons are taken up from the opposite sides of the membrane, the reaction results in a net charge
separation across the membrane which, together with the coupled proton pumping, corresponds to the overall
translocation of two positive charges across the membrane per electron transferred to O2 from the negative (N)
side to the positive (P) side of the membrane.
O2 þ 8HN þ þ 4cp 2þ / 2H2 O þ 4HN þ þ 4c3þ
While the structure of the mammalian enzyme, with between 8 and 13 different subunits has been determined,
it presents a very much more complex problem than the enzyme from Rhodobacter sphaeroides. This contains
only the two catalytic subunits e subunit I with 3 redox-active centres, containing haem a, and the catalytic site
made up of haem a3 and CuB, where dioxygen is reduced, and subunit II with the CuA redox centre made up of two
copper ions, together with two other non-catalytic subunits (Qin et al., 2006). Its overall structure and the location
of the different electron transfer components are shown in Figure 1.11a.
Electrons coming from cytochrome c enter the CcO complex via the dinuclear copper centre (CuA), and are
then transferred consecutively one at a time to haem a, and then to the catalytic site of CcO, the dinuclear haemcopper centre (haem a3-CuA). This is the primary oxygen-binding site, involving a haem iron, haem a3, together
with a copper ion, CuB, and it is at this dinuclear metal site that dioxygen is reduced. A tyrosine residue, Y(I-288),
which is covalently cross-linked to one of the CuB ligands (His 240), is also a part of the active site. A more
detailed view of the redox-active cofactors and amino acid residues involved in the proton transfer pathways is
shown in Figure 1.11b (Brzezinski and Johansson, 2010).
The individual steps of oxygen binding and its subsequent reduction by CcO are presented in Figure 1.12
(Brzezinski and Johansson, 2010). In oxidised CcO (designated O0), both haem a3 and CuB are oxidised. Transfer
þ
of the first and second electrons to the catalytic site results in the formation of states E1 and R2 (Fe2þ
a3 and CuB ),
each step associated with proton uptake to the catalytic site and proton pumping, and O2 then binds to haem a3 in
the R2 state. In the next step, the O-O bond is broken, with four electrons being donated, two from haem a3,
2À
forming an oxo-ferryl state Fe4þ
a3 ]O2 , one from CuB with a hydroxide ion bound, and one electron (and a proton)

from residue Tyr288, located within the catalytic site, which forms a tyrosyl radical, Tyr288,. Formation of this P2
state is not linked to any proton uptake from solution and, both the protons and the electrons are only relocated
locally within the catalytic site resulting in oxidation of haem a3 and CuB. In the next step, an electron is
transferred to the Tyr radical, accompanied by proton uptake to form state F3 and again, protons are pumped. In
the final step, the last electron is transferred to the catalytic site, forming state O4 also accompanied by proton
uptake and proton pumping. The O4 state is equivalent to O0 as the enzyme becomes fully oxidised when four
electrons have been transferred to O2.

Ni AND Co e EVOLUTIONARY RELICS
Both nickel and cobalt, together with iron, have the characteristic that they are electron-rich. They are further
distinguished by the fact that in lower oxidation states some of their 3d electrons are forced into exposed s-(or pÀ)
orbitals: the outcome is that tetragonal Co(II) or Ni(III) are reactive-free radicals, able to give or take an odd
electron, like s-organic-free radicals. So, like iron, cobalt functions in free-radical reactions, such as the transformation of ribonucleotides into their corresponding deoxy derivatives. When one examines the kinds of reactions
catalysed by nickel and cobalt enzymes and their evolutionary distribution, one arrives at the conclusion that these


14

Practical Approaches to Biological Inorganic Chemistry


Chapter j 1

An Overview of the Roles of Metals in Biological Systems

15

FIGURE 1.12 The catalytic cycle of cytochrome c oxidase. The electrons (eÀ) and protons at the arrows (in green) are those transferred to the
catalytic site, while the protons indicated by arrows (in red) perpendicular to the reaction arrows indicate pumped protons. Y is Tyr288 (see
text), whereas Y Â O indicates the tyrosyl radical. The reaction pathway along the blue arrows is that observed during reaction of the fully

reduced CytcO (with four electrons) with O2. (From Brzezinski and Johansson, 2010. Copyright 2010, with permission from Elsevier).

two elements were particularly important in the metabolism of chemicals like methane, carbon monoxide and
hydrogen, all particularly abundant in the pre-oxygen evolutionary era. This is reflected in the high levels of both
elements in a number of anaerobic bacteria. In contrast, the level of both metals in mammalian serum is less than
100-fold that of zinc, iron or copper. Nonetheless, cobalt, through its involvement in a number of important vitamin
B12-dependent enzymes continued to be used in higher organisms, including mammals. In contrast, with the
exception of the plant enzyme urease, nickel proteins are virtually unknown in higher eukaryotes.
The Ni-Fe hydrogenases which play an important role in microbial energy metabolism catalyse the reversible
oxidation of hydrogen:
H2 # 2Hþ þ 2eÀ
Whereas in some anaerobic microorganisms, production of hydrogen serves as a mechanism to get rid of excess
reducing potential, in many others hydrogen consumption is coupled to the reduction of carbon dioxide, oxygen,
sulfate, or other electron acceptors while simultaneously generating a proton gradient for use in ATP production.
[NiFe] hydrogenases have an unusual Ni-Fe active site (Figure 1.13) which required a combination of both spectroscopic and crystallographic studies to identify the three non-protein diatomic ligands e a good example of why
one must use as many techniques as possible when studying metal ions in proteins. The unusual coordination of
cyanide and carbon monoxide ligands to the 2Fe subcluster could only be established by spectroscopic methods,
since the electron density of carbon, nitrogen and oxygen does not permit their differentiation by X-ray crystallography. It is thought that the single CO and two CNÀ ligands maintain iron in its low spin ferrous state.
Vitamin B12 is a tetrapyrrole cofactor in which the central hexacoordinate cobalt atom is coordinated by four
equatorial nitrogen ligands donated by the pyrroles of the corrin ring (Figure 1.14). The fifth Co ligand is

=

FIGURE 1.11 (a) The structure of cytochrome c oxidase from R. sphaeroides (PDB code 1M56). The four subunits of the enzyme are
coloured as indicated in the figure. Haems a and a3 are shown in red and the copper centres CuA and CuB in yellow. The red spheres are water
molecules resolved in the structure. Residues Glu286, Asp132, Lys362, all in SU I, and Glu101 in SU II, are shown in the figure (the subscript
indicates the subunit number). The approximate position of the membrane is indicated by the solid lines, where the p- and n-sides are the more
positively and negatively charged sides of the membrane, respectively. The purple sphere is a non-redox-active Mg2þ ion found in the structure.
(b) The D and K proton pathways shown in more detail. Also the haem a3 propionates are indicated. (From Brzezinski and Johansson, 2010.
Copyright 2010, with permission from Elsevier).

16

Practical Approaches to Biological Inorganic Chemistry

FIGURE 1.13 Structure of the NiFe active site of the hydrogenase from Desulfomicrobium baculatum. (From Garcin et al., 1999, PDB
1CC1).

a nitrogen atom from a 5,6-dimethylbenzimidazole nucleotide (Dmb) covalently linked to the corrin D ring. The
sixth ligand in vitamin B12 is eCN. In the coenzyme B12 (AdoCbl), this ligand is 50 -deoxyadenosine, while in the
other biologically active alkylcobalamine (MeCbl), it is a methyl group. This sixth ligand is unusual in that it
forms a C-Co bond e carbon-metal bonds are rare in biology.
The principal role of essentially all AdoCbl-dependent enzymes is to facilitate the interchange of a group X and
a hydrogen atom (H) on adjacent carbon atoms of the substrate. The identity of the migrating species X can be a small
carbon-skeleton fragment or a small heteroatom-containing group like OH or NH2, depending on the enzyme.
Figure 1.15 outlines the generally accepted mechanism for these reactions and shows how radical intermediates play
a crucial mechanistic role. In the first step, the substrate (1) binding induces homolytic cleavage of the CoÀC bond of
AdoCbl generating the 50 -deoxyadenosyl radical (Ado,) plus cob(II)alamine. Hydrogen abstraction by Ado, from 1
then occurs to form 50 -deoxyadenosine (Ado-H) plus a substrate-derived radical 2 (step A). The rearrangement of 2
gives the product-related radical 3 (step B), which is followed by H-atom transfer from Ado-H to 3 to form the
product 4 and to regenerate Ado, (step C), which is able to recombine with cob(II)alamine, thereby completing the
catalytic cycle. In some cases, elimination of H2O or NHþ
4 from 4 results in production of an aldehyde (5, step D).

Mn e WATER SPLITTING AND OXYGEN GENERATION
The particular biological importance of manganese might be considered to reside in the tetranuclear Mn cluster
which is involved in oxygen production in photosynthetic plants, algae and cyanobacteria. However, it also plays
a key role in a number of mammalian enzymes like the key enzyme of the urea cycle, arginase and the mitochondrial superoxide dismutase. Most of manganese biochemistry can be explained on the one hand by its redox
activity, and on the other by its analogy to Mg2þ. Mn has an extraordinarily important role in the photosystem II

(PSII), which uses solar energy to power the oxidation of water to oxygen in photosynthetic plants, algae and
cyanobacteria. The overall reaction catalysed by PSII is:
light

2Q þ 2 H2 O ƒƒ! O2 þ 2 QH2
where Q represents plastoquinone and QH2 represents plastoquinol. The electrons required to convert the oxidised
quinone to the reduced quinol are extracted from the oxidation of two molecules of water, generating molecular
dioxygen. This latter reaction takes place at a special centre, often called the oxygen evolving complex, which
contains a tetranuclear Mn complex.
Organisms which produce oxygen, use chlorophyll a in their PSII reaction centre. This can generate a redox
potential as oxidizing as þ1 V, allowing the evolution of machinery that can oxidise water (redox potential


Chapter j 1

An Overview of the Roles of Metals in Biological Systems

17

FIGURE 1.14 Ball and stick representation of adenosylcobalamin. (From Reed, 2004. Copyright 2004 with permission from Elsevier).

FIGURE 1.15 Mechanism for the rearrangements catalysed by AdoCbl-dependent enzymes (adapted from Sandala et al., 2010).


18

Practical Approaches to Biological Inorganic Chemistry

þ0.9 V), by extracting four electrons from two water molecules to yield a molecule of dioxygen. The pathway of
electron transfer in PSII is generally agreed to be:

H2 O/½Mn4 CaC1Š/Yz =Yz à / P680=P680þ /Pheoa =Pheoa À /QA =QA À /QB =QB À
where [Mn4CaCl] is the manganese cluster, YZ is the tyrosine residue that mediates electron transfer between the
manganese cluster and the chlorophyll pair P680, Pheoa is a pheophytin, and QA and QB are plastoquinones
(Figure 1.16).
The photochemistry of PSII begins when the special pair of chlorophyll molecules in PS II, often called P680,
absorbs light at 680 nm and transfers an electron to a nearby pheophytin molecule, from where it is transferred
through other electron carriers to an exchangeable plastoquinone pool (Figure 1.16). A long-lived charge separation between the positively charged species which is formed on the special pair, P680þ (a powerful oxidant) and
the plastoquinone QB, some 26A away, means that each time a photon of light kicks an electron out of P680,
P680þ extracts an electron from water molecules bound at the Mn centre, which is transferred through the redoxactive TyrZ to reduce P680þ back to P680 for yet another photosynthetic cycle. In classic experiments using an
oxygen electrode and short flashes of light, it was established that four flashes were required for every molecule of
oxygen that was released, and the features of this were rationalised into a kinetic model, known as the S-state
cycle. In this model, five states, designated Sn, of the enzyme are proposed to exist, with n 0e4, where each state
corresponds to a different level of oxidation of the tetra-Mn centre. When the most oxidised state, S4 is generated,
it reacts in less than a microsecond to release dioxygen and return to the most reduced form of the enzyme, S0.
While the structure of PSII from the cyanobacterium T. elongatus has been elucidated by X-ray crystallog˚ resolution (Ferreira et al., 2004; Guskov et al., 2009), the precise positions of the Mn ions
raphy at 3.5 and 2.9 A
and water molecules in the photosynthetic water-splitting Mn4Ca2þ cluster remain uncertain. This is due to the
low resolutions of the crystal structures, and the possibility of radiation damage at the catalytic centre. Very
˚ resolution
recently, the structure of PSII from another cyanobacterium T. vulcanus has been determined at 1.9 A
(Kawakami et al., 2011), which has yielded a detailed picture of the Mn4CaO5-cluster for the first time. In the
high-resolution structure (Figure 1.17), the Mn4CaO5-cluster is arranged in a distorted chair form, with a cubanelike structure formed by 3 Mn and 1 Ca, four oxygen atoms as the distorted base of the chair, and 1 Mn and 1
oxygen atom outside of the cubane as the back of the chair. In addition, four water molecules were associated with
the cluster, among which, two are associated with the terminal Mn atom and two are associated with the Ca atom.
Some of these water molecules may therefore serve as the substrates for water-splitting.

Mo AND V e NITROGEN FIXATION
With the exception of bacterial nitrogenase, whose Fe-Mo-cofactor will be discussed in detail below, all other Mo
enzymes contain the molybdenum pyranopterindithiolate cofactor (MoCo), which is the active component of their
catalytic site (and of tungsten enzymes, in organisms which do not use molybdenum). They can be divided into

three families, the xanthine oxidase, sulfite oxidase and the DMSO reductase families.
A relatively limited number of anaerobic microorganisms are capable of converting atmospheric dinitrogen
into ammonia which can then be incorporated into amino acids glutamate and glutamine, and from there into other
nitrogen-containing molecules. This represents about 108 tons/year, about the same as is produced by the
industrial Haber-Bosch process e which functions at both high pressures (150e350 atm) and high temperatures
(350e550  C).
All nitrogenases consist of two types of subunit, one of which contains a special Fe-S cluster, known as the
P-cluster, and a second, which contains an iron- and sulfur-containing cofactor which includes a heterometal. The
heterometal is usually molybdenum, hence the cofactor is known as FeMoCo. In some species, under conditions of
particular metal bio-availability, Mo can be replaced by vanadium or iron. These “alternative” nitrogenases
contain vanadium instead of molybdenum (when Mo levels are low and V is available) and another which contains


Chapter j 1

An Overview of the Roles of Metals in Biological Systems

19

FIGURE 1.16 A structural model of the cyanobacterial PSII. (a) View from the cytoplasm of the PSII monomer. The model was constructed
according to Guskov et al., PDB 3BZ1. Green, chlorophylls; cyan, all the other pigments and prosthetic groups. ETC is coloured as in (b).
(b) Side view of the electron transfer components and the oxygen-evolving complex. Green, chlorophylls; magenta, pheophytin; blue, plastoquinone; red, iron; blue, YZ d tyrosine 161/D1; cyan, chloride; yellow, calcium; purple, manganese. (From Nelson, 2011. Copyright 2011
with permission from Elsevier).


20

Practical Approaches to Biological Inorganic Chemistry

˚ resolution. (a) Structure of the metal cluster with oxo-bridges and water

FIGURE 1.17 Structure of the Mn4CaO5 cluster determined at 1.9 A
˚ . Hydrogen bonds were depicted as dashed lines. (b) Hydrogen-bond network linking the Mn4CaO5
ligands. The bond distances are shown in A
cluster and YZ, and further from YZ to the opposite side of PSII. (From Kawakami et al., 2011. Copyright 2011 with permission from Elsevier).

only iron (when both Mo and V levels are low). However, by far the greatest advances in our understanding of the
structure and mechanism of nitrogenases have come from studies on the MoFe-nitrogenases from free-living
nitrogen-fixing bacteria like Azotobacter, Clostridium and Klebsiella.
The overall reaction catalysed by nitrogenases is:
N2 þ 8Hþ þ 8e- þ 16ATP þ 16H2O ➙ 2NH3 þ H2 þ 16ADP þ 16Pi
Nitrogen fixation is extremely energy-intensive, requiring both large amounts of ATP and of reducing equivalents. The nitrogenase is made up of two proteins (Figure 1.18), termed the MoFe protein and the Fe protein. The
a2b2 heterotetrameric MoFe protein contains both the FeMo-cofactor and the so-called P-cluster, with the functional unit constituted by ab dimer, containing one FeMo-cofactor and one P-cluster. In contrast, the Fe protein is
a homodimer, which binds a single [4Fe-4S] cluster at the interface between the two subunits. Unlike many other
multipleeelectron transfer reactions in biochemistry, each individual electron transfer between the Fe protein and
the MoFe protein requires the binding and hydrolysis of at least two ATP molecules. The basic mechanism of
nitrogenase involves the following steps:


Chapter j 1

An Overview of the Roles of Metals in Biological Systems

21

FIGURE 1.18 Structures of the nitrogenase MoFe and Fe proteins. The MoFe protein is an a2b2 tetramer, with the alpha subunits shown in
magenta and the beta subunits shown in green. The Fe protein is a g2 dimer, with each subunit shown in blue. A MoFe protein binds two Fe
proteins, with each ab unit being a catalytic unit. One Fe protein is shown associating with one ab unit of the MoFe protein. The relative
positions and structures of two bound MgADP molecules, the Fe protein [4Fe-4S] cluster, and MoFe protein P-cluster (8Fe-7S), and FeMo
cofactor (7Fe-Mo-9S-homocitrate-X) are shown. Each is highlighted to the right. The flow of electrons is from the [4Fe-4S] cluster to the
P-cluster to the FeMo cofactor. The element colour scheme is C grey, O red, N blue, Fe rust, S yellow, and Mo magenta. Graphics were

generated with the program Pymol using the Protein Data Base (PDB) files 1M1N for the MoFe protein and 1FP6 for the Fe protein. (From
Seefeldt et al., 2009. Copyright 2009 with permission from Annual Reviews, Inc.).

(i) complex formation between the MoFe protein and the reduced Fe protein with two molecules of ATP
bound, (ii) electron transfer between the two proteins coupled with hydrolysis of ATP, (iii) dissociation of
the oxidised Fe protein from the complex accompanied by its re-reduction and exchange of the 2ADPs for
ATPs, (iv) repetition of this cycle of association, reduction, ATP hydrolysis and dissociation to transfer one electron at a time to the MoFe protein. Once a sufficient number of electrons and protons have been accumulated,
available substrates can be reduced. Usually, when eight reducing equivalents have been accumulated, and 16
molecules of ATP hydrolysed, the enzyme can bind and reduce the very stable triple bond of a dinitrogen molecule to two molecules of ammonia. Concomitantly, two protons and two electrons are converted to gaseous
hydrogen. Electrons derived from photosynthesis or from the mitochondrial electron transport chain are transferred to the Fe protein.