1
Introductory note
This chapter is intended to provide a concise summary of the principles with which
you should be familiar in order to understand the structures and functions of
proteins. If you have already studied chemistry in some depth, you may well not
need to study this chapter in detail; a good test would be to see how well you can
tackle the problems at the end of the chapter. If you are in any doubt, however, it
is probably worthwhile taking the time to read it carefully and master its contents!
Aims of this chapter
Proteins are complex three-dimensional structures with a very wide variety of
functions in nature. This chapter is designed to help you acquire or consolidate
your understanding of the structure and function of molecules, with a particular
focus on those relevant to the study of proteins. It is not meant to be a ‘smallscale’
chemistry course, but to build on a basic knowledge of the Periodic Table of
the elements and the ways in which ionic and covalent compounds can be formed.
It starts with the amino acids as the basic building blocks of proteins and then
deals with the various levels of protein structure. The forces involved in stabilizing
the three-dimensional structures of proteins and in the interactions between
proteins and other binding partners are described. At the end of this chapter,
there is a set of chemical structures that you should aim to learn, as they will help
to deepen your understanding of the ways in which proteins behave. There is also
a set of problems that you can use to check your understanding of the principles
in this chapter.
. .
If you wish to read more
about the molecular
principles outlined in this
chapter you will find that
there is good coverage of
these in most of the standard
textbooks on biochemistry
or introductory chemistry,
such as Berg et al. (2007),
Crowe et al. (2006), Jones
(2005), Mathews et al.
(2000), Price et al. (2001),
and Voet et al. (2006)
Conceptual toolkit: the
molecular principles for
understanding proteins
1.1 The different aspects of protein structure
KEY CONCEPT
■ Understanding the four levels at which protein structure can be defined
To understand how proteins function it is necessary to know their structure.
Whereas for small molecules such as ethanol, benzene, or glucose we can gain very
useful insights into the properties from simple two-dimensional representations
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 3
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
4 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
C CO2HNH2
R
H
Fig. 1.1 General structure of
an amino acid.
of their covalent structure, the same cannot be said for the much more complex
molecules of proteins, which contain many thousands of atoms. In this case we
have to extend the description of structure into three dimensions.
It is convenient to discuss the structure of proteins at four levels:
A. Primary structure, which refers to the sequence of amino acids in the
polypeptide chain, i.e. the covalent structure of the protein. This includes any
post-translational modifications such as glycosylation, phosphorylation, etc.
B. Secondary structure, which refers to the local folding of the polypeptide chain,
such that segments of the chain may form helices, strands of sheet or turns.
C. Tertiary structure, which refers to the long-range folding of the polypeptide
chain so that portions of the chain that are remote in terms of sequence are
brought close together in space.
D. Quaternary structure, which refers to the association of the individual
polypeptide chains (or subunits) in a multi-subunit protein.
Before we discuss these aspects of structure we should review the properties of the
amino acids, which represent the building blocks of proteins.
1.2 The constituents of proteins, the amino acids
KEY CONCEPTS
■ Knowing the general structure of an amino acid and understanding its
stereochemistry
■ Knowing the structures of the 20 common amino acids found in proteins
■ Understanding the basis of amino acid classifications
■ Explaining the behaviour of amino acids in chemical terms
■ Understanding the properties of water as a solvent
There are 20 different amino acids commonly found in proteins. Of these one
(proline) is actually a secondary amino acid (sometimes termed an imino acid).
The general structure of an amino acid is shown in Fig. 1.1, where NH2– is the
amino group, –CO2H is the acid group, and R is the side chain. The central C atom
to which R is attached is referred to as the α-carbon atom (denoted Cα). If the side
chain consists of a chain of C atoms, these are usually referred to by Greek letters
. .
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 4
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
1.2 THE CONSTITUENTS OF PROTEINS, THE AMINO ACIDS 5
Many of the amino acids
were named for what now
seem rather obscure reasons,
e.g. glycine comes from the
Greek glykeros (sweet)
because pure glycine was
found to have a sweet taste.
Leucine is named from the
Greek leukos (white) because
it was first isolated as a white
crystalline solid; at the time
whiteness was considered to
be the defining property of
a pure compound. Serine is
derived from the Greek
serikon and Latin sericum
(silk) because it is found in
significant quantities in the
hydrolysis products of silk.
It should be noted that the
classification of a few of the
amino acids (particularly
Cys and Gly) is difficult; they
may be placed in different
categories in other books.
The zwitterionic (from the
German zwitter, meaning
hybrid) form of an amino
acid is one which is neutral
overall but carries both a
positive and a negative
charge.
C CO2
–+
NH3
R
HFig. 1.2 The zwitterionic
form of an amino acid in
which the carboxyl group has
lost a proton and the amino
group has gained a proton.
going away from the α-carbon atom, i.e. β, γ, δ, ε. Thus the side chain of glutamic
acid (Glu) (see section 1.2.1) contains a γ-carboxyl group and of lysine (Lys) contains
an ε-amino group.
In reality, in aqueous solution at pH values around neutrality, the amino group
gains a proton and the acid group loses a proton, giving rise to the so-called zwitterionic
form, shown in Fig. 1.2.
It will greatly help your understanding of proteins if you can learn the names,
abbreviations, and chemical structures of the amino acids, i.e. of their side chains,
because then you will be able to account for the structure and properties of these
proteins in molecular terms. A list of the full structures of the amino acids is
included in the Compendium of Structures in section 1.8.
1.2.1 The variety of amino acids
The amino acids are listed in alphabetical order in Table 1.1, with details of some
of their important properties.
1.2.2 Classification of the amino acids in terms of polarity
The different amino acids can be classified in a number of ways, but probably the
most useful in terms of understanding protein structure is that based on the polarity
of the side chain. This can be thought of as a measure of whether the side chain
promotes (polar) or discourages (non-polar) solubility in water. The polar amino
acids are often further sub-divided into charged and uncharged side chains.
Non-polar side chain
Ala, Gly, Ile, Leu, Met, Phe, Pro, Trp, Val
Polar, uncharged side chain
Asn, Cys, Gln, Ser, Thr, Tyr
Polar, charged side chain
Arg, Asp, Glu, His, Lys
. .
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 5
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
6 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
A quantitative measure of polarity can be provided by a so-called hydrophobicity
scale. There are several ways in which this can be set up; that due to Engelman et al.
(1986) corresponds to the free energy change for transfer of a given amino acid
in an α-helix from the interior of a membrane to an aqueous medium (Table 1.2).
In the table, the amino acids are ranked in order from most non-polar to most
polar.
1.2.3 General properties of the amino acids
In this section we shall explore two general properties of amino acids before looking
at specific chemical characteristics in the next section.
. .
The sign and magnitude
of the free energy change
indicates the tendency of a
process or reaction to occur
(see Chapter 4, section 4.1).
A process with a large
positive free energy change
will have little tendency to
proceed; conversely a
process with a large negative
free energy change will have
a high tendency to proceed.
Table 1.1 Properties of the amino acids commonly found in proteins
Name Abbreviation Massa
pKa of Frequency of
side chainb
occurrence (%)c
Three-letter One-letter
Alanine Ala A 71.08 – 7.83
Arginine Arg R 156.19 12.5 5.35
Asparagine Asn N 114.10 – 4.18
Aspartic acid Asp D 115.09 3.9 5.32
Cysteine Cys C 103.14 8.4 1.52
Glutamine Gln Q 128.13 – 3.95
Glutamic acid Glu E 129.12 4.1 6.64
Glycine Gly G 57.05 – 6.93
Histidine His H 137.14 6.0 2.29
Isoleucine Ile I 113.16 – 5.91
Leucine Leu L 113.16 – 9.64
Lysine Lys K 128.17 10.5 5.93
Methionine Met M 131.20 – 2.38
Phenylalanine Phe F 147.18 – 4.00
Proline Pro P 97.12 – 4.83
Serine Ser S 87.08 – 6.86
Threonine Thr T 101.11 – 5.42
Tryptophan Trp W 186.21 – 1.15
Tyrosine Tyr Y 163.18 10.5 3.06
Valine Val V 99.13 – 6.71
a
The mass in Da (Daltons) is given minus that of water. The molecular mass of a protein can be obtained by
adding the values shown for the masses of the constituent amino acids plus that of one molecule of H2O
(18.02). This mass would be adjusted if necessary for the effect of any post-translational modifications, e.g.
the addition of one phosphate group would add 79.98 Da.
b
The pKa value is that of the side chain in the free amino acids; in a protein the value of the pKa can be
markedly influenced by the precise environment of the side chain. The pKa values of the carboxyl and αamino
groups for most amino acids are typically about 2.2 and 9.5, respectively.
c
The frequency is taken from the occurrence of the amino acids in all sequences deposited in the Swiss-Prot
database (release 49.0 7 Feb 2006; representing 75438310 amino acids in 207132 sequences).
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 6
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
1.2.3.1 Stereochemistry
Since the α-carbon atom has four different substituents (except in the case of Gly
for which R = H), all amino acids, except Gly, will display chirality. Only one
of the enantiomers (mirror image forms) occurs in proteins; this is the l form of
the amino acids (see Fig. 1.3). The d/l system for designating the stereochemistry
of the amino acids is based on the d and l forms of glyceraldehyde as a reference
compound, i.e. related to other compounds by a series of reactions of known
stereochemistry. The more recent and absolute R/S system for describing the
stereochemistry of chiral compounds is based on ranking the atoms or groups
attached to the asymmetric carbon atom primarily in terms of atomic number and
then establishing the direction of rotation of the ranked groups when looking
along the bond from the carbon atom to the lowest ranked group (see Fig. 1.4).
Almost all the l-amino acids have the S configuration at the α-carbon atom.
The side chains of Ile and Thr each contain a second chiral centre; these can be
readily designated in the R/S system as S and R for Ile and Thr, respectively.
d-amino acids are found occasionally in nature, for example d-Ala and d-Gln in
the peptidoglycan component of some bacterial cell walls, and d-Phe in gramicidin
S, a cyclic decapeptide synthesized by the bacterium Bacillus brevis. Peptides
containing these amino acids are made by special enzyme systems distinct from
the normal ribosome-based protein synthesizing machinery.
1.2 THE CONSTITUENTS OF PROTEINS, THE AMINO ACIDS 7
. .
Table 1.2 A hydrophobicity scale for amino acids
Amino acid Transfer free energy (kJ mol−−1
)
Phe (F) 15.5
Met (M) 14.2
Ile (I) 13.0
Leu (L) 11.7
Val (V) 10.9
Cys (C) 8.4
Trp (W) 7.9
Ala (A) 6.7
Thr (T) 5.0
Gly (G) 4.2
Ser (S) 2.5
Pro (P) −0.8
Tyr (Y) −2.9
His (H) −12.5
Gln (Q) −17.1
Asn (N) −20.1
Glu (E) −34.3
Lys (K) −36.8
Asp (D) −38.5
Arg (R) −51.4
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 7
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
8 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
1.2.3.2 Ionization
In Table 1.1, the average pKa values of the α-amino and carboxyl groups of amino
acids are given (9.5 and 2.2, respectively). These allow us to see why at neutral pH
amino acids will exist in the zwitterionic form (NH3
+
–CHR–CO2
−
), since at pH 7
the amino group is below its pKa and will thus be protonated, and the carboxyl
group will be above its pKa and thus be deprotonated. The actual charge carried
by an amino acid at any particular pH will also depend on the nature and state of
ionization of the side chain (see Table 1.1).
The isoelectric point (denoted by pI) of an amino acid is the pH at which it carries
no net charge, i.e. where there is an exact balance of positive and negative charges.
At the pI, the amino acid will not move in an electric field. For an amino acid with
. .
H
L-Glyceraldehyde
1
2
3
+ +
C
CH2OH
HO
CHO
OH
D-Glyceraldehyde
C
CH2OH
H
CHO
H
L-Alanine
C
CH3
H3N
COO–
NH3
D-Alanine
C
CH3
H
COO–
Fig. 1.3 The d/l system for denoting the stereochemistry of amino acids. The upper panels show the structures of
l- and d-glyceraldehyde, which is the reference compound; the lower panels show the l and d forms of alanine. The
solid arrows show atoms pointing towards the reader; the dotted arrows show atoms pointing away from the reader.
The l and d forms (enantiomers) of each compound are mirror images of each other and are non-superimposable.
4 4
Clockwise
(R)
Counterclockwise
(S)
3
2
1
2
3
1
C C
Fig. 1.4 The R/S system for denoting the stereochemistry at a chiral carbon atom. In this system the atoms linked
to the carbon atom are ranked in terms of priority (1 being the highest ranked and 4 the lowest). When viewed
looking from the carbon atom to atom 4, the direction in which the other atoms are ranked 1→2→3 is either
clockwise (denoted R) or counterclockwise (denoted S). The priorities of some common substituent groups are
–OCH2X > –OH > –NH2 > –COOH > –CHO > –CH2OH > –CH3 > –H.
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 8
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
no ionizable side chain, the pI is equal to the average of the pKa values of the carboxyl
and α-amino groups (e.g. for Gly, pI = (2.4 + 9.8)/2 = 6.1). For an amino acid with
an ionizable side chain, the pI is equal to the average of the two pKa values on either
side of the zwitterionic form. Thus for Asp, where the pKa values for the α-carboxyl,
the side chain carboxyl, and the α-amino group are 2.0, 3.9, and 9.9, respectively,
the pI = (2.0 + 3.9)/2 = 2.95. For a protein, the actual pI will depend on the balance
between the numbers of the acidic and basic amino acids in that protein.
1.2 THE CONSTITUENTS OF PROTEINS, THE AMINO ACIDS 9
The calculation of the pI
value for amino acids with
ionizing side chains is
discussed in Problems 1.6
and 1.7 at the end of this
chapter.
Aliphatic compounds are
those in which carbon atoms
are linked in straight or
branched chains, rather
than rings.
Aromatic compounds
are those which contain
benzene-type rings. Their
enhanced stability arises
from the delocalized
π-electron systems.
. .
KEY INFO
It should be remembered that the ionization states of the side chains of amino
acids in proteins can be significantly different from those of the free amino
acids. For example, the pKa of Glu 35 in lysozyme is raised from 4.1 in the
free amino acid to about 6.0 because the side chain is in a rather non-polar
environment. This has the effect of favouring the neutral (protonated) form
rather than the negative (deprotonated) form. Conversely, the pKa of Tyr 28
in type II dehydroquinase is lowered from 10.5 in the free amino acid to about
8.0 by the presence of a neighbouring Arg side chain. The positive charge on
the Arg will favour the deprotonated form of the Tyr side chain relative to
the protonated form. In the case of Ser 195 of chymotrypsin, the pKa is lowered
dramatically (by over 7 units) due to the charge relay system. It is therefore
always important to examine the actual environment of an amino side chain in
a protein before assessing its likely function.
!
1.2.4 Chemical characteristics of the amino acids
It is convenient to discuss the various amino acids as groups which display similar
chemical features. Again, knowing the structures of the side chains involved will
help to understand the molecular basis of the property.
1.2.4.1 Aliphatic side chains (Ala, Gly, Ile, Leu, Val)
These side chains are chemically unreactive. They are important in hydrophobic
interactions (section 1.7.4) with other non-polar side chains or parts of other
molecules.
Because of its small side chain, Gly has a large degree of conformational flexibility,
which is important when the three-dimensional structures of proteins are
considered (section 1.4).
1.2.4.2 Aromatic side chains (Phe, Tyr, Trp)
These side chains absorb radiation in the near-ultraviolet (the absorbance maxima
for Phe, Tyr, and Trp are 258 nm, 275 nm, and 280 nm, respectively).
The side chain of Phe is chemically unreactive but can participate in hydrophobic
interactions (section 1.7.4).
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 9
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
10 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
The side chain of Tyr is more reactive (due to the hydroxyl group) and will
undergo electrophilic substitution reactions. One such example is iodination,
which occurs naturally to give the hormone thyroxine. The hydroxyl group
has a high pKa (typically about 10.5) so will not be significantly ionized under
normal physiological circumstances; it can, however, participate in hydrogen
bonding (section 1.7.2) and be phosphorylated by the action of kinases (section
1.2.4.5).
The side chain of Trp is of limited chemical reactivity, but the N–H group can
participate in hydrogen bonding (section 1.7.2).
1.2.4.3 Basic side chains (Arg, Lys)
These side chains are grouped together because at neutral pH they carry a positive
charge and can therefore be involved in ionic interactions with negatively charged
side chains (Asp and Glu) or with negatively charged parts of other molecules,
such as phosphate groups.
The pKa of Lys is around 10.5 for the free amino acid. The –NH3
+
group is unreactive,
but the deprotonated form (–NH2) can act as a powerful nucleophile.
The pKa of Arg is around 12.5, and the guanidine part of the side chain is therefore
essentially positively charged under all physiological conditions.
1.2.4.4 Acidic side chains (Asp, Glu)
For both free amino acids, the pKa values of the side chains are around 4.0, and
hence would generally be negatively charged at neutral pH. These side chains can
be involved in ionic interactions with positively charged side chains, or in binding
to metal ions such as Ca2+
or Zn2+
.
1.2.4.5 Hydroxyl side chains (Ser, Thr)
The –OH group of the side chain of both amino acids can take part in hydrogen
bonding (section 1.7.2). The pKa of the side chains is very high (about 15), so they
are in the protonated state under physiological conditions and act as very weak
nucleophiles. However, in certain cases, such as the Ser proteases (see Chapter 9,
section 9.9.3), the Ser can lose its proton by the charge relay system and become a
very powerful nucleophile. Ser and Thr side chains (along with Tyr side chains,
section 1.2.4.2) in a number of proteins can become phosphorylated by the action
of specific enzymes (kinases); such processes are often involved in regulation of
the activity of these proteins.
1.2.4.6 Amide side chains (Asn, Gln)
The amide group has only very weak acidic and basic properties so the side chains
of these amino acids are always uncharged. They can participate in hydrogen
bonding (section 1.7.2) as either donors or acceptors.
. .
An electrophile is a species
or part of a molecule with a
positive or partial positive
charge which will seek an
electron-rich species or
centre. Examples include
the Zn2+
ion or the C atom
in the >C=O bond. In the
case of tyrosine, reagents
that bring about electrophilic
substitution include I2 and
the NO2
+
(nitronium) ion.
Thyroxine, an iodinated
derivative of tyrosine, is a
hormone which is produced
in the thyroid gland. It
stimulates the synthesis
of specific proteins and is
essential for normal growth
and development. A number
of isotopes of iodine
(particularly 129
I and 131
I)
are radioactive and can be
dangerous because they will
be concentrated in the
thyroid gland, leading to
thyroid cancer.
A nucleophile is a species or
a centre in a molecule with a
negative or partial negative
charge which will seek an
electron-deficient species
or centre. Examples include
OH−
ions and the side chains
of Cys (–CH2–S−
), Lys
(–NH2), and Ser (–CH2–O−
)
with one or more lone pairs
of electrons.
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 10
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
1.2.4.7 Sulphur-containing side chains (Cys, Met)
The sulphur atom of the Cys side chain is a highly reactive nucleophile, especially
in the deprotonated state (–CH2S−
). The pKa of the side chain (8.4 for the free
amino acid) is sufficiently low that at neutral pH there will be a small, but
significant, fraction in the deprotonated form.
Under oxidizing conditions, two appropriately positioned Cys side chains can
form a disulphide bond (–CH2–S–S–CH2–), which constitutes the amino acid
cystine. Disulphide bonds are found in extracellular, secreted proteins such as
antibodies, digestive enzymes, etc.
In the presence of oxidizing agents, the Cys side chain can be oxidized to a
sulphonic acid (–CH2–SO3H) or the intermediate sulphenic (–CH2–SOH) and
sulphinic (–CH2–SO2H) acid forms. The sulphur atom in Cys can act as a ligand
to a number of metal ions, e.g. Zn2+
or Fe2+
, and is a major site of inhibition of proteins
by heavy metals such as Hg, Cd, or Pb.
The sulphur atom of Met is much less reactive, although it can function as a
nucleophile towards adenosine-5′-triphosphate (ATP) to form S-adenosylMet
(adoMet) in a reaction involving the release of the three phosphate groups, which
is catalysed by methionine adenosyl transferase. The positive charge on the sulphur
atom of adoMet makes it susceptible to nucleophilic attack, and hence adoMet is
a powerful methylating agent, e.g. of cytosine bases in DNA. The side chain of Met
can also be modified by strong oxidizing agents to give Met sulphoxide.
1.2.4.8 Proline
Proline is a special amino acid since its side chain is a cyclic ring. This structural
feature means that there are rigid geometrical constraints on the peptide bonds to
Pro (section 1.3.1). The side chain of Pro is chemically rather unreactive, although
it can be hydroxylated in the 4 position by the action of prolyl-4-hydroxylase.
4-Hydroxyproline is found in the protein collagen, which occurs extensively in
bone, skin, and connective tissue.
1.2.4.9 Histidine
The pKa of the imidazole ring is about 6.0 in the free amino acid. Thus, at
near-neutral pH, there is balance of the protonated (positively charged) and
deprotonated (neutral) forms. His side chains are often found at the active sites of
enzymes as they can play a key role in acid–base catalysis. The neutral form of His
is a powerful nucleophile and can participate in hydrogen bonding (section 1.7.2);
in addition it can act as a ligand to metal ions such as Zn2+
or Fe2+
.
1.2.5 The structure of water
Since the majority of biological processes take place in an aqueous medium, it is
appropriate to review the structure of liquid water. The geometry of the water
molecule is well known, with the H–O–H bond angle = 104.5° and the O–H bond
1.2 THE CONSTITUENTS OF PROTEINS, THE AMINO ACIDS 11
Disulphide bonds are only
found in secreted proteins.
In intracellular proteins
(i.e. those which remain
inside cells) the cysteine
side chains remain in the
reduced (–CH2–SH) form.
The oxidation of secreted
proteins occurs in the lumen
(interior compartment) of
the endoplasmic reticulum
in eukaryotic cells and in the
periplasmic space of
prokaryotic cells.
Oxidative damage to
proteins, particularly to the
sulphur-containing amino
acids, is thought to be an
important factor in ageingrelated
diseases. Reaction of
Cys side chains in proteins
with reactive oxygen species
such as H2O2 or the
superoxide anion can yield
the sulphenic acid form. This
reaction is usually reversible,
but further oxidation cannot
generally be reversed.
Prolyl-4-hydroxylase
requires ascorbic acid
(vitamin C) as a cofactor.
A deficiency of this vitamin
leads to the condition of
scurvy, a disease of the skin,
joints, etc. caused by
disruption to the correct
synthesis and assembly of
collagen fibres.
. .
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 11
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
12 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
H
HH
H
H
H
H
H
H
O
O
O
O
O
H
Fig. 1.5 A water molecule
hydrogen bonded to its four
nearest neighbours. The
covalent O–H bonds are
shown as solid lines; hydrogen
bonds are shown as dotted
lines. The four nearestneighbour
water molecules
are arranged in a tetrahedral
fashion. The arrangement is
found in one of the most
common structures of ice,
where there are effectively
layers of water molecules.
length = 0.096 nm. There is a separation of charge due to the higher electronegativity
(electron attracting power) of the O atoms, leading to partial positive charges
on the hydrogens and a partial negative charge on the oxygen (about 33% ionic
character). The two lone pairs of electrons on the O atom can each act as acceptors
in hydrogen bonds and the two hydrogen atoms can act as donors (section 1.7.2).
Thus, each water molecule can have up to four nearest-neighbour hydrogen-bonded
water molecules (Fig. 1.5). This network is found in ice, which has a significant
amount of empty space and thus a relatively low density (ice floats on water).
In liquid water, this regular hydrogen-bonded network is broken down and instead
there are fluctuating clusters of water molecules with, on average, each water
having about 3.5 hydrogen-bonded nearest-neighbour molecules. Because of the
polar nature of the water molecule there will be strong interactions with charged or
other polar species, making water an excellent solvent for these types of molecules.
However, water is a very poor solvent for non-polar molecules; this is discussed
further in the context of hydrophobic interactions in section 1.7.4.
1.3 The primary structure of proteins
KEY CONCEPTS
■ Drawing the structure of a peptide chain
■ Understanding the cis/trans isomerism of the peptide bond unit
■ Understanding the information available from the primary structure of
a protein, including molecular mass, extinction coefficient, isoelectric
point, and any post-translational modifications
The term ‘primary structure’ refers to the sequence of amino acids in a protein, and
is dictated by the nucleotide sequence of the gene encoding the protein, taking into
. .
Water is unusual in that its
solid form (ice) is less dense
than its liquid form. This is
of considerable importance
for aquatic organisms in the
winter!
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 12
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
account (a) the splicing out of any intervening sequences (introns) to produce the
mature messenger RNA and (b) any post-translational modifications which might
remove amino acids by proteolysis or modify amino acids, e.g. by addition of
carbohydrate or phosphoryl groups.
1.3.1 The peptide bond
The peptide bond is formed between the carboxyl group of one amino acid and the
amino group of a second amino acid, with the loss of a molecule of H2O. By convention,
a peptide chain is written such that the amino acid with the free amino
group (the amino (or N)-terminal acid) is at the left-hand end of the chain and the
amino acid with the free carboxyl group (the carboxyl (or C)-terminal acid) is at
the right-hand end (Fig. 1.6). Peptide chains in proteins are unbranched.
The peptide bond itself is not adequately represented by the structure shown in
Fig. 1.7, as shown by the fact that the C–N bond distance (0.132 nm) is between
1.3 THE PRIMARY STRUCTURE OF PROTEINS 13
The formation of a dipeptide
from two amino acids is
thermodynamically
unfavourable. In nature,
ATP is used as an energy
source to activate one of the
amino acids and drive the
reaction to the side of
synthesis.
CH
O
CNH2
R1
H
Amide
bond
R2
CH CO2HN
Fig. 1.7 Structure of a
peptide bond with the C–N
bond shown as a single bond.
NH2 CH
CH3
CH3 CH3
N – terminus C – terminus
CO
CH2 CH2 CH2
CH2
S
CH3
CH
NH CH CO NH
α – carbon atoms
Side chains
OH
CH CO NH CH CO2H
α
Fig. 1.6 Structure of a tetrapeptide showing the α-carbon atoms, side chains, and Nand
C-termini. The tetrapeptide is written as Ala–Leu–Phe–Met (ALFM in the oneletter
code). In this diagram the terminal amino and carboxyl groups are shown in
the uncharged forms; in practice they would carry a positive charge and negative
charge from the gain and loss of a proton, respectively.
. .
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 13
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
14 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
that of a C–N single bond (0.145 nm) and a C=N double bond (0.125 nm). This is
usually explained by resonance stabilization, with electron density in the lone pairs
on the O atom being transferred to the N atom, giving rise to partial double bond
character (Fig. 1.8).
This resonance has two consequences. Firstly, it establishes a permanent dipole
(separation of charge) associated with the peptide bond. The H and N atoms carry
positive and negative charges of about 0.28 electron charges respectively and the C
and O atoms carry positive and negative charges of about 0.39 electron charges
respectively (Fig. 1.8).
Secondly, it means that peptide bonds are planar units in which the two αcarbon
atoms are on the same side of the C–N bond (cis form) or on opposite sides
of the C–N bond (trans form) (Fig. 1.9).
The repulsion between non-bonded atoms linked to the two α-carbon atoms
is greater in the cis form than in the trans form. This means that the trans form
. .
Since the N atom carries a
net negative charge, there
must be extensive backdonation
of σ electrons from
C to N to compensate for the
π electron movement.
Fig. 1.9 The cis and trans
forms of the peptide bond.
(a) and (b) show the cis and
trans forms, respectively, of
the peptide bond (Xaa–Zaa),
where Xaa is any amino acid
and Zaa is any amino acid
other than proline. (c) and (d)
show the cis and trans forms,
respectively, of the peptide
bond (Xaa–Pro).
C N¨
C H
O
C N
H
δ−
δ+
δ+
δ−
O
α
Cα
Cα
Cα
Fig. 1.8 Resonance stabilization of the peptide bond. The left-hand panel shows
delocalization of the lone pair of electrons on the N atom. The right-hand panel
shows the partial double bonds in the peptide bond unit and the partial charges on
the C, O, N, and H atoms.
C
(a)
N
HO
C
(b)
N
H
O
C
(c)
N
O
C
(d)
N
O
Cα
CαCα Cα
Cα
CαCαCα
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 14
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
is more stable than the cis; in proteins it can be estimated that this is by a margin
of about 20 kJ mol−1
.
An exception to this statement is in the case when the peptide bond is to proline
(Fig. 1.9), which as we have seen is an unusual amino acid with the side chain
forming a cyclic ring. This has the effect that the repulsion between non-bonded
atoms is considerably reduced and the balance between cis and trans is much
closer. The trans/cis ratio is now much lower (about 20), i.e. about 5% of Xaa–Pro
bonds in proteins are in the cis form (Jabs et al., 1999).
1.3.2 Information available from the amino acid sequence of
a protein
Knowledge of a protein sequence gives us a good deal of information, as set out
below. Further details are given in the section on bioinformatics (see Chapter 5,
section 5.8).
1.3.2.1 Exact molecular mass
From the numbers of each amino acid in a protein, we can calculate the exact mass
using the values given in Table 1.1 for the masses of each amino acid and adding
the mass of H2O. For example, the calculated molecular mass of the polypeptide
chain of type II dehydroquinase from Streptomyces coelicolor is 16550.6 Da. This
can be compared with the very precise measurements of mass available from mass
spectrometry to confirm the identity and integrity of the protein. In this case the
measured mass of the purified enzyme is 16549.7 ± 1.2 Da, in excellent agreement
with the predicted value.
1.3.2.2 Isoelectric point (pI)
As mentioned in section 2.3, the pI of a protein is the pH where it carries no net
charge. The way that the charge on the protein will change with pH is illustrated in
Fig. 1.10.
The pI of a protein will reflect a balance between the numbers of positively
charged groups (N-terminal amino group and the side chains of Arg, His, and Lys)
and negatively charged groups (C-terminal carboxyl group and the side chains of
Asp and Glu). The predicted pI is based on the numbers of these charged groups
and their average pKa values in proteins. If the positive groups are predominant,
the pI will be high (e.g. for pig lysozyme, a value of 9.06 is predicted); if the negative
groups predominate, the pI will be low (e.g. for pig pepsin, a value of 3.24 is
predicted). It should be noted that these are only rough estimates of the pI value
since the pKa values of side chains in proteins can depend markedly on their environment.
Nevertheless, the predicted value is a useful guide to the behaviour of the
1.3 THE PRIMARY STRUCTURE OF PROTEINS 15
It should be noted that there
is a considerable activation
energy barrier (of the order
of 80 kJ mol−1
) to rotation
about the peptide bond
because of its partial double
bond character. This
isomerization of the peptide
bonds to Pro can represent
one of the slow steps in the
folding of proteins and there
is an enzyme, peptidyl-prolyl
isomerase, which can
catalyse this process.
Electrophoresis refers to
the movement of a charged
molecule in an electric field.
Electrophoresis of proteins
is often performed in a crosslinked
polyacrylamide gel
(see Chapter 6, section 6.2,
and Chapter 8, sections 8.2.1
and 8.2.2). Ion-exchange
chromatography is a widely
used technique for
separating proteins on the
basis of the charge that
they carry (see Chapter 7,
section 7.2.2).
. .
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 15
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
16 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
protein on electrophoresis (see Chapter 6, section 6.2.3) or ion-exchange chromatography
(see Chapter 7, section 7.2.2), for example.
1.3.2.3 Absorption coefficient
The absorption of radiation at 280 nm depends on the aromatic amino acids Tyr
and Trp; there is also a small contribution from any disulphide bonds which
may be present. The molar absorption coefficient (also known as the extinction
coefficient) of the protein at 280 nm can be derived by adding together the contributions
from the numbers of these different amino acids; these are based on their
numbers in the protein and their known absorption coefficients at 280 nm.
Division by the molecular mass of the protein gives the absorption coefficient for a
1 mg mL−1
solution of the protein (see Chapter 6, section 6.1.5).
1.3.2.4 Hydrophobicity
The balance between polar and non-polar amino acids in the protein will give
a guide to the overall polarity of the protein and can point to whether it may
be membrane-associated, for example. The aliphatic index is a measure of the
frequency of occurrence of the bulky aliphatic side chains (Ala, Ile, Leu, and Val).
. .
The calculation of the
absorption coefficient
depends on the assumption
that there is no other
chromophore (absorbing
group) present, such as the
haem group in haemoglobin.
If such a group is present,
its contribution to the
absorption at 280 nm must
be added.
Fig. 1.10 The effect of pH
on the charge of a protein.
At low pH, the protein has
an overall positive charge.
As the pH is raised, acidic
side chains
(Asp and Glu) become
deprotonated. When the
number of negatively charged
side chains equals the number
of positively charged side
chains, the protein has no
net charge; this pH equals
the isoelectric point (pI).
As the pH is raised further,
the protein has an overall
negative charge as a result of
deprotonation of basic side
chains.
COOH
pH < pI pH > pIpH = pI
NH3
+
NH3
+
3
+2
+1
−1
−2
0
4 5 6 7 8 9 10 11 pH
COOH
Net charge
Isoelectric point (pI)
COO−
NH3
+
NH3
+
COO−
COO−
NH2
NH2
COO−
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 16
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
1.3.2.5 Post-translational modifications
Many proteins undergo changes in covalent structure after translation of the
mRNA on the ribosome. These changes are collectively termed post-translational
modifications and some of the more important of these are listed in Table 1.3.
The occurrence of post-translational modifications can be deduced either by
direct analysis, e.g. testing for the presence of saccharide units in the case of glycosylation,
or by careful measurements of the molecular mass of the protein using
mass spectrometry. If the measured molecular mass of a protein does not agree
with the value calculated on the basis of the amino acid sequence, then it can be
concluded that either the sequence is incorrect or (more likely) post-translational
modification has occurred. From the observed mass differences it should be
possible to draw firm conclusions about the types of modifications which have
occurred. In the case of horse lysozyme the measured mass is 14644 Da, indicating
a loss of 8 Da from the predicted value (14652 Da); this can be accounted for by the
formation of four disulphide bonds from the eight Cys residues in the sequence,
which corresponds to the loss of eight hydrogen atoms.
1.3.2.6 Structural and functional motifs
Analysis of the sequence can reveal the way certain parts of the protein may play
distinct roles. For example, a stretch of 20 amino acids which is very likely to form
a membrane-spanning (or transmembrane) α-helix (see section 1.4) can be
1.3 THE PRIMARY STRUCTURE OF PROTEINS 17
In a number of cases,
extensive post-translational
modification (e.g.
glycosylation) may hinder
the analysis of a protein by
mass spectrometry, as the
protein cannot readily form
gaseous phase ions.
. .
Table 1.3 Some important post-translational modifications of proteins
Type of modification Possible effects on function
Proteolysis Removal of targetting sequences
Generation of several new products, e.g. hormones
Activation of proteins, e.g. enzymes
Disulphide bond formation Stabilization of structure of secreted proteins
Hydroxylation Formation of hydroxy–Lys or –Pro increases the stability of
the triple helix of collagen
Glycosylation Many cell surface proteins are involved in cell–cell
recognition
Attachment of glycosyl-phosphatidyinositol groups
anchors proteins to membranes
The polar nature of proteins can be enhanced
Phosphorylation Phosphorylation of Ser, Thr or Tyr side chains can regulate
the activity of proteins, especially in signalling pathways
N-terminal acylation Attachment of C14 (myristoylation) or C16 (palmitoylation)
chains will enhance the association of the protein with
membranes
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 17
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
18 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
identified by constructing a hydropathy plot in which the free energy for transferring
this stretch of amino acids from the membrane to water (obtained by adding
the appropriate values of the amino acids as listed in Table 1.2) is plotted as a function
of the first residue in the stretch. If the free energy is above a certain threshold
value (+84 kJ mol−1
) this indicates that a transmembrane helix is likely. Protein
sequences can be scanned automatically to locate such elements, which often
occur many times in a single chain, for example the G-protein coupled receptors
have seven transmembrane helices and the family of glucose transporters in mammalian
tissues have 12 transmembrane helices.
Other types of motifs that can be revealed by sequence analysis are listed below
(Xaa equals any amino acid):
Targetting sequences, e.g. –Ser–Lys–Leu at the carboxyl terminus and
–Lys–Asp–Glu–Leu at the carboxyl terminus act as sequences which direct
proteins to the peroxisome or to be retained in the endoplasmic reticulum
respectively.
Metal binding, e.g. –Cys–Xaa4–Cys–Xaa2–Cys– is a consensus site for Fe binding
in the 2Fe–2S cluster iron sulphur proteins.
Phosphorylation sites, e.g. –Arg–Xaa1–2–Ser/Thr– or –Arg–Arg–Xaa–Ser/Thr–
are consensus sites for phosphorylation at the Ser or Thr by protein kinase A.
Glycosylation sites, e.g. –Asn–Xaa–Ser/Thr– (where Xaa is any amino acid but is
rarely Pro or Asp) is a consensus sequence for N-glycosylation at Ser/Thr.
1.3.2.7 Sequence relationships between proteins
Analysis of the huge number of sequences in the databases has provided powerful
insights into many aspects of proteins. For example, comparisons between the
analogous proteins from different organisms has given clues about evolutionary
relationships and helped in classification of species. When many such proteins are
compared, it is seen that some amino acids are totally conserved between species,
some are partially conserved (i.e. are of similar chemical types) and others are
freely variable. The totally conserved residues are likely to play essential roles
in the structure and/or the function of the protein. The comparisons between related,
but distinct, proteins within the same organism, for example haemoglobin
and myoglobin, trypsin and chymotrypsin, point to gene duplication followed
by independent (divergent) evolution as the likely mechanism. Comparisons
between sequences have also been used to look for clues to possible functions of
proteins, or of segments of proteins. For example, analysis of the sequences of a
number of steroid hormone receptor proteins shows that they consist of three
domains (independent folded and functional units), namely a transactivation
domain which regulates transcription, a DNA binding domain, and a hormone
binding domain. For further details about the use of bioinformatics-based
approaches, see Chapter 5, section 5.8.
A commonly used tool
for identifying structural
motifs is Prosite (http://
www.expasy.org/prosite).
For further details see
Chapter 5, section 5.8.5.
It should be noted that the
choice of the free energy
threshold for identification
of a membrane-spanning
element of a protein is
somewhat arbitrary; the
quoted value has been found
to predict such elements
with reasonable accuracy.
. .
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 18
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
1.4 THE SECONDARY STRUCTURE OF PROTEINS 19
The precise allowed areas of
the Ramachandran plot will
depend on the nature of the
side chains involved. Thus
glycine, with its very small
side chain, can occupy a
considerably greater fraction
of the total Ramachandran
plot than can an amino acid
with a bulky side chain such
as valine or tryptophan.
1.4 The secondary structure of proteins
KEY CONCEPTS
■ Defining the torsional (dihedral) angles in a peptide bond
■ Drawing the hydrogen bonding patterns in the α-helix, β-sheet, and
β-turn structures
■ Explaining why proline cannot be accommodated in regular α-helix
and β-sheet structures
Because of the partial double bond character of the peptide bond, the six atoms of
Cα–CO–NH–Cα can be considered to lie in a plane (Fig. 1.11). Rotations around
the bonds to the linking Cα atoms between peptide units will define the path traced
and hence the structure adopted by the polypeptide chain. The two angles (known
as torsion or dihedral angles) are defined as φ (phi) and ψ (psi) for the bonds
between N and Cα and between Cα and C, respectively. Values of 180° refer to the
fully extended polypeptide chain. Positive values of φ and ψ refer to clockwise
rotation, negative values to counter-clockwise rotation.
. .
Fig. 1.11 Structure of
a dipeptide element of a
protein showing the planar
peptide bond units. The
angles of rotation (φ and ψ)
about the bonds to the central
α-carbon atom are shown.
Values of φ = ψ = 180°
correspond to the fully
extended polypeptide chain.
O
O
C
N H
φ
ψ
R
N
C
H
H
Cα
Cα
Cα
Rotations about the bonds to the Cα atoms will alter the distances between nonbonded
atoms of the side chains and hence the energy of the system. Calculations
of the energy as a function of the angles φ and ψ have been made and the resulting
plot (known as a Ramachandran diagram, Fig. 1.12) shows that there are certain
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 19
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
Fig. 1.12 The Ramachandran
diagram showing the
secondary structures of
proteins formed by successive
residues with identical
conformations. The allowed
values of the angles (φ and ψ)
for a polypeptide chain
consisting of alanine residues
are shown by the shaded
areas. The positions of a
number of secondary
structures are indicated.
collagen helix
310 helix
right-handed
-helix
-helix
anti-parallel -sheet
+180°
+180°
ψ
φ
β
α
left-handed
-helix
(not observed)
α
π
parallel
-sheetβ
−180°
−180°
0°
0°
20 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
well-defined structures adopted by peptide chains that are of greatest stability and
are termed ‘allowed’.
Some of the more common structural elements adopted by proteins are listed
below.
1.4.1 The α-helix
The α-helix is a right-handed helix (i.e. the direction of twist from the N- to the
C-terminus is that of a right-handed corkscrew). Hydrogen bonds are formed
between the carbonyl group of the amide bond between amino acids n and n + 1,
and the amino group of the amide bond between amino acids n + 3 and n + 4
(Fig. 1.13).
For model peptides of regular structure (polyamino acids) the dimensions of
the α-helix are well characterized. There are 3.6 amino acids per turn, the pitch
of the helix is 0.54 nm, and the angles φ and ψ are −57° and −47°, respectively.
The hydrogen bonds are parallel to the axis of the helix and there is a linear
arrangement of nuclei with the OjN distance 0.286 nm. There is a helix dipole
with the N-terminal and C-terminal ends of the helix having positive and negative
charges, respectively; this dipole can be important in stabilizing interactions
between adjacent helices. Since there are 13 atoms in the ring containing the
hydrogen bond (Fig. 1.13) and 3.6 amino acids per turn, an α-helix is sometimes
referred to as a 3.613 helix.
. .
Electrostatic interactions
between the carbonyl group
of amino acid n and the
carbonyl groups of amino
acids n + 3 and n + 4 also
contribute to the stability
of the α-helix. For l-amino
acids, the right-handed
α-helix is more stable than
a left-handed helix.
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 20
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
Because of the steric constraints imposed by its cyclic side chain, Pro cannot be
accommodated in the regular α-helix structure and will therefore either act as a
helix breaker or will cause a major kink in the helix.
In proteins the α-helices are often considerably distorted from the idealized
structures seen in model polypeptides with, for example, the hydrogen bond geometry
less than ideal. The average length of α-helices in proteins is quite short,
about 12 amino acids or 3.5 turns.
The representation of the α-helix in Fig. 1.13 does not show the side chains;
these project out from the helix into the solvent. For some purposes it is useful to
depict the arrangement of side chains in a projection known as the helical wheel
(Fig. 1.14).
This might show, for example, that one face of an α-helix is polar and the other
non-polar, in which case the helix is said to be amphipathic; such a helix might be
involved in anchoring the protein to a membrane. If both sides of the helix were
1.4 THE SECONDARY STRUCTURE OF PROTEINS 21
Note that with 3.6 amino
acids per turn of the α-helix,
the angle between successive
α-carbon atoms in the
helical wheel plot is 360/
3.6 degrees, i.e. 100 degrees.
. .
Fig. 1.13 The structure
of the α-helix showing the
arrangement of hydrogen
bonds. (a) The path of the
polypeptide chain with the αcarbon
atoms and hydrogen
bonds labelled. The arrows
show the direction of the
chain (N→C). (b) The
hydrogen bonding
arrangement; there are
13 atoms in the ring
formed by the hydrogen
bond.
N
N
N
N
N
N
H
H
C
C
α
C
C
O
O
O
O
O
O
O
O
O
O
α
C
C
C
α
Cα
C
C
α
C
C
α
C
C
α
C
C
C
α
Cα
C
C
(a) (b)
α
C
C
α
H
H
N
N
N
N
H
H
H
N
H
H
H
0.54 nm
C
O
N
H
CN
C
n n + 2
n + 1 n + 3
OH
C
O
N
N
H
CN
n + 5
n + 4
OH H
O
Cα Cα Cα
Cα Cα Cα
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 21
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
22 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
non-polar, the helix would be very likely to be buried in the interior of a protein or
might be a membrane-spanning element of a protein. From the dimensions of
the α-helix, 6 turns would be 6 × 3.6 = 21.6 amino acids in length and would span
a distance of 6 × 0.54 nm = 3.24 nm, which is a typical value for the thickness of
a phospholipid bilayer.
Other types of helix include the 310 helix (three amino acids per turn, ten atoms
in the hydrogen bond ring), in which the hydrogen bond is between the carbonyl
group of the amide bond between amino acids n and n + 1, and the amino group of
the amide bond between amino acids n + 2 and n + 3. This more tightly folded helix
is sometimes found in short stretches in proteins, particularly near the ends of
helical segments or of polypeptide chains. Another possible type is the π-helix,
which is more loosely coiled than the α-helix, with the hydrogen bond between
the carbonyl group of the amide bond between amino acids n and n + 1, and the
amino group of the amide bond between amino acids n + 4 and n + 5.
1.4.2 The β-sheet
A β-sheet is made up from strands of polypeptide chain in an extended structure,
with the side chains of the amino acids projecting alternately above and below the
plane of the sheet. The strands can run parallel or anti-parallel to each other with
hydrogen bonds forming between them; the bonds are more distorted in the case of
parallel strands (Fig. 1.15). The values of the angles φ and ψ for idealized versions of
(a) parallel β-sheet are −119° and +113°, respectively, and (b) anti-parallel β-sheet
−139° and +135°, respectively. In both cases, the span of the sheet is about 0.33 nm
per amino acid, i.e. more than twice that of the α-helix. When the structures of
proteins are analysed, it is found that sheets composed of anti-parallel strands are
more common than those formed from parallel strands, with generally at least four
of the latter needed to form a sheet, compared with only two of the former. There
. .
The α-helix is not the only
secondary structure found in
membrane proteins; several
membrane-spanning
proteins are β-sheet in
character. A good example
is the family of porins found
in the outer membranes of
Gram-negative bacteria,
mitochondria, and
chloroplasts, which form
channels for small polar
molecules. Porins consist
of 16 strands of β-sheet
arranged in an anti-parallel
fashion to form a pore
through the membrane.
The π-helix has been found
very occasionally in proteins
(Weaver, 2000).
Electrostatic interactions
between the carbonyl groups
of adjacent chains contribute
significantly to the stability
of β-sheet structures. In
actual protein structures
(rather than model
compounds) there appears
to be little difference
between the dihedral angles
for the parallel and antiparallel
sheet patterns.
Fig. 1.14 The helical wheel
projection of helix structure.
The positions of 11 amino
acid residues are plotted at
100° intervals in a clockwise
direction looking down the
helix from the N-terminus. In
this sequence all the amino
acids except Ser261 and
Asn267 are non-polar. This
helix would be expected to be
located in the non-polar
interior of a protein.
Asn 267 Leu 260
Ala 263 Ala 264
Ala 270
Met 266
Phe 262
Leu 269
Ala 265
Ser 261
Gly 268
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 22
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
is a right-handed twist between the strands, which appears to be a result of the electrostatic
interactions between the carbonyl groups of adjacent peptide bonds in each
strand; these are much more favourable with a right-handed than a left-handed
twist. The average length of β-strands in proteins is about six amino acids.
1.4.3 Other structural features in proteins
In addition to α-helices and β-sheets, the most abundant structural element found
in proteins is the β-turn. A β-turn is a sequence of amino acids in which the
polypeptide chain folds back on itself by nearly 180°, allowing proteins to adopt
globular, rather than extended, shapes. There are at least eight different types of
1.4 THE SECONDARY STRUCTURE OF PROTEINS 23
About 15% of residues in
proteins not in α-helices or
β-sheets can be in β-turns
(Panasik et al., 2005).
. .
Fig. 1.15 Structure of
β-sheets showing the
arrangement of hydrogen
bonds. (a) Hydrogen
bonding in the parallel
β-sheet; (b) hydrogen
bonding in the anti-parallel
β-sheet. In three dimensions,
the sheet structures are in fact
corrugated with the amino
acid side chains projecting
perpendicularly to the plane
of the sheet.
C
O
N
NC
O H O H O
(a)
(b)
C
C N
N
H
C
C N
NC
O H
C
O
C N
N
H
C
O
O H HO
O H H
C N
NC
O H
C
O
C N
N
H
C
O
C N
N
H
C
O
O
HO
OH
CN
C N
HO
OH
CN
C N
H
0.7 nm
(i.e. 0.35 nm per amino acid)
O
Cα
Cα
Cα
Cα
Cα
Cα
C N
N
H
C
O
HO
OH
O H
CN
C N
HO
OH
CN
C N
HO
Cα
Cα
Cα
Cα
Cα
Cα
Cα
Cα
Cα
Cα
Cα
Cα
Cα
Cα
Cα
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 23
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
24 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
turn which differ in the number of amino acids in the loop, the types of hydrogen
bonds formed, and the values of the angles φ and ψ for the various amino acids.
The most frequently occurring β-turns consist of four amino acids (Fig. 1.16).
As previously noted, Pro cannot be accommodated in regular secondary structural
elements such as the α-helix and β-sheet. Pro residues are often found in βturns
or at the ends of helices or strands of sheets. Chains of Pro residues can also
adopt regular structures, thus poly-Pro can exist in two structures poly-Pro I (all
peptide bonds cis) and poly-Pro II (all peptide bonds trans). The angles φ and ψ for
poly-Pro I are −83° and +158°, respectively, and for poly-Pro II are −78° and +149°,
respectively. Poly-Pro I is a right-handed helix with 3.3 residues per turn and a
span of 0.19 nm per residue, whereas poly-Pro II is a highly extended left-handed
helix with three residues per turn and a span of 0.31 nm per residue.
1.4.4 Structural preferences of the different amino acids
The 20 amino acids have side chains which differ in terms of their bulkiness,
polarity, and charge properties. These will lead in turn to different preferences for
the various types of secondary structures in proteins. These preferences have been
quantified in two main ways, firstly from an analysis of the structures of actual proteins
and secondly from studies of the effect of incorporating different amino acids
into regular polymers (e.g. introducing a Val residue into poly-Ala, which has a
strong tendency to form an α-helix). From analysis of the structures of proteins,
tables of preferences for the amino acids have been compiled. Met, Glu, Leu, and
Ala have the highest tendency to be found in α-helices, whereas Pro, Gly, and Tyr
have the lowest. Val, Ile, and Phe have the highest tendencies to be found in βsheets;
Pro and Asp the lowest. Pro, Gly, and Asp have the highest tendencies to be
found in β-turns; Met, Val, and Ile the lowest. While detailed structural explanations
for these preferences are not always clear, such data are useful in formulating
‘rules’ for predicting secondary structural features from protein sequences. It
should be remembered that these rules are only guidelines; in general helices can
be predicted with more confidence than sheets or turns, but even so the accuracy is
relatively modest (of the order of 70%).
. .
An analysis of the crystal
structures of a number
of proteins of different
structural types showed
that a significant number
(approximately 5%) of the
amino acid residues could
be assigned to poly-Pro II
structures which were at
least three amino acids in
length (Sreerama and
Woody, 1994).
A commonly used tool for
predicting helix, strand,
and loop regions of a protein
from the amino acid
sequence is PHDsec
(PredictProtein@EMBLHeidelberg.de),
which is
stated to achieve 72%
accuracy when applied to
water-soluble, globular
proteins.
Fig. 1.16 Structure of a
typical β-turn. Turns allow
the formation of an antiparallel
sheet structure; in this
turn there are ten atoms in
the ring formed by the
hydrogen bond.
N
C
C
C
C4
O
O
O
H
H
H
N
N
α
C1
α
C3
α
C2
α
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 24
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
1.5 The tertiary structure of proteins
KEY CONCEPTS
■ Understanding the general principles governing the formation of the
tertiary structure of proteins
■ Describing how protein structures can be classified
The tertiary structure of a protein refers to its long-range folding so that parts of
the polypeptide chain remote in sequence are brought into close proximity.
1.5.1 General principles
Although proteins adopt a very wide of three-dimensional structures, a number of
general principles have emerged from the analysis of the many thousands of
known structures of proteins. These are summarized below.
■ Close packing Proteins generally adopt very closely packed globular structures
with only a small number of internal cavities, which are generally filled by water
molecules
■ Elements of secondary structure The structural elements (helices, strands, turns
etc.) are generally similar to those seen in small model compounds, but there can
be deviations. For example, there is a tendency for α-helices to be distorted
towards 310 helices, and for strands to adopt a twist in forming sheets
■ Distributions of side chains Non-polar side chains tend to be buried in the
interior of the protein away from the solvent, and polar side chains exposed to
the solvent. Polar side chains which are buried will often play key functional
roles. This tendency would, of course, be reversed for membrane-spanning
elements of a protein, where the non-polar side chains of a helix will project into
the non-polar interior of the membrane
■ Pairing of polar groups Polar groups which are internal are paired in hydrogen
bonds; this can include carbonyl and amino groups of the main chain and
pairing with internal water molecules
■ Formation of domains Larger protein molecules tend to form structural domains,
which are independent folded globular units, typically of the order of 100–
150 amino acids in size. These domains are usually associated with particular
functions, e.g. binding or catalysis, so that it is possible to assemble a multifunctional
protein in this modular fashion.
1.5.2 Classification of protein structures
A number of classification schemes have been devised to characterize the tertiary
structures of proteins. These generally place the structures into one of three main
categories, namely all-α, all-β, and αβ. (In some schemes, the αβ class is further
divided into α/β and α + β, depending whether the helix and sheet elements
1.5 THE TERTIARY STRUCTURE OF PROTEINS 25
In many cases it is possible
to prepare the individual
domains either by controlled
proteolysis of the multidomain
protein or by
expression of the appropriate
part of the gene coding for
that domain.
. .
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 25
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
26 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
The TIM barrel is
named after the enzyme
triosephosphate isomerase
in the glycolytic pathway,
the first protein structure
shown to possess the
repeating (αβ)8 pattern.
Fig. 1.17 Examples of
protein superfolds. (a), (b),
and (c) show examples of the
TIM barrel, α/β doubly
wound, and Greek key
structures respectively;
these are three of the most
commonly occurring
superfolds. α-Helices and βsheets
are shown as ribbons
and arrows respectively.
(a)
(c)
(b)
C
C
C
N
N N
alternate along the chain or are in separate regions of the sequence). One of the
most useful tools for the classification of protein is the SCOP (Structural Classification
Of Proteins) database, which classifies proteins at three levels: (a) protein
folds, where there are similar arrangements of secondary structural elements, but
no particularly close relationship in terms of function or overall structure, (b) protein
superfamilies, where there are probable relationships between the proteins, as
shown by similar functions, such as binding nucleotides, and (c) protein families,
where there is very strong evidence for close relationships with a high (>30%) level
of amino acid sequence identity between members.
The number of different protein folds has been estimated to be of the order
of 1000. Analysis shows that only nine fold families contain proteins with no
sequence or functional similarity to each other. These so-called superfolds account
for over 30% of all known structures; they include the TIM barrel, α/β doubly
wound, and Greek key structures (see Fig. 1.17).
1.5.3 Forces involved in stabilizing tertiary structures
The forces involved in stabilizing tertiary structures of proteins are (apart from
any disulphide bonds which may be formed in extracellular proteins) of the weak
non-covalent type and are discussed in section 1.7. The same types of forces are
involved in the interaction of proteins with each other or with other ligands.
. .
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 26
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
1.6 THE QUATERNARY STRUCTURE OF PROTEINS 27
A protein consisting only of
multiple copies of the same
polypeptide chain is termed
a ‘homooligomer’, e.g.
lactate dehydrogenase is a
homotetramer (denoted α4).
If there are different types
of polypeptide chain present
the protein is termed a
heterooligomer, e.g. the
G-proteins involved in
signal transduction are
heterotrimers (denoted
αβγ). Haemoglobin is a
heterotetramer (α2β2),
where the α. and β. chains
are very similar to each
other. The Greek letters are
used in a generic sense to
distinguish different types
of polypeptide chains in
a protein, rather than to
denote specific sequences
of amino acids.
The proteasome is a large
complex (1.4 MDa) that
contains a barrel-shaped
proteolytic unit (700 kDa)
consisting of 14 copies of
each of two types of subunit.
The proteasome degrades
proteins which have been
tagged with the small protein
ubiquitin for destruction.
Chaperones are proteins
which assist the folding
of proteins in the cell by
preventing unwanted
interactions between
hydrophobic surfaces, which
would otherwise lead to
the formation of protein
aggregates. In the cpn60
class of chaperones, which
contains 14 copies of a
57 kDa subunit, there is a
large internal cavity that
allows the correct folding of
polypeptide chains to occur.
1.6 The quaternary structure of proteins
KEY CONCEPT
■ Understanding the significance of association of polypeptide chains
(subunits) in terms of the function of proteins
As noted in section 1.5, there is a tendency in polypeptide chains of more than
100–150 amino acids to form domains. For larger proteins, there is a tendency
to exist as multiple subunits (polypeptide chains); this generates the quaternary
structure of proteins. Proteins of molecular mass <30 kDa tend to be monomers
(single subunits) and those of molecular mass >50 kDa tend to be oligomers
(several subunits), although these figures should be viewed as general guidelines only.
In general the arrangement of the subunits maximizes the number of inter-subunit
contacts, for example in a tetrameric protein (four subunits) the preferred geometrical
arrangement will be tetrahedral rather than square planar or linear. The
subunits in a multi-subunit protein are generally held together by the weak noncovalent
forces of the type described in section 1.7, although it should be noted that
in a few cases, such as immunoglobins (Chapter 7, section 7.3.6) and bovine seminal
ribonuclease, a disulphide bond is involved in holding the subunits together.
The association of several subunits in a protein can offer possibilities of novel
properties to a protein. These include (a) regulation of activity, by communication
between the binding sites or active sites on different subunits (e.g. in the case of O2
binding to haemoglobin), (b) increased stability as a result of favourable packing of
subunits which might be prone to unfolding as monomers, and (c) generation
of novel structures such as found in chaperone proteins and proteasomes, where
internal ‘cavities’ can be created to allow protein folding or protein degradation,
respectively.
1.7 Forces contributing to the structures and
interactions of proteins
KEY CONCEPTS
■ Understanding the contribution of weak forces to the structures and
interactions of proteins
■ Describing the basis of the principal types of weak forces (ionic
interactions, hydrogen bonds, van der Waals interactions, and
hydrophobic interactions)
■ Knowing the range of energies involved in protein interactions
The amino acids in a protein are linked together by strong covalent peptide bonds
(section 1.3.1). In some cases the three-dimensional structures will also be stabilized
by covalent bonds, for example disulphide bonds, which can form between
pairs of Cys side chains, and the pairs of Lys side chains, which (after oxidative
. .
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 27
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
28 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
deamination) can form strong cross-links in collagen. These covalent bonds have
energies of the order of several hundred kJ mol−1
. However, the forces which stabilize
the folded structures of the vast majority of proteins, and which underpin the
interactions between proteins and other molecules, are of a non-covalent type and
hence much weaker and usually short lasting. These forces can be described under
the following categories.
1.7.1 Ionic (electrostatic) interactions
These refer to the electrostatic forces between oppositely charged groups, e.g. between
negatively charged side chains (Asp, Glu) and positively charged side chains
(Arg, Lys, and (under certain circumstances) His). The energy of interaction (E)
between charges of magnitude q1 and q2, separated by a distance r is given by
E = q1q2/Dr, where D is the dielectric constant. By definition, D = 1 for a vacuum;
the value of D for water = 78.5. Non-polar solvents have low values of D (e.g.
benzene, D = 2.3; ethanol, D = 25). Since the majority of charged amino acid side
chains in a protein are on the surface and therefore significantly exposed to the
aqueous medium, ionic interactions are generally weak (a high value of D means
a low value for the energy), typically of the order of 5 kJ mol−1
. However, if ionic
interactions occur in the much less polar interior of a protein (as in the case of the
α-amino group of Ile 16 and the side chain of Asp 194 in chymotrypsin), the forces
will be considerably stronger (possibly up to about 20 kJ mol−1
).
1.7.2 Hydrogen bonds
A hydrogen bond is essentially a form of electrostatic interaction involving partial
charges. It occurs when an H attached to an electronegative atom (almost always
O or N in biological systems) interacts with a second electronegative atom. For
example, if we call these atoms A (donor) and B (acceptor), we would have the
arrangement shown in Fig. 1.18, where the δ indicates a partial charge. There is an
attractive force between the partial charges on the H and on the acceptor atom.
This is a weak interaction (typically of the order of 5–10 kJ mol−1
), but when a
number of hydrogen bonds are involved they can contribute significantly to local
stability, as is the case in secondary structural features such as the α-helix or βsheet
(section 1.4). The particular importance of hydrogen bonds in terms of the
structure and function of proteins is that they impose geometrical constraints, and
thus specificity, on interactions. For maximum stability of the hydrogen bond, the
three nuclei (A, H, and B) should be in a straight line and the separation between
the nuclei A and B should be within fairly narrow limits (0.30 ± 0.05 nm depending
on the atoms concerned). (This is how hydrogen bonds are identified in X-ray
crystal protein structures, since the H atom cannot be ‘seen’ by X-rays as it has so
little electron density compared with the heavier atoms C, N, and O). In proteins,
. .
The effective dielectric
constant (D) in the interior
of proteins has been
estimated from the effects of
mutations on the pKa values
of amino acid side chains. In
the case of proteins such as
cytochrome c and subtilisin,
it has been estimated that D
is in the range 30 to 60, but
in staphylococcal nuclease,
D is estimated to be 12. The
value of D appears to depend
on the extent of water
penetration into the interior
of the protein. In a truly
non-polar region in the
protein interior, D may be as
low as about 4 (Dwyer et al.,
2000).
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 28
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
hydrogen bonds can involve interactions between particular side chains (see
section 1.2.4) or main chain atoms (the N–H or C=O groups of the peptide bond).
It should also be remembered there is often a compensating mechanism whereby
groups on the protein may equally well form hydrogen bonds with water, rather
than with other parts of the protein, thereby reducing the energetic contribution of
the interaction to the structural or binding properties of the protein.
1.7.3 van der Waals’ interactions
This term is applied collectively to various weak forces involving interactions
between dipoles (separations of charge) in molecules or parts of molecules. In
addition to permanent dipoles reflecting differences in electronegativity (e.g.
between C and O in the C=O bond), there are transient dipoles which arise from the
fact that electrons occupy fluctuating positions in space. These dipoles can in turn
induce dipoles in other molecules or groups. The van der Waals’ forces include
dipolejdipole, dipolejinduced dipole, and induced dipolejinduced dipole
interactions. Although the forces are weak individually (of the order of 5 kJ mol−1
),
there are usually many such interactions and thus collectively they can contribute to
the stability of a protein structure or the strength of a protein–ligand interaction.
1.7.4 Hydrophobic interactions
Water is a uniquely poor solvent for non-polar molecules or groups such as the
aliphatic or aromatic side chains in proteins. As mentioned in section 1.2.5, liquid
water consists of fluctuating clusters of hydrogen-bonded water molecules. When
a non-polar group is introduced into this system, the water molecules tend to form
more ordered hydrogen bonded ‘cage’ structures around the non-polar group.
This process has a favourable (negative) change in enthalpy, but there is a highly
unfavourable (negative) change in entropy due to the ordering of the solvent,
thus the overall free energy change is unfavourable (positive) (see Chapter 4,
section 4.1). The term ‘hydrophobic interactions’ is used to describe the tendency
of non-polar molecules to be excluded from water and associate with each other,
although it should be noted that such interactions are in fact very weak, being
largely due to forces of the van der Waals type. For this reason, some workers prefer
the term ‘hydrophobic effect’.
1.7 FORCES CONTRIBUTING TO THE STRUCTURES AND INTERACTIONS OF PROTEINS 29
The van der Waals’ forces
between atoms or molecules
represent a balance between
the longer range attractive
forces (which show a 1/r6
dependence, where r is a
measure of internuclear
distance) and the shorter
range repulsive forces
(which show a 1/r12
dependence).
The strengths of individual
hydrogen bonds vary
considerably; they are
stronger if a charged group
is involved, e.g. if –NH+
3 acts
as a donor.
It should be pointed out that
the term ‘hydrophobic’ is
something of a misnomer
since it implies that the nonpolar
groups ‘hate water’,
whereas in fact it is the water
which hates them!
. .
A H B
δ− δ+ δ−
Fig. 1.18 Formation of a hydrogen bond. Atoms A and B represent electronegative
atoms (generally O and/or N in biological systems). The partial charges on the atoms
are shown.
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 29
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
30 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
The hydrophobic interactions make the most important contribution in energy
termstothestabilityofthefoldedstructureofaproteininwater.Sincethereisaposi-
tivechangeinenthalpyassociatedwiththeexclusionofnon-polargroupsfromwater,
hydrophobic interactions become weaker at lower temperatures (at least over a restricted
range); this can account for the cold-induced denaturation of some proteins.
The strength of hydrophobic interactions can be altered by various agents.
For example, urea, and guanidinium chloride (GdmCl) decrease the hydrophobic
interactions; since they can participate in hydrogen bonds, they presumably disrupt
the hydrogen-bonded network of water molecules. By contrast, the hydrophobic
interactions can be strengthened by the addition of various salts; the order
of effectiveness follows the Hofmeister series:
Cations Mg2+
> Li+
> Na+
> NH4
+
Anions SO4
2−
> HPO4
2−
> acetate > Cl−
> ClO4
−
> SCN−
1.7.5 Balance of energy contributions
Many of the forces involved in stabilizing the three-dimensional structures of proteins
and the interactions between proteins and other molecules are relatively weak
and represent a balance between competing processes. Thus, for example, hydrophobic
interactions represent a balance between an unfavourable enthalpy term
but a favourable entropy term, and an α-helix is stabilized by hydrogen bonding
between the main chain –N–H and C=O groups, but these are formed at the
expense of hydrogen bonds that might be formed to the solvent water if the protein
were unfolded. The stability of the folded structure of a protein represents a very
delicate balance between two opposing tendencies, namely the large (unfavourable)
negative change in entropy when a compact globular structure is formed
from a disordered polypeptide chain and the large (favourable) negative change in
enthalpy due to the favourable interactions which occur within the folded state of
the protein. For most proteins the (thermodynamic) stability of the folded state is
very small, in the range 20–60 kJ mol−1
(see Chapter 8, section 8.6), compared with
the entropy and enthalpy terms, each of which is of the order of several hundred
kJ mol−1
. Of course, there may be a large kinetic barrier (activation energy) to
unfolding of the folded state of a protein, but it is important to realize that,
in thermodynamic terms at least, life hangs by a slender thread!
The balance of the forces can be altered by addition of certain agents. As
indicated in section 1.7.4, urea and GdmCl weaken hydrophobic interactions. This
is shown by the fact that different amino acid side chains are much more nearly
equally soluble in 8 M urea or 6 M GdmCl than would be the case in water. Since
the hydrophobic interactions make the largest contribution in energy terms to the
stability of the folded states of proteins, it is easy to understand how these agents
promote the unfolding of proteins.
Proteins can also be unfolded by exposure to extremes of pH (reflecting changes
in the ionization states of amino acid side chains and hence disruption of
The Hofmeister series was
derived in the 1880s from
studies of the effectiveness
of different salts in
precipitating proteins from
solution.
In concentrated solutions of
urea and GdmCl, proteins
unfold because the entropy
gain in unfolding is
dominant.
. .
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 30
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
1.7 FORCES CONTRIBUTING TO THE STRUCTURES AND INTERACTIONS OF PROTEINS 31
The dissociation constant Kd
for an interaction represents
the ratio of the rate constants
koff/kon (see Chapter 4,
section 4.1). Since there
is an upper limit for the
association rate constant kon
(set by the rate of diffusion)
of about 109
M−1
s−1
, an
estimate can be made of the
dissociation rate constant,
koff, and hence the life-time
of the complex.
Table 1.4 Typical dissociation constants for some interactions involving proteins
Interaction Typical dissociation constant (Kd) (M) ΔΔG0
310 (kJ mol−−1
)*
Avidin–biotin 10−15
89
Protein–protein 10−10
59
Antibody–antigen 10−9
53
Receptor–hormone 10−7
42
Enzyme–substrate 10−5
30
*The free energy values are positive as they refer to the dissociation of the complex (PL j P + L),
which would be unfavourable under standard state conditions (1 M concentrations of PL, P, and L).
Dissociation would become progressively more favourable as the concentration of PL is lowered. In the
cell the concentrations of many complexes are likely to be in the micromolar range.
electrostatic forces and hydrogen bonds) or temperature (reflecting alterations
in the balance between polar and non-polar interactions); see also Chapter 5, section
5.6.1.
1.7.6 The range of energies involved in protein interactions
The function of the vast majority of proteins is to undertake specific interactions
with other molecules (ligands). It is instructive to examine the strengths of these
interactions in the context of the discussion of the forces involved. Some representative
types of interactions are listed in Table 1.4, together with a note of the
corresponding standard free energy change at 37°C (ΔG°310), derived using the
relationship −ΔG° = RT ln Keq (eqn. 4.4; see Chapter 4, section 4.1).
The magnitudes of the free energy changes are such that only a relatively small
number of the weak interactions (hydrogen bonds, van der Waals’ forces, etc.) are
required to account for them. It should be remembered, however, that many
interactions involve a balance between forces, as described in section 1.7.5, thus
a protein and a ligand may well each form favourable interactions with the
solvent that are lost when the protein and ligand form a complex. The net free
energy change can therefore be relatively small.
The values of the dissociation constants can be correlated with the functions of
the complexes. For example, the extremely tight interaction between the vitamin
biotin (which is produced by bacteria) and the egg white protein avidin may play
a role in anti-bacterial defence for the developing chick embryo. The interactions
involved in forming oligomeric proteins have to be very favourable in order to
generate the stability required for a protein to survive and function under cellular
conditions. Antibodies and receptors must bind their respective partners sufficiently
tightly to allow subsequent events such as complement activation or signal
transduction to occur. On the other hand, it is important that enzyme–substrate
(and enzyme–product) interactions are not too strong since otherwise the catalytic
cycle (consisting of binding of substrate(s), structural changes in the enzyme, conversion
of substrate(s) to product(s), and release of product(s)) would be slowed
and the enzyme would become an unacceptably inefficient catalyst.
. .
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 31
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
32 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
1.8 Compendium of chemical structures
It is a good idea to try to learn the structures of a number of important molecules
that are connected with the study of proteins; a list is given below. Once you know
the structure of a compound, it will help you to understand its properties. For
example, you should be able to recognize the following features of a molecule from
its structure:
■ Which parts are polar and which are non-polar
■ Which parts may be involved in weak interactions such as hydrogen bonds,
ionic bonds or hydrophobic interactions
■ Which parts may be involved in ionization processes and over which pH range
different charged forms may predominate
■ Which parts may have important chemical roles, such as acting as a
nucleophile or coordinating a metal ion.
When looking at these structures, bear in mind the following:
■ Chains of carbon atoms are often represented as zig-zag lines, rather than each
individual carbon atom being displayed
■ In complex structures, hydrogen atoms are often omitted for the sake of clarity
■ In ring structures the ring atoms are assumed to be C unless otherwise
indicated (e.g. N, O, S, etc.)
■ The actual charged states of compounds will depend on the prevailing pH
and the pKas of ionizing groups such as –NH2 and –CO2H (see Chapter 3,
section 3.7). They may not be in exactly the forms shown below.
Amino acids
. .
CH3
Alanine
(CH2)3
NH
C NH2
NH2
Arginine
CH2
C
O NH2
Asparagine Aspartic acid
CH2
C
O O
CH2
SH
Cysteine
C
H
C
R
H2N
O
OH
The general structure of an amino acid
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 32
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
1.8 COMPENDIUM OF CHEMICAL STRUCTURES 33
. .
CH2
C
O O
CH2
Glutamic acid
−
CH2
C
H2N O
CH2
Glutamine
H
Glycine
HN
NH
Histidine
CH2
Isoleucine
CHH3C
CH2
CH3
Leucine
CH
CH2
H3C CH3
Lysine
CH2
CH2
CH2
CH2
NH3
Methionine
CH2
CH2
S
CH3
Phenylalanine
CH2
H
Proline
CH2H2C
H2N COO−
H2
C
Serine
CH2
OH
Threonine
CHOH
CH3
N
H
Tryptophan
CH2 CH2
OH
Tyrosine
CH
H3C CH3
Valine
H2N C
H
R
C
O
N
H
C
H
R
C
O
OH
The peptide bond
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 33
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
34 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
Bases, nucleotides, and related compounds
. .
N
N
NH2
N
N
H
Adenine Cytosine
O
N
NH2
N
H
H3C
N
O
H
Thymine
ON
H
N
O
H
Uracil
ON
H
O
OH
H
H
OH
H
OH
HOH2C
H
Ribose
HOH2C
O
H
H
H
H
OH
H
2-deoxyribose
OH NH2
N
N N
N
O
OHOH
AMP
Adenosine monophosphate
CH2OPOOPO
OO
O O O
P
O
ADP
Adenosine diphosphate
ATP
Adenosine triphosphate
Cyclic AMP
N
N
NH2
N
N
H
O
H
OH
H
OO
CH2O
H
P
O
O
Base
Phosphodiester
bond
The 5′–3′ bond between nucleotides
O CH2OP
O
O
O
O
Base
O CH2OP
O
OH
NH2N
N
O
N
H
N
H
Guanine
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 34
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
Glycolytic intermediates
1.8 COMPENDIUM OF CHEMICAL STRUCTURES 35
. .
C
C
C
C
C
H O
H OH
HHO
OHH
OHH
CH2 OH
Glucose
O
HOH2C
HO
OH
OH
OH
Glucose-6-phosphate
C
C
C
C
C
H O
H OH
HHO
OHH
OHH
CH2 OPO3
2
2
O3POH2C
O
HO
OH
OH
OH
Fructose-6-phosphate
O
OH
HO
CH2OH
CH2 OH
C
C
C
C
O
HHO
OHH
OHH
CH2OPO3
2
2−
O3OH2C
OH
Fructose-1,6-bisphosphate
O
OH
HO
CH2OPO3
2
CH2OPO3
2
C
C
C
C
O
HHO
OHH
OHH
CH2OPO3
2
2
O3OH2C
OH
Dihydroxyacetone phosphate
CH2OPO3
2
C
CH2OH
O
Glyceraldehyde-3-phosphate
C
C
CH2OPO3
2
H O
H OH
3-phosphoglycerate
C
C
CH2OPO3
2
O O
H OH
2-phosphoglycerate
C
C
CH2OH
O O
H OPO3
2
Phosphoenolpyruvate
C
C
CH2
O O
OPO3
2
Pyruvate
C
C
CH3
O O
O
Lactate
C
C
CH3
O O
H OH
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 35
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
36 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
. .
Reagents
O OH
N
N
HO
O
O
OH
OHO
EDTA
HC C
OH2C
NH2
Acrylamide
C
H2
H2
C
HS CH
CH SH
OH
OH
Dithiothreitol
NH2
NH2
Cl−
H2N
Guanidinium chloride
HS
C
H2
H2
C
OH
2-mercaptoethanol
CH2H2C
O
N N
H H
Methylene bis-acrylamide
O
O O Na
O O
SDS
S
NH2
HO
HO
HO
Tris
NH2H2N
C
O
Urea
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 36
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
1.8 COMPENDIUM OF CHEMICAL STRUCTURES 37
. .
Redox cofactors
FADH2
H
N
O
O
OH
HO
N N
H
NH
OH
O
P
O−
O O
P
O−
O O
O
HO OH
N
N
N
N
NH2
FAD
N
O
O
OH
HO
N N
NH
OH
O
P
O
O O
P
O
O O
O
HO OH
N
N
N
N
NH2
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 37
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
38 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
. .
CONH2
HH
NAD(P)H
This phosphate group is absent from NADH
but present in NADPH
O
O
OH
C
H
H
P
O
O O
P
O
O O
OH
O Adenine
H
OH
C
H
H
OPO3
2
N
NAD(P)
This phosphate group is absent from NAD+
but present in NADP+
O
O
OH
C
H
H
P
O
CONH2
−
O O
P
O
−
O O
OH
O Adenine
H
OH
C
H
H
OPO3
2−
N
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 38
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
1.8 COMPENDIUM OF CHEMICAL STRUCTURES 39
. .
Citrate
CHO COO
H2C COO
H2C COO
Isocitrate
CH COO
H2C
OH
COO
HC COO
2-oxoglutarate
CH2
H2C
O
COO
C COO
Succinate
H2C COO
H2C COO
C
C
H
H COO
OOC
Fumarate
Malate
HC
OH
H2C COO
COO
Oxaloacetate
O
C COO
H2C COO
TCA cycle intermediates
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 39
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
. .
40 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
1.9 Problems
Full solutions to odd-numbered problems are available to all in the student section of
the Online Resource Centre at www.oxfordtextbooks.co.uk/orc/price/. Full solutions
to even-numbered problems are available to lecturers only in the lecturer section of
the Online Resource Centre.
1.1 The amino acid ornithine (which is an intermediate in the synthesis of arginine)
has the side chain –(CH2)3–NH2. What would you expect the principal chemical
characteristics of this amino acid to be?
1.2 The synthetic amino acid norleucine has the side chain –(CH2)3–CH3. How
would you expect incorporation of this amino acid in place of leucine to affect
the properties of a protein?
1.3 The sequence of the tripeptide glutathione is usually depicted as
(γ)Glu–Cys–Gly. Draw the full chemical structure of glutathione.
What might happen to glutathione under oxidizing conditions?
1.4 Calmodulin is a member of a family of small (17 kDa) proteins that can bind Ca2+
ions tightly. What types of amino acids might allow such proteins to achieve this?
1.5 Compared with Ca2+
ions, Zn2+
ions have a preference to coordinate with
nitrogen and sulphur as compared with oxygen atoms. Which amino acids
might you expect to find in Zn-binding sites in proteins?
1.6 For aspartic acid, the pKa values for the α-carboxyl, the side chain carboxyl and
the α-amino group are 2.0, 3.9, and 9.9, respectively. Draw the structures of the
predominant charged forms of the amino acid at pH values 1, 3, 6, and 11.
Explain why the pI is obtained by averaging the pKa values of the α-carboxyl
and the side chain carboxyl groups.
1.7 For lysine, the pKa values of the α-carboxyl, the α-amino, and the side chain
amino groups are 2.2, 9.1, and 10.5, respectively. What is the pI for lysine?
1.8 Iodoacetamide (I–CH2–CO–NH2) is a commonly used reagent to react with
nucleophilic side chains in proteins. Using structural formulae, show how it
reacts with the side chains of cysteine and lysine. How might you monitor the
extent of modification of a protein by the reagent?
1.9 The pKa of the side chain amino group of lysine is 10.5 in the free amino acid.
Explain how the pKa of a lysine side chain in a protein could be affected by the
presence of (a) a neighbouring Arg side chain and (b) a neighbouring Asp side
chain.
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 40
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
. .
PROBLEMS 41
1.10 The following sequences of 20 amino acids occur in the protein
glycophorin, which is found in the membrane of red blood cells:
(a) YPPEEETGERVQLAHHFSEP and (b) EITLIIFGVMAGVIGTILLI. What
would you predict about where these parts of the protein would be located?
1.11 The free energy of the trans form of a peptide bond is estimated to be 20 kJ mol−1
lower than that of the cis form. If a peptide bond could freely interconvert
between trans and cis forms, what would be the equilibrium constant for the
trans j cis equilibrium at 37°C? (The gas constant R = 8.31 J K−1
mol−1
).
1.12 In a study of protein structures, it is found that approximately 5% of the
peptide bonds to proline (Xaa–Pro) occur in the cis form. What is the free
energy difference at 37°C between the trans and cis forms in the case of the
Xaa–Pro bond?
1.13 The molar absorption coefficients at 280 nm for small model compounds of Trp
and Tyr are 5690 and 1280 M−1
cm−1
respectively. The polypeptide chain of yeast
alcohol dehydrogenase has a molecular mass of 36712 Da and contains 5 Trp and
14 Tyr. Calculate the molar absorption coefficient of alcohol dehydrogenase at
280 nm. (You can assume that the values for the model compounds apply to
these amino acids in the protein). What would be the absorbance at 280 nm
of a 0.32 mg mL−1
solution of the protein in a cuvette of 1-cm pathlength?
1.14 The polypeptide chain of the chaperone protein GroEL from Escherichia coli
has a molecular mass of 57200 Da and does not contain tryptophan. The A280
of a 1 mg mL−1
solution of the purified protein in a 1-cm pathlength cuvette is
0.160. Assuming that the molar absorption coefficient at 280 nm for a small
model compound of Tyr is 1280 M−1
cm−1
, calculate the number of Tyr in the
polypeptide chain.
1.15 The following amino acid sequences are found in α-helical structures
in three folded proteins. Use the helical wheel projection to predict where
each of these helical segments may be located in the structure of the protein:
(a) LSFAAAMIGLA (citrate synthase), (b) INEGFDLLRSG (alcohol
dehydrogenase), and (c) KEDNKGKSEEE (troponin C).
1.16 The calculated molecular mass of the polypeptide chain of type II
dehydroquinase from Streptomyces coelicolor is 16550.6 Da. What is
the expected mass of a mutant form of the enzyme in which Tyr 28 has
been replaced by Phe? Compare your answer with the observed mass
(16535.1 ± 1.4 Da). In a preparation of a second mutant in which Ser 108 is
replaced by Ala, there is a significant amount of material (30% of the total) with
a mass some 227 Da lower than the expected value. How could this lower
molecular mass material have arisen?
1.17 Draw chemical structures to show the hydrogen bonding which could arise
in each case between a Tyr side chain and (a) a second Tyr side chain, (b) the
carbonyl group of a peptide bond, and (c) the α-amino group of an N-terminal
amino acid.
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 41
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
. .
42 1. CONCEPTUAL TOOLKIT: THE MOLECULAR PRINCIPLES FOR UNDERSTANDING PROTEINS
1.18 From the X-ray structure of the complex between two proteins (X and Y), the
following interactions have been identified: (a) Arg 45 of X with Glu 13 of Y,
(b) Ile 53 of X with both Val 84 and Phe 85 of Y. Explain the probable basis of
these interactions.
1.19 Explain how the strength of ionic interactions and hydrophobic interactions
would be affected by changes in the ionic strength of the solution. Use your
answers to explain how changes in ionic strength can be used to elute adsorbed
proteins in ion-exchange chromatography and hydrophobic interaction
chromatography.
1.20 The folded form (F) of ribonuclease T1 is estimated to be 31.4 kJ mol−1
more
stable than the unfolded form (U) at 25°C. Calculate the equilibrium constant
for the F j U equilibrium. In the presence of 5.6 M urea, 58.4% of the protein
is unfolded; what is the free energy difference between the two forms under
these conditions?
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 42
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods
. .
REFERENCES FOR CHAPTER 1 43
References for Chapter 1
Berg, J.M., Tymoczko, J.L., and Stryer, L. (2007) Biochemistry, 6th edn. Freeman, New York,
1026 pp.
Crowe, J., Bradshaw, A., and Monk, P. (2006) Chemistry for the Biosciences: The Essential
Concepts. Oxford University Press, Oxford, 496 pp.
Dwyer, J.J., Gittis, A.G., Karp, D.A., Lattman, E.E., Spencer, D.S., Stites, W.E., and GarciaMoreno,
E.B. (2000) Biophys. J. 79, 1610–20.
Engelman, D.M., Steitz, T.A., and Goldman, A. (1986) Ann. Rev. Biophys. Biophys. Chem.
15, 321–53.
Jabs, A., Weiss, M.S., and Hilgenfeld, R. (1999) J. Mol. Biol. 286, 291–304.
Jones, A. (2005) Chemistry: An Introduction for Medical and Health Sciences. John Wiley
and Sons, Chichester, 260 pp.
Mathews, C.K., van Holde, K.E., and Ahern, K.G. (2000) Biochemistry, 3rd edn. Benjamin/
Cummings, San Francisco, California, 1186 pp.
Panasik, N., Jr., Fleming, P.J., and Rose, G.D. (2005) Protein Sci. 14, 2910–14.
Price, N.C., Dwek, R.A., Ratcliffe, R.G., and Wormald, M.R. (2001) Principles and Problems
in Physical Chemistry for Biochemists, 3rd edn. Oxford University Press, Oxford, 401 pp.
Sreerama, N. and Woody, R.W. (1994) Biochemistry 33, 10022–25.
Voet, D., Voet, J.G., and Pratt, C.W. (2006) Fundamentals of Biochemistry, 2nd edn. John
Wiley and Sons, Hoboken, New Jersey, 1130 pp.
Weaver, T.M. (2000) Protein Sci. 9, 201–6.
9780199205707_001_043_CH01.qxd 14/01/2009 18:28 Page 43
© Oxdord University Press 2009. Nicholas C. Price and Jacqueline Nairn: Exploring proteins: a student's guide to experimental skills and methods