FTIR Analysis of Protein Structure

Warren Gallagher

A. Introduction to protein structure

The first structures of proteins at an atomic resolution were determined in the late
1950’s.1 From that time to the early 1990’s approximately 300 protein structures were
added to the list, primarily using X-ray crystallography. Currently over 20,000 structures
are solved,2 some of these using the newer technique of multidimensional nuclear
magnetic resonance spectroscopy (NMR). While X-ray crystallography and NMR
spectroscopy provide the greatest level of detail about a protein’s structure, there are
many situations where these techniques cannot be applied. Other techniques, such as
FTIR, may not provide the same level of structural detail, but can be applied readily to
further our understanding of how proteins function

Before describing the structural information that FTIR spectroscopy provides, a

brief introduction to protein structure is given. Proteins are linear biological polymers for
which the monomeric units are amino acids. (Figure 1). Twenty different amino acids
are used to make proteins, each distinguished by the identity of the “R” group. The
amino acids are linked to form a polymer by linking the amino group on one amino acid
with the carboxylic acid group on another amino acid to form an amide bond (Figure 2).

amino group carboxylic acid

H O group

H N C C OH

H R alpha carbon

"R" group

Figure 1: The structure of an amino acid; the basic building block for making proteins.

Protein chemists call the amide bond a peptide bond. When two amino acids are linked
together by an amide bond it is called a dipeptide; and when many amino acids are linked
together in this fashion it is called a polypeptide. A polypeptide consists of a backbone
and sidechains. The backbone comprises the amide nitrogen, the alpha carbon and the
carbonyl carbon that are contributed by each amino acid unit. The sidechains comprise
the “R” groups, and dangle from the backbone like charms from a bracelet.

1 Kendrew et al. (1958) Nature 18, 662-666.

2 Protein Data Bank, Brookhaven National Laboratory, “http://ftp.pdb.bnl.gov”, May 16, 1997

1
 amide bond

H O H O H O H O

H N C C OH + H N C C OH H N C C N C C OH + HOH
H CH3 H CH2 H CH3 H CH2
OH OH

alanine serine a dipeptide water

Figure 2: Linking two amino acids together by an amide bond to form a dipeptide.

Proteins are distinguished from one another by the number of amino acids units
they contain and by the identity and sequential order of the amino acids. The numbers
vary from 50 to hundreds of amino acid units, making the number of possible sequences
astronomical. Both the number of amino acids in a protein and their sequence are
genetically determined. Few polypeptides are considered proteins. For a polypeptide to
be considered a protein it must be able to fold into a well-defined 3-dimensional
structure. This is usually a requirement for protein function.

When a protein folds to form a well defined 3-dimensional structure it exhibits three
levels of structure: primary, secondary and tertiary. The genetically determined
sequence of amino acids is the primary structure. The primary structure is often modeled
as beads on a string, where each bead represents one amino acid unit. The tertiary
structure might be modeled as a tightly-packed snowball, where each atom in the protein
has a well defined location. Protein folding might be likened to scrunching up the string
of beads (the primary structure) into a tightly-packed ball (the tertiary structure) (Figure
3a). A problem that proteins have in forming a tertiary structure is that many of the
amino acids find themselves on the inside of the tightly-packed ball and no longer in
direct contact with water. However, amides like to be exposed to water, because they can
form “hydrogen bonds” with water (Figure 3b). This problem is solved by providing
alternative ways for the buried amides to hydrogen bond. Primary among these is to have
the amide bonds hydrogen bond to each another (Figure 4).

H O H O
hydrogen bond H
with water H
H O H O

N C C N C C
H CH3 H CH2
H O H O OH
H H

a. b.

Figure 3: a. The crudest of protein folding models: scrunching up a string of beads. b. The hydrogen
bonds that form between the amide bond and water.

2
 H O H O

N C C N C C N
hydrogen bonds H R H R H
between amides
O H O H O

C N C C N C C

H R H R

Figure 4: The hydrogen bonds that form between amide bonds buried inside a folded protein.

This leads to an intermediate level of protein structure called secondary structure.

The types of secondary structure includes the αhelices and βsheets, which allow the
amides to hydrogen bond very efficiently with one another. Both are periodic structures.
In an α-helix the polypeptide backbone is coiled in a right-handed helix where the
hydrogen bonding occurs between successive turns of the helix. Figure 5(left) shows a
segment of αhelix from the small protein bovine pancreatic trypsin inhibitor (BPTI). In
βsheets, the strands of polypeptide are stretched out and lie either parallel or antiparallel
to one another. The hydrogen bonds form between the strands. Figure 5(right)
illustrates this with a piece of antiparallel β-sheet from BPTI. The other elements of
secondary structure include βturns and unordered structure. βturns are sharp turns that
connect the adjacent strands in an antiparallel βsheet. Unordered strucature is generally a
catch-all for regions that do not fall into one of the other categories. These are often
loops which form near the surface of proteins and join the other elements of secondary
structure.

Tertiary structure arises when various elements of secondary structure pack tightly
together to form the well-defined 3dimensional structure. The tertiary structure of a
protein is defined by the spacial coordinates of all its atoms. held together by favorable
interactions between the sidechains. These interactions are weak, which is why proteins
are easily “denatured” by heat and exposure to certain chemicals. The sidechains buried
inside a folded protein are packed tightly together and must interact favorably in order to
remain folded. Few

3
Figure 5: Segments of an α-helix (left), and a β-sheet (right) taken from BPTI. The hydrogens and
sidechains have been omitted for clarity. Carbons are black, oxygens are red and nitrogens are blue.
The hydrogen bonds between the amide oxygens and nitrogens are shown as dotted lines.

of the possible polypeptide sequences meet these stringent requirements for folding into
the well-defined 3-dimensional structures. Figure 6 shows both a spacefilling and a
ribbon model for BPTI. In the space-filling model the atoms are represented by spheres
scaled to their true relative sizes, illustrating the tightly packed folded protein. The
ribbon model shows the polypeptide backbone tracing its way through the folded
structure. It highlights the various types of secondary structure.

B. Obtaining structural information about proteins using FTIR spectroscopy

FTIR spectroscopy provides information about the secondary structure content of

proteins, unlike X-ray crystallography and NMR spectroscopy which provide information
about the tertiary structure. FTIR spectroscopy works by shining infrared radiation on a
sample and seeing which wavelengths of radiation in the infrared region of the spectrum
are absorbed by the sample. Each compound has a characteristic set of absorption bands
in its infrared spectrum. Characteristic bands found in the infrared spectra of proteins
and polypeptides include the Amide I and Amide II. These arise from the amide bonds
that link the amino acids. The absorption associated with the Amide I band leads to
stretching vibrations of the C=O bond of the amide, absorption associated with the Amide
II band leads primarily to bending vibrations of the N—H bond (Figure 7). Because both
the C=O and the N—H bonds are involved in the hydrogen bonding that takes place
between the different elements of secondary structure, the locations of both the Amide I
and Amide II bands are sensitive to the secondary structure content of a protein. Studies
with proteins of known structure have been used to correlate systematically the shape of
the Amide I band to secondary structure content.3,4 The Amide II band, though sensitive

3 Byler, D.M. and Susi, H. (1986) Biopolymers 25, 469-487.

4 Surewicz, W.K. and Mantsch, HH. (1988) Biochim. Biophys. Acta 952, 115-130.

4
to secondary structure content, is not as good a predictor for quantitating the secondary
structure of proteins. One difficulty with analyzing the Amide I band for secondary
structure is that the shifts in

Figure 6: Bovine pancreatic trypsin inhibitor. Left, spacefilling model; right, ribbon model. Colors are
used to label α-helix (purple), β−sheet (yellow), and loops (white).

the Amide I band are small compared to the intrinsic width of the band. Instead of a
series of nicely resolved peaks for each type of secondary structure, one broad lumpy
peak is observed (Figure 8a). Several numerical methods are used to increase the
apparent resolution of the Amide I band so that estimates can be made of the secondary
structure content. Figure 8b shows results we obtained for the Amide I band of BPTI
using the Fourier self-deconvolution method developed by Kauppenin et al..5

Amide I O O
vibration
C N C N
Amide II
H H vibration

Figure 7: The vibrations responsible for the Amide I and Amide II bands in the infrared spectra of proteins
and polypeptides. The Amide I band is due to carbonyl stretching vibrations while the Amide II is due
primarily to NH bending vibrations.

C. Monitoring hydrogen isotope exchange by FTIR spectroscopy

When a protein is dissolved in water the hydrogens attached to the amide nitrogen
can readily exchange for hydrogens attached to water molecules. For unfolded

5 Kauppinen, J.K.; Moffatt, D.J.; Mantsch, H.H. and Cameron, D.G. (1981) Appl. Spectr. 35, 271-276.

5
polypeptides this exchange can occur thousands of times a second. When a polypeptide
folds to form the well-defined 3dimensional structure of a protein, many of the amide
hydrogens become buried in the interior of the protein and no longer come into direct
contact with the solvent water. This causes a marked decrease in the rate of exchange for
these hydrogens. The most deeply buried amide hydrogens can take weeks, even months,
to exchange. The fact that they do exchange eventually is taken as evidence that folded
proteins are somewhat flexible, i.e., their structures fluctuate with time to allow buried
regions eventual access to the solvent. Hydrogen isotope exchange is becoming a
powerful tool for studying the protein folding process.6

60 60

Amide I band Amide II band

50 50 Unord.
?

40 40 α-helix

Turns
30 β-sheet
30
β-sheet
20 β-sheet
20 Sidechain
Sidechain
Turns
10
10

0
0

1800 1700 1600 1500 1400 1300 1720 1700 1680 1660 1640 1620 1600 1580
-1 -1
Wavenumber (cm ) Wavenumber {cm }

a. b.

Figure 8: a. FTIR spectrum of bovine pancreatic trypsin inhibitor showing the Amide I and Amide II
bands. b. Results of a Fourier self-deconvolution analysis on the Amide I band for bovine pancreatic
trypsin inhibitor. The assignments of the various peaks to the different elements of secondary
structure are shown.

To observe the rate of hydrogen exchange a different isotope of hydrogen is used

for the solvent water than that initially present on the protein. For our experiments we
use heavy water, which contains the deuterium isotope of hydrogen (2H) instead of the
more abundant 1H isotope of hydrogen. The 2H isotope has nearly identical chemical
properties to that of the 1H isotope, but twice the mass. Like adding mass to the end of a
pendulum, the exchange of the 1H isotope for the 2H isotope affects the vibrational
frequencies of the amide bond and therefore affect the Amide I and Amide II bands of
the infrared spectrum of a protein. Figure 9 shows the results of a hydrogen isotope
exchange experiment on BPTI, monitored using FTIR..

We are carrying out a systematic study to see if we can resolve hydrogen isotope
exchange for the different types of secondary structure using FTIR spectroscopy. We
chose to do our initial study on BPTI for several reasons: it is a very stable protein that is
easy to work with; it contains representative examples of each of the major elements of

6 Kunihiro, K., Kim, P. and Baldwin, R.L. (1984) Biopolymers 22, 59-67.

6
secondary structure (see Figure 6(right)); and extensive hydrogen isotope exchange
experiments were carried out on this protein using NMR spectroscopy, so that exchange
rates are available for nearly every amide hydrogen in the protein.7 Results of an
experiment carried out at pH 6.04 and 25°C are show in Figure 10a. To see which
elements of secondary structure were affected the most during this experiment, Figure 11
plots the relative intensity of each peak as a function of time along with assignments for
each. A logarithmic scale is used for time scale because the exchange is an exponentially
decaying process. From this experiment we made peak assignments for both the
hydrogenated and deuterated forms of each of the major elements of secondary structure.
As shown in Figure 11, the assigned peaks mirror each other’s behavior during the
exchange process. The elements exhibiting the greatest change in this experiment are the
unordered and α-helical elements. The turns and β-sheet elements show the least change.
This is expected for BPTI. The turns are located at the surface of the protein and most
likely completed their exchange before the first time point was taken. Conversely, the
β-sheet in BPTI is buried in the core of the protein, so most of the β-sheet hydrogens
should have exchanged very little by the end of this experiment. Under the conditions of
pH and temperature used in this experiment these hydrogens should take years to
exchange. However, if the temperature is raised to near the boiling point, exchange from
all sites is complete within minutes. Many

Figure 9: Time course for hydrogen/deuterium from BPTI, from 3 minutes to 374 minutes after initiating
the exchange. The Amide II band at 1450 cm1 is for deuterated BPTI and is increasing with time. The

7 Hilton, B.D., and Woodward, C.K. (1978) Biochemistry 17, 3325-3332; Hilton, B.D., and Woodward,
C.K. (1979) Biochemistry 18, 5834-5841; Wüthrich, K. and Wagner, G. (1979) J. Mol. Biol. 130, 1-18;
Richarz, R., Sehr, P., Wagner, G. and Wüthrich, K. (1979) J. Mol. Biol. 130, 19-30; Wüthrich, K. and
Wagner, G. (1979) J. Mol. Biol. 130, 31-37; Tüchsen, E. and Woodward, C. (1985) J. Mol. Biol. 185,
405-419; Tüchsen, E. and Woodward, C. (1985) J. Mol. Biol. 185, 421-430; Roder, H., Wagner, G. and
Wüthrich, K. (1985) Biochemistry 24, 7396-7407; Tüchsen, E. and Woodward, C. (1987) J. Mol. Biol.
193, 793-802; Gallagher, W.H., Tao, F., and Woodward, C. (1992) Biochemistry 31,4673-4680

7
 Amide II band at 1550 cm1 is for hydrogenated BPTI and is decreasing with time. The bottom red
spectrum at 1550 cm1 was obtained after complete exchange; the residual absorbency in this region is
due to groups on the protein other than the amide group.

changes occur between the last time point taken in the experiment carried out at pH 6.04
and 25°C and an infinite time when the exchange will be complete (see Figure 10b).
During this time hydrogen isotope exchange is expected to be primarily from the β-sheet
region of the BPTI.
We propose to extend these studies to look at the effects that hydrogen isotope exchange
from the β-sheet of BPTI has on the Amide I band. This will be done by carrying out the
experiments at higher pH values. At pH values above 5, hydrogen isotope exchange is
base catalyzed. Exchange rates are expected to increase 10fold for every 1 unit increase
in the pH. This will allow us to bring exchange from the β-sheet into the time window
accessible with our procedures.