Document Text (Pages 21-30) Back to Document

Application of ultra high hydrostatic pressure for investigating the binding of flavor compounds to ß-lactoglobulin via headspace solid phase microextraction-gas chromatography

by 1977- Hoang, Tinyee Arden, PhD

Page 21

There are several intrinsic fluorophores in proteins (mainly aromatic amino acids:
tryptophan, tyrosine, and phenylalanine) that have been used to explore protein structurefunction
relationships or molecule interactions in association with other physical
parameters such as temperature and pressure (Lakowicz 1983). Among these
fluorophores, the tryptophan residue (Trp) is most often used. The characteristics of its
fluorescence emission spectra depend on its environment. The emission maximum
wavelength (λBmaxB) of free Trp in aqueous solution is at about 352 nm and becomes shorter
(blue shift) when Trp is in an nonpolar solvent or in a hydrophobic environment such as
inside the protein core. The more nonpolar the environment, the larger the Trp λB

shift. When the protein is gradually unfolded, the Trp residue becomes more and more
exposed to the bulk aqueous solution. Then the λB

shifts to the red and finally reaches
352 nm (free Trp in water) when Trp is totally solvated, i.e. when the protein is fully
denatured. Therefore protein Trp fluorescence can be used as a probe related to the
medium polarity allowing for detection and following protein conformational changes,
from which information about protein structure-function, unfolding and molecule
interaction, can be obtained. In comparison with Trp, tyrosine and phenylalanine residues
in proteins have not been often used for such studies, mainly because their fluorescence
quantum yields are lower and less sensitive to the environment (Lakowicz 1983), i.e. less
sensitive to conformational changes. To measure the Trp fluorescence emission coming
only from Trp residues, without disturbance from the excitation of tyrosine and
phenylalanine residues, the protein is often excited at 295 nm or at longer wavelength,
where tyrosine and phenylalanine do not absorb (Lakowicz 1983).
Extrinsic Fluorescence

Page 22

Some extrinsic fluorescence probes including 1-anilinonaphthalene-8-sulfonic
acid (ANS) and cis-parinaric acid (CPA) are widely used in BLG structure-function
studies, especially for determining protein hydrophobicity. Upon binding of the probes to
accessible hydrophobic regions of proteins, an increase in fluorescence is observed,
which is used as a measure of protein surface hydrophobicity. However, due to the
possible contribution of both electrostatic and hydrophobic interactions to the binding of
these anionic probes, the interpretation based on these probes has not been easy.
PRODAN (6-propionyl-2-dimethylaminonaphthalene), introduced by Weber and
Farris (1979), is a fluorescent probe that is very sensitive to the polarity of the
environment. The binding of PRODAN to protein results in a fluorescence enhancement
and a blue shift that provides for easy separation of the fluorescence contribution from
the free and bound probe (Moreno and others 1999). PRODAN is suitable for studies of
polarity of many protein cavities by spectroscopic techniques (McGregor and Weber
1986). Other uses for PRODAN were described for the study of pressure effects on
ligand-protein complexes (Chong and Weber 1983) and on the dielectric constant in
phosphatidylcholine lipid bilayers (Chong 1988).
Using an uncharged probe (PRODAN) for measuring protein surface
hydrophobicity, especially under conditions of varying pH, is advantageous over using
anionic probes ANS and CPA. Alizadeh-Pasdar and Li-Chan (2000) reported low surface
hydrophobicity of BLG when using ANS or CPA at pH 3.0 as opposed to relatively high
surface hydrophobicity when using PRODAN for unheated and heated (80°C) BLG. The
presence or absence of a permanent charge results in electrostatic interactions that may
overestimate protein hydrophobicity determined under selected pH conditions (Alizadeh-


Page 23

Pazdar and Li-Chan 2000). Haskard and Li-Chan (1998) reported that increasing the
ionic strength of buffers to 1.0 M NaCl had a negative effect on surface hydrophobicity
determination of bovine serum albumin when using ANS and no significant effect when
using PRODAN.
Flavor-Protein Interactions
Flavor is considered one of the most important attributes of food in the
determination of acceptance by a consumer. The mixture of odorous compounds that is
present in food is often very complex. Individual components may be present in
extremely small concentrations (µg to sub-pg/Kg) and may be unstable or volatile. As a
consequence, it is difficult to concentrate flavor compounds from foods representative of
the original flavor and free of contaminants and artifacts.The quantities of flavor
compounds perceived by the human nose are determined by the release of flavor
compounds from the food matrix. The rate of release is based on the affinity of the flavor
compounds for biopolymers that make up the food matrix.
The predominant influence of proteins on flavor release and perception is caused
by interactions of flavor components with the protein. The interactions that occur
between proteins and flavor compounds are either reversible binding, including hydrogen
bonds, hydrophobic interactions, and ionic bonds, or irreversible binding via covalent
linkages and the condensation of aldehydes with amino (NHB2B) and sulfhydryl (SH)
groups (Izzo and Ho 1993; Fischer and Widder 1997; Lubbers and others 1998; Adams
and others 2001).
One of the most widely used proteins in flavor binding studies is BLG, which is
known to interact with many flavor compounds, such as ketones, ionones, aldehydes, and

Page 24

esters (van Ruth and Villeneuve 2002). Increases in percentage of retention of BLG with
increasing chain length for a series of alkanones and ethyl esters suggest hydrophobic
interactions (Andriot and others 2000). Most of the binding that occurs between the
flavor compound and BLG is reversible through hydrophobic affinity and hydrogen
binding (Tromelin and Guichard 2003; Guichard 2006). Any factor affecting
hydrophobic affinity or surface hydrophobicity of BLG influences flavor binding.
Salting-in-type salts, such as BaClB2B, CaClB2B, MgClB2B and MnClB2B, destabilizes hydrophobic
interactions thereby decreasing flavor binding, whereas salting-out type salts, such as
MgSOB4B and NaB2BSOB4B, increase flavor binding through preferential hydration of proteins
(Arakawa and Timasheff 1984). Two opposing factors, the surface tension effect
contributing to an unfavorable free energy change and the salt binding to peptide bonds
and some side chains as major sources of favorable free energy, are responsible for
protein preferential interactions with salts (Arakawa and Timasheff 1984). The binding of
salting-out type salts to proteins is greater than the binding of salting-in type salts to
proteins is due to the surface tension effect that leads to significant preferential hydration
of proteins.
BLG is a possible carrier for flavor compounds and may be effective in delivering
or delaying release of flavor compounds. BLG can be altered to bind to a wide range of
volatile flavor compounds during food manufacturing or to release these flavor
compounds in a more or less controlled way upon modifications through chemical or
thermal means (Kuhn and others 2006).
Ultra High Hydrostatic Pressure of Foods


Page 25

Introduced in 1899 by Bert H. Hite for the preservation of milk, meats, and juices
by high pressure processing (Johnston 1995), ultra-high pressure (UHP), or high
hydrostatic pressure (HHP), treatment of foods and other biological tissues affects the
structural stability of chemical constituents, including proteins, vitamins, lipids,
saccharides and pigments in such a way that may improve their intrinsic functional
properties in terms of color, flavor, and texture retention. UHP treatment involves
subjecting food materials to a high pressure (up to 87,000 pounds per square inch or 600
MPa) and holding (isobarically) the food under constant pressure for a specific period of
time (less than 30 min) before pressure release. UHP is sufficient in bringing about the
necessary level of treatment in terms of protein conformational change, microbial
deactivation, and extended shelf-life (Tedford and others 1999). There is a growing body
of literature concerning the effects of UHP on selected food ingredients. Reversible
effects such as dissociation of polymeric structures or partial unfolding are observed
below 200 MPa (Iametti and others 1997). Structural changes are induced in proteins
when pressures greater than 200 MPa are employed. Pressures greater than 500 MPa
result in irreversible unfolding of monomeric proteins, aggregation of monomers
stabilized by thiol/disulfide interchange reactions (Futenberger and others 1997), and
formation of gel structures of stable polymers made of denatured monomers, yet only
occurs in 10% of the total protein as observed by circular dichroism (Iametti and others
1997). However, temperature induces irreversible denaturation of the whole protein
through the hydrolysis of covalent bonds and/or aggregation of denatured proteins
(Tedford and others 1999).


Page 26

Effects on protein structure
The effects of pressure on proteins can be understood in terms of the Le Chatelier
principle which states that any reaction, conformational change, or phase transition that is
accompanied by a decrease in volume will be enhanced when pressure increases, while
reactions involving an increase in volume will be inhibited (Johnston 1995; Ledward
2000). For pressure-induced protein unfolding, the equilibrium constant at atmospheric
pressure (KBatmB) is a function of the equilibrium constant at a given pressure (KBpB) by the
Kp = KBatmBexp(-pVBunfB/RT) ,
where p is pressure, VBunfB is the standard volume change of unfolding, R is the gas
constant and T is the absolute temperature. As the equilibrium constant at a given
pressure is directly related to the degree of unfolding (Kp = α /1- α), where α is the
degree of denaturation, it follows that (Botelho and others 2000):
ln(α/1- α) = lnKBatmB pVBunfB/RT
The Gibbs energy that determines thermodynamic equilibrium among conformers
(stereoisomers) of a protein in solution is driven by pressure according to the following
relation (Lasalle and others 2003),

where pB0B

0 0

is the atmospheric pressure (1 bar), GP


and VP


are differences in the Gibbs
energy and partial volume at 1 bar, respectively, and ∆κ denotes the difference in
compressibility. Pressure changes the conformational equilibrium by acting on
volumetric properties, while denaturants such as urea directly perturb the interaction


energy and entropy embedded in GP


(Wu and Wang 1999). That is, pressure drives the

Page 27

equilibrium to increase the population of the lower volume conformer to the higher
volume conformer (Weber and Drickamer 1983).
Proteins in solution may adopt a variety of conformations between the native state
and the fully denatured state. The molten globule state is one of the conformations
observed in globular proteins that is thermodynamically stable (Ptitsyn 1995) and is a
compact denatured state under mild denaturing conditions with native-like secondary
structure, with loss of tight native tertiary structure (Yang and others 2001). While in the
molten globule state, the amino acid residues exhibit increased mobility compared to the
native state (Ptitsyn 1995). Molten globules have increased exposure and hydration of
hydrophobic residues compared with the native state (Dagget and Levitt 1992), as
evidenced by increased binding of hydrophobic probes, especially ANS (Yang and others
2001), decreased solubility in water, increased susceptibility to aggregation, and
increased heat capacity (Ptitsyn 1987).
The molten globule state can be induced for many proteins using chemical and
physical methods such as acid denaturation, chemical denaturants (urea or guanidine
hydrochloride), heating, high pressure treatment, or ethanol (Kumar and others 2004)
(Engelhard and Evans 1995; Yang and others 2001; Mazon and others 2004; Paci and
others 2005).
Pressure may affect the secondary, tertiary, and quaternary structure of proteins.
Moderate pressure does not disrupt secondary structures because hydrogen bonds that
stabilize the secondary structure are not affected by the little effect of pressure (Masson
and Clery 1996). Belloque and others (2000) showed that at pressure treatments of 300
and 400 MPa, the BLG structure exhibits a high degree of flexibility through the

Page 28

exposure of a thiol group near β-strands FGH, and exists in a completely unfolded state.
However, the conformation of the BLG core after pressure release is identical to the
native structure. The preservation of the core implies that most of the β-barrel is
structured. Belloque and others (2000) suggested that after pressure treatment of 400
MPa, complete refolding of BLG may occur when BLG is returned to ambient pressure
or during pressurization BLG did not unfold completely. Iametti and others (1997) used
circular dichroism and fluorometry and observed that only 10% of the BLG structure was
lost at both 600 and 900 MPa. Moller and others (1998) reported that the exposure of the
thiol group after pressure treatment of 150 MPa decreased after 2 days, indicating the
refolding of BLG. Hydrophobic amino acids are found in discrete patches on the surface
of proteins. Iametti and Bonomi (1993) reported that physical denaturing agents can
modify the number and properties of these surface hydrophobic patches. After pressure
treatment of 900 MPa, a decrease in the number of surface hydrophobic sites was
observed with a concomitant decrease of the affinity of BLG for hydrophobic probes
(Iametti and others 1997).
Effects on flavor-protein binding
UHP treatment changes the structure of BLG, thus affecting the protein-flavor
interactions. Liu and others (2005) have reported effects of UHP treatment on whey
protein-flavor binding using fluorescence spectroscopy and headspace analysis.
Depending on the structure of the flavor compounds, benzaldehyde and methyl ketones,
the flavor concentration, and UHP treatment times, the number of binding sites and
apparent dissociation constants were either unaffected or increased upon UHP treatment
(Liu and others 2005).


Page 29

Solid Phase Microextraction
Sample preparation techniques based on adsorption have been applied for the
extraction of volatiles for instrumental analysis (Yang and Peppard 1994). Solid phase
microextraction (SPME), developed by Pawliszyn and co-workers (Arthur and Pawliszyn
1990; Zhang and others 1994), involves the extraction of compounds onto a chemically
modified fused silica optical fiber. It is a convenient and solvent-free method that is
suitable for the analysis of food, beverages, oils (Page and Lacroix 1993; Yang and
Peppard 1994; Steffen and Pawliszyn 1996; Ai 1997; Kataoka and others 2000), and
many dairy products, including the detection of flavor compounds in headspace of β-
lactoglobulin (Jung and Ebeler 2003). SPME is used in combination with gas
chromatography (GC) (Rocha and others 2001), GC-mass spectrometry (GC-MS) (Jelen
and others 2003), and GC-olfactometry (GC-O) (Rega and others 2003), and introduced
for direct coupling with HPLC and LC-MS for the analysis of weak volatile or thermally
labile compounds not amenable to GC or GC-MS (Pawliszyn 1999).
SPME is a multiphase equilibrium process. Most extraction systems are complex
consisting of an aqueous phase with suspended particles having various adsorption
interactions with analytes, and a gaseous headspace. During extraction, analytes migrate
between all three phases until equilibrium is reached. A number of factors must be
considered to ensure good accuracy and precision in the development of a SPME method.
The SPME fiber coating is determined by the chemical nature of the target analyte, and
its polarity and volatility characteristics. Polydimethylsiloxane (PDMS) is often
considered first for its rugged liquid coating to withstand high injector temperatures, up
to about 300 °C. PDMS is a nonpolar phase that extracts nonpolar analytes very well.

Page 30

However, it can be successfully applied to extract more polar analytes, particularly after
optimizing extraction conditions. It is important to consider the use of mixed phase
coatings for extraction of volatile compounds, and perform experiments with different
fibers to identify the optimal coating type for a broad range of compound characteristics
(Pawliszyn 1997).
The extraction mode is selected by considering the sample matrix, analyte
volatility, and its affinity to the matrix. Headspace extraction is preferred due to faster
equilibration times and for medium to high volatile analytes. For very polar analytes,
such as acids and bases, which can have high affinity toward the matrix, changing the pH
of the matrix can improve extraction.
Efficient desorption and rapid transfer of the analytes from the injector of the GC
to the column require high linear flow rates of the carrier gas around the coating, and can
be accomplished by using a narrow bore injector insert. Temperature also plays an
important role in accelerating the desorption process. When the maximum allowable
coating temperature is used as the injector temperature, adjustment of the desorption time
facilitates quantitative desorption in a single injection.
The equilibration time is defined as the time after which the amount of analyte
extracted remains constant and corresponds within the limits of experimental error to the
amount extracted at infinite extraction time (Pawliszyn 1999). When equilibration times
are excessively long, shorter extraction times can be used. Care must be taken when
determining the equilibration time in HS-SPME determinations, especially when a rapid
rise of the equilibration curve is followed by a very slow increase that is related to the
mass transfer of the analyte from water through the headspace to the fiber. At

© 2009 All Rights Reserved.