Document Text (Pages 121-130) 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 121

by K = C C , where CBsB
phase and CBgB

s g

the concentration of the flavor compound in the aqueous
is the concentration of the flavor compound in the gas phase. The partition
coefficients represent the quantity of flavor compounds in the gas phase and are useful in
the quantitative determination of the availability of flavor compounds to be detected by
sensory receptors.
Flavor perception in foods, which is an important determinant of food acceptance,
is influenced by flavor binding and release between flavor compounds and a variety of
food components, such as protein, carbohydrates, and lipids. Proteins in foods often
decrease the volatility of flavor compounds. Proteins interact with volatile flavor
compounds, both reversibly and irreversibly, with a decrease in flavor perception for the
latter (Fabre and others 2002). In general, proteins retain volatile compounds by
reversible hydrophobic adsorption or absorption, resulting in chemical affinities of
various strengths. BLG is one of the best known and most studied of proteins. The whey
protein is extensively characterized, exhibits good emulsification, and can interact with
many flavor compounds such as aldehydes and ketones (van Ruth and Villeneuve 2002),
ionones (Jung and Ebeler 2003), and hydrocarbons (Wishnia and Pinder 1966).
Analytical techniques, including exclusion chromatography, equilibrium dialysis,
static headspace, fluorometry, affinity chromatography, as well as sensory evaluation, are
used to detect the binding of proteins to flavor compounds (Damodaran and Kinsella
1980; Steffen and Pawliszyn 1996; Pelletier and others 1998; Guichard and Langourieux
2000; Fabre and others 2002; Apps and Tock 2005). Fluorescence methods are important
tools to investigate the structure, function and reactivity of proteins and other biological
molecules. Fluorescence quenching is defined as a process which decreases the

Page 122

fluorescence intensity of a given substance (Lakowicz 1983), and is widely used to study
the accessibility and localization of intrinsic fluorophores of proteins and the
permeability of membranes to quenchers (Lakowicz 1983). Busti and others (2002) have
used external quenching of intrinsic protein fluorescence by acrylamide to monitor BLG
denaturation, and reported that BLG denatures with the dissociation of dimers preceding
the unfolding of monomers at both pH 6.8 and pH 2.5. Equilibrium headspace analysis of
volatile components in the headspace equilibrated over a food using solid-phase
microextraction (SPME) gives the association or binding of flavor to food constituents
(Damodaran and Kinsella 1980; Guichard and Langourieux 2000). However, collection
of headspace volatile flavors compounds using a syringe is not sensitive enough to detect
small concentrations resulting in irreproducible chromatography (Rouseff and
Cadwallader 2001).
Studies have been done to understand the effects of thermal denaturation on
binding of flavors to whey proteins (Burova and others 2003; Patel and others 2006).
Damodaran and Kinsella (1980) reported that the affinity of ketones for bovine serum
albumin increased with chain length, the binding is hydrophobic in nature, and the
thermodynamic driving force for binding is entropic. Whey protein was found to have
higher affinity to vanillin than soy or casein proteins, and binding was enthalpy driven
(Li and others 2000). There are many research studies that try to understand the structural
changes BLG undergoes during ultra high pressure treatment (UHP). Increases in the
surface hydrophobicity (Knudsen and others 2004) and aggregation (Futenberger and
others 1995) of BLG were observed when BLG was treated with UHP between 200 and
400 MPa. Yang and others (2001) reported that the native ordered state of BLG contains

Page 123

an intrinsic pocket that binds 1-anilinonaphthalene-8-sulfonate (ANS). An increase in the
extrinsic fluorescence of ANS after high pressure treatment (600 MPa, 50C) of BLG
follows the transition from the ordered to molten globular state. Also, a decrease in near-
UV circular dichroism intensity from aromatic side chains occurs during the transition of
BLG from native to molten globule states (Dolgikh and others 1981). At pH 2.0,
pressure-induced denaturation of BLG is reversible (Ikeuchi and others 2001). Proteins in
the molten globule state usually retain the secondary structure of the native state and
exhibit a compact tertiary structure, but exhibit increased mobility and looser packing of
the protein chain (Ptitsyn 1995). Kolakowski and others (2001) reported that using low
pressure (300 MPa) and low temperature (4 °C) minimizes the loss of native structure,
decreases undesirable aggregation, and enhances the exposure of the hydrophobic regions
of BLG to the solvent. Limited research has been done on the effects of ultra high
pressure (UHP) treatment of BLG on binding of flavor compounds with the purpose of
enhancing food flavors and perception of reduced-fat foods. Liu and others (2005)
studied the effects of UHP on the flavor-binding properties of whey protein concentrate
(WPC), and reported that there is not one pressure treatment for WPC that increases the
number of binding sites or apparent dissociation constant of any flavor compound. A
pressure treatment of 600 MPa for 0, 10, or 30 min holding time on WPC increased the
number of binding sites and apparent dissociation constant for benzaldehyde, but only a
0-min hold time was required to increase the number of binding sites and apparent
dissociation constant for heptanone and octanone (Liu and others 2005). Currently there
is no research that focuses on pre-incubating a mixture of flavor compounds and proteins,


Page 124

subjecting the mixture to UHP treatment, and then analyzing the mixture through
extraction and detection procedures.
The objectives of this study were to investigate the binding properties of selected
flavor compounds with UHP-treated β-lactoglobulin. Intrinsic fluorescence quenching of
tryptophan and headspace SPME-GC analysis were employed in the present research.
β-Lactoglobulin (BLG) (90% pure) was purchased from Sigma Chemical Co. (St.
Louis, MO, no. L-0130). BLG with the same lot number was used throughout the
δ-Decalactone, diacetyl, ethyl lactate and 2-methylbutyraldehyde were flavor
compounds selected to bind with BLG. These flavor compounds were selected because
they are associated with fat-rich foods, such as dairy products (Badings and Neeter 1980;
Milo and Reineccius 1997), chocolate, or cream-based foods , and also for their
characteristic physical properties. Each flavor compound (in 10% aqueous ethanol) was
added to BLG solutions to reach final concentrations of 1-200 ppm and held at 4 °C prior
to UHP treatment.
Ultra High Pressure Treatment
BLG solutions, at a concentration of 0.35 mg/mL in sodium phosphate buffer
(0.01 M, pH 7), that contained selected flavor compounds were treated with UHP of 600
MPa for holding times of 0 (come-up time), 8, or 32 min in the presence of ice to reduce
effects of temperature. Temperature was measured before and after UHP treatment was

Page 125

5°C and 8°C, respectively. The come-up time (5.45 min) was the compression time
required to reach a pressure of 600 MPa. After exposure to UHP, BLG solutions were
studied immediately or stored at 4 °C for fewer than 10 days, otherwise the protein
Flavor Compound Fluorescence Binding Assay
Stern-Volmer analysis
Flavor binding experiments were performed in 0.01 M sodium phosphate buffer
(pH 7.0, adjusted with 0.01 N HCl or NaOH) with selected flavor concentrations.
Intrinsic tryptophan fluorescence was assayed using an excitation wavelength of 287 nm
and emission at a wavelength of 332 nm. The quantitative binding of ligands was
determined from the decrease of the protein tryptophan fluorescence at 332 nm (Dufour
and Haertle 1991). Since the flavor-BLG complex is responsible for the fluorescence
quenching of tryptophan in BLG, we assume that static quenching is responsible for the
decrease of BLG fluorescence upon addition of flavor compounds. Stern-Volmer analysis
for static quenching predicts that the ratio FB0B/F between the fluorescence of BLG in the
absence of selected flavor compounds (FB0B) and the fluorescence of BLG in the presence
of selected flavor compounds (F) is, for small amounts of flavor compounds, a linear
function of the concentration of the flavor compound (Q) (Tian and others 2006).



[ ]+1

= K


Q (1)

The slope obtained from the linear regression provides the static quenching constant, a
relevant binding parameter.


Page 126

Double logarithmic analysis
Double logarithmic analysis predicts a linear relationship between

F0 F
Log ⎢ ⎥ and Log[Q], where F is the fluorescence of BLG in the presence of

F Finf
selected flavor compounds, FB0B is the fluorescence of BLG in the absence of selected
flavor compounds and FBinfB is the residual fluorescence of BLG saturated with selected
flavor compounds. The slope of the double Log plot yields the number of binding sites in
F0 F
BLG and the intercept at Log⎢ ⎥ = 0 provides the dissociation constant (Tian and
F Finf
others 2006).
Headspace Analysis
δ-Decalactone, diacetyl, ethyl lactate and 2-methylbutyraldehyde were chosen to
investigate the effects of UHP on flavor retention by BLG. Analyses were conducted in
duplicate in amber flasks (4 mL) with a flavor concentration of 5 ppm in 10% ethanol and
a BLG concentration of 0.35 mg/mL in sodium phosphate buffer (0.01 M, pH 7.0).
Analyzed solutions containing 0.65g NaCl were stirred at 400 rpm on a magnetic hot stir
plate (model PC-220, Corning, NY) for 5 min at 37 °C to pre-equilibrate volatiles in the
headspace. A polydimethylsiloxane-divinylbenzene (PDMS/DVB) solid-phase
microextraction (SPME) needle was used to adsorb volatiles in the headspace of the
flavor-BLG solution. The PDMS/DVB SPME fiber was used because the fiber can
adsorb the high molecular weight (170.25 MW) δ-decalactone flavor compound. The
SPME fiber was allowed to extract volatiles for 10 min. Volatile flavor compounds
adsorbed to the SPME fiber were thermally desorbed into the injection port of a Hewlett-
Packard HP 5890A (Agilent Technologies, Palo Alto, CA, USA) gas chromatograph

Page 127

connected to a flame ionization detector, equipped with a SPME inlet liner, and a DB-
WAX column (60 m x 0.32 mm i.d., 0.5 µm film thickness, J&W Scientific, Folsom,
CA). Helium was used as the carrier gas. The injector and detector temperatures were the
same at 250 °C according to Adahchour and others (2005). The column oven program
temperature was initially maintained at 33 °C for 5 min before increasing to 50 °C at a
rate of 2 °C/min and then to 200 °C at a rate of 5 °C/min and maintained at 200 °C for 45
Statistical Analysis
Data were analyzed by 2-way ANOVA followed by Tukey’s pairwise comparison
using SAS 9.1, and p ≤ 0.05 was defined as significance.
Quenching of BLG by Selected Flavor Compounds
UHP treatment of BLG solutions pre-incubated with selected flavor compounds at
selected UHP hold times produced quenching of the BLG intrinsic fluorescence. Figures
1-3 show that for diacetyl, ethyl lactate, or 2-methylbutyraldehyde, at all UHP treatment
times, the Stern-Volmer plots were linear in the 0-2 mM range. The Stern-Volmer plot
for the quenching of native or UHP treated BLG by δ-decalactone was not linear
suggesting that the number of replications were not adequate to justify linearity (Figure
4). The quenching constant of BLG for diacetyl increased as UHP treatment increased
from come-up time to 32 min (Table 1). The quenching constants of native BLG for δ-
decalactone, 2-methylbutyraldehyde, or ethyl lactate were low. Following UHP treatment
of 32 min, the quenching constant showed a 500%, 33%, and 78% increase, respectively,


Page 128


indicating that the unfolding of BLG structure and exposure of the hydrophobic interior
allowed for the binding of the selected compounds.
In the present study, the dissociation constants (KBdB) determined from the double
logarithmic plots for native BLG with diacetyl, ethyl lactate, 2-methylbutyraldehyde, and

-3 -3 -3 -3

δ-decalactone were 2.1 x 10P


M, 1.6 x 10P


M, 0.8 x 10P


M, and 0.1 x 10P


respectively. The affinity of BLG for a flavor compound or a ligand is dependent upon
the molecular structure of the flavor compound or ligand (Damodaran and Kinsella
1980). The high KBdB by native BLG for diacetyl results from the high polar
nature of the two carbonyl groups present. Ethyl lactate may also be excluded from
binding to BLG because of a hydroxyl group and carbonyl group present. On the other
hand, 2-methylbutyraldehyde and δ-decalactone have short hydrocarbon tails that may fit
in the hydrophobic pockets on native BLG while the polar heads (carbonyl groups) are
exposed to the solvent.
The highest dissociation constants of UHP treated BLG occurred after 32 min for
δ-decalactone and 2-methylbutrladehyde (Table 1). Since Yang and others (2001) have
reported that pressure and temperature of 600 MPa and 50 °C induce BLG into the
molten globule state, it is possible that the increased mobility and looser packing of the
protein chain and further exposure of hydrophobic sites may prevent these flavor
compounds from further binding to BLG. Liu and others (2005) reported apparent
dissociation constants for benzaldehyde, heptanone, octanone, and nonanone on the order


of 1.8-6.2 x10P


M of whey protein concentrate for aliphatic methyl ketones. These
findings indicated that BLG has smaller affinity for the selected flavor compounds in this
study than aliphatic methyl ketones selected by Liu and others (2005), probably due to

Page 129

the hydrophobic binding sites on interior of the proteins favoring nonpolar flavor
compounds. However, small KBdB values were observed after UHP come-up time for


diacetyl and ethyl lactate, 1.2 and 0.9 x 10P


M, respectively, suggesting that a short UHP
treatment may improve the retention of certain flavor compounds.
The fractional number of binding sites determined by the slope of the double log
plot shows an increase in binding sites of BLG after UHP come-up time for diacetyl, δ-
decalactone, and ethyl lactate. A decrease in the fractional number of binding sites of
native BLG following UHP was observed with 2-methylbutyraldehyde (Table 1). The
fractional number of binding sites did not increase with increasing pressure treatment
indicating that ultra high pressure does not improve binding.
Headspace Analysis
The effect of UHP on the retention of flavor by BLG was studied by HS-SPME
GC analysis. Static headspace analysis determines the quantity of volatiles contained in
the gas phase above a sample, usually in a sealed system at equilibrium. The composition
of the headspace depends on the partitioning of volatiles between the gas phase and
aqueous phase. The headspace composition may be influenced by many other factors,
such as temperature, the addition of salts, headspace volume, and vial shape (Pawliszyn
The quantity of flavor compounds detected in the headspace significantly
decreased (p < 0.05) in the presence of UHP treated BLG compared to the quantity of 2-
methylbuyraldehyde (Figure 5), δ-decalactone (Figure 6), diacetyl (Figure 7), and ethyl
lactate (Figure 8) in the presence of native BLG, resulting from hydrophobic and polar
interactions between the flavor compounds and BLG (Guichard 2006). The retention of

Page 130

the selected flavor compounds by native BLG in decreasing order is: δ-decalactone >
ethyl lactate > diacetyl > 2-methylbutyraldehyde. The retention of the selected flavor
compounds after UHP-treatment of BLG in decreasing order is: diacetyl > ethyl lactate >
δ-decalactone > 2-methylbutyraldehyde. The percentage of retention of 2-
methylbutyraldehyde, δ-decalactone, diacetyl, and ethyl lactate were 30.6%, 40.9%,
55.8%, and 56.9%, respectively. These findings demonstrate larger retentions with a preincubation
step than previously reported when selected flavor compounds were added to
BLG already treated with UHP (Yang and others 2002; Liu and others 2005). The
increase in BLG retention of selected flavor compounds was attributed to the preincubation
period of flavor compounds with BLG prior to UHP treatment. Significant
increases of 88-97% in retention were observed with UHP treatment of BLG over native
BLG with diacetyl, ethyl lactate and 2-methylbutyraldehyde. Small, yet significant (p ≤
0.05), increases in retention (56-65%) were observed for δ-decalactone, consistent with
previously reported results for the retention of ketones in BLG solutions (Jouenne and
Crouzet 2000). Although decreases in headspace concentration of selected flavor
compounds may not be reflected in the results from fluorescence quenching experiments,
flavor compounds may bind to sites on BLG far from tryptophan not detected by the
fluorescence quenching experiments.
Fluorescence quenching data that were used to generate double logarithmic plots
for the determination of dissociation constants suggested that UHP-treated BLG had low
affinity for the selected flavor compounds. However, HS-SPME GC analysis exhibited a
high degree of retention of BLG following UHP treatment. It may be that the

© 2009 All Rights Reserved.