The Pennsylvania State University
The Graduate School
College of Agricultural Sciences
BITTER: UNDERSTANDING HOW PROTEIN, LIPIDS, AND OIL-IN-WATER
EMULSIONS INFLUENCE PERCEPTION
A Thesis in
Food Science
by
Kelsey Tenney
© 2016 Kelsey Tenney
Submitted in Partial Fulfillment
of the Requirements
for the Degree of
Master of Science
May 2016
The thesis of Kelsey Tenney was reviewed and approved* by the following:
John N. Coupland
Professor of Food Science
Thesis Adviser
John E. Hayes
Associate Professor of Food Science
Ryan J. Elias
Associate Professor of Food Science
Robert F. Roberts
Professor and Head of Food Science
*Signatures are on file in the Graduate School
ii
ABSTRACT
Drugs often have a bitter taste which can be unpleasant and aversive. The risk is
greatest for pediatric patients who are more sensitive to bitter yet require liquid drug
delivery systems. Removal of the bioactive ingredients that are the source of the bitter
taste would cause loss of biological function of the medication. Suppression of the bitter
taste has therefore become a goal of medication development. For pediatric
formulations this currently involves the use of sweeteners and flavors to mask
bitterness. This approach is not fully effective in suppressing bitterness, and the large
amounts of sugar may be nutritionally undesirable. Additionally, foods with added
phytochemicals are also rather bitter, and development of more palatable functional
foods has been a challenge. One under-utilized approach to flavor masking is
encapsulation—using physical means to prevent interaction between the tastant and
receptor.
It has been asserted in the literature that tastants must diffuse through the aqueous
saliva to be perceived, and thus, the freely available aqueous phase will determine
bitterness perception. Utilizing the characteristics of ingredients such as protein binding
behavior and lipid phase volume would prevent bitterant interaction with taste
receptors and, theoretically, prevent perception. The majority of published studies
regarding bitterness-masking have been limited by their focus on single bitter
compounds in poorly defined food matrices, and by a lack of either adequate analytical
measurements or adequate sensory measurements. Addressing those weaknesses, the
goal of this study is to investigate (i) the binding and partitioning of bitterants (quinine
and caffeine) by whey protein isolate (WPI) and lipids respectively and (ii) the
suppression of bitterness by proteins and o/w emulsions. I hypothesize that the degree
of binding will differ for quinine and caffeine and that the perceived bitterness will
depend on the unbound, aqueous concentration rather than total bitterant
concentration.
Protein binding was assessed for a WPI solution of either caffeine or quinine followed by
filtration and collection of supernatant. Partitioning (Log Kow) was measured for caffeine
and quinine using vegetable oil and water in a shake flask method. Bitterant
concentration was measured by HPLC. Consumer sensory tests (n≥100) were used to
measure scaled sample bitterness.
The first study investigated the effect of protein binding on bitterness of caffeine and
quinine. WPI binding curves (0-4.5%) were established for both bitterants. Caffeine was
largely unaffected by the presence of WPI while quinine was mostly bound. A sensory
test was conducted for WPI at 0% and 1% with three levels of caffeine (1.8, 5.7, 18 mM)
and quinine (0.056, 0.10, 0.18 mM). There was no significant effect from the WPI on
caffeine bitterness, but WPI did decrease the bitterness of quinine with significant
differences at the low and medium bitterant levels. This is consistent with the
hypothesis that higher binding results in lower bitterness, but the magnitude of
bitterness reduction was not large. Additionally, the aqueous concentration does not
seem to determine the bitterness perception indicating that something else is
iii
happening in the mouth that was previously not taken into account. Mucin, a salivary
protein, was investigated as an explanation for this discrepancy as a possible
competitive binding protein. Binding curves were constructed just as with the WPI
testing for caffeine and quinine, and the results were similar to the WPI on a per unit
mass of protein basis. This indicates that mucin does not have an unusually high affinity
to quinine, but the interaction suggests the oral environment may somehow change the
distribution of the bitterant.
The second study investigated the effect of oil partitioning on bitterness of caffeine and
quinine through an o/w emulsion. Caffeine and quinine had Log Kow values of -1.32 and
2.97 respectively, establishing quinine as a hydrophobic bitterant. Fat was assessed in
an o/w emulsion with WPI (at a constant 0.125%) as the emulsifier. An increase in fat
from 0.5% to 2% had no effect on the perceived bitterness of caffeine (18 mM) and
caused a significant decrease (p < 0.05) in bitterness perception of quinine (0.1 mM).
The emulsion sensory results correlate with aqueous concentration of bitterant in the
formulations, though the bitterness decrease was modest. The complete phase
separation of saliva and oil in the emulsion may be why the aqueous concentration in
this test was indicative of bitterness perception as opposed to the protein test.
However, the modest suppression indicates that this solution may have to be used with
other suppression techniques for practical success.
In conclusion, partitioning of bitter compounds by oil reduces the perceived bitterness
at a level indicative of aqueous concentration decreases, but protein binding does not.
This finding has implications for the formulations of oral pediatric medications and
foods enhanced with phytochemicals. More research needs to be completed in order to
determine how saliva influences the system once in the mouth and more accurately
defines the binding interactions.
iv
TABLE OF CONTENTS
List of Figures………………………………………………………………………………………………………………..vii
List of Tables…..……………………………………………………………………………………………………………..ix
Acknowledgements…………………………………………………………………………………………………….....x
Chapter 1. Factors Affecting Bitter Taste………………………………..……………..……………….………1
1.1 The Basic Tastes……………………………………..……………………………………………………1
1.2 The Mechanics of Taste………………………………………..……………………………………..2
1.3 Effects of Saliva…………………………..……………………………………………………….………5
1.4 Altering Taste Perception…..………………………………………………………………………..7
1.5 Considering Protein……………………………………………………………………………………..8
1.5.1 Protein Function and Binding Characteristics.....…………………………..8
1.5.2 The Influence of Protein on Taste…………………………………………….…10
1.6 Considering Lipids……………………………………………………..……………………………….12
1.6.1 Lipid Chemistry and Partitioning…………………………………………………12
1.6.2 The Influence of Lipids on Taste………………………………………………….13
1.7 References……..………………………………………………………………………………………….18
Chapter 2. Significance and Goals…………………………………………………………………………………25
2.1 Significance………………………………………………………………………………………………..25
2.2 Goals and Hypotheses……………………………………………………………………………….25
2.3 References…………………………………………………………………………………………………27
Chapter 3. Materials and Methods………………………………………………….……………………………29
3.1 General………………………………………………………………………………………………………29
3.2 Techniques Used Specifically in Chapter 4………………………………………………….30
3.3 Techniques Used Specifically in Chapter 5………………………………………………….32
3.4 References…………………………………………………………………………………………………35
Chapter 4. Effect of Protein on Perception of Bitter Compounds…….……………………………36
4.1 Introduction……………………………………………………………………………………………….36
4.2 Methods…………………………………………………………………………………………………….37
4.3 Protein-Binding and Sensory Test Results and Discussion………………………….38
4.3.1 Binding by Proteins…………………………………………………………………….38
4.3.2 Effect of Protein on Bitterness………………………….………………………..40
4.4 Considering Saliva…………..…………………………………………………………………………43
4.5 Mucin Binding of Quinine……..…………………………………………………………………..45
4.6 Conclusions……………………………..………………………………………………………………..48
4.7 References…………………………………………………………………………………………………49
Chapter 5. Effect of Lipids on Perception of Bitter Compounds………………….…………………53
5.1 Introduction……………………………………………………………………………………………….53
v
5.2 Methods…………………………………………………………………………………………………….54
5.3 Oil Partitioning and Sensory Test Results and Discussion……………………………54
5.3.1 Oil/Water Partitioning………………………………………………………………..54
5.3.2 Bitterant Partitioning in Emulsion Systems…………………………………57
5.3.3 Effect of Emulsions on Bitterness……………………………………………….58
5.4 Conclusions………………………………………………………………………………………………..62
5.5 References………..………………………………………………………………….……………………63
Chapter 6. Conclusions and Future Directions…..…………………………………….……………………65
6.1 Study Significance………………………………………………………………………………………65
6.1.1 Summary of Results……………………………………………………………………65
6.1.2 Implications and Potential Applications……………………………………..67
6.2 Limitations and Recommendations for Future Work………………………………...68
6.3 References………………………………………………………………………………………………..71
Appendix: Chapter 1 Copyright Permissions…………………………………………………………………73
vi
LIST OF FIGURES
Figure 1.1: From Mennella et al. (2013). How bitter works: the process of bitter
perception…………………………………………………………………………………………………………….……….4
Figure 1.2: From Matsuo (2000). Schematic diagram of a taste bud (A) and model of the
initial events in taste perception (B)……………………………………………………………………………….5
Figure 1.3: Effect of oil volume fraction on the aqueous concentration of a hydrophilic
(Kow = 0.001), neutral (Kow = 1), and hydrophobic (Kow = 1000) compound. From Coupland
and Hayes (2014)………………………………………………………………………………………………………….13
Figure 4.1: Proportion of caffeine (0.51 mM) and quinine (0.13 mM) bound as a function
of (a) native and (b) denatured WPI solution concentration. Error bars indicate standard
error. A logarithmic model shown alongside the data……………………………………………………39
Figure 4.2: Mean protein sensory test results for (a) caffeine (n=105) and (b) quinine
(n=119). Bitterness intensity, represented by the bars, is plotted on the primary y-axis
(left). Low, medium, and high concentration samples refer to nominal total
concentrations of 1.8 mM, 5.7 mM, and 18 mM for caffeine and 0.056 mM, 0.1 mM, and
0.18 mM for quinine, respectively, while the experimentally measured aqueous
concentrations of each bitterant are shown on the secondary y-axis (right). Different
letters indicate significant differences in bitterness rating (p<0.05). Error bars indicate
standard error……………………………………………………………………………………………………………...42
Figure 4.3: Relationship between aqueous (unbound) concentration of quinine and
perceived bitterness in water () and in 1% WPI (). Total concentration of quinine
(bound and unbound) in 1% WPI is also presented for comparison (). Logarithmic fit
shown alongside the data. Error bars indicate standard error.………………………………………43
Figure 4.4: Proportion of caffeine (0.51 mM) and quinine (0.13 mM) bound as a function
of mucin solution concentration (a) and proportion of quinine bound as a function of WPI
or mucin solution concentration (b) Error bars indicate standard error. Logarithmic fit
shown alongside the data.…………………………………………………………………………………………….47
Figure 5.1: Proportion of caffeine and quinine unbound as a function of oil concentration.
Original aqueous concentration was 0.51 mM and 0.13 mM for caffeine and quinine,
respectively. Log KOW of caffeine and quinine are -1.32 and 2.97 respectively. ………………55
Figure 5.2: Effect of lipid concentration in an o/w emulsion (d32~0.225 m) on the relative
aqueous concentration of (a) caffeine (0.51 mM) and (b) quinine (0.13 mM). The
emulsions prepared with caffeine were stabilized with 1% WPI while those containing
quinine were prepared with 0.125% WPI. Error bars indicate standard error.………………58
Figure 5.3: Mean o/w emulsion sensory test results (n = 200) at 0.5% and 2% oil with
constant 0.125% WPI. Bitterness intensity, represented by the bars, is plotted on the
primary y-axis (left) with bitterant aqueous concentration on the secondary y-axis (right)
(caffeine = 18 mM, quinine 0.1 mM). Note: scale is shown from 40 to 65 to better show
vii
differences. Different letters indicate significant differences (p<0.05). Error bars indicate
standard error…….……………………………………………….……………………………………………………….61
Figure 5.4: Flow curves for (a) 2% oil and (b) 0.5% oil emulsions. The best fit of the
Bingham-Plastic model is shown alongside the data. (c) Sensory thickness data for
vegetable oil/water emulsions at 0.5% and 2% oil with constant 0.125% WPI. The
thickness scale is shown from 0 to 50 to better show differences and different letters
indicate significant differences (p<0.05). Error bars indicate standard error………….………61
Figure 5.5: Relationship between aqueous (unbound) concentration of quinine and
perceived bitterness in water () and in emulsion systems with 0.5% and 2% fat ().
Lines are best fits of a logarithmic model. Error bars indicate standard error……………….62
viii
LIST OF TABLES
Table 1.1: Protein composition of milk. Adapted from Morr and Ha (1993)…………………9
Table 1.2: Physicochemical properties of whey proteins and composition in commercial
whey protein products. All values compiled from Morr and Ha (1993) unless otherwise
noted………………………………………………………………………………………………………………………….10
Table 5.1: Partition coefficients for caffeine and quinine from the literature………………56
ix
ACKNOWLEDGEMENTS
To Dr. Coupland and my committee. Thank you for providing guidance and answering
my many questions. This work truly could not have been done without your help. I know
those bitter solutions were not so pleasant to taste… In particular I would like to thank
Dr. Coupland for pushing me to do my very best and completely expanding my view of
food science. I cannot begin to explain how much I have learned from you.
To the members of the Coupland, Elias, and Hayes lab. Thank you for your support and
feedback. Alyssa and Cori, you have especially been instrumental in my progress here.
Thank you for taking the time to brainstorm solutions with me and helping me design
my sensory experiments.
To Charlene. Thank you for being my best friend throughout my time here at Penn State.
You are an amazing, strong role model and I don’t know what I would have done
without our television binges and classy cocktails. Long-live Movie Tuesday! To Erin.
Thank you for being a fabulous friend to me and the voice of reason even from New
Jersey. The Mars candy you provided me literally fueled this thesis.
To my family. I am eternally grateful for the guidance and love you give me every day.
My mom, thank you for being my rock and confidante. You are my partner in crime and
the woman I strive to emulate. My dad, thank you for encouraging me to attain my
Master’s degree and providing advice all the way through. DJ, thank you for teaching me
to take life less seriously and enjoy the ride.
x
CHAPTER 1
Factors Affecting Bitter Taste
1. Introduction and Review of Related Literature
1.1 The Basic Tastes
It is broadly accepted that there are five prototypical tastes: sweet, sour, salty, umami,
and bitter. The last of these – bitterness – is the primary focus of this thesis. However, it
should also be noted that in addition to the classical five, there are several additional
qualities that have demonstrated the necessary characteristics to be considered
prototypical tastes, including metallic and fat (Keast and Costanzo 2015). The taste of fat
is relevant here due to its presence as an independent variable. Specifically, fat is used
to reduce the perception of bitter in several experiments, but it is assumed that that the
unique taste of fat (‘oleogustus’) does not have an effect on bitter in and of itself (Keast
and Breslin 2002), though fatty acids have demonstrated some reduction of tastant
intensity (Mattes 2007).
The origination of basic tastes is most typically framed in terms of evolution (Shi 2005;
Chaudhari and Roper 2010). According to this theory, each taste detects a class of
evolutionarily important stimuli in the environment. Sweet detects carbohydrates
yielding some measure of energy density; salty taste detects ions such as Na+ that play
an important role in water balance and blood circulation; umami detects L-amino acids,
particularly L-glutamate, which gives an indication of protein content; sour detects free
hydrogen atoms from dietary acids preventing ingestion of unripe fruits, and bitter is
innately aversive to protect against ingestion of poisonous compounds (Chaudhari and
Roper 2010).
Bitter, in particular, is most commonly referenced as being important to the evolution of
vertebrates via a poison-detection role (Glendinning 1994). Potentially harmful
compounds like some secondary plant metabolites, rancid fats, and most importantly
almost all naturally-occurring environmental toxins have a very bitter taste for every
animal phylum (Glendinning 1994). Because of this, scientists have attributed the bitter
taste to a defense mechanism against poisons. It makes sense, then, that some plants
have evolved to become more bitter discouraging their consumption (Breslin 2013).
Avoiding deleterious compounds was more evolutionarily favorable to survival than
consuming enough calories; therefore, the taste of bitter is much more sensitive than
that of pleasant tastes such as sweet and umami. Based on the poison detection theory,
it is unsurprising that compounds that are recognized as bitter are abundant and
extremely diverse in their structure (Coupland and Hayes 2014). The main argument
against this theory is that if bitterness intensity is correlated with poison, it would mean
that greater degrees of toxicity should been associated with lower detection thresholds.
Accordingly, threshold of poisonous compounds should correlate with LD50. Empirical
evidence suggests this is not the case, although there is an explanation that sensitivity
1
may have adjusted over time for likelihood of encountering in the diet (Glendinning
1994).
1.2 The Mechanics of Taste
Please refer to Figure 1.1 throughout this section. The prototypical tastes are perceived
through taste buds in papillae on the tongue and in the throat and soft palate. There are
four classes of papillae on the tongue, but only three contain taste buds. The fungiform
papillae are found on the front portion of your tongue; foliate papillae are located on
the sides of the tongue; and circumvallate papillae are located at the back of the tongue
(Matsuo 2000). The taste buds, a group of 50-100 epithelial cells, are situated on the
sides of deep grooves in the foliate and circumvallate papillae. In fungiform papillae,
taste buds sit on top of the surface of the tongue (Matsuo 2000).
To be perceived, a tastant must first dissolve in the saliva (Figure 1.2) and then diffuse
across a mucus layer to reach taste cells that are located within taste buds (Matsuo
2000; Chaudhari and Roper 2010). The importance of saliva in taste has been
demonstrated such that individuals with decreased salivary production experienced
reduced overall taste perception (Matsuo 2000). Additionally, there are two main
channels involved in oral processing—one of which heavily depends on saliva. 1)
Chewing wherein the food is broken down and 2) impregnation by saliva. This creates a
cohesive mixture ready to be swallowed (Salles et al. 2011). Specific salivary
components will be discussed later.
There are three major types of taste cells. Type I, also called a glial-like cell, is
differentiated by a potassium ion channel that is believed to regulate the potassium
levels in the taste bud (Dvoryanchikov et al. 2009). It is the most prevalent taste cell, yet
the least is known about Type I cells. Salty taste recognition is attributed to Type I cells
due to the ionic currents important to their function (Vandenbeuch et al. 2008). When a
bitter, sweet, or umami stimulus is ingested, it is recognized by a Type II cell — a taste
receptor cell that expresses G protein-coupled receptors (GPCRs). This class of taste cells
is further split into two distinct groups. Type 1 (T1R) and Type 2 (T2R) differentiate
between receptors that recognize sweet and umami stimuli, T1R, or bitter stimuli, T2R
(Chaudhari and Roper 2010). A Type 1 or 2 cell can only express one GPCR specific to a
single taste quality. In other words, one taste receptor cell cannot recognize a
compound as bitter and savory, only one or the other. Bitter is believed to occur
exclusively via individual receptors expressed on Type 2 cells whereas sweet and umami
occur via heterodimers expressed on Type 1 cells (Meyerhof et al. 2005). Type III cells
are called presynaptic cells. They are characterized by proteins that form direct
connections with nerve terminals and calcium ion channels that signal neurotransmitter
release (Yang et al. 2000; Yee et al. 2001; DeFazio 2006). Type III cells have been
implicated in sour taste perception; much remains unknown about this class of cells, but
a decrease in cytoplasmic pH from organic acid triggers the calcium ion transport
channel to release serotonin (5-HT) and/or epinephrine (NE) (Chaudhari and Roper
2010). Type III cells are also involved in sweet, umami, and bitter taste. Type II receptor
2
cells release ATP after stimuli is detected which simulates nerve fibers sending the taste
quality to the brain; however, it also excites nearby Type III presynaptic cells that
release 5-HT and/or NE as a result (Chaudhari and Roper 2010).
Because this thesis focuses specifically on bitter taste, transduction of the bitter taste
stimuli is outlined in greater detail below. The Type 2 taste receptor cell is a family of
GPCRs. While T1Rs are large complexes that feature a “Venus Flytrap” structure to bind
stimuli, T2Rs are proteins that resemble helices with binding sites on the interior
(Floriano et al. 2006; Chaudhari and Roper 2010). The hidden structure of T2Rs makes
the understanding of how they work less clear than the T1Rs.
Adding to the complexity of T2Rs, 3 functional genes code for T1Rs in humans as
compared to 25 functional and 11 pseudogenes that code for T2Rs (Shi 2005). In
addition, T2Rs have different molecular receptive ranges. Some receptors are
exclusively tuned to one ligand whereas others recognize many, with the latter being
called ‘promiscuous’ receptors (Slack et al. 2010). Different patterns of receptor cells are
expressed across cells which means that we can discriminate between bitter compounds
(Behrens et al. 2007). This heterogeneity was investigated in situ using Ca2+ direct
imaging in which most taste receptor cells were activated by multiple compounds, and
of those, there was no significant association between any two stimuli activating a
specific pattern of taste receptor cells (Caicedo and Roper 2001).
Following ligand recognition of a bitter compound, a phosphoinositide pathway is
activated. G protein subunits (α and βγ) are freed from the GPCR and interact with
phospholipase (PLCβ2). IP3 is then synthesized—a molecule that stimulates an ion
channel in the cell. This ion channel releases Ca2+ into receptor cells, targeting two
locations: 1) TRPM5, a taste-selective ion channel which creates an action potential in
the receptor cells and 2) a gap junction hemichannel. This causes the release of ATP that
stimulates nerve fibers in the taste bud sending the signal of bitter to the brain.
Secondarily, the ATP travels to nearby Type III cells for further action as reviewed above
(Chaudhari and Roper 2010).
It should be noted that while genetic differences in sensitivity of bitter compounds do
exist, this should not affect this particular study where caffeine and quinine are used as
model bitterants. There is no known genetic variation for caffeine sensitivity, and there
is weak evidence for quinine sensitivity genetic variation.
3
Figure 1.1 From Mennella et al. (2013). How bitter works: the process of bitter
perception. The generation of bitter taste starts when a bitter compound enters the oral
cavity, where the ligand binds to one of a class of G protein-coupled receptors (GPCRs).
These receptors, called TAS2Rs, are expressed in the apical membrane of receptor cells
found in the taste buds, triggering a cascade of signaling events, leading to the release
of neurotransmitter that activates an afferent nerve fiber that transmits the signal via
the cranial nerve to the brain. Taste buds are distributed in distinct fields in the oral,
pharyngeal, and laryngeal epithelia, with each field innervated by a different cranial
nerve branch. Only the taste buds on the tongue are depicted in the figure. The taste
signals course through the brain and provide input to the circuits that subserve various
functions.
The figure illustrates the complexity of the mechanisms intervening between the
application of the bitter stimulus and the generation of the behavioral response,
providing a variety of potential targets for strategies to modulate the bitterness of
medications. DAG = Diacylglyceral; Gαgus = G-protein subunit α-gusducin; Gβγ = Gprotein subunits β and γ; IP3R = inositoltriphosphate receptor; PIP2 =
phosphatidylinositol 4,5-biphosphate; PLCβ2 = phospholipase C β2; TrpM5 = transient
receptor potential ion channel subfamily M member 5; VPMPC = ventral posterior medial
nucleus, parvicellular subdivision. *The insula/operculum is actually lateral to the
sagittal plane of section shown.
4
Figure 1.2. From Matsuo (2000). Schematic diagram of a taste bud (A) and model of the
initial events in taste perception (B). (A) Microvilli extend from the apical portion of
taste cells into the taste pore. Taste stimulants must enter and diffuse through the fluid
layer to come into contact with receptor sites on the microvilli. (B) Taste sensitivity is
affected by the solubility of taste substances in saliva and in the taste pore material and
by the chemical interaction with various components of saliva, before coming into
contact with the receptor site. Taste receptors normally adapt to saliva, resulting in a
decrease or increase of their sensitivity.
1.3 Effects of Saliva
Saliva is a critical part of the oral environment; while it is mostly water (~99%),
electrolytes, proteins, and urea are also some of many minor components that
contribute greatly to the function of saliva (Humphrey and Williamson 2001).
Electrolytes and urea provide saliva with some buffering capacity. Large proteins such as
mucins produce lubricating, protective mucus layers in the mouth (Duxbury et al. 1989).
And immunoglobulins, enzymes, and various proteins act as antibacterial agents
(Humphrey and Williamson 2001). Additionally, components of saliva have been
identified as being important to how we sense taste active molecules. At a fundamental
level, low levels of glucose in saliva creates a hypotonic environment which actually
enhances taste perception (Humphrey and Williamson 2001).
Proteins, in particular, have been implicated in taste responsiveness. These proteins
make up a majority of the remaining 1% of non-water saliva composition (Dsamou et al.
2012). A common way of studying the connection between salivary protein and taste
perception performance is through correlating panelist salivary profiles with detection
thresholds of various compounds. For example, groups of caffeine-hypersensitive
individuals possessed saliva with higher levels of amylase, immunoglobulin, and serum
albumin, along with lower levels of cystatin SN, a protease inhibitor, as compared to
caffeine-hyposensitive individuals (Dsamou et al. 2012). Cystatin SN is a protein specific
to saliva that prevents cysteine proteolysis, so the authors concluded proteolysis within
the oral cavity was a perireceptor factor associated with caffeine bitterness sensitivity
5
(Dsamou et al. 2012). Relatedly, salivary protein profiles of young infants are often used
in the literature because of the lack of desensitization to tastes. In a study by Morzel et
al. (2014), 73 infants were profiled at 3 and 6 months and intake ratios of bitter urea
solution were collected. At 3 months, the most significant component of saliva that
predicted intake ratios was, again, cystatin. Higher amounts of cystatin were correlated
with higher bitterness acceptance, which was interpreted as lower sensitivity to
bitterness. A separate, smaller group of infants saw this same trend, though it was not
significant. At 6 months, there was no significant pattern in salivary profiles (Morzel et
al. 2014).
Proline-rich proteins (PRPs) are an excellent example of salivary components that are
incredibly important to sensations elicited by foods. There is strong support for the
theory that interactions with PRPs are primarily responsible for astringent sensations
(Baxter et al. 1997). In addition, basic PRPs (bPRPs) have been observed to influence
bitter taste sensitivity. Panelists with a 6-n-propylthiouracil (PROP) “supertaster” status
secreted bPRPs in response to PROP introduction into the mouth (Cabras et al. 2012).
Oral supplementation of PROP “nontaster” panelists with peptides (PS-1 and II-2) from
the bPRP family resulted in an increase of PROP bitterness intensity (Melis et al. 2013).
H-NMR confirmed interaction between PROP and PS-1. The researchers concluded that
bPRPs could be responsible for orienting the PROP molecule to optimize its binding in
the TRC binding pocket, though this was not specifically observed or measured (Melis et
al. 2013). It was also theorized that a strong genetic correlation between PRPs and
quinine bitterness perception indicated that PRPs were important for bitterness
perception (Lush and Holland 1988; Spielman 1990). Subsequent binding experiments
between rabbit PRP and quinine showed no interaction between the two (Spielman
1988; Spielman 1990). This suggests that PRPs do not directly influence the bitterness
perception of quinine.
One of the most prominent salivary proteins in the literature relevant to taste
perception is the von Ebner’s gland (VEG) protein—later renamed lipocalin-1. The VEG
produces saliva to fill the furrows of circumvallate and foliate papillae where most taste
receptors are located (Matsuo 2000). Lipocalin-1 is rather large, 18 kDa, and highly
expressed in the VEG (Schmale et al. 1990). Its structure is rather similar to a protein
responsible for delivering odorants called the odorant binding protein. It is for that
reason and the fact that it sits directly inside of the trough of papillae that researchers
hypothesized lipocalin-1 was responsible for delivering hydrophobic bitterants to the
taste receptors (Schmale et al. 1990). In order for a protein to be responsible for
delivering bitterants, it needs to have a broad range of specificity. Studies following the
discovery and sequencing of lipocalin-1 revealed that many bitterants do not show
interaction with lipocalin-1 (Creuzenet and Mangroo 1998). Additionally, mice do not
secrete lipocalin-1, yet they have very similar response patterns as rats, who do secrete
lipocalin-1, to bitter substances (Schmale et al. 1993). Nonetheless, lipocalin-1 still
appears to be an important salivary protein for taste. It is one of the main arguments
being made for the taste of fat. Lipocalin-1 has demonstrated binding behavior to fatty
acids (Tucker et al. 2014).
6
Finally, no discussion of salivary proteins and perception would be complete without
mentioning mucin. Mucin is a very large, heavily glycosylated protein that is extremely
important for maintaining lubrication in the mouth (Duxbury et al. 1989). 16% of the
proteins in saliva are mucins (Rayment et al. 2000). While less-prominently studied in
the taste literature, mucins have been hypothesized to play a role in astringency (Fábián
et al. 2015). They have also improved function of polyphenols as antioxidative agents by
improving their solubility in water (Ginsburg et al. 2012), suggesting mucins may be
doing something similar with bitterants in saliva.
1.4 Altering Taste Perception
Currently, there are three accepted methods in sensory literature that can be used to
suppress taste (Galindo-Cuspinera 2011). The first is peripheral interactions in which an
added compound interacts with a tastant or taste receptor, thereby altering interaction
between the tastant and the receptor (Keast and Breslin 2002). The most commonly
researched uses of peripheral interactions are salts like ZnSO4 or MgSO4 in which taste
receptor cell integrity is altered (Gaudette and Pickering 2013). The second is central
cognitive suppression in which a strong taste or aroma reduces the perception of
another taste in the brain—this effect is not caused by any physical interactions
between the compounds (Lawless 1979). Using sugar to suppress bitterness of a wide
range of compounds with different structures is a great example of this method
(Mennella et al. 2014). The third mechanism is encapsulation—preventing interaction
between the compound and taste receptors through physical means (Douroumis 2007;
Galindo-Cuspinera 2011; Hoang Thi et al. 2012). It is a focus on this third technique that
is the goal of this research—specifically protein and fat as components in foods. It is
worth noting that there is some overlap of these techniques, and it is likely that the
most effective bitter taste suppression will use a combination of them (Coupland and
Hayes 2014). For example, NaCl has been shown to be effective in reducing bitterness at
low intensities via primarily peripheral interactions through taste receptor cells, but as
much as 22% of the suppression is caused by central cognitive effects (Kroeze and
Bathoshuk 1985). Because bitterness suppression is so important to product
development in the food and pharmaceutical industries, the majority of the proposed
technology is documented in the patent literature (Ley 2008). This is problematic for
advancing our understanding, as the patent literature tends to document effects
without exploring the specific mechanisms that are involved.
As encapsulation is the main mode of taste suppression that has been investigated to
date, a brief review of current research in the area is relevant. Three important
parameters for tastant encapsulation are solubility, partitioning, and binding
phenomena as outlined by McClements (2015). Cyclodextrins, particularly γ- and βcyclodextrins, are one of the most effective encapsulation techniques (Brewster and
Loftsson 2007). The cyclic oligosaccharide ring structure creates a hydrophobic interior
and hydrophilic exterior increasing the solubility of hydrophobic ingredients via noncovalent interactions (Brewster and Loftsson 2007). The inclusion complex that is
formed by β-cyclodextrin shields the ingredient inside from interacting with taste
7
receptor cells, making it a popular approach in pharmaceutical applications (Gaudette
and Pickering 2012). The hydrophobic method of this interaction translates to γ- and βcyclodextrin being the most effective for primarily hydrophobic ingredients (Gaudette
and Pickering 2012). For example, γ-cyclodextrins reduced the bitterness of ginseng
from a 4.6 rating on a 6-point scale to a 1.0—the lowest bitterness rating on the scale
(Tamamoto et al. 2010). β-Cyclodextrin also demonstrated significant reduction of
catechin bitterness by 60% and was the most effective of all of the treatments including
other bitter blockers (Gaudette and Pickering 2012).
This thesis uses protein binding and fat partitioning methods of encapsulation to reduce
bitterness. Binding phenomenon and lipid partitioning will be reviewed separately with
considerations and background for each in the following sections.
1.5 Considering Protein
1.5.1 Protein Function and Binding Characteristics
Proteins are a large class of macromolecules responsible for a wide variety of biological
functions. Their primary structure is comprised of amino acid residues that vary in their
hydrophobicity. That hydrophobicity influences how the protein spontaneously folds
because the aqueous environment surrounding it will favor interaction with the more
hydrophilic components of the protein. Therefore, protein folding will tend to maximize
the amount of hydrophilic residues on the exterior by folding as many of the
hydrophobic residues into the interior of the protein although specific interactions
between amino acids such as hydrogen bonds also play a role (Coupland 2014).
Fluid milk is comprised of approximately 30-36 g/L protein (Kimpel and Schmitt 2015) of
which there are two main classes: caseins and whey proteins. The percentages of each
of these are outlined in Table 1.1. Casein is the majority of protein in milk, and, while
they have been sequenced, their secondary, tertiary and quaternary structures remain
ill-defined or unknown. A widely accepted model for casein in milk is the proteins form
a large “micelle” with an open interior that are stabilized by hydrophobic interactions
and phosphate components forming bridges and nanoclusters (Livney 2010). κ-Casein, a
subclass of casein proteins, is located on the exterior of these micelles and creates steric
hindrance which prevents aggregation of casein micelles (Kimpel and Schmitt 2015). The
open interior of a casein micelle is hydrophobic and can serve as a carrier of
hydrophobic small molecules (e.g., Vitamin D, calcium, and phosphorous) (Livney 2010;
Cheema et al. 2015). Additionally, the casein micelle has also been studied as a delivery
system for hydrophobic compounds such as bioactive ingredients in pharmaceuticals
(Semo et al. 2007; Cheema et al. 2015).
Whey proteins make up approximately 20% of milk proteins and consist of βlactoglobulin (BLG), α-lactalbumin, bovine serum albumin (BSA), and various
immunoglobulins (Table 1.1, Table 1.2). Whey protein isolates (WPI) and concentrates
(WPC) are powdered food ingredients that are made from the whey fraction of milk.
They are used in many food and pharmaceutical formulations due to their high protein
8
content, emulsifying behavior, and gelling ability (de Wit et al. 1988; Morr and
Foegeding 1990). The percentages of each major whey protein in commercial WPI or
WPC is given in Table 1.2.
Whey proteins are typical globular proteins in structure, with the majority of their
hydrophobic residues on the interior as outlined above. The hydrophobic parts of the
protein are capable of binding hydrophobic small molecules. BLG, has been shown to
bind fatty acids (Wu et al. 1999), esters (Pelletier et al. 1998), ketones (O’Neill and
Kinsella 1987), aldehydes, alcohols, and lactones (Guichard and Langourieux 2000) more
strongly when the small molecules have longer hydrocarbon chain lengths. The
strongest evidence for the location and mechanism of binding is that a β-barrel pocket
in the interior of the protein binds hydrophobic molecules with their most hydrophobic
portion sliding into the barrel space (Wu et al. 1999; Kontopidis et al. 2004). For
example, computational modelling has shown the fatty acid tail of palmitic acid fits
directly within the barrel confirming that it is the main hydrophobic binding pocket of
BLG (Wu et al. 1999). Additionally, Guichard and Langourieux (2000) showed that aroma
perception decrease was correlated with BLG binding strength increases. Those aroma
compounds with the greatest affinity for BLG had the most significant decreases in
intensity (Guichard and Langourieux 2000). BSA, another whey protein component, also
binds small molecules via hydrophobic interactions. For example, NMR was used to
confirm a binding complex between BSA and the high-intensity sweetener,
Rebaudioside A (RebA) (Mudgal et al. 2016).
Table 1.1. Protein composition of milk. Adapted from Morr and Ha (1993).
Component
Total milk protein
Caseins
Whey proteins
β-Lactoglobulin (BLG)
α-Lactalbumin
Bovine serum albumin
(BSA)
Immunoglobuins
Minor proteins
Fat globule membrane
proteins
Concentration
in milk (g/L)
33.0
26.0
6.3
3.2
1.2
0.4
Percentage of
total protein (%)
100
79.5
19.3
9.7
3.7
1.2
Percentage of
whey protein (%)
100
56-60
18-24
6-12
0.8
0.8
0.4
2.4
2.4
1.2
6-12
-
9
Table 1.2. Physicochemical properties of whey proteins and composition in commercial
whey protein products. All values compiled from Morr and Ha (1993) unless otherwise
noted.
Isoelectric Molecular Average
point
weight
hydrophobicity
(daltons) (kcal/residue)
5.2
18,000
1075
4.2-4.5
14,000
1020
4.7-4.9
66,000
995
β-Lactoglobulin
α-Lactalbumin
Bovine serum
albumin
Immunoglobulins 5.5-8.3
≥ 146,000
a
Values from Morr and Foegeding (1990).
NA
Mean
composition
in WPC (%) a
67.3 ± 10.6
17.5 ± 3.0
8.6 ± 4.2
Mean
composition
in WPI (%) a
70.2 ± 3.3
14.3 ± 4.3
8.6 ± 1.6
7.2 ± 4.4
6.9 ± 0.7
1.5.2 The Influence of Protein on Taste
Proteins themselves do have reported intrinsic tastes, though these vary. Umami, for
example, is one of the basic tastes that is elicited by the amino acid L-glutamate as
touched on in the basics of taste section (Chapter 1.1). Umami is believed to provide a
sense related to the protein content in foods (Shi 2005). Additionally, peptides are
commonly perceived as having a bitter taste—particularly after hydrolysis (Tavano
2013). Separately, some proteins have an intrinsic sweet taste, though the mechanism
of this taste is largely unknown (Masuda and Kitabatake 2006). However, large intact
proteins, such as whey proteins, do not have a detectable intrinsic taste in isolation but
they can have an effect on taste indirectly by influencing other tastants.
Suppression of bitterness and other tastes can be accomplished by preventing
interaction of the bitterant with the taste receptors through physical means such as
encapsulation (Matsuo 2000; Coupland and Hayes 2014). Proteins, particularly dairy
proteins, are self-assemblers and form encapsulation complexes naturally (Livney 2010).
Therefore, the non-tastant ingredients in a food or drink can be very important to how
one experiences the taste. For example, proteins have the ability to bind small
molecules and have demonstrated the capacity to alter taste and aroma perception
(Bohin et al. 2013).
Soy protein was shown to bind a small fraction of the sodium ions in solution (5.7%) but
there was no effect on perceived saltiness as compared to in plain water (Mosca et al.
2015). In another saltiness study, tomato soup prepared with added skim milk was less
salty than soup prepared with added water (Rosett et al. 1997), but this is not
necessarily due to the proteins. Sugar, present as lactose in the milk, has been shown to
influence the other basic tastes (Keast and Breslin 2002). Due to this ambiguity that food
products present, more controlled systems will provide clearer relationships between
protein binding and sensory perception. I will briefly review some of the important
studies that investigate the effects of protein binding on taste, with an emphasis on
bitter.
10
Pripp et al. (2004) demonstrated that caseinate reduced the bitterness of phenolics
from olive oil. In addition, Bohin et al. (2013) reported a large bitterness-masking effect
from β-casein on EGCG after establishing its strong interaction and incorporation into
the micelle (Bohin et al. 2012). The researchers also measured ‘bitterness” in vitro with
an hTAS2R39 taste receptor assay. Interestingly, BLG had poor binding capacity with
EGCG, but still significantly reduced the activation of the receptor (Bohin et al. 2013). In
addition, thermally denatured BLG has enhanced binding with EGCG and has been
shown to suppress bitterness and astringency (Shpigelman et al. 2010; Shpigelman et al.
2012).
The sweetener rebaudioside A (RebA) also has a bitter off-taste believed to be due to
the presence of a diterpene group. BSA binds RebA hydrophobically via this group and,
while no sensory work was done on the complex, it was hypothesized that BSA would
suppress the bitterness of RebA (Mudgal et al. 2016).
While binding remains the dominant mechanism proposed for the effects of protein on
bitter taste, alternatives have been proposed. for example, Maehashi et al. (2008)
showed riboflavin-binding protein (RBP), an egg glycoprotein, bound quinine but not
caffeine yet depressed the bitterness of both compounds to same degree. They
hypothesized that RBP interacts directly with the taste receptors—possibly through
competitive binding – to suppress bitterness and complexation with the bitterant is not
required. The results with RBP have not been repeated with other proteins and the
interaction of a protein with a bitter taste receptor has not been demonstrated directly.
Lastly, as discussed Section 1.3, proteins in saliva have been shown to enhance taste.
Panelists sensitive to the taste of 6-n-propylthiouracil (PROP) were observed to secrete
more basic proline-rich proteins (bPRPs) than those who could not taste PROP (Cabras
et al. 2012). Non-tasters of PROP supplemented with these bPRPs reported statistically
higher sensitivities to PROP than the bitterant solution on its own suggesting that
proteins can play a role in enhancing the bitter taste of a compound (Melis et al. 2013).
The proposed mechanism for this function is that bPRPs bind the bitterant and transport
it to the taste receptor however interaction with the taste receptor and in-mouth
modeling have not been reported.
Proteins have exhibited significant effects on taste and flavor and the primary
mechanism proposed is binding of the stimulus molecules. However, other
considerations such as direct interaction of the protein or complex with taste receptors,
the presence of salivary protein, and the nature of the binding that may influence any
effect.
11
1.6 Considering Lipids
1.6.1 Lipid Chemistry and Partitioning
Lipids are a large class of compounds that are generally soluble in organic solvents. Food
lipids are broadly classified as polar (e.g. phospholipids) or nonpolar (e.g. triglycerides)
which indicates differences in their properties and functionality. The bulk of food oils
are nonpolar triglycerides – tri-esters of glycerol with various fatty acids. The
physicochemical characteristics of each triglyceride are dependent on the fatty acids in
the molecule. For example, polyunsaturated fatty acids have low melting points leading
to liquid oils and are prone to oxidation while saturated fatty acids have high melting
points leading to solid fats (depending on the temperature) and are not prone to
oxidation. Other factors such as chain length or cis/trans conformation of the double
bonds will also lend different characteristics to the lipid.
A stable dispersion of oil and water is called an emulsion. Either the oil or water is
broken down into small droplets through mechanical action and suspended in the other
phase. The two phases are differentiated by defining which substance is contained
within the other creating the continuous and dispersed (or discontinuous) phases. For
example, an oil-in-water (o/w) emulsion consists of an oil phase dispersed in a
continuous water phase. Triglycerides are immiscible in water because of the
hydrophobic effect, so a stabilizer (aka an emulsifier) must be used to reduce the
interfacial tension between the phases. The emulsifier absorbs onto the surface of
dispersed droplets and prevents the phase separation of oil and water.
At equilibrium, the activity, but not necessarily the concentration, of a molecule in a
system is the same in all phases in which it occurs. Emulsions are multi-phase materials,
and solutes in the mixture will have different activity coefficients depending on their
structure. That will lead the solutes to concentrate in either the lipid phase or the water
phase. For example, hydrophobic molecules will tend to favor the oil phase in an
emulsion. In order to determine how hydrophobic (organic solvent-soluble) a molecule
of interest is, a partition coefficient is used. For octanol-water partitioning the ratio of
concentrations is the partition coefficient P (often reported as log P):
𝑃=
[𝑐𝑜𝑚𝑝𝑜𝑢𝑛𝑑] 1−𝑜𝑐𝑡𝑎𝑛𝑜𝑙
[𝑐𝑜𝑚𝑝𝑜𝑢𝑛𝑑]𝑤𝑎𝑡𝑒𝑟
[1.1]
Where [compound] represents the concentration of compound in each respective
phase. If there are no chemical interactions between the molecule and either phase,
there will be an equal concentration in both the octanol and in the water and P=1 (log P
= 0). If the molecule is more hydrophobic than hydrophilic, it will tend to partition more
favorably in the non-polar phase (i.e., positive log P) and vice versa for more hydrophilic
molecules. Generally, water and octanol are used as the two phases as octanol was
thought to be representative of a membrane lipid phase, making log P a useful
parameter for pharmaceutical studies. However other phases can be used so the ratio is
more representative of the system under consideration. For example, in an emulsion,
the particular oil can be substituted for octanol. The partition coefficient (Kow) is then:
12
[𝑐𝑜𝑚𝑝𝑜𝑢𝑛𝑑] 𝑜𝑖𝑙
𝐾𝑜𝑤 = [𝑐𝑜𝑚𝑝𝑜𝑢𝑛𝑑]
𝑤𝑎𝑡𝑒𝑟
[1.2]
Equation 1.2 can be used to calculate how changing the phase volume of an emulsion
affects the aqueous concentration of a compound (Figure 1.3). For the hydrophobic
molecule (Kow = 1000), aqueous concentration decreases rapidly as the oil increases in
the mixture until it is almost exclusively in the oil phase at equal parts oil and water.
Alternatively, the hydrophilic molecule (Kow =0.001) increases in aqueous concentration
as it becomes concentrated in the smaller water volume. A small amount of added fat
would therefore have a huge effect on the aqueous concentration of a hydrophobic
molecule but only a very small effect on a hydrophilic molecule.
Figure 1.3. Effect of oil volume fraction on the aqueous concentration of a hydrophilic
(Kow = 0.001), neutral (Kow = 1), and hydrophobic (Kow = 1000) compound. From Coupland
and Hayes (2014).
1.6.2 The Influence of Lipids on Taste
As described in Section 1.1, there is a growing body of evidence that fat has a taste
(Keast and Costanzo 2015). However, it was assumed in this work that the possible taste
of fat does not have a direct effect on the perception of bitter (Keast and Breslin 2002).
While fatty acids have been shown to affect the intensity of several tastants, bitter
intensity was unaffected (Mattes 2007).
Suppression of taste intensity can be accomplished by preventing interaction of the
bitterant with the taste receptors through physical means such as binding or
partitioning (Matsuo 2000; Coupland and Hayes 2014). Consequently, the composition
of the food or drink ingested can be very important to how a consumer experiences the
molecules responsible for taste and flavor. In the most extreme case, a molecule that is
so tightly bound by food that it cannot be released into the saliva will not be tasted.
More likely though, weaker binding may serve to suppress the concentration in aqueous
saliva and thus the taste. In the case of an oil-water emulsion, a bitterant that is
13
extremely nonpolar (e.g. the Kow = 1000 compound in Figure 1.3 above) will minimize its
interaction with saliva, and have a less strong taste. In this section I will describe
examples of how this idea has been explored in various studies.
Mackey (1958) showed more panelists could correctly identify quinine, caffeine and
saccharin tasted in water rather than in corn oil and the difference was greatest with
the most hydrophobic stimulant, quinine. Similarly, oil mouthcoating has an effect on
bitterness; panelists demonstrated significantly lower bitterness perception of quinine
sulfate after pre-rinsing with oil (Lynch et al. 1993). These papers indicate that oil
hinders the perception of hydrophobic tastants, presumably due to partitioning effects.
Generally, the effects of fat on taste are not encountered in bulk oils but rather in food
emulsions because of their positive sensory appeal. Simple emulsions are particularly
useful in experimental studies of taste as the composition can be systematically varied
and because they avoid possible interactions with other food components. In one such
study, Metcalf and Vickers (2001) compared the effect on taste intensity of diluting
tastant solutions with either water or oil in the form of emulsions (final oil
concentration 9% or 17%) with Tween 80 as the emulsifier. In this way, the researchers
could compare total and aqueous concentration of tastants to the intensity perceived by
panelists. For example, adding an amount of water to a tastant solution will dilute the
tastant concentration and result in a weaker taste. Adding an equivalent amount of oil
to a tastant solution will dilute the overall concentration of the tastant, but a hydrophilic
tastant will presumably remain in the water phase of an oil/water emulsion and the
aqueous concentration will not change. Therefore, the researchers were able to
determine whether the total or aqueous concentration of a tastant was more indicative
of taste perception by comparing the original tastant control solution to the oil and
water dilution samples. If the taste intensity of an oil-diluted solution does not differ
significantly from the water-diluted solution, then the total concentration of the tastant
is indicative of taste perception. If the taste intensity of an oil-diluted solution does not
differ significantly from the control solution, then the tastant is remaining in the
aqueous phase of the emulsion and aqueous concentration is determining taste
perception.
The hydrophilic tastants—sucrose (sweet), citric acid (sour), sodium chloride (salty),
monosodium glutamate (umami)—all had no significant increase or decrease in intensity
when comparing the oil dilutions to the control. However, the taste intensity did
decrease for the water-diluted samples indicating that aqueous concentration, not total
concentration, is most important for how these tastes are perceived. In other words,
adjusting for the overall dilution returns the tastants to their original intensities. This
can be related back to Figure 1.3. The Kow = 0.001 compound is the most similar to these
hydrophilic tastants. As the lipid fraction increases in the emulsions, the aqueous
concentration gradually increases as the phase volume of water decreases. However,
the levels of oil used in these experiments (9% and 17%) would likely not have created
enough of a concentration effect to significantly increase the intensities.
The bitterness of hydrophobic quinine sulfate, was more significantly reduced with an
addition of oil than with an addition of water. This indicates that the effect of oil on the
14
bitterant was greater than the intensity decrease caused by dilution alone. Additionally,
the bitterness suppression caused by the oil suggests that quinine sulfate has a stronger
affinity to oil such that the aqueous concentration in the emulsion is lower than the
control. It is worth noting that the increase in oil from the 9% to the 17% oil emulsion
did not cause significant further reductions in bitterness (Metcalf and Vickers 2001).
Literature reports the Log P of quinine as being at least 3 (i.e., P=1000) (Hansch et al.
1995; Zissimos et al. 2002; Barzanti et al. 2007); this means that the molecule is
favorably partitioning into the oil phase of an emulsion such that increasing the oil
following an initial introduction as small as the 9% used here does not have a further
partitioning effect. Again referring to Figure 1.3, quinine can be represented by the Kow =
1000 curve, and this partitioning effect can be seen. Oil at very small amounts
dramatically decreases the aqueous concentration. However, the aqueous
concentration at 9% and 17% oil on the curve are not very different.
The overall conclusion from this paper was that aqueous concentration, not total
emulsion concentration, determines overall perception which agrees with the assertions
put forth by other taste researchers (Matsuo 2000; Chaudhari and Roper 2010).
However, this model may not tell the whole story. Suzuki and others (2014) used a
comparable methodology for the study of saltiness of NaCl in emulsions with different
fat contents (0-40%) (Suzuki et al. 2014). The researchers began with aqueous solutions
of salt and added emulsified oil so that the total salt concentration decreased as the oil
fraction increased. However, because NaCl is so hydrophilic, the aqueous concentration
would theoretically remain unchanged, though the researchers did not measure this.
For example, the 0.4% [NaCl]aq 40% emulsion has a 0.24% NaCl total concentration and
an assumed 0.4% aqueous concentration. The researchers observed that at high NaCl
concentrations, the addition of fat decreased the saltiness intensity. This effect becomes
less prominent at lower NaCl concentrations and goes away altogether at the lowest
NaCl concentration of 0.03% NaCl. While Metcalf and Vickers (2001) observed no
difference between the control and oil dilution saltiness intensities, Suzuki et al. (2014)
saw a suppression effect. Rather than a direct effect from oil on the salt concentration,
this suppression effect may have been due to higher viscosity of the higher volume
fraction emulsions; increases in thickness have been shown to suppress the intensity of
tastes (Rosett et al. 1997; Arancibia et al. 2011).
Similarly, Yamamoto and Nakabayashi (1999) studied saltiness in 35% and 70% oil
emulsions. Using the magnitude estimation scale, 0.9% NaCl in water scored 100, but
135 and 158 in 35% and 70% emulsions respectively, i.e., the total salt concentration
was the same in the three systems but the aqueous concentration and the taste
increased as the fat level increased. These workers went on to measure the aqueous
concentration of NaCl (by centrifugal separation of the phases) and found that in the
70% oil emulsion the aqueous phase contained 80% of the NaCl or 2.66 wt % NaCl (the
rest of the salt was presumably bound to the droplet surfaces). However, a 70% oil
emulsion was much less salty than an aqueous solution at the same concentration (158
vs 270) suggesting that the presence of fat droplets suppresses the taste of salt
(Yamamoto and Nakabayashi 1999). This suppression could be caused by the high level
of oil in the emulsion coating the tongue and preventing access of the hydrophilic ions
15
to the ion channels (Lynch et al. 1993). Additionally, studies show that emulsion
depletion flocculation can occur in the mouth due to salivary protein interaction
(Vingerhoeds et al. 2005). The Suzuki et al. (2014) and Yamamoto and Nakabayashi
(1999) studies indicate that the use of oil to study changes in taste can be more nuanced
than changes in concentration and partitioning alone.
In another well-controlled emulsion study, Thurgood and Martini (2010) used a 20% oilin-water emulsion system to measure the effect of lipid on the threshold of various
tastants, though the aqueous concentration was not measured, nor was it controlled for
as in the Metcalf and Vickers study (2001). Both quinine hydrochloride and citric acid
elicited higher thresholds of bitterness and sour tastes respectively. Overall, the
perception of these two tastants was suppressed in the presence of fat. While the
quinine result was expected due to its hydrophobicity, the citric acid suppression is
surprising due to its hydrophilic nature. On the other hand, MSG likely became
concentrated in the aqueous phase by the fat because it had higher intensities recorded
in the emulsion (Thurgood and Martini 2010). The slightly higher fat percentage of 20%
compared to the 17% used in Metcalf and Vickers (2001) study could explain why the
concentration effect was perceivable in this study and not the other, though this is
unlikely. At 20% oil, the aqueous concentration of the hydrophilic Kow = 0.001 compound
in Figure 1.3 is slightly higher than the concentration at 0% fat. Additionally, this paper
compared animal and plant-based oils and found no significant differences in the effects
which further indicates that these results were caused by the presence of a fat phase
and partitioning—not any differences in chemical composition of the fats.
The effect of fat on taste has been seen in other, more confounded, studies. With food
emulsions, studies using various levels of fat in milk are common because fat in dairy is
easy to manipulate. For example, Ares et al. (2009) concluded that milk fat was effective
at reducing the bitterness and astringency of the antioxidant extract from achyrocline
satureioides as compared to the extract in water, however there was no difference
between skim and whole milk. The extract is a mixture of both moderately hydrophobic
and amphiphilic molecules (Schwingel et al. 2008; Zorzi et al. 2015) yielding an overall
mixture that is neither predominantly hydrophobic or hydrophilic, so this lack of a fat
effect makes sense and it is possible that other components in the milk (e.g., protein
and lactose) played a role (Ares et al. 2009).
Conversely, Keast (2008) observed a concentration effect as caused by increasing fat
with the hydrophilic bitterant, caffeine. Milk fat at levels of 0, 2, and 4%, was evaluated
for its effect on bitterness intensity. The 4% fat milk significantly increased the
bitterness of caffeine. The increasing phase volume of milk fat decreased the overall
phase volume of water presumably concentrating it in the aqueous phase, though these
results are unexpected because of the very small amounts of fats used means the
change in aqueous phase volume is also small. As a comparison, Metcalf and Vickers
(2001), did not observe a significant difference in taste intensity of hydrophilic tastants
at levels of 9% and 17% oil. The caffeine can be treated as the Kow = 0.001 curve in
Figure 1.3 above (Klebanov et al. 1967), and following that curve, the oil amounts should
16
exceed levels of at least 20% before significant changes in aqueous concentration are
likely to be perceived.
While Keast (2008) measured an increase in hydrophilic caffeine bitterness with
increasing fat (0-4%) in milk samples (see above), a simple emulsion system showed the
opposite effect. A 20% canola oil emulsion increased the bitterness recognition
threshold and modestly suppressed the bitterness intensity of caffeine compared to a
water control (Torrico et al. 2015a; Torrico and Prinyawiwatkul 2015). These effects,
while significant, were very small and an increase of oil content (20% to 40% oil) showed
no significant differences to the 20% emulsion sample. The small effects at these fat
levels are expected due to the partitioning curve in Figure 1.3 for the Kow = 0.001
compound. However, the direction of effect is unexpected. A gum-based thickener was
used in the emulsions but not in the water control which may have confounded the
results by contributing to binding or changing the viscosity.
In another dairy fat study, sweetness in vanilla custards with higher levels of fats (3, 6,
and 12%) was investigated (Hoppert et al. 2012). Keeping viscosity and flavor constant
across samples, the sensitivity to sugar content increased with increasing fat content at
high sugar percentages. Presumably because it is a hydrophilic molecule, the sugar is
being concentrated in the aqueous phase of the system as the oil content increases.
However, the aqueous concentration increase by volume is unlikely to be important
enough to play a role in perception.
While not strictly partitioning, some workers have reported the formation of bitterant
complexes with fatty acids. For example, a study identified fatty acid fractions as the
most important in masking bitterness of quinine as an added bitterant in cheese
(Homma et al. 2012). Isothermal calorimetry identified a complex forming between the
fatty acids and quinine. NMR was later used to clarify the nature of binding (Ogi et al.
2015). Accessible nitrogen atom-containing bitterants were reported to show hydrogen
bonding with the fatty acid carboxyl group in those fatty acids with carbon chains longer
than 12. The researchers hypothesize that these interactions form insoluble complexes
that mask bitterness. Quinine complexed further with sodium laureate and crystallized.
Subsequent sensory testing claimed the bitterness of these crystals was significantly
lower than quinine hydrochloride in isolation. However, this experiment was not
explained clearly, and the actual results were not reported (Ogi et al. 2015).
Lipids have significantly affected taste acuity in many studies. The primary mechanism
causing these effects is likely partitioning. If partitioning is the sole factor responsible,
then measuring the partition coefficient of a bitterant would allow an effective
prediction of its taste in a given emulsion system. However, there are other factors that
lipids may influence (e.g., viscosity, mouthcoating, and complexation) that may
complicate the effect.
17
1.7 References
Arancibia C, Jublot L, Costell E, Bayarri S (2011) Flavor release and sensory
characteristics of o/w emulsions. Influence of composition, microstructure and
rheological behavior. Food Res Int 44:1632–1641. doi:
10.1016/j.foodres.2011.04.049
Ares G, Barreiro C, Deliza R, Gámbaro A (2009) Alternatives to reduce the bitterness,
astringency and characteristic flavour of antioxidant extracts. Food Res Int 42:871–
878. doi: 10.1016/j.foodres.2009.03.006
Barzanti C, Evans R, Fouquet J, et al (2007) Potentiometric determination of octanolwater and liposome-water partition coefficients (log P) of ionizable organic
compounds. Tetrahedron Lett 48:3337–3341. doi: 10.1016/j.tetlet.2007.03.085
Baxter NJ, Lilley TH, Haslam E, Williamson MP (1997) Multiple interactions between
polyphenols and a salivary proline-rich protein repeat result in complexation and
precipitation. Biochemistry 36:5566–5577. doi: 10.1021/bi9700328
Behrens M, Foerster S, Staehler F, et al (2007) Gustatory Expression Pattern of the
Human TAS2R Bitter Receptor Gene Family Reveals a Heterogenous Population of
Bitter Responsive Taste Receptor Cells. J Neurosci 27:12630–12640. doi:
10.1523/JNEUROSCI.1168-07.2007
Bohin MC, Roland WSU, Gruppen H, et al (2013) Evaluation of the bitter-masking
potential of food proteins for EGCG by a cell-based human bitter taste receptor
assay and binding studies. J Agric Food Chem 61:10010–10017. doi:
10.1021/jf4030823
Bohin MC, Vincken JP, Van Der Hijden HTWM, Gruppen H (2012) Efficacy of food
proteins as carriers for flavonoids. J Agric Food Chem 60:4136–4143. doi:
10.1021/jf205292r
Breslin PAS (2013) An Evolutionary Perspective on Food and Human Taste. Curr Biol
23:R409–R418. doi: 10.1016/j.cub.2013.04.010
Brewster ME, Loftsson T (2007) Cyclodextrins as pharmaceutical solubilizers. Adv Drug
Deliv Rev 59:645–666. doi: 10.1016/j.addr.2007.05.012
Cabras T, Melis M, Castagnola M, et al (2012) Responsiveness to 6-n-propylthiouracil
(PROP) is associated with salivary levels of two specific basic proline-rich proteins
in humans. PLoS One. doi: 10.1371/journal.pone.0030962
Caicedo A, Roper SD (2001) Taste Receptor Cells That Discriminate Between Bitter
Stimuli. Science (80- ) 291:1557–1561.
Chaudhari N, Roper SD (2010) Review series: The cell biology of taste. J Cell Biol
190:285–296. doi: 10.1083/jcb.201003144
Cheema M, Mohan MS, Campagna SR, et al (2015) The association of low-molecularweight hydrophobic compounds with native casein micelles in bovine milk. J Dairy
Sci 98:5155–63. doi: 10.3168/jds.2015-9461
18
Coupland JN (2014) An Introduction to the Physical Chemistry of Food. Springer
Coupland JN, Hayes JE (2014) Physical approaches to masking bitter taste: lessons from
food and pharmaceuticals. Pharm Res 31:2921–39. doi: 10.1007/s11095-014-14806
Creuzenet C, Mangroo D (1998) Physico-chemical characterization of human von Ebner
gland protein expressed in Escherichia coli: implications for its physiological role.
Protein Expr Purif 14:254–260. doi: 10.1006/prep.1998.0960
de Wit J, Backx-Hontelez E, Adamse M (1988) Evaluation of functional properties of
whey protein concentrates and whey protein isolates, 3: functional properties in
aqueous solution. Netherlands Milk Dairy J 42:155–172.
DeFazio RA (2006) Separate Populations of Receptor Cells and Presynaptic Cells in
Mouse Taste Buds. J Neurosci 26:3971–3980. doi: 10.1523/JNEUROSCI.051506.2006
Douroumis D (2007) Practical approaches to taste masking technologies in oral solid
forms. Expert Opin Drug Deliv 4:417–426. doi: 10.1038/2091178c0
Dsamou M, Palicki O, Septier C, et al (2012) Salivary protein profiles and sensitivity to
the bitter taste of caffeine. Chem Senses 37:87–95. doi: 10.1093/chemse/bjr070
Duxbury AJ, Thakker NS, Wastell DG (1989) A double-blind cross-over trial of a mucincontaining artificial saliva. Br Dent J 166:115–120.
Dvoryanchikov G, Sinclair MS, Perea-Martinez I, et al (2009) Inward rectifier channel,
ROMK, is localized to the apical tips of glial-like cells in mouse taste buds. J Comp
Neurol 517:spc1. doi: 10.1002/cne.22202
Fábián T, Beck A, Fejérdy P, et al (2015) Molecular Mechanisms of Taste Recognition:
Considerations about the Role of Saliva. Int J Mol Sci 16:5945–5974. doi:
10.3390/ijms16035945
Floriano WB, Hall S, Vaidehi N, et al (2006) Modeling the human PTC bitter-taste
receptor interactions with bitter tastants. J Mol Model 12:931–941. doi:
10.1007/s00894-006-0102-6
Galindo-Cuspinera V (2011) Taste Masking : Trends and Technologies. Prep Foods 51–
56.
Gaudette NJ, Pickering GJ (2013) Modifying bitterness in functional food systems. Crit
Rev Food Sci Nutr 53:464–81. doi: 10.1080/10408398.2010.542511
Gaudette NJ, Pickering GJ (2012) The efficacy of bitter blockers on health-relevant
bitterants. J Funct Foods 4:177–184. doi: 10.1016/j.jff.2011.10.003
Ginsburg I, Koren E, Shalish M, et al (2012) Saliva increases the availability of lipophilic
polyphenols as antioxidants and enhances their retention in the oral cavity. Arch
Oral Biol 57:1327–1334. doi: 10.1016/j.archoralbio.2012.04.019
Glasgow BJ (1995) Tear lipocalins bind a broad array of lipid ligands. Curr Eye Res
14:363–372.
19
Glendinning JI (1994) Is the bitter rejection response always adaptive? Physiol Behav
56:1217–1227. doi: 10.1016/0031-9384(94)90369-7
Guichard E, Langourieux S (2000) Interactions between Beta-lactoglobulin and flavour
compounds. Food Chem 71:301–308. doi: 10.1016/S0308-8146(00)00181-3
Hansch C, Leo A, Hoekman D (1995) Exploring QSAR - Hydrophobic, Electronic, and
Steric Constants. American Chemical Society, Washington, DC.
Hoang Thi TH, Morel S, Ayouni F, Flament MP (2012) Development and evaluation of
taste-masked drug for paediatric medicines - Application to acetaminophen. Int J
Pharm 434:235–242. doi: 10.1016/j.ijpharm.2012.05.047
Homma R, Yamashita H, Funaki J, et al (2012) Identification of bitterness-masking
compounds from cheese. J Agric Food Chem 60:4492–4499. doi:
10.1021/jf300563n
Hoppert K, Zahn S, Puschmann A, et al (2012) Quantification of sensory difference
thresholds for fat and sweetness in dairy-based emulsions. Food Qual Prefer
26:52–57. doi: 10.1016/j.foodqual.2012.03.008
Humphrey SP, Williamson RT (2001) A review of saliva: normal composition, flow, and
function. J Prosthet Dent 85:162–169.
Keast RS, Costanzo A (2015) Is fat the sixth taste primary? Evidence and implications.
Flavour 4:5. doi: 10.1186/2044-7248-4-5
Keast RSJ (2008) Modification of the bitterness of caffeine. Food Qual Prefer 19:465–
472. doi: 10.1016/j.foodqual.2008.02.002
Keast RSJ, Breslin P a S (2002) An overview of binary taste – taste interactions. Food
Qual Prefer 14:111–124.
Kimpel F, Schmitt JJ (2015) Review: Milk Proteins as Nanocarrier Systems for
Hydrophobic Nutraceuticals. J Food Sci 80:R2361–R2366. doi: 10.1111/17503841.13096
Klebanov GS, Mednikova LN, Ovcharova AD (1967) Extraction of Caffeine from Aqueous
Solutions. Pharm Chem J 1:221–223.
Kontopidis G, Holt C, Sawyer L (2004) Invited Review: β-Lactoglobulin: Binding
Properties, Structure, and Function. J Dairy Sci 87:785–796. doi:
10.3168/jds.S0022-0302(04)73222-1
Kroeze J, Bathoshuk L (1985) Bitterness Suppression as Revealed by Split-Tongue Taste
Stimulation in Humans. Physiol Behav 35:779–783.
Lawless HT (1979) Evidence for Neural Inhibition in Bittersweet Taste Mixtures. J Comp
Physiol Psychol 93:538-547.
Ley JP (2008) Masking Bitter Taste by Molecules. Chemosens. Percept. 1:58–77.
Livney YD (2010) Milk proteins as vehicles for bioactives. Curr Opin Colloid Interface Sci
15:73–83. doi: 10.1016/j.cocis.2009.11.002
20
Lush IE, Holland G (1988) The genetics of tasting in mice. V. Glycine and cycloheximide.
Genet Res 52:207–212.
Lynch J, Liu Y-H, Mela DJ, MacFie HJH (1993) A time-intensiy study of the effect of oil
mouthcoatings on taste perception. Chem Senses 18:121–129.
Mackey A (1958) Discernment of taste substances as affected by solvent medium. Food
Res 23:580–583.
Maehashi K, Matano M, Nonaka M, et al (2008) Riboflavin-binding protein is a novel
bitter inhibitor. Chem Senses 33:57–63. doi: 10.1093/chemse/bjm062
Masuda T, Kitabatake N (2006) Developments in biotechnological production of sweet
proteins. J Biosci Bioeng 102:375–389. doi: 10.1263/jbb.102.375
Matsuo R (2000) Role of Saliva in the Maintenance of Taste Sensitivity. Crit Rev Oral Biol
Med 11:216–229. doi: 10.1177/10454411000110020501
Mattes RD (2007) Effects of linoleic acid on sweet, sour, salty, and bitter taste
thresholds and intensity ratings of adults. Am J Physiol Gastrointest Liver Physiol
292:G1243–G1248. doi: 10.1152/ajpgi.00510.2006
McClements DJ (2015) Encapsulation, protection, and release of hydrophilic active
components: Potential and limitations of colloidal delivery systems. Adv Colloid
Interface Sci 219:27–53. doi: 10.1016/j.cis.2015.02.002
Melis M, Aragoni MC, Arca M, et al (2013) Marked increase in PROP taste
responsiveness following oral supplementation with selected salivary proteins or
their related free amino acids. PLoS One 8:e59810. doi:
10.1371/journal.pone.0059810
Mennella JA, Reed DR, Mathew PS, et al (2014) “A Spoonful of Sugar Helps the Medicine
Go Down”: Bitter Masking by Sucrose Among Children and Adults. Chem Senses.
doi: 10.1093/chemse/bju053
Mennella JA, Spector AC, Reed DR, Coldwell SE (2013) The bad taste of medicines:
Overview of basic research on bitter taste. Clin Ther 35:1225–1246. doi:
10.1016/j.clinthera.2013.06.007
Metcalf KL, Vickers ZM (2001) Taste intensities of oil-in-water emulsions with varying fat
content. J Sens Stud 17:379–390.
Meyerhof W, Behrens M, Brockhoff A, et al (2005) Human bitter taste perception. In:
Chemical Senses. pp 14–15
Morr CV, Foegeding EA (1990) Composition and Functionality of Commercial Whey and
Milk Protein Concentrates and Isolates: A Status Report. Food Technol 44:100–112.
Morr CV, Ha EYW (1993) Whey protein concentrates and isolates: processing and
functional properties. Crit Rev Food Sci Nutr 33:431–476.
Morzel M, Chabanet C, Schwartz C, et al (2014) Salivary protein profiles are linked to
bitter taste acceptance in infants. Eur J Pediatr 173:575–82. doi: 10.1007/s00431013-2216-z
21
Mosca AC, Andriot I, Guichard E, Salles C (2015) Binding of Na+ ions to proteins: Effect
on taste perception. Food Hydrocoll 51:33–40. doi: 10.1016/j.foodhyd.2015.05.003
Mudgal S, Keresztes I, Feigenson GW, Rizvi SSH (2016) Controlling the taste receptor
accessible structure of rebaudioside A via binding to bovine serum albumin. Food
Chem 197:84–91. doi: 10.1016/j.foodchem.2015.10.064
O’Neill TE, Kinsella JE (1987) Binding of alkanone flavors to beta-lactoglobulin: effects of
conformational and chemical modification. J Agric Food Chem 35:770–774.
Ogi K, Yamashita H, Terada T, et al (2015) Long-Chain Fatty Acids Elicit a BitternessMasking Effect on Quinine and Other Nitrogenous Bitter Substances by Formation
of Insoluble Binary Complexes. J Agric Food Chem 63:8493–8500. doi:
10.1021/acs.jafc.5b03193
Pelletier E, Sostmann K, Guichard E (1998) Measurement of Interactions between BetaLactoglobulin and Flavor Compounds (Esters, Acids, and Pyrazines) by Affinity and
Exclusion Size Chromatography. J Agric Food Chem 46:1506–1509.
Pripp AH, Busch J, Vreeker R (2004) Effect of viscosity, sodium caseinate and oil on
bitterness perception of olive oil phenolics. Food Qual Prefer 15:375–382. doi:
10.1016/S0950-3293(03)00084-3
Rayment S a, Liu B, Offner GD, et al (2000) Immunoquantification of human salivary
mucins MG1 and MG2 in stimulated whole saliva: factors influencing mucin levels.
J Dent Res 79:1765–1772. doi: 10.1177/00220345000790100601
Rosett TR, Kendregan SL, Klein BP (1997) Fat, protein, and mineral components of added
ingredients affect flavor qualities of tomato soups. J Food Sci 62:190–193. doi:
10.1111/j.1365-2621.1997.tb04397.x
Salles C, Chagnon M-C, Feron G, et al (2011) In-mouth mechanisms leading to flavor
release and perception. Crit Rev Food Sci Nutr 51:67–90. doi:
10.1080/10408390903044693
Schmale H, Ahlers C, Blaker M, et al (1993) Perireceptor events in taste. Ciba Found
Symp 179:167–185.
Schmale H, Holtgreve-Grez H, Christiansen H (1990) Possible role for salivary gland
protein in taste reception indicated by homology to lipophilic-ligand carrier
proteins. Nature 343:366–369. doi: 10.1038/346183a0
Schwingel L, Fasolo D, Holzschuh M, et al (2008) Association of 3-O-methylquercetin
with β-cyclodextrin: Complex preparation, characterization and ex vivo skin
permeation studies. J Incl Phenom Macrocycl Chem 62:149–159. doi:
10.1007/s10847-008-9450-4
Semo E, Kesselman E, Danino D, Livney YD (2007) Casein micelle as a natural nanocapsular vehicle for nutraceuticals. Food Hydrocoll 21:936–942. doi:
10.1016/j.foodhyd.2006.09.006
Shi P (2005) Contrasting Modes of Evolution Between Vertebrate Sweet/Umami
Receptor Genes and Bitter Receptor Genes. Mol Biol Evol 23:292–300. doi:
22
10.1093/molbev/msj028
Shpigelman A, Cohen Y, Livney YD (2012) Thermally-induced β-lactoglobulin–EGCG
nanovehicles: Loading, stability, sensory and digestive-release study. Food
Hydrocoll 29:57–67. doi: 10.1016/j.foodhyd.2012.01.016
Shpigelman A, Israeli G, Livney YD (2010) Thermally-induced protein-polyphenol coassemblies: beta lactoglobulin-based nanocomplexes as protective nanovehicles
for EGCG. Food Hydrocoll 24:735–743. doi: 10.1016/j.foodhyd.2010.03.015
Slack JP, Brockhoff A, Batram C, et al (2010) Modulation of bitter taste perception by a
small molecule hTAS2R antagonist. Curr Biol 20:1104–1109. doi:
10.1016/j.cub.2010.04.043
Spielman AI (1990) Interaction of saliva and taste. J Dent Res 69:838–843. doi:
10.1177/00220345900690030101
Spielman AI (1988) Isolation and Characterization of the Proline-rich Proteins from
Rabbit Parotid Saliva.
Suzuki a. H, Zhong H, Lee J, Martini S (2014) Effect of Lipid Content on Saltiness
Perception: A Psychophysical Study. J Sens Stud 29:404–412. doi:
10.1111/joss.12121
Tamamoto LC, Schmidt SJ, Lee S-Y (2010) Sensory properties of ginseng solutions
modified by masking agents. J Food Sci 75:S341–7. doi: 10.1111/j.17503841.2010.01749.x
Tavano OL (2013) Protein hydrolysis using proteases: An important tool for food
biotechnology. J Mol Catal B Enzym 90:1–11. doi: 10.1016/j.molcatb.2013.01.011
Thurgood JE, Martini S (2010) Effects of three emulsion compositions on taste
thresholds and intensity ratings of five taste compounds. J Sens Stud 25:861–875.
doi: 10.1111/j.1745-459X.2010.00311.x
Torrico DD, Prinyawiwatkul W (2015) Psychophysical Effects of Increasing Oil
Concentrations on Saltiness and Bitterness Perception of Oil-in-Water Emulsions. J
Food Sci 80:S1885–S1892. doi: 10.1111/1750-3841.12945
Torrico DD, Sae-Eaw A, Sriwattana S, et al (2015) Oil-in-Water Emulsion Exhibits
Bitterness-Suppressing Effects in a Sensory Threshold Study. J Food Sci 80:S1404–
S1411. doi: 10.1111/1750-3841.12901
Vandenbeuch A, Clapp TR, Kinnamon SC (2008) Amiloride-sensitive channels in type I
fungiform taste cells in mouse. BMC Neurosci 9:1. doi: 10.1186/1471-2202-9-1
Vingerhoeds MH, Blijdenstein TBJ, Zoet FD, van Aken G a. (2005) Emulsion flocculation
induced by saliva and mucin. Food Hydrocoll 19:915–922. doi:
10.1016/j.foodhyd.2004.12.005
Wu SY, Pérez MD, Puyol P, Sawyer L (1999) β-Lactoglobulin binds palmitate within its
central cavity. J Biol Chem 274:170–174. doi: 10.1074/jbc.274.1.170
Yamamoto Y, Nakabayashi M (1999) Enhancing Effect of an Oil Phase on the Sensory
23
Intensity of Salt Taste of Nacl in Oil/Water Emulsions. J Texture Stud 30:581–590.
doi: 10.1111/j.1745-4603.1999.tb01409.x
Yang R, Crowley HH, Rock ME, Kinnamon JC (2000) Taste cells with synapses in rat
circumvallate papillae display SNAP-25-like immunoreactivity. J Comp Neurol
424:205–215. doi: 10.1002/1096-9861(20000821)424:2<205::AID-CNE2>3.0.CO;2-F
Yee CL, Yang R, Bottger B, et al (2001) “Type III” cells of rat taste buds:
Immunohistochemical and ultrastructural studies of neuron-specific enolase,
protein gene product 9.5, and serotonin. J Comp Neurol 440:97–108. doi:
10.1002/cne.1372
Zissimos AM, Abraham MH, Barker MC, et al (2002) Calculation of Abraham descriptors
from solvent-water partition coeffcients in four different systems; evaluation of
different methods of calculation. J Chem Soc Perkin Trans 2 470–477. doi:
10.1039/b110143a
Zorzi GK, Caregnato F, Moreira JCF, et al (2015) Antioxidant Effect of Nanoemulsions
Containing Extract of Achyrocline satureioides (Lam) D.C.-Asteraceae. AAPS
PharmSciTech Ahead of Print. doi: 10.1208/s12249-015-0408-8
24
CHAPTER 2
Significance and Goals
2.1 Significance
Many drugs and phytochemicals are bitter, and aversion to the unpleasant taste has
been shown to reduce compliance with a treatment regimen (Shahiwala 2011), or the
selection of certain healthy foods in a diet respectively (Drewnowski and GomezCarneros 2000). This is particularly true in children, who are less able to weigh the longterm benefits over the short-term discomfort (Negri et al. 2012) and evolutionary
biologists have argued the stronger aversion of children to bitter is based on lack of
environmental learning (Steiner et al. 2001). Whatever the reason, pediatric patients are
at risk for harm due to avoidance of medication consumption (Mennella and
Beauchamp 2008). In fact, this problem is so persistent, the European Medicines Agency
will require a pediatric development plan to control the bitter/unpleasant tastes caused
by the active ingredients in oral medications in the coming years (Davies and Tuleu
2008; Mennella and Beauchamp 2008). The bitterness effect in medications is further
compounded by the fact that young children are generally believed to be unable or
unwilling to swallow tablets, so liquid formulations or chewable tablets become
necessary (Nunn and Williams 2005). This keeps the product in the mouth for a
prolonged period of time increasing the bitter taste intensity. Food product
development directly parallels this challenge in creation of healthy foods that utilize
plant-derived compounds. Because these pharmaceuticals or phytochemicals are
essential to the function of the medication or food product, their removal is not
possible. Suppression of the bitter taste should therefore be the focus.
One way of accomplishing bitter taste suppression is through physical separation of the
bitterant from the taste receptors (Douroumis 2007; Coupland and Hayes 2014). This
has been demonstrated via various encapsulation techniques (Brewster and Loftsson
2007; Gaudette and Pickering 2013). Currently, the majority of published bitternesssuppression research is unclear because of confounding food matrices (Keast 2008;
Bennett et al. 2012; Homma et al. 2012) and the generalization of results from only one
bitterant use throughout the study (Metcalf and Vickers 2001; Mattes 2007; Keast 2008;
Thurgood and Martini 2010). However, the major limitation in the literature is the lack
of effective chemical measurements used in conjunction with effective human sensory
measurements (Metcalf and Vickers 2001; Keast 2008; Ogi et al. 2015; Mudgal et al.
2016) ensuring that the mechanism of change in bitterness is uncertain or making an
assumption about how human subjects will perceive the sample respectively.
2.2 Goals and Hypotheses
My main goal is to investigate different concentrations of two food systems (protein
complexes and oil-in-water emulsions) that create non-aqueous environments for the
bitterant. I will use two bitterants (quinine and caffeine) that differ in hydrophobicity.
My overarching hypothesis is that the concentration of free (unbound), aqueous
25
bitterant will determine the intensity of perceived bitterness. Within this framework, I
have two specific objectives:
1. Bitterness suppression with whey protein isolate (Chapter 4): I hypothesize
that:
a. WPI will bind quinine more strongly than caffeine due to whey protein
components’ demonstrated hydrophobic binding behavior (Wu et al.
1999; Kontopidis et al. 2004).
b. Thermal denaturation of WPI will cause an increase in binding capacity
of quinine while caffeine binding behavior remains unchanged.
c. The bitterness of caffeine will not be affected while quinine will
significantly decrease in the presence of whey protein isolate due to
protein complexation.
d. The bitterness intensity for quinine will be proportional to the log
aqueous concentration caused by protein binding. This result would
indicate that aqueous, rather than total, concentration is more
indicative of perception.
2. Bitterness suppression with an oil-in-water emulsion (Chapter 5): I hypothesize
that:
a. Caffeine will have a negative log Kow value indicating a hydrophilic
molecule while quinine will have a highly positive log Kow value
indicating a hydrophobic molecule.
b. As the oil fraction of an oil-in-water emulsion increases, the aqueous
concentration of the hydrophilic caffeine will remain the same or
marginally increase with large increases in oil while the hydrophobic
quinine will decrease.
c. The bitterness intensity of caffeine will remain unchanged while quinine
bitterness will decrease as the oil fraction increases due to oil
partitioning preventing quinine from accessing the aqueous saliva and
therefore the taste receptors.
d. Quinine’s bitterness intensity will be proportional to the log aqueous
concentration. This result would indicate that aqueous, rather than
total, concentration is more indicative of perception.
This research will clarify how bitterants are affected by protein and lipid ingredients and
what mechanism is responsible. This will directly inform the development of oral
pediatric medication formulations to improve patient compliance without adding excess
sugar (Mennella and Beauchamp 2008; Mennella et al. 2014). In addition, this research
will apply to the creation of more palatable foods with added phytochemicals (Gaudette
and Pickering 2013).
26
2.3 References
Bennett SM, Zhou L, Hayes JE (2012) Using milk fat to reduce the irritation and bitter
taste of ibuprofen. Chemosens Percept 5:231–236. doi: 10.1007/s12078-012-91286
Brewster ME, Loftsson T (2007) Cyclodextrins as pharmaceutical solubilizers. Adv Drug
Deliv Rev 59:645–666. doi: 10.1016/j.addr.2007.05.012
Coupland JN, Hayes JE (2014) Physical approaches to masking bitter taste: lessons from
food and pharmaceuticals. Pharm Res 31:2921–39. doi: 10.1007/s11095-014-14806
Davies EH, Tuleu C (2008) Medicines for Children: A Matter of Taste. J Pediatr 153:599–
604. doi: 10.1016/j.jpeds.2008.06.030
Douroumis D (2007) Practical approaches to taste masking technologies in oral solid
forms. Expert Opin Drug Deliv 4:417–426. doi: 10.1038/2091178c0
Drewnowski a, Gomez-Carneros C (2000) Bitter taste, phyonutrients, and the consumer:
a review. Am J Clin Nutr 72:1424–1435.
Gaudette NJ, Pickering GJ (2013) Modifying bitterness in functional food systems. Crit
Rev Food Sci Nutr 53:464–81. doi: 10.1080/10408398.2010.542511
Homma R, Yamashita H, Funaki J, et al (2012) Identification of bitterness-masking
compounds from cheese. J Agric Food Chem 60:4492–4499. doi:
10.1021/jf300563n
Keast RSJ (2008) Modification of the bitterness of caffeine. Food Qual Prefer 19:465–
472. doi: 10.1016/j.foodqual.2008.02.002
Kontopidis G, Holt C, Sawyer L (2004) Invited Review: β-Lactoglobulin: Binding
Properties, Structure, and Function. J Dairy Sci 87:785–796. doi:
10.3168/jds.S0022-0302(04)73222-1
Mattes RD (2007) Effects of linoleic acid on sweet, sour, salty, and bitter taste
thresholds and intensity ratings of adults. Am J Physiol Gastrointest Liver Physiol
292:G1243–G1248. doi: 10.1152/ajpgi.00510.2006
Mennella JA, Beauchamp GK (2008) Optimizing oral medications for children. Clin Ther
30:2120–2132. doi: 10.1016/j.clinthera.2008.11.018
Mennella JA, Reed DR, Mathew PS, et al (2014) “A Spoonful of Sugar Helps the Medicine
Go Down”: Bitter Masking by Sucrose Among Children and Adults. Chem Senses.
doi: 10.1093/chemse/bju053
Metcalf KL, Vickers ZM (2001) Taste intensities of oil-in-water emulsions with varying fat
content. J Sens Stud 17:379–390.
Mudgal S, Keresztes I, Feigenson GW, Rizvi SSH (2016) Controlling the taste receptor
accessible structure of rebaudioside A via binding to bovine serum albumin. Food
Chem 197:84–91. doi: 10.1016/j.foodchem.2015.10.064
27
Negri R, Di Feola M, Di Domenico S, et al (2012) Taste Perception and Food Choices. J
Pediatr Gastroenterol Nutr 54:624–629. doi: 10.1097/MPG.0b013e3182473308
Nunn T, Williams J (2005) Formulation of medicines for children. Br J Clin Pharmacol
59:674–676. doi: 10.1111/j.1365-2125.2005.02410.x
Ogi K, Yamashita H, Terada T, et al (2015) Long-Chain Fatty Acids Elicit a BitternessMasking Effect on Quinine and Other Nitrogenous Bitter Substances by Formation
of Insoluble Binary Complexes. J Agric Food Chem 63:8493–8500. doi:
10.1021/acs.jafc.5b03193
Shahiwala A (2011) Formulation approaches in enhancement of patient compliance to
oral drug therapy. Expert Opin Drug Deliv 8:1521–1529. doi:
10.1517/17425247.2011.628311
Steiner JE, Glaser D, Hawilo ME, Berridge KC (2001) Comparative expression of hedonic
impact: Affective reactions to taste by human infants and other primates. Neurosci
Biobehav Rev 25:53–74. doi: 10.1016/S0149-7634(00)00051-8
Thurgood JE, Martini S (2010) Effects of three emulsion compositions on taste
thresholds and intensity ratings of five taste compounds. J Sens Stud 25:861–875.
doi: 10.1111/j.1745-459X.2010.00311.x
Wu SY, Pérez MD, Puyol P, Sawyer L (1999) β-Lactoglobulin binds palmitate within its
central cavity. J Biol Chem 274:170–174. doi: 10.1074/jbc.274.1.170
28
CHAPTER 3
Materials and Methods
3.1: General
Chemicals
Materials were obtained from commercial sources. Quinine hydrochloride (food grade),
caffeine (food grade), methanol (HPLC grade), and triethylamine (HPLC grade) were
purchased from Fisher Scientific (Pittsburgh, PA, USA). Acetonitrile (HPLC grade),
phosphoric acid (HPLC grade), and glacial acetic acid (HPLC grade) were purchased from
VWR International (Radnor, PA, USA). Mucin from porcine stomach, type III was
obtained from Sigma Aldrich (St. Louis, MO, USA). BiPRO whey protein isolate was
donated by Davisco Food International, Inc. (Eden Prairie, MN, USA). Crisco Vegetable
Oil was purchased from Weis grocery. Millipore water was used throughout the
experiments.
Nitrogen Analysis Testing
Protein content of the WPI and Mucins was analyzed via a LECO FP-528 Nitrogen
analyzer. 4 calibration standards of EDTA were used prior to analysis. WPI was
measured in solid form with samples measuring 0.2026 ± 0.0002 g. Mucins were
measured in solution form (100 mg/4 mL) measuring 0.2045 ± 0.0010 g. All was
performed in triplicate.
Storage of Emulsion/Protein Solutions
All samples were stored at 2⁰C (in the refrigerator) for no longer than 48 hours prior to
laboratory or sensory analysis. Before all measurements, the solutions were left to
equilibrate to room temperature.
High Performance Liquid Chromatography
HPLC was used to measure the amount of bitterant in the aqueous phase of all of the
experiments using an Agilent 1220 Affinity LC Manual Injection instrument from Agilent
Instruments with a Zorbax SB-C18 4.6X50mm separation column. A UV-Vis detector set
to 254 nm was used for quantification with a mobile phase flow rate of 1.5 ml/min. For
caffeine, a mobile phase of 94% water, 5.5% acetonitrile, 0.2% triethylamine, and 0.2%
glacial acetic acid at pH = 5 was used for analysis. The mobile phase for quinine was 50%
water and 50% methanol adjusted to pH = 2.5 with phosphoric acid for better peak
resolution. Prior to analysis, all samples were filtered with a PTFE membrane syringe
filter from VWR International with a pore size of 0.45μm.
Statistical Analysis
Statistical analysis was performed with Minitab Software (Minitab Inc. PA, USA) and
Compusense Cloud (Compusense Inc. ON, Canada) Software with a significance of 0.05.
Initial differences among sensory samples were determined using Two Way ANOVA and
Tukey’s HSD. Subsequent model development and analysis was performed on Minitab
29
Software. All other benchtop testing differences (two sample t-test) and summary data
(mean, standard error, sample distribution) were analyzed via Minitab as well.
Ethics Statement
Testing was performed in two sessions (protein effect testing) or a single session
(emulsion effect testing) in the Sensory Evaluation Center in the Department of Food
Science at The Pennsylvania State University. Procedures were exempted from
Institutional Review Board review by professional staff in the Penn State University
Office of Research Protections under the wholesome foods/approved food additives
exemption 6 in the 45 CFR 46. 101(b). Participants provided informed, implied consent
and were paid for their time.
3.2: Techniques Used Specifically in Chapter 4
Protein Binding by WPI
Protein binding assessed for both bitterants with WPI ranging from 0.001% (g protein/g
water) to 4.5% or gastric pig mucin ranging from 0.001% to 0.5%. A stock solution of
protein was prepared and diluted to the lower concentrations in parallel. Either quinine
HCL or caffeine stock solutions were added to each set of protein solutions at levels of
0.13 mM or 0.51 mM respectively. The protein was separated with EMD Millipore
Amicon Ultra-4 centrifugal filter units with a size cutoff of 3 kDa. 4 mL of the protein
solution was added to the filter unit and centrifuged in a Fisher Scientific Centrifuge
model 228 at 3400 RPM for 30 minutes. The bottom filtrate was collected and measured
via HPLC. The experiments were conducted in triplicate. Buffer was not needed as the
pH for all samples never deviated from natural saliva conditions (6.8-7.2).
Denatured WPI Binding
Protein binding tests with denatured WPI were used to determine the method of
protein binding. WPI solution were prepared as described above. While stirring
constantly, the protein solutions were heated to 85⁰C on a hot plate and held there for
5 minutes to ensure denaturation. Temperature was monitored with a thermocoupler.
Immediately following heating, the protein solution was placed in an ice water bath to
cool quickly and prevent protein coagulation. The WPI solution was then diluted with
water and addition of bitterant of performed in parallel with caffeine (0.51 mM) or
quinine HCL (0.13 mM) as described above. The solutions were filtered as described
above, and the bottom filtrate was measured via HPLC.
Testing the Effect of Protein on Perceived Bitterness.
Sensory testing was conducted with two levels of WPI (0% and 1%) and three levels of
bitterant (low, medium, and high concentration). After bitterant concentrations were
determined in informal pilot testing, three bitterant levels were used: quinine HCL =
0.056 mmol, 0.10 mmol, 0.18 mmol; caffeine = 1.8 mmol, 5.7 mmol, 18 mmol.
Untrained panelists were recruited from an opt-in participant database maintained by
the Sensory Evaluation Center at Penn State. Potential participants were screened for
30
and excluded based on contraindications related to dairy protein, caffeine, and quinine.
Pregnant women, smokers, and individuals with tongue piercings were excluded. Tests
were conducted on two separate days (quinine on one day, caffeine on the second) with
two separate sets of panelists in parallel. Data were collected using Compusense Cloud
software (Compusense Inc. ON, Canada). In each test session, the low, medium, and
high bitterant levels were tested with and without protein. Controls of water and a WPI
blank (to assess the effect of protein on the sample rating) were also included. Written
consent for the sensory test was obtained, and $15 compensation was given to each
participant before tasting anything. All 8 samples were equilibrated to room
temperature and presented under red light in 1 oz clear, plastic sample cups in a
William’s design to reduce position, order, and carryover effects. Randomly generated
three-digit blinding codes were used on all of the samples.
Prior to rating any samples in isolated sensory testing booths, a brief orientation to
familiarize the panelists with bitter taste was conducted in a common area; no more
than 4 panelists participated at a time. The orientation sample used for this warm up
task was the low concentration of the opposing bitterant (e.g. when quinine was to be
tested, caffeine was used in the orientation). All participants were orientated by a single
researcher (KNT), using the following script:
“Today, you will be rating the bitterness intensity of several samples. Before you go into
the booth, we would like you to taste this sample as an example of what we mean when
we say bitterness. If you have trouble sensing it while swishing, look for it in the
aftertaste. Please make your booth ratings while swishing, however. Throughout the
test, when you rate bitterness, please keep this reference sample in mind.”
A 10 mL orientation sample was then taken into the panelist’s mouth, swished for 10
seconds, and expectorated.
For the formal testing in isolated testing booths, participants were asked to take a 10 mL
solution in their mouth and swish for 10 seconds. They then rated the bitterness
intensity on an unstructured line scale (0 = low -100 = high) and liking on a nine-point
hedonic scale with the anchors of 1 = Dislike Extremely, 9 = Like Extremely while the
sample was swished in the mouth. Participants then expectorated the sample and
rinsed with filtered water as needed. An interstimulus interval (ISI) of 2 minutes
between samples was enforced via software. 105 participants provided ratings for
caffeine and 119 provided ratings for quinine. A panelist could only participate in one of
the bitterness tests so that brief desensitization would not occur and to avoid learning
effects or biases. The concentration of the unbound bitterant in each of the samples
was verified via HPLC in order to correlate aqueous concentration to bitterness
intensity.
31
3.3: Techniques Used Specifically in Chapter 5
Determination of Oil-Water Partition Coefficients
Partition behavior for both caffeine and quinine HCL were measured via bulk oil/water
experiments. Soybean oil was used as a readily-available commercial oil. Additionally,
the source of oil source does not significantly affect the sensory perception in emulsions
(Thurgood and Martini 2010). Soybean oil is composed of triglycerides of long-chain
fatty acids: primarily oleic, linoleic, and linolenic (Holcapek et al. 2003). Amount of
bitterant (0.51 mM caffeine and 0.13 quinine HCL) and water (10 mL) was kept constant
while the volume of oil was varied from 1 mL to 512 mL. The containers were inverted
gently several times, covered, and left to equilibrate for at least 36 hours. The aqueous
phase was collected and measured via HPLC. The experiments were conducted in
triplicate. In all experiments, plain filtered water was used as the aqueous phase. Buffer
was not needed as the pH for all samples never deviated from natural saliva conditions
(6.8-7.2). The coefficients were calculated with the following equation:
𝐾𝑂𝑊 =
[𝑏𝑖𝑡𝑡𝑒𝑟𝑎𝑛𝑡]𝑜𝑖𝑙
[𝑏𝑖𝑡𝑡𝑒𝑟𝑎𝑛𝑡]𝑤𝑎𝑡𝑒𝑟
[𝑏𝑖𝑡𝑡𝑒𝑟𝑎𝑛𝑡]
= [[𝑏𝑖𝑡𝑡𝑒𝑟𝑎𝑛𝑡] 𝑖𝑛𝑖𝑡𝑎𝑙 − 1] ∗
𝑤𝑎𝑡𝑒𝑟
𝑉𝑤𝑎𝑡𝑒𝑟
𝑉𝑜𝑖𝑙
[1]
Emulsion Preparation
Emulsions were prepared as follows. 160 g of 1% aqueous WPI solution was filtered and
mixed with 40 g of vegetable oil to create a 20% oil emulsion. Initial homogenization
with a high speed blender Brinkmann polytron from Brinkmann Instruments, inc. was
performed for 30 seconds. Immediately, this mixture was passed through a M110Y
Microfluidizer at 80 psi for at least 5 passes.
Particle Size Analysis
Emulsion droplet particle size was measured directly following emulsion preparation
with a Horiba LA-920 Laser Scattering Particle Size Distribution Analyzer (Horiba
Instruments, Inc.). The refractive index of the dispersed phase was estimated at 1.47 for
vegetable oil with the dispersant refractive index of water being 1.33.
Emulsion Partitioning Measurement
Partitioning of bitterants within the emulsions was conducted in triplicate with the HPLC
procedure described above. The emulsion was diluted with either water or WPI solution
in order to obtain various fat and protein contents for analysis of bitterant behavior.
Caffeine was tested with 5%, 10%, and 20% vegetable oil emulsions with constant 1%
protein. Quinine was tested with 0.5%, 1%, and 2% vegetable oil emulsions with
constant 0.125% protein. Following emulsion preparation, either caffeine or quinine HCL
stock solution was added to create a final concentration of 0.51 mM or 0.13 mM
respectively. In order to measure the aqueous concentration, the fat was separated
from the emulsions with a Sorvall MTX-150 Micro Ultracentrifuge from Thermo
Scientific. 15 mL of an emulsion was placed in a thick-walled ultracentrifuge tube and
spun at 43,000 RPM in a S50-A Fixed Angle Rotor for 20 minutes at room temperature.
The bottom aqueous layer was then filtered with the protein filtration procedure
described above. The aqueous bottom layer was collected and measured via HPLC.
32
Food-Grade Emulsion Preparation
Emulsion preparation for sensory testing was conducted in The Pennsylvania State
University Department of Food Science pilot plant. 14.88 pounds of 1% WPI solution
was mixed with 3.72 pounds of vegetable oil to create 18.6 pounds of 20% oil emulsion.
All equipment was cleaned and sanitized immediately prior to use. A Lightnin high shear
mixer was used as the preliminary mixing step for the emulsion. The impellor was
emerged in the mix and operated at medium/high speed for approximately 5 minutes.
The mixture was then poured directly into a two-stage dairy homogenizer at 3000 psi.
The emulsion was circulated for several minutes to begin homogenization. Then the
emulsion was fully passed through the system in 3 batches, collected, and stored at
34⁰F. Preliminary testing batches were measured for particle size and performance of
bitterant partitioning just like the emulsion testing protocol above. There was no
significant difference (p > 0.05) between the two emulsion preparations in either of the
measurements.
Testing the Effect of Fat on Perceived Bitterness.
Sensory testing with emulsions was conducted with untrained participants, who were
recruited from an opt-in participant database maintained by the Sensory Evaluation
Center at Penn State. The exclusion criteria were the same as in the protein testing
described previously. Consent for the sensory test was obtained and $10 compensation
was given to each participant before samples were administered. Two levels of fat (0.5%
and 2%) were used to maximize potential differences in quinine bitterness; a constant
amount of WPI (0.125%) was used as an emulsifier. In pilot testing, concentrations of
quinine HCL (0.10 mM) and caffeine (18 mM) were determined to be sufficiently bitter
to avoid potential floor effects that might obscure any reduction in perceived intensity.
All data were collected using Compusense Cloud software (Compusense Inc. ON,
Canada).
A 2x2 design (bitterant by fat level) was used, with replicates. These 8 stimuli were
presented in a counterbalanced serving order across participants in a William’s design to
reduce position, order, and carryover effects; no attempt was made to block by
bitterant or fat level in the design (i.e., all stimuli were treated independently for
counterbalancing purposes). All samples were equilibrated to room temperature and
presented in 1 oz black, plastic sample cups. Randomly generated three-digit blinding
codes were used on all of the samples. A warm-up emulsion sample not included in the
design was used to prepare the participants for the solutions in an attempt to prevent a
first position bias (e.g. Bennet et al. 2012). This warm up sample consisted of the higher
fat level and a 50:50 blend of the caffeine and quinine solutions, such that the final
caffeine and quinine concentrations were half of the concentrations used in the tested
samples. That is, the warm up sample contained 0.05 mM QHCL and 9 mM caffeine, 2%
fat, and 0.125% WPI. All data were collected using Compusense Cloud software
(Compusense Inc. On, Canada).
For the formal testing in isolated testing booths, participants were asked to take a 10 mL
solution in his or her mouth and swish for 10 seconds. Participants then rated the
33
bitterness intensity on an unstructured line scale (0 = low – 100 = high) and liking on a
nine-point hedonic scale (anchors of 1 = Dislike Extremely, 9 = Like Extremely). To
reduce potential dumping effects from the inclusion of fat in the sample, participants
were also asked to rate thickness (thin and thick as anchors) and vegetable oil aroma
(weak and strong as anchors) on unstructured line scales (0-100). Participants then
expectorated the sample and rinsed with filtered water as needed. An ISI of 2 minutes
between samples was enforced via software. Red light and opaque containers were
used to remove any visual cues related to stimuli. 100 participants completed this study.
The concentration of the aqueous bitterant in each of the samples was confirmed via
HPLC in order to correlate aqueous concentration to bitterness intensity.
Viscosity Measurement
Analysis of the viscosity of the emulsion samples were measured with a rheometer
instrument (DHR3 rheometer, TA Instruments, New Castle DE) using a concentric
cylinder probe at constant, controlled temperature (37⁰C, mouth temperature in order
to mimic the sensory experience) with shear rates from 0 to 80 s-1. A flow curve was
used to approximate what type of flow behavior the panelist experiences. Yield,
viscosity, and model fit (R2) were measured. All samples were conducted in triplicate.
34
3.4 References
Bennett SM, Zhou L, Hayes JE (2012) Using milk fat to reduce the irritation and bitter
taste of ibuprofen. Chemosens Percept 5:231–236. doi: 10.1007/s12078-012-91286
Holcapek M, Jandera P, Zderadicka P, Hrubá L (2003) Characterization of triacylglycerol
and diacylglycerol composition of plant oils using high-performance liquid
chromatography-atmospheric pressure chemical ionization mass spectrometry. J
Chromatogr A 1010:195–215. doi: 10.1016/S0021-9673(03)01030-6
Thurgood JE, Martini S (2010) Effects of three emulsion compositions on taste
thresholds and intensity ratings of five taste compounds. J Sens Stud 25:861–875.
doi: 10.1111/j.1745-459X.2010.00311.x
35
CHAPTER 4
Effect of Protein on Perception of Bitter Compounds
4.1: Introduction
Bitterness is one of five commonly recognized prototypical tastes and it tends to be
innately aversive. Because bitterness is common in many drugs and phytochemicals,
bitter-masking strategies are of great interest in order to improve patient compliance
with medication regimens, as well as encouraging consumption of healthier, functional
foods (Gaudette and Pickering 2013; Mennella et al. 2014). One of the strategies to
suppress the taste of bitter is preventing the interaction of the bitterant with taste
receptors via physical means such as encapsulation (Coupland and Hayes 2014).
Proteins in particular can be utilized in bitterness-masking because they have the ability
to bind small molecules and have demonstrated capacity to alter taste and aroma
perception (Bohin et al. 2013).
Whey proteins were chosen for this study because of their good emulsifying properties
and widespread use as food ingredients (de Wit et al. 1988). Whey proteins make up 1820% of the protein in milk and are the fraction soluble at pH 4.6 (Morr and Foegeding
1990; Jovanović et al. 2005). They are a mixture of proteins including β-lactoglobulin
(BLG) (60%), bovine serum albumin (BSA, 10%), α-lactalbumin (20%), and
immunoglobulins (5%) (Morr and Foegeding 1990). Whey protein isolate (WPI) is a
powdered food ingredient made from the whey fraction of milk containing about 90%
protein.
BLG is known to bind small molecules predominantly via hydrophobic interactions.
Ketones (O’Neill and Kinsella 1987), esters (Pelletier et al. 1998), fatty acids (Wu et al.
1999), aldehydes, alcohols, and lactones (Guichard and Langourieux 2000) showed
greater binding with BLG the greater their hydrocarbon chain lengths. While there is still
some debate about the mechanism of binding (Kontopidis et al. 2004), the strongest
evidence supports the theory that a β-barrel pocket in the interior of the protein binds
hydrophobic molecules with their most hydrophobic portion sliding into the barrel
space (Wu et al. 1999; Kontopidis et al. 2004). In silico modeling has shown the fatty
acid tail of palmitic acid fits directly within the barrel supporting that it is the main
hydrophobic binding pocket of BLG (Wu et al. 1999). While most binding studies have
been done with BLG, it is worth remembering that WPI is actually a mixture of proteins,
and BSA has also shown hydrophobic binding behavior. For example, NMR was used to
suggest that the binding of the hydrophobic Rebaudioside A (RebA) bitter ligand to BSA
could suppress the bitter taste of RebA, though no sensory testing was performed
(Mudgal et al. 2016). In other work, BLG complexes (i.e., nanoparticles) created by
thermal denaturation showed high binding of epigallocatechin gallate (EGCG) and a
subsequent reduction of bitterness and astringency (Shpigelman et al. 2010; Shpigelman
et al. 2012). For example, 70% of the EGCG present in a preparation was bound by the
1% BLG protein, which reduced bitterness perception from 3 to 1.7 on a 5 point scale
(Shpigelman et al. 2012).
36
Modulation of salty taste perception has also been investigated in protein binding
studies. These studies are not directly applicable to bitter, as salty taste is not received
by G-protein coupled taste receptors (Chaudhari and Roper 2010) but, the broader
hypothesis that bound tastants should be unable to interact with taste-sensitive cells is
relevant. In one study, skim milk decreased the saltiness of tomato soup compared to a
water control (Rosett et al. 1997). This may have been due to sodium binding by milk
proteins, but this was not measured. In other work, Mosca et al. (2015) measured
sodium binding to soy and milk proteins by NMR, but the interaction was relatively
weak and did not lead to reduced perception of saltiness in sensory testing.
Studies on the effect of protein on aroma are also relevant to the present work, as
greater binding would be expected to depress the free-aqueous and headspace volatile
concentration. For example, whey protein concentrate (0.5%), a less purified whey
protein fraction, decreased the flavor intensity of vanillin, benzaldehyde, and dlimonene by at least 50% (Hansen, 1997). The authors attributed this to hydrophobic
binding of the aroma compounds by BLG, but this was not measured directly.
Additionally, other researchers correlated a decrease in headspace concentration via
BLG-binding with a reduction in aroma intensity (Guichard and Langourieux 2000). The
magnitude of that reduction varied between the compounds. 2-heptanone had a strong
agreement between analytical and sensory results with a 20% reduction in headspace
concentration in the presence of BLG and a similar reduction of aroma perception on an
8-point rank-scale. On the other hand, 2-octanone and 2-nonanone had reductions in
headspace concentration of approximately 42% and 50% with only small reductions in
aroma perception (Guichard and Langourieux 2000).
The phenomenon of bitterness-masking by proteins has been investigated, but the
mechanism is unclear due to limitations in existing studies. First, the widespread use of
complex food matrices make it difficult for researchers to conclude which interaction is
causing any effects (Keast 2008; Bennett et al. 2012; Homma et al. 2012). Second, the
use of only one bitterant in most studies prevents any generalization of results (Metcalf
and Vickers 2001; Mattes 2007; Keast 2008). Lastly, many studies do not combine
effective chemical measurements with sensory measurements (Metcalf and Vickers
2001; Mattes 2007; Keast 2008; Thurgood and Martini 2010). In this work I will use
caffeine and quinine as model bitterants as they are commonly used in sensory studies
as well as in real foods yet have very different chemical structures. I add them to a
simple protein solution and measure both binding and perceived bitterness.
4.2: Methods
Experiments were conducted as described in Chapter 3.
37
4.3: Protein-Binding and Sensory Test Results and Discussion
4.3.1: Binding by Proteins
The WPI used throughout these tests was 14.106% nitrogen, or, using the dairy
conversion factor of 6.38, 90.0% protein, which is typical. Figure 4.1 shows the binding
behavior of caffeine and quinine in WPI solutions. Caffeine interacted minimally with
WPI (e.g., 88.5% of caffeine remained unbound in 1% WPI) while quinine interacted
strongly (e.g., 21.8% of quinine remained unbound in 1% WPI). A logarithmic function
was used to model the binding behavior of bitterants with WPI. Because there was only
very limited interaction between WPI and caffeine, the model did not fit well (p =
0.236), with an R2 value of 0.4124. Quinine interacted strongly with WPI and the model
gave a good fit (p < 0.001), with an R2 value of 0.9707. This observation is consistent
with previous work showing a linear relationship between log P and log Kb (a measure
of protein binding) for a wide range of compounds (Guichard and Langourieux 2000).
The data also support my hypothesis that WPI binds hydrophobic quinine much more
strongly than hydrophilic caffeine.
38
Figure 4.1. Proportion of caffeine (0.51 mM) and quinine (0.13 mM) bound as a function
of (a) native and (b) denatured WPI solution concentration. Error bars indicate standard
error. A logarithmic model is shown alongside the data.
If quinine binds to WPI via hydrophobic interactions, then a more hydrophobic protein
would be expected to bind more quinine while still not binding caffeine. The simplest
way to increase the surface hydrophobicity of WPI is by denaturing with heat (AlizadehPasdar and Li-Chan 2000; Jovanović et al. 2005; Loveday 2016) and Perez et al. (2014)
observed complexes between linoleic acid and BLG after heating to 85⁰C.
The protein-binding experiment was repeated, but prior to adding bitterant, the WPI
solutions were heated (85⁰C for 5 minutes) in order to cause irreversible denaturation
without gelling effects (Figure 4.1b). Caffeine interacted less with the heat-treated WPI
than with native WPI (e.g. 98.8% free at 1% denatured WPI vs. 88.5% free at 1% native
WPI). In contrast, quinine showed a dramatic increase in binding to denatured WPI (e.g.,
5.2% free at 1% denatured WPI vs. 21.8% free at 1% native WPI).
39
The binding was modeled using the same logarithmic function used for the native
proteins. Once again, the model fit was poor for caffeine (p = 0.316) while quinine gave
a good fit (p < 0.001, R2 = 0.9491), duplicating the patterns observed in the first binding
experiment. On the other hand, a steeper decrease for WPI binding to quinine as
evidenced by a smaller equation constant (24.44 versus 8.0601 for native and denatured
WPI respectively) in the logarithmic model indicates stronger binding, and this
difference was statistically significant using a z-test (p < 0.001). This suggests stronger
binding of the hydrophobic quinine when more hydrophobic groups were exposed in
the protein, which is consistent with the hydrophobic binding model described in the
literature of WPI and clarifies the nature of ingredient interaction.
4.3.2: Effect of Protein on Bitterness.
Regardless of the precise mechanism, differences in binding are useful for this study. It
is generally understood that in order for something to reach the taste receptor and be
perceived, it must first dissolve in saliva (Matsuo 2000; Coupland and Hayes 2014). I
hypothesize that by preventing the bitterant from accessing the saliva in the mouth,
binding by WPI will cause a reduction of bitterness. Specifically, added protein will
suppress the bitterness of quinine but not caffeine.
My original goal was to do sensory testing on bitterants with denatured proteins where I
had hypothesized greater binding of quinine would lead to lower bitterness compared
to the native protein. However, in pilot work thermal treatment led to off-odors and
small but noticeable changes in texture that created confounding issues and practical
challenges with any subsequent sensory testing. Accordingly, I decided not to proceed
with a denatured protein model.
Here, human psychophysical testing was used to determine how WPI (0% or 1%)
influences bitterness of several concentrations of caffeine and quinine (Figure 4.2).
Liking scores were inversely correlated with the bitterness results in all of these
experiments, and this data is not shown. The measured aqueous concentrations of
caffeine and quinine in the same samples are presented on the secondary y-axis.
For both bitterants in 0% WPI (water), bitterness intensity increased with the logarithm
of concentration (Figure 4.3), as would be expected. The observed dose-response
relationships also correspond well with previously reported data (Keast and Roper
2007).
The 1% WPI blank (i.e., no caffeine or quinine present) did not differ significantly in
bitterness from water, suggesting there is no contribution to perceived bitterness from
the WPI, at least at the concentration used here (Figure 4.2). There was no significant
change in caffeine bitterness with the addition of 1% WPI (p = 0.508). This was
expected, because there was little chemical interaction between caffeine and WPI, as
shown in Figure 4.1a, and the measured aqueous concentration of caffeine did not
substantially change with the addition of WPI, as shown in Figure 4.2.a.
40
In contrast, WPI reduced the bitterness of quinine solutions. The perceived bitterness of
both the low and medium quinine concentrations were significantly lower in 1% WPI
compared to the same concentrations in water (p < 0.05). However, WPI did not cause a
significant reduction in bitterness for the highest concentration of quinine, although the
pattern was in the same direction of the two lower concentrations. The substantial
reduction in aqueous concentration of quinine in the WPI samples, shown by the
squares in Figure 4.2b, supports the hypothesis that strong protein binding is involved in
the reduced bitterness observed for quinine.
It should be noted that while possible, it is highly unlikely that the protein chemically
transforms the quinine into a less bitter, completely different molecule. There was no
difference in the HPLC analysis with the protein solutions as compared to quinine in
water, so the bitterness decrease is solely due to protein-binding effects.
Linear modeling of the effects of WPI on the perception of bitterness from caffeine was
not conducted given the absence of an effect in the ANOVA model (Figure 4.2a). The
effect of WPI on quinine bitterness was modeled via regression using Minitab 17 to
determine which parameters were significant. The participant effect was significant (p <
0.001) as expected due to person-to-person variation in scale usage. Protein level was
also a significant predictor (p < 0.001), indicating the importance of protein on
bitterness perception. While aqueous concentration was not significant (p = 0.099), the
interaction between protein and aqueous concentration was significant (p = 0.024). This
suggests the aqueous concentration, as influenced by WPI, significantly affects the
bitterness of quinine.
While it was expected WPI would decrease the bitterness of quinine due to its strong
binding behavior, the decrease in perceived bitterness was rather modest compared to
the large reduction observed in aqueous phase concentration. This point is illustrated in
Figure 4.3, where the open points are the dose-response functions for quinine in water,
and the filled points are the aqueous concentrations in the presence of 1% WPI. The
slope of the dose response function for the aqueous concentration of quinine in protein
solutions is not significantly different than the slope of the dose response function in
water without protein (p = 0.765). However, the protein samples are much more bitter
than would be expected for a given aqueous concentration (i.e., the curve is shifted left
on the x-axis). Normally, a leftward shift would imply higher potency (e.g., Antenucci
and Hayes 2014). However, the filled circles are the aqueous concentration, not the
total concentration. If the aqueous quinine concentration was solely responsible for
bitter taste, then both sets of data—the response curves in water and in WPI solution—
should fit on the same trend line. This suggests the protein-bound fraction must still
somehow contribute to taste.
As a comparison, if the bitterness versus total concentration of the 1% protein samples
(i.e., bound plus unbound quinine) is plotted, these stimuli (filled squares) are less bitter
than the water (0% protein) samples, as shown by the downward shift relative to the
open circles. (In the 0% protein samples (i.e., water), the aqueous concentration is, by
definition, equal to the total concentration). This downward shift indicates that total
bitterant concentration is also not indicative of the perceived bitterness experienced by
41
participants, although it is much closer than the free bitterant concentration. This
evidence contradicts my hypothesis that aqueous concentration of bitterant (i.e., the
amount that remains unbound) predicts perceived bitterness. In the next section I will
consider some possible explanations.
Figure 4.2. Mean protein sensory test results for (a) caffeine (n=105) and (b) quinine
(n=119). Bitterness intensity, represented by the bars, is plotted on the primary y-axis
(left). Low, medium, and high concentration samples refer to nominal total
concentrations of 1.8 mM, 5.7 mM, and 18 mM for caffeine and 0.056 mM, 0.1 mM, and
0.18 mM for quinine, respectively, while the experimentally measured aqueous
concentrations of each bitterant are shown on the secondary y-axis (right). Different
letters indicate significant differences in bitterness rating (p<0.05). Error bars indicate
standard error.
42
Figure 4.3. Relationship between aqueous (unbound) concentration of quinine and
perceived bitterness in water () and in 1% WPI (). Total concentration of quinine
(bound and unbound) in 1% WPI is also presented for comparison (). Logarithmic fit
shown alongside the data. Error bars indicate standard error.
4.4: Considering Saliva
The bitterness of a quinine-WPI mixture is slightly less than would be predicted from the
total quinine concentration, but much greater than would be predicted from the
unbound quinine concentration. Similar discrepancies have been noted elsewhere.
Bohin et al. (2012, 2013) evaluated the masking of bitterness of EGCG by different
proteins. EGCG is typically considered to be a ligand for the bitter receptor hTAS2R39.
This receptor was used in an in vitro assay to evaluate receptor activation, and these
results were compared to in vivo sensory tests. Casein had the strongest binding
behavior with EGCG, reduced the activation of hTAS2R39 the most, and was rated as the
least bitter by panelists. There was good agreement between the reduction in in vitro
receptor activation in the presence of casein and reduction in perceived bitterness
(38.5% and 34.3% respectively). However, predictions from the binding curve (i.e.,
based on the free, non-casein bound, EGCG concentration) suggested bitter receptor
activation should have been reduced by even more (51.9%) in the presence of protein
(casein). This observation is consistent with present data: bound bitterant still has the
ability to interact with the receptors in some capacity. However, the magnitude of the
effect reported by Bohin and colleagues (2013) was much smaller than seen here.
Two possible explanations for this discrepancy are that either a) the bound bitterant can
still be tasted or that b) the in vitro measurement of binding is not representative of the
situation in the mouth.
However, the first of these (the bitterant-protein complex can stimulate the bitter
receptor on the tongue) seems improbable. Kontopidis et al. (2004) showed that when
43
hydrophobic small molecules are bound by BLG, they insert deeply into the β-barrel
structure where they would be unavailable for the delicate docking required with the
receptor. Additionally, other researchers have asserted that a strong complex between
BSA and RebA would shield the bitter receptor ligand sufficiently enough to suppress
the bitterness but not the sweetness of the sweetener, though no sensory testing was
used to confirm this (Mudgal et al. 2016). The alternative explanation may therefore
involve some role for saliva.
When the WPI-bitterant solution is taken into the mouth, it is diluted somewhat by
saliva. Assuming the saliva behaves simply as water, then changing the relative phase
volumes would alter the proportion of bitterant bound. For example, if the sample was
diluted in, say an equal volume of saliva, then a 1% WPI solution would become a 0.5%
WPI solution. According to the binding curve (Figure 4.1b) this would increase the
proportion of free quinine from 24% to 34%. However, that addition of saliva would
increase the volume of water diluting the concentration of quinine in the solution. The
theoretical concentration of aqueous quinine would then become 0.022 mM as
compared to the 0.13 mM original concentration in the binding studies—an equivalent
measure to approximately 17% free quinine. This, in turn, cancels out any effect coming
from the reduced protein binding and therefore cannot explain the large discrepancy
between aqueous concentration and bitterness perception.
However, saliva is more than just water, and other components have been implicated in
taste perception (Matsuo 2000; Humphrey and Williamson 2001; Dsamou et al. 2012;
Melis et al. 2013). For example, saliva contains approximately 0.5-0.9% protein
(Dsamou et al. 2012), and it is possible these proteins competitively bind tastants and
affect taste perceptions (Fábián et al. 2015). Additionally, it has been proposed that
several salivary proteins are important in delivering tastants to the taste receptors
(Melis et al. 2013; Tucker et al. 2014). If this is the case, then the delivery system (i.e.
water versus WPI solution) may not be as important for perception as I have
hypothesized in this chapter. Evolutionarily, reduced bitterness of toxins due to
complexation in the food matrix would presumably reduce the protective effect of
bitterness. Thus, it is not unreasonable to speculate that salivary proteins may
conceivable bind and release bitterants as a means to recover function that would
otherwise be lost.
As a side note, saliva composition and flow is influenced by stimulation and gland source
(Humphrey and Williamson 2001). Therefore, the exact percentages of protein content
as referenced here may vary slightly. This brief review is not meant to be
comprehensive, and these differences in saliva are not as important to the thought
process.
An example of salivary proteins increasing taste perception are the basic proline-rich
proteins (bPRPs) are secreted in response to PROP in supertaster panelists (Cabras et al.
2012). When non-tasters were given these proteins (PS-1 and II-2) or related free amino
acids along with the PROP, their perception of PROP bitterness greatly increased (Melis
et al. 2013). The researchers suggested that the increased tasting ability of PROP in nontasters implicates bPRPs in orienting the PROP molecule in order to optimize its binding
44
in the taste receptor cell binding pocket. It should be noted, though, that interaction
between the bitterant and the taste receptor was not monitored, so this mechanism
remains conjecture. Nonetheless, this example suggests that salivary proteins increased
bitter response rather than suppressing it. As mentioned above, these proteins may
have an active role in overcoming any pre-existing binding of bitterants to proteins
native to the food matrix. Indeed Spielman (1990) observed a correlation between PRP
content in saliva and quinine sulfate tasting sensitivity in humans; however when
binding assays were performed on rabbit PRPs with quinine sulfate, no interaction was
observed (Spielman 1988).
Salivary proteins have also been implicated in the mechanism for oleogustus, the unique
taste of some fatty acids. Lipocalin-1 or Von Ebner’s glad (VEG) protein has a
demonstrated interaction with fatty acids (Glasgow 1995). Lipocalin-1 is only expressed
in the VEG and is one of the only large salivary proteins that interacts directly with the
taste receptor environment (Schmale et al. 1990; Matsuo 2000). Because of this, it is
hypothesized that lipocalin-1 helps solubilize and deliver fat to the taste receptors
(Tucker et al. 2014). However, with regard to bitterness, a role for lipocalin-1 has been
largely discounted. Mice, who do not secrete the protein, have similar responses to a
broad range of bitter substances as compared to rats, who do secrete the protein
(Schmale et al. 1993). Binding curves have also been attempted with various bitterants
including strychnine, sucrose octaacetate, aloin, cholic acid (Schmale et al. 1993),
caffeine, and sodium benzoate (Creuzenet and Mangroo 1998) with little interaction. In
addition, retinol, cholesterol, and various odorants had no measureable binding capacity
with lipocalin-1, and the only substances that showed measureable binding activity
were fatty acids (Schmale et al. 1993).
Therefore, PRPs and lipocalin-1 are not likely candidates for salivary proteins that would
increase the taste perception of quinine as explained above, but their roles in PROP and
fatty acid taste, respectively suggest that another salivary protein may be responsible
for delivering quinine to the taste receptors. One potential candidate protein is salivary
mucin. Mucins are glycosylated proteins that play a role in several functions throughout
the body (Rayment et al. 2000) and make up about 16% of the salivary proteins. They
are known to be effective in binding hydrophobic small molecules (Ginsburg et al. 2012)
and can affect interactions with taste receptors (Hutteau and Mathlouthi 1998). For
example, salivary mucin proteins increase the solubility of polyphenols and therefore
increased the oxidant scavenging abilities of polyphenols both in fruit beverages and
reagents (Ginsburg et al. 2012). If mucins interact strongly with quinine, it could go
some way to explain the discrepancy between simple binding data and perceived
bitterness in vivo.
4.5: Mucin Binding of Quinine
As a preliminary study, gastric mucin sourced from pigs was used to study the binding
behavior to caffeine and quinine. I decided not to use human salivary mucin because the
source does not significantly affect the structure or aggregation behavior of the mucin
45
proteins (Koop et al. 1990; Teubl et al. 2013), and porcine mucin is more economical for
preliminary research. The nitrogen content of the mucin was measured as 2.77%, or
17.31% protein using the standard 6.25 conversion factor.
The binding of quinine and caffeine by mucin solutions (0-0.5%) was measured as
described above, and is shown in Figure 4.5a. The mucin percentage was chosen as a
range that is commonly used in the literature and representative of saliva (Levine 1993;
Hutteau and Mathlouthi 1998). Similar to their interactions with WPI, hydrophilic
caffeine was weakly bound by mucins and strongly bound with hydrophobic quinine.
The quinine interaction with WPI and with mucins are plotted together on the same axis
to allow better comparison (Figure 4.5b). The logarithmic equations used to model the
two sets of data are similar on a mass basis, suggesting that the nature of the binding is
quite similar. It is not clear if such binding may be important in vivo or whether that
could play a role on taste perception but it does represent one factor not considered in
my original experimental design that may contribute to a difference between the in
vitro and in vivo work. Other factors may also be important.
46
Figure 4.4. Proportion of caffeine (0.51 mM) and quinine (0.13 mM) bound as a function
of mucin solution concentration (a) and proportion of quinine bound as a function of
WPI or mucin solution concentration (b) Error bars indicate standard error. Logarithmic
fit shown alongside the data.
Another consideration for saliva and its effect on taste is salivary protease. Groups of
caffeine-hypersensitive individuals possessed saliva with higher levels of amylase,
immunoglobulin, and serum albumin, along with lower levels of cystatin SN, a protease
inhibitor, as compared to caffeine-hyposensitive individuals (Dsamou et al. 2012).
Cystatin SN is a protein specific to saliva that prevents cysteine proteolysis, and it was
concluded that proteolysis within the oral cavity was a perireceptor factor associated
with caffeine bitterness sensitivity (Dsamou et al. 2012). Relatedly, in a study by Morzel
et al. (2014), 73 infants were profiled at 3 and 6 months and intake ratios of bitter urea
solution were collected. At 3 months, the most significant component of saliva the
predicted intake ratios was, again, cystatin. Higher amounts of cystatin were correlated
with higher bitterness acceptance, which was interpreted as lower bitterness sensitivity.
47
Individuals with high bitterness sensitivity have higher salivary protease activity.
However, when assessing panelists’ salivary effects on liquid model dairy systems, there
was no effect on sample characteristics caused by protease activity (Drago et al. 2011).
4.6: Conclusions.
The key conclusions of this work are (i) WPI binds quinine but not caffeine, (ii) WPI
suppresses the bitterness of quinine but not caffeine, (iii) the degree of bitterness
suppression of quinine by WPI is less than expected given the degree of binding. While
to my knowledge, there are no published binding isotherms of caffeine or quinine to
WPI, the first conclusion is not surprising given the difference in hydrophobicity of the
two bitterants. Similarly, while a new observation, the second conclusion is consistent
with the commonly accepted perspective in the literature. The third conclusion
however, was not expected. Dilution in salivary water is not enough to explain the
discrepancy, and while mucin does bind quinine, it is not clear if this interaction is
important to bitter taste perception.
48
4.7: References
Alizadeh-Pasdar N, Li-Chan EC (2000) Comparison of protein surface hydrophobicity
measured at various pH values using three different fluorescent probes. J Agric
Food Chem 48:328–334. doi: 10.1021/jf990393p
Antenucci RG, Hayes JE (2014) Non-nutritive sweeteners are not super-normal stimuli.
Int J Obes 1-6. Doi: 10.1038/ijo.2014.109
Bennett SM, Zhou L, Hayes JE (2012) Using milk fat to reduce the irritation and bitter
taste of ibuprofen. Chemosens Percept 5:231–236. doi: 10.1007/s12078-012-91286
Bohin MC, Roland WSU, Gruppen H, et al (2013) Evaluation of the bitter-masking
potential of food proteins for EGCG by a cell-based human bitter taste receptor
assay and binding studies. J Agric Food Chem 61:10010–10017. doi:
10.1021/jf4030823
Bohin MC, Vincken JP, Van Der Hijden HTWM, Gruppen H (2012) Efficacy of food
proteins as carriers for flavonoids. J Agric Food Chem 60:4136–4143. doi:
10.1021/jf205292r
Cabras T, Melis M, Castagnola M, et al (2012) Responsiveness to 6-n-propylthiouracil
(PROP) is associated with salivary levels of two specific basic proline-rich proteins
in humans. PLoS One. doi: 10.1371/journal.pone.0030962
Chaudhari N, Roper SD (2010) Review series: The cell biology of taste. J Cell Biol
190:285–296. doi: 10.1083/jcb.201003144
Coupland JN, Hayes JE (2014) Physical approaches to masking bitter taste: lessons from
food and pharmaceuticals. Pharm Res 31:2921–39. doi: 10.1007/s11095-014-14806
Creuzenet C, Mangroo D (1998) Physico-chemical characterization of human von Ebner
gland protein expressed in Escherichia coli: implications for its physiological role.
Protein Expr Purif 14:254–260. doi: 10.1006/prep.1998.0960
de Wit J, Backx-Hontelez E, Adamse M (1988) Evaluation of functional properties of
whey protein concentrates and whey protein isolates, 3: functional properties in
aqueous solution. Netherlands Milk Dairy J 42:155–172.
Drago SR, Panouillé M, Saint-Eve A, et al (2011) Relationships between saliva and food
bolus properties from model dairy products. Food Hydrocoll 25:659–667. doi:
10.1016/j.foodhyd.2010.07.024
Dsamou M, Palicki O, Septier C, et al (2012) Salivary protein profiles and sensitivity to
the bitter taste of caffeine. Chem Senses 37:87–95. doi: 10.1093/chemse/bjr070
Fábián T, Beck A, Fejérdy P, et al (2015) Molecular Mechanisms of Taste Recognition:
Considerations about the Role of Saliva. Int J Mol Sci 16:5945–5974. doi:
10.3390/ijms16035945
Gaudette NJ, Pickering GJ (2013) Modifying bitterness in functional food systems. Crit
49
Rev Food Sci Nutr 53:464–81. doi: 10.1080/10408398.2010.542511
Ginsburg I, Koren E, Shalish M, et al (2012) Saliva increases the availability of lipophilic
polyphenols as antioxidants and enhances their retention in the oral cavity. Arch
Oral Biol 57:1327–1334. doi: 10.1016/j.archoralbio.2012.04.019
Glasgow BJ (1995) Tear lipocalins bind a broad array of lipid ligands. Curr Eye Res
14:363–372.
Guichard E, Langourieux S (2000) Interactions between Beta-lactoglobulin and flavour
compounds. Food Chem 71:301–308. doi: 10.1016/S0308-8146(00)00181-3
Homma R, Yamashita H, Funaki J, et al (2012) Identification of bitterness-masking
compounds from cheese. J Agric Food Chem 60:4492–4499. doi:
10.1021/jf300563n
Humphrey SP, Williamson RT (2001) A review of saliva: normal composition, flow, and
function. J Prosthet Dent 85:162–169.
Hutteau F, Mathlouthi M (1998) Physicochemical properties of sweeteners in artificial
saliva and determination of a hydrophobicity scale for some sweeteners. Food
Chem 63:199–206. doi: 10.1016/S0308-8146(98)00007-7
Jovanović S, Bara M, Ma O (2005) Whey proteins-Properties and Possibility of
Application. Mljekarstvo 55:215–233.
Keast RSJ (2008) Modification of the bitterness of caffeine. Food Qual Prefer 19:465–
472. doi: 10.1016/j.foodqual.2008.02.002
Keast RSJ, Roper J (2007) A complex relationship among chemical concentration,
detection threshold, and suprathreshold intensity of bitter compounds. Chem
Senses 32:245–253. doi: 10.1093/chemse/bjl052
Kontopidis G, Holt C, Sawyer L (2004) Invited Review: β-Lactoglobulin: Binding
Properties, Structure, and Function. J Dairy Sci 87:785–796. doi:
10.3168/jds.S0022-0302(04)73222-1
Koop HM, Valentijn-Benz M, Nieuw Amerongen a V, et al (1990) Aggregation of oral
bacteria by human salivary mucins in comparison to salivary and gastric mucins of
animal origin. Antonie Van Leeuwenhoek 58:255–263.
Levine MJ (1993) Development of artificial salivas. Crit Rev Oral Biol Med 4:279–286.
doi: 10.1177/10454411930040030401
Loveday SM (2016) β-Lactoglobulin heat denaturation: A critical assessment of kinetic
modelling. Int Dairy J 52:92–100. doi: 10.1016/j.idairyj.2015.08.001
Matsuo R (2000) Role of Saliva in the Maintenance of Taste Sensitivity. Crit Rev Oral Biol
Med 11:216–229. doi: 10.1177/10454411000110020501
Mattes RD (2007) Effects of linoleic acid on sweet, sour, salty, and bitter taste
thresholds and intensity ratings of adults. Am J Physiol Gastrointest Liver Physiol
292:G1243–G1248. doi: 10.1152/ajpgi.00510.2006
50
Melis M, Aragoni MC, Arca M, et al (2013) Marked increase in PROP taste
responsiveness following oral supplementation with selected salivary proteins or
their related free amino acids. PLoS One 8:e59810. doi:
10.1371/journal.pone.0059810
Mennella JA, Reed DR, Mathew PS, et al (2014) “A Spoonful of Sugar Helps the Medicine
Go Down”: Bitter Masking by Sucrose Among Children and Adults. Chem Senses.
doi: 10.1093/chemse/bju053
Metcalf KL, Vickers ZM (2001) Taste intensities of oil-in-water emulsions with varying fat
content. J Sens Stud 17:379–390.
Morr CV, Foegeding EA (1990) Composition and Functionality of Commercial Whey and
Milk Protein Concentrates and Isolates: A Status Report. Food Technol 44:100–112.
Morzel M, Chabanet C, Schwartz C, et al (2014) Salivary protein profiles are linked to
bitter taste acceptance in infants. Eur J Pediatr 173:575–82. doi: 10.1007/s00431013-2216-z
Mudgal S, Keresztes I, Feigenson GW, Rizvi SSH (2016) Controlling the taste receptor
accessible structure of rebaudioside A via binding to bovine serum albumin. Food
Chem 197:84–91. doi: 10.1016/j.foodchem.2015.10.064
O’Neill TE, Kinsella JE (1987) Binding of alkanone flavors to beta-lactoglobulin: effects of
conformational and chemical modification. J Agric Food Chem 35:770–774.
Pelletier E, Sostmann K, Guichard E (1998) Measurement of Interactions between BetaLactoglobulin and Flavor Compounds (Esters, Acids, and Pyrazines) by Affinity and
Exclusion Size Chromatography. J Agric Food Chem 46:1506–1509.
Perez AA, Andermatten RB, Rubiolo AC, Santiago LG (2014) β-Lactoglobulin heatinduced aggregates as carriers of polyunsaturated fatty acids. Food Chem 158:66–
72. doi: 10.1016/j.foodchem.2014.02.073
Rayment S a, Liu B, Offner GD, et al (2000) Immunoquantification of human salivary
mucins MG1 and MG2 in stimulated whole saliva: factors influencing mucin levels.
J Dent Res 79:1765–1772. doi: 10.1177/00220345000790100601
Rosett TR, Kendregan SL, Klein BP (1997) Fat, protein, and mineral components of added
ingredients affect flavor qualities of tomato soups. J Food Sci 62:190–193. doi:
10.1111/j.1365-2621.1997.tb04397.x
Schmale H, Ahlers C, Blaker M, et al (1993) Perireceptor events in taste. Ciba Found
Symp 179:167–185.
Schmale H, Holtgreve-Grez H, Christiansen H (1990) Possible role for salivary gland
protein in taste reception indicated by homology to lipophilic-ligand carrier
proteins. Nature 343:366–369. doi: 10.1038/346183a0
Shpigelman A, Cohen Y, Livney YD (2012) Thermally-induced β-lactoglobulin–EGCG
nanovehicles: Loading, stability, sensory and digestive-release study. Food
Hydrocoll 29:57–67. doi: 10.1016/j.foodhyd.2012.01.016
51
Shpigelman A, Israeli G, Livney YD (2010) Thermally-induced protein-polyphenol coassemblies: beta lactoglobulin-based nanocomplexes as protective nanovehicles
for EGCG. Food Hydrocoll 24:735–743. doi: 10.1016/j.foodhyd.2010.03.015
Spielman AI (1988) Isolation and Characterization of the Proline-rich Proteins from
Rabbit Parotid Saliva.
Teubl BJ, Absenger M, Frohlich E, et al (2013) The oral cavity as a biological barrier
system: Design of an advanced buccal in vitro permeability model. Eur J Pharm
Biopharm 84:386–393.
Thurgood JE, Martini S (2010) Effects of three emulsion compositions on taste
thresholds and intensity ratings of five taste compounds. J Sens Stud 25:861–875.
doi: 10.1111/j.1745-459X.2010.00311.x
Tucker RM, Mattes RD, Running CA (2014) Mechanisms and effects of fat taste in
humans. BioFactors 40:313–326. doi: 10.1002/biof.1162
Wu SY, Pérez MD, Puyol P, Sawyer L (1999) β-Lactoglobulin binds palmitate within its
central cavity. J Biol Chem 274:170–174. doi: 10.1074/jbc.274.1.170
52
CHAPTER 5
Effect of Lipids on Perception of Bitter Compounds
5.1: Introduction
The aversive taste of bitter compounds creates noncompliance with medication
regimens (Shahiwala 2011) and avoidance of healthful foods (Drewnowski and GomezCarneros 2000). This necessitates development of better methods to suppress
bitterness. One approach is to bind the bitter compound to polymers so they cannot
interact with taste receptors in the oral cavity (see previous Chapter). In this chapter I
will examine a related method – using an oil to dissolve a hydrophobic bitter compound
to again reduce the concentration in aqueous saliva, under the assumption that this will
reduce perceived taste intensity. Mackey (1958) showed that bitterants (i.e., caffeine
and quinine) dissolved in oil tasted less bitter than similar compounds in water;
subsequent work suggested an oil mouthwash prior to bitter aqueous sample ingestion
is capable of reducing taste (Lynch et al. 1993). However, emulsions offer a much more
flexible system for studying the effect of lipid composition, as well as representing a
more realistic model for bitter foods.
Increasing the fat content of a food emulsion has, in some cases, been shown to reduce
perceived bitterness (Metcalf and Vickers 2001; Thurgood and Martini 2010). The
mechanism most commonly proposed is the partitioning of a hydrophobic bitterant out
of the aqueous phase into the oil phase, presumably preventing interaction with saliva
(Matsuo 2000; Coupland and Hayes 2014). Conversely however, one would expect that
for a hydrophilic bitterant increasing the lipid concentration in an emulsion would
increase the concentration in the aqueous phase, thereby increasing bitterness. For
example, Keast (2008) reported that increasing the fat content of milk from 0 to 4%
increased the bitterness of caffeine, which is hydrophilic. However, this model is not
universally accepted. In a pair of 2015 studies, the opposite effect was seen with
caffeine bitterness in 20% vs. 40% canola oil emulsions (Torrico and Prinyawiwatkul
2015; Torrico et al. 2015b). The recognition threshold, but not the detection threshold,
was significantly increased for caffeine at 20% oil compared to water (Torrico et al.
2015b), and both 20% and 40% emulsions decreased the bitterness intensity of caffeine
(Torrico and Prinyawiwatkul 2015). However, it should be noted that that there was no
difference between the ratings of the 20% and 40% emulsions, and that the magnitude
of intensity decrease was small.
In one of the most complete studies on the effect of lipid content on taste, Metcalf and
Vickers (2001) compared the effect on taste intensity of diluting aqueous tastant
solutions with either water or oil. When oil was added, the entire mixture was
emulsified to create the final o/w emulsions (9% and 17% oil). If a tastant did not
partition into the lipid phase at all, then dilution with water would decrease the
concentration and taste intensity while dilution with oil would have no effect on the
aqueous concentration and taste intensity (though aqueous concentration was not
measured). The hydrophilic tastants (sucrose, NaCl, citric acid, MSG) all lacked
significant interaction with the oil at either level so the taste intensity of the initial
53
undiluted control solution did not differ significantly from that of the emulsions. In
contrast, when hydrophobic quinine sulfate was used as a model bitterant, dilution with
oil reduced the bitterness intensity as compared to the undiluted control solution. This
suggests that some of the bitterant went into the oil droplets, reducing the aqueous
concentration and perceived intensity. Also the sample diluted with oil was less bitter
than the sample diluted with the same volume of water because of quinine partitioning.
Many of the studies on bitter taste in emulsion are limited for the similar reasons as
outlined in Chapter 4.1, i.e., the use of uncontrolled food matrices (Keast 2008; Bennett
et al. 2012; Homma et al. 2012), use of one bitterant, which limits the generalizability of
the conclusions (Metcalf and Vickers 2001; Mattes 2007; Keast 2008), and the lack of
adequate chemical and sensory measurements (Metcalf and Vickers 2001; Keast 2008;
Thurgood and Martini 2010). In this thesis, I use caffeine and quinine as model
bitterants because they are commonly used in sensory studies, and are GRAS additives
found in real foods, yet have very different hydrophobicity/philicity. Here, I address gaps
in prior studies by adding them to oil-in-water emulsions (with whey protein isolate
(WPI) as an emulsifier), and measure both aqueous concentration instrumentally and
perceived bitterness psychophysically.
5.2: Methods
Experiments were conducted as described in Chapter 3.
5.3: Oil Partitioning and Sensory Test Results and Discussion
5.3.1: Oil/Water Partitioning
As a first step in a study of taste suppression in emulsions, I determined the oil-water
partitioning behavior of caffeine and quinine using a shake flask method. The volume of
the water phase was kept constant while the amount of vegetable oil was varied and
the aqueous concentration was measured to determine the oil-water partition
coefficient, KOW:
𝐾𝑂𝑊 =
[𝑏𝑖𝑡𝑡𝑒𝑟𝑎𝑛𝑡]𝑜𝑖𝑙
[𝑏𝑖𝑡𝑡𝑒𝑟𝑎𝑛𝑡]𝑤𝑎𝑡𝑒𝑟
[𝑏𝑖𝑡𝑡𝑒𝑟𝑎𝑛𝑡]
= [[𝑏𝑖𝑡𝑡𝑒𝑟𝑎𝑛𝑡] 𝑖𝑛𝑖𝑡𝑎𝑙 − 1] ∗
𝑤𝑎𝑡𝑒𝑟
𝑉𝑤𝑎𝑡𝑒𝑟
𝑉𝑜𝑖𝑙
[5.1]
Where [bitterant]initial and [bitterant]water represent the concentration of bitterant added
to the system and in the water phase respectively. Vwater and Voil represent the volumes
of water and oil phases respectively.
The results are presented in Figure 5.1. The oil-water partition coefficient of caffeine
was 0.0479 (log KOW = -1.32; that is, relatively hydrophilic) while quinine was 932 (log
KOW = 2.97; that is, relatively hydrophobic). In a 3% oil mixture, more than 99.8% of the
caffeine will be present in the aqueous phase. In contrast, in a 3% oil mixture, 98% of
the quinine will be present in the lipid phase.
54
Figure 5.1. Proportion of caffeine and quinine unbound as a function of oil
concentration. Original aqueous concentration was 0.51 mM and 0.13 mM for caffeine
and quinine, respectively. Log KOW of caffeine and quinine are -1.32 and 2.97
respectively.
The measured partition coefficients are compared to literature values in Table 5.1.
Much of the published work is based on either theoretical calculations or empirical
measurements of log P values (i.e., octanol-water partition coefficients). Octanol is
chemically quite different and less polar than vegetable oil, so it is unsurprising there is
poor agreement with present data. However, despite the chemical differences between
octanol and vegetable oil, the experimental measurements for octane-water
partitioning of caffeine (-1.52, Mednikova, and Ovcharova 1967) were in reasonable
agreement to the measurements made here (-1.32). Conversely, the measurement of
quinine obtained here (2.97) does not agree well with previously reported values using
octanol as the non-polar solvent. However, measurements by Barzanti (2007) and
Zissimos et al. (2002) obtained via a potentiometric method are in more similar
agreement (log P’s = 3.20 and 3.47, respectively).
55
Table 5.1. Partition coefficients for caffeine and quinine from the literature.
Bitterant
Citation
Experimental
/ Calculation
Solvent/
Program
Caffeine
(Machatha
and
Yalkowsky
2005)
(Harris and
Logan 2014)
Calculation
Calculation
Calculation
Caffeine
Caffeine
Caffeine
Caffeine
Caffeine
Quinine
Quinine
Quinine
Quinine
Quinine
Quinine
Other
Considerations
KowWin
ACD
ClogP
Log
Partition
Coefficient
0.16
-0.13
-0.06
Experimental
Octanol
-0.21
(Klebanov et
al. 1967)
Experimental
Octane
-1.52
(Yalkowsky
et al. 1983)
(Hansch et
al. 1995)
Present
work
Experimental
Octanol
-0.20
Room temperature,
neutral pH, shake
flask method
Room temperature,
pH = 5.0, shake flask
method
30C, neutral pH,
shake flask method
Calculation
QSAR
-0.07
Experimental
Vegetable
oil
-1.32
(Machatha
and
Yalkowsky
2005)
(Hansch and
Anderson
1967)
(Zissimos et
al. 2002)
Calculation
Calculation
Calculation
KowWin
ACD
ClogP
3.29
3.44
2.79
Experimental
Octanol
1.73
Experimental
3.47
(Hansch et
al. 1995)
(Barzanti et
al. 2007)
Calculation
pKa
measurem
ent/
calculation
QSAR
3.20
Potentiometry
method
Present
work
Experimental
pKa
measurem
ent/
calculation
Vegetable
oil
2.97
Room temperature,
neutral pH, shake
flask method
Experimental
56
Room temperature,
neutral pH, shake
flask method
Room temperature,
neutral pH, shake
flask method
Potentiometry
method
3.44
5.3.2: Bitterant Partitioning in Emulsion Systems
It is generally understood that in order for something to reach and activate a taste
receptor to initiate the signal transduction events that ultimately result in perception,
the tastant must first dissolve in saliva (Matsuo 2000; Coupland and Hayes 2014). I
hypothesize that by preventing the bitterant from accessing the saliva in the mouth,
partitioning by oil will cause a reduction of bitterness. That is, added oil should suppress
the bitterness of quinine, but not caffeine, due to differences in their respective
partition coefficients. The following experiment uses an oil-in-water emulsion as an
experimental model system representative of pharmaceutical or food formulations.
However, a stable food emulsion requires an emulsifier, and emulsifier-bitterant
interactions would also influence the results. Here, I prepared emulsions that were
stabilized with WPI, and the WPI interacts more with quinine than caffeine (Figure 4.1).
Thus, with the inclusion of WPI as the emulsifier, the emulsion becomes a three phase
partitioning model (i.e., bitterant can be present in aqueous solution, dissolved in the
oil, or bound to the protein). I hypothesize that the aqueous concentration of the
bitterant (i.e., not in the lipid phase or bound to protein) is responsible for perceived
intensity.
Because caffeine does not have a strong affinity for oil (Figure 5.1), a large range of oil
concentrations were used to observe an effect in the emulsions (0-20% oil, constant 1%
WPI). Conversely, quinine has a much stronger affinity for oil so more dilute emulsion
were expected to provide an effect (0-2% oil, constant 0.125% WPI). The WPI was held
constant at the lowest level possible to minimize effects of protein binding while
maintaining emulsion stability at all oil levels (data not shown).
Emulsion droplet diameter was similar for all samples (d32=0.21-0.24 m) and did not
change over the course of the experiments. Increasing the lipid content in emulsions (020%) led to a non-significant increase on the aqueous concentration of caffeine, while a
much smaller change in the lipid content (0-2%) led to a large decrease in aqueous
quinine (Figure 5.2). These results are expected and consistent with the differences in
oil-water partitioning seen in Figure 5.1. The reduced aqueous concentration in the 0%
lipid samples (12 and 43% for caffeine and quinine respectively) is due to the different
levels of binding by the protein present. Based on the emulsion partitioning behavior
shown in Figure 5.2, I hypothesize that an increase in fat will have no effect on caffeine
bitterness or aqueous concentration and decrease the aqueous concentration and
perceived bitterness of quinine.
57
Figure 5.2. Effect of lipid concentration in an o/w emulsion (d32~0.225 m) on the
relative aqueous concentration of (a) caffeine (0.51 mM) and (b) quinine (0.13 mM). The
emulsions prepared with caffeine were stabilized with 1% WPI while those containing
quinine were prepared with 0.125% WPI. Error bars indicate standard error.
5.3.3: Effect of Emulsions on Bitterness
Human psychophysical testing was used to determine how an o/w emulsion influences
bitterness of caffeine and quinine. Participants were asked to rate the in-mouth
bitterness of 0.5% and 2% oil emulsions containing either caffeine (18 mM) or quinine
HCL (0.1 mM); the emulsifier 0.125% WPI was held constant. These concentrations of
bitterant were selected based on results described in the previous chapter: they were
chosen to be moderately, but not exceedingly, bitter and no attempt was made to
precisely match them in terms of perceived intensity (Figure 4.2). Because even much
higher levels of fat caused no significant change in aqueous caffeine concentration
(Figure 5.2a), the caffeine samples were used as a negative control with the same fat
58
levels as those tested with quinine. The aqueous concentration of bitterants in the
samples was also determined via HPLC to allow direct comparison of chemical and
sensory data. It should be mentioned that emulsions can be destabilized by salivary
proteins and by oral processing (Dresselhuis et al. 2008); however the binding model
here is based on phase volumes, and not microstructure, so effects of oral processing
would not be expected to affect the results described here.
The effects of fat and aqueous bitterant concentration on perceived bitterness intensity
and aqueous concentration are shown in Figure 5.3. The results of the sensory test
support my hypotheses: the emulsions containing caffeine failed to show a significant
difference (p = 0.397) in bitterness between 0.5% oil and 2% oil, whereas in the
containing quinine emulsions, the 2% oil emulsion was significantly less bitter than the
0.5% oil emulsion (p < 0.05). These data are entirely consistent with the instrumental
HPLC data on partitioning. Specifically, the aqueous concentration of caffeine was
unchanged by the addition of oil (18.0 mM and 17.2 mM; the total caffeine in each
emulsion was 18 mM) while the aqueous concentration of quinine was much lower in
the 2% oil emulsion than in the 0.5% emulsion (0.038mM and 0.0193 mM respectively;
the total quinine concentration in each emulsion was 0.1 mM).
Because of the high multicollinearity between the sample formulation factors (fat,
aqueous concentration) and the low number of samples (two), a partial least squares
regression was used. Partial least squares is frequently used to analyze the magnitude
and sign of correlation between the predictors (factors) and the response (bitterness
intensity). The correlation between fat content and aqueous quinine concentration was
significantly correlated to the bitterness response (p = 0.016) with fat content being
highly negative in direction and aqueous concentration being highly positive in
direction. This implies that as fat increases, aqueous concentration decreases, and the
perception of quinine bitterness intensity is highly likely to decrease, as predicted by our
hypothesis.
Previously, viscosity and thickness perception have been reported to alter the perceived
taste intensity of liquid foods (Rosett et al. 1997; Arancibia et al. 2011; Torrico et al.
2015b); thus the higher viscosity of a more concentrated emulsion might be expected to
confound the bitterness results. To investigate this possibility, the viscosity was
measured for the two emulsions as a function of shear rate (Figure 5.4a and 5.4b). The
curves were modeled as Bingham-Plastics:
𝜏 = 𝜇𝛾 + 𝜏0
[5.2]
Where τ and τ0 are the stress and yield stress respectively, γ is the shear rate, and μ the
viscosity. This model gave a good fit at both fat contents (r2 > 0.98). The viscosity
(calculated as the slope of the flow curve with shear rates from 0-80 s-1) for the 0.5%
and 2% oil emulsions were 9.19 x 10-4 Pa.s and 9.36 x 10-4 Pa.s respectively, which were
not significantly different (n = 3, p = 0.184).
However, instrumental measurements do not necessarily correlate well with sensory
viscosity in emulsions and any perceived differences might bias the sensory
measurements. In order to prevent dumping effects from any thickness differences of
59
the different fat percentages, participants were also given the opportunity to rate
perceived thickness on a line scale, in addition to rating perceived bitterness. These data
are presented in Figure 5.4c. There was a small apparent drop in perceived thickness for
the 0.5% fat samples as compared to the 2% samples, but this difference did not reach
criterion for statistical significance (p = 0.152). Taken together, these results suggest
that viscosity does not confound the bitterness measurements in this work.
Once viscosity was eliminated as a potential cause of the significant decrease in the
bitterness of quinine in the 2% fat emulsion as compared to the 0.5% emulsion, binding
by the oil remains as the most plausible explanation. Perceived bitterness from quinine
decreased with an increase in oil content and was proportional to the aqueous
concentration (Figure 5.5). The open points show the dose-response function for
quinine obtaining in the previous chapter (i.e., for WPI in water without oil), while the
filled points show the aqueous concentrations obtained the emulsion system.
While the emulsions appear to be slightly more bitter than predicted for their aqueous
concentrations, the results were close to the line obtained from the aqueous solutions.
This suggests that aqueous concentration is a much better predictor of quinine
bitterness in emulsions than it was for the bitterness in protein solutions (Figure 4.3).
Also, it should be noted the concentration of WPI used here for these emulsions was
much lower than was used previously in water (0.125% and 1%, respectively). While 1%
protein bound 80% of the quinine and caused only a modest decrease in bitterness
(Figure 4.2), 0.125% protein bound approximately 56% of the quinine (Figure 4.1) and
therefore can be expected to have an even smaller effect. The small amount of protein
used here suggests the bulk of the bitterness suppression effect in the emulsion samples
is caused by fat, not protein binding.
The close agreement between the emulsion data points and the quinine dose-response
function in a water system indicates that the aqueous concentration is correlated with
what can be tasted in emulsion systems, but perhaps not in protein complexes. One
explanation could be that there is a more complete physical separation of the bitterant
in the oil phase from the saliva in the mouth whereas the protein-bound bitterant is still
available to diffuse into aqueous saliva. For example, chapter 4.5 outlines some possible
explanations for the limited suppression effect of protein-bound bitterants.
Previously, Metcalf and Vickers (2001) reported that diluting an emulsion with 9% oil
reduced the bitterness more than dilution with the same amount of water. This is
consistent with present data. However, when Metcalf and Vickers increased the oil
concentration to 17%, there was no further reduction in bitterness. As a fairly
hydrophobic compound, the proportion of quinine dissolved in oil increases rapidly with
oil volume fraction so this additional amount of fat would be expected to lead to a
relatively small change in aqueous concentration (Figure 5.1). Effective reduction of
quinine bitterness/aqueous concentration can be achieved by adding a relatively small
amount of fat and further increases in fat concentration may not lead to further
benefits.
60
Figure 5.3. Mean o/w emulsion sensory test results (n = 200) at 0.5% and 2% oil with
constant 0.125% WPI. Bitterness intensity, represented by the bars, is plotted on the
primary y-axis (left) with bitterant aqueous concentration on the secondary y-axis (right)
(caffeine = 18 mM, quinine 0.1 mM). Note: scale is shown from 40 to 65 to better show
differences. Different letters indicate significant differences (p<0.05). Error bars indicate
standard error.
Figure 5.4. Flow curves for (a) 2% oil and (b) 0.5% oil emulsions. The best fit of the
Bingham-Plastic model is shown alongside the data. (c) Sensory thickness data for
vegetable oil/water emulsions at 0.5% and 2% oil with constant 0.125% WPI. The
thickness scale is shown from 0 to 50 to better show differences and different letters
indicate significant differences (p<0.05). Error bars indicate standard error.
61
Figure 5.5. Relationship between aqueous (unbound) concentration of quinine and
perceived bitterness in water () and in emulsion systems with 0.5% and 2% fat ().
Lines are best fits of a logarithmic model. Error bars indicate standard error.
5.4: Conclusions
In summary, prior studies have implicated fat in reducing the perception of bitterness.
Here we confirm this, and extend it by showing that the suppression of bitterness by fat
depends on the hydrophobicity of the bitterant. Furthermore, the bitterness rating is
proportional to the aqueous phase concentration of an emulsion delivery system—a
clarification of previous sensory research from Metcalf and Vickers (2001). These
findings provide evidence that a bitterant must reside in the aqueous phase in order to
be perceived. Furthermore, the use of fat, rather than large amounts of sugar, as a
bitterness suppression technique in medication formulations should be pursued.
However, it remains unclear why the approach of fat in an emulsion system is more
effective at reduction of bitter than protein binding explored in the previous chapter.
62
5.5: References
Arancibia C, Jublot L, Costell E, Bayarri S (2011) Flavor release and sensory
characteristics of o/w emulsions. Influence of composition, microstructure and
rheological behavior. Food Res Int 44:1632–1641. doi:
10.1016/j.foodres.2011.04.049
Barzanti C, Evans R, Fouquet J, et al (2007) Potentiometric determination of octanolwater and liposome-water partition coefficients (log P) of ionizable organic
compounds. Tetrahedron Lett 48:3337–3341. doi: 10.1016/j.tetlet.2007.03.085
Bennett SM, Zhou L, Hayes JE (2012) Using milk fat to reduce the irritation and bitter
taste of ibuprofen. Chemosens Percept 5:231–236. doi: 10.1007/s12078-012-91286
Coupland JN, Hayes JE (2014) Physical approaches to masking bitter taste: lessons from
food and pharmaceuticals. Pharm Res 31:2921–39. doi: 10.1007/s11095-014-14806
Dresselhuis DM, de Hoog EH a, Cohen Stuart M a., et al (2008) The occurrence of inmouth coalescence of emulsion droplets in relation to perception of fat. Food
Hydrocoll 22:1170–1183. doi: 10.1016/j.foodhyd.2007.06.013
Drewnowski a, Gomez-Carneros C (2000) Bitter taste, phyonutrients, and the consumer:
a review. Am J Clin Nutr 72:1424–1435.
Hansch C, Anderson S (1967) The effect of intramolecular hydrophobic bonding on
partition coefficients. J Org Chem 32:2583–2587. doi: 10.1021/jo01283a049
Hansch C, Leo A, Hoekman D (1995) Exploring QSAR - Hydrophobic, Electronic, and
Steric Constants. American Chemical Society, Washington, DC.
Harris MF, Logan JL (2014) Determination of log Kow values for four drugs. J Chem Educ
91:915–918. doi: 10.1021/ed400655b
Homma R, Yamashita H, Funaki J, et al (2012) Identification of bitterness-masking
compounds from cheese. J Agric Food Chem 60:4492–4499. doi:
10.1021/jf300563n
Keast RSJ (2008) Modification of the bitterness of caffeine. Food Qual Prefer 19:465–
472. doi: 10.1016/j.foodqual.2008.02.002
Klebanov GS, Mednikova LN, Ovcharova AD (1967) Extraction of Caffeine from Aqueous
Solutions. Pharm Chem J 1:221–223.
Lynch J, Liu Y-H, Mela DJ, MacFie HJH (1993) A time-intensiy study of the effect of oil
mouthcoatings on taste perception. Chem Senses 18:121–129.
Machatha SG, Yalkowsky SH (2005) Comparison of the octanol/water partition
coefficients calculated by ClogP??, ACDlogP and KowWin?? to experimentally
determined values. Int J Pharm 294:185–192. doi: 10.1016/j.ijpharm.2005.01.023
Matsuo R (2000) Role of Saliva in the Maintenance of Taste Sensitivity. Crit Rev Oral Biol
Med 11:216–229. doi: 10.1177/10454411000110020501
63
Mattes RD (2007) Effects of linoleic acid on sweet, sour, salty, and bitter taste
thresholds and intensity ratings of adults. Am J Physiol Gastrointest Liver Physiol
292:G1243–G1248. doi: 10.1152/ajpgi.00510.2006
Metcalf KL, Vickers ZM (2001) Taste intensities of oil-in-water emulsions with varying fat
content. J Sens Stud 17:379–390.
Rosett TR, Kendregan SL, Klein BP (1997) Fat, protein, and mineral components of added
ingredients affect flavor qualities of tomato soups. J Food Sci 62:190–193. doi:
10.1111/j.1365-2621.1997.tb04397.x
Shahiwala A (2011) Formulation approaches in enhancement of patient compliance to
oral drug therapy. Expert Opin Drug Deliv 8:1521–1529. doi:
10.1517/17425247.2011.628311
Thurgood JE, Martini S (2010) Effects of three emulsion compositions on taste
thresholds and intensity ratings of five taste compounds. J Sens Stud 25:861–875.
doi: 10.1111/j.1745-459X.2010.00311.x
Torrico DD, Prinyawiwatkul W (2015) Psychophysical Effects of Increasing Oil
Concentrations on Saltiness and Bitterness Perception of Oil-in-Water Emulsions. J
Food Sci 80:S1885–S1892. doi: 10.1111/1750-3841.12945
Torrico DD, Sae-Eaw A, Sriwattana S, et al (2015) Oil-in-Water Emulsion Exhibits
Bitterness-Suppressing Effects in a Sensory Threshold Study. J Food Sci 80:S1404–
S1411. doi: 10.1111/1750-3841.12901
Yalkowsky SH, Valvani SC, Roseman TJ (1983) Solubility and partitioning VI: octanol
solubility and octanol-water partition coefficients. J Pharm Sci 72:866–870. doi:
10.1002/jps.2600720808
Zissimos AM, Abraham MH, Barker MC, et al (2002) Calculation of Abraham descriptors
from solvent-water partition coeffcients in four different systems; evaluation of
different methods of calculation. J Chem Soc Perkin Trans 2 470–477. doi:
10.1039/b110143a
64
CHAPTER 6
Conclusions and Future Directions
6.1 Study Significance
Bitter, one of the five basic tastes, is typically unpleasant and aversive. Many active
medicinal compounds and secondary plant metabolites used in pharmaceutical
formulations and functional foods are predominantly bitter. The bitter taste has been
shown to reduce compliance with treatment regimens (Shahiwala 2011) or reduce the
selection of more healthful foods respectively (Drewnowski and Gomez-Carneros 2000).
Individuals that are especially sensitive to bitterness, like children, are at a greater risk
of these concerns (Steiner et al. 2001; Mennella and Beauchamp 2008; Negri et al.
2012). Currently, artificial flavors and high levels of sugar are often used to suppress
bitterness in pediatric liquid oral medications, but these techniques have proved to be
less-than-perfect thus far (Mennella and Beauchamp 2008). Additionally, patients with
substantial medication regimens consume a large amount of sugar as a result of these
formulations. If, the active pharmaceutical ingredient or phytochemical in the product
imparting the primary beneficial effects is the source of bitterness, then removal of the
bitter compound is not possible. Therefore, development and research should be
focused on masking bitterness.
While there are several approaches to this problem, this work is focused on
encapsulation/binding of the tastant. Previous research in this area is limited for several
reasons. First, uncontrolled food matrices confound the results (Keast 2008; Bennett et
al. 2012; Homma et al. 2012). Second, the use of one bitterant limits the generalizability
of the results (Metcalf and Vickers 2001; Mattes 2007; Keast 2008). And third, most
studies lack the combination of adequate sensory with adequate chemical
measurements (Metcalf and Vickers 2001; Keast 2008; Thurgood and Martini 2010). This
research provides pharmaceutical researchers with the tools to use proteins and/or
small amounts of fat to optimize bitterness suppression—perhaps in conjunction with
sugar but at a lower level.
6.1.1 Summary of Results
I approached this research with the central hypothesis that the concentration of free
(unbound), aqueous bitterant will determine the intensity of perceived bitterness. I used
two food systems, protein solutions and oil-in-water emulsions, to investigate this
hypothesis. Utilizing the binding characteristics and lipid phase volume of these systems
and preventing interaction with taste receptors would, theoretically, prevent
perception. Within this framework, the main goals were to create binding curves of
either whey protein isolate (WPI) or soybean oil with model bitterants caffeine and
quinine, and determine if the unbound fraction of bitterant is proportional to the
bitterness intensity. By combining analytical measurements with sensory analysis, this
work can be used to clarify previous research on bitter suppression.
65
The specific hypotheses and results for the protein study were.
a.
b.
c.
d.
Hydrophobic binding pockets present in whey proteins would not bind
hydrophilic caffeine while hydrophobic quinine would bind strongly. Indeed,
Caffeine did not interact with WPI while quinine was strongly bound as
expected.
WPI binds small molecules in a hydrophobic manner. When the binding
assay was repeated with thermally denatured WPI, the caffeine behavior
remained unchanged while quinine binding dramatically increased. The
hydrophobic binding of WPI agrees with reports in the literature (Wu et al.
1999; Kontopidis et al. 2004).
The bitter taste of caffeine would not be affected by the presence of protein
due to lack of interaction, while quinine bitterness would be significantly
reduced in the presence of protein because of the strong binding behavior.
The 1% WPI did not affect the bitterness of caffeine which was expected. On
the other hand, compared to water, 1% WPI reduced the bitterness of all of
the quinine solutions and significantly reduced the bitterness in the two
lower (0.056 mM and 0.1 mM) quinine concentrations although not the
higher concentration (0.18 mM).
The logarithmic concentration of unbound, aqueous quinine should be
proportional to the bitterness intensity. While there was a reduction of
bitterness intensity caused by protein, the effect was rather modest and not
proportional to log aqueous concentration as was hypothesized. The
salivary environment, especially proteins present, was considered one of
the most likely variables that influenced the discrepancy between bitterness
and unbound quinine concentration. Mucin proteins did not show an
especially strong affinity for quinine which suggests that mucins may not be
the salivary protein responsible, but much is unknown about the
interaction, so that could not be said for sure.
The specific hypotheses and results for the emulsion study were:
a.
b.
Hydrophilic caffeine would have a negative log Kow (partition coefficient)
value while hydrophobic quinine would have a highly positive log K ow value.
This hypothesis was confirmed using a bulk-phase shake flask method;
caffeine had a partition coefficient of -1.32, and quinine had a partition
coefficient of 2.97.
In an oil-in-water emulsion, the aqueous concentration of caffeine should
not be affected as oil concentration changes, and quinine should decrease in
aqueous concentration as oil concentration increases. Caffeine aqueous
concentration did not significantly change with changes in oil percentages in
the emulsion system while quinine aqueous concentration decreased with
increases in oil—even at rather low oil percentages (0.5-2%). These results
agreed with the partitioning data although the free aqueous concentrations
were lower due to the presence of protein.
66
c.
d.
Caffeine bitterness would not be affected by the changes in oil percentage in
an emulsion due to lack of oil partitioning while quinine bitterness would be
significantly reduced at the higher oil percentage because of the partitioning
behavior. The caffeine bitterness was not significantly affected with changes
in oil percentage as expected. The quinine bitterness intensity was
significantly lower in the 2% oil emulsion than the 0.5% emulsion.
The logarithmic concentration of unbound, aqueous quinine should be
proportional to the bitterness intensity. The hypothesis was supported. The
better agreement between aqueous concentration and bitterness rating of
emulsion samples as compared to the WPI samples is likely due to the
complete separation of phases created by the oil in the emulsions. This
makes the oil-bound quinine less susceptible to interactions with
components of saliva.
6.1.2 Implications and Potential Applications
These results have implications for pharmaceutical development. Measuring
hydrophobicity should allow researchers to make decisions about how best to suppress
the bitterness. For example, a very hydrophobic bitterant with a highly positive Log K ow
would need only a small amount of oil to reduce the bitter perception. Even with the
lowest percentage of 0.5% oil used in this work, quinine bitterness was significantly
lower than the bitterness in water. If the partition coefficient was measured to be
hydrophilic, or even less hydrophobic, then the use of oil to suppress bitterness would
not be as desired in realistic oil quantities for pharmaceutical use.
As for the use of protein, WPI did significantly reduce the bitterness of quinine at a
relatively low level, but it was much less effective than expected. This work suggests
that more research needs to be conducted on how protein conformation and binding is
affected in the mouth and whether or not bound tastants can be received. Perhaps
protein can be utilized to reduce the bitterness of strongly-bound bitterants, but the
limitations and maximum effect is unknown. These results are not enough to positively
eliminate protein binding as a mechanism for bitterness suppression, but the topic does
require further study.
In real foods, the system is much more complex, but fat still has the possibility of
suppressing bitterness. In fact, bitter compounds in foods are likely already distributed
into any fat present without the need for added ingredients or processing steps. For
example, the fatty fraction of natural white mold cheese reduced the bitterness of
added quinine hydrochloride, leading researchers to make the conclusion that the fat
and fatty acids in cheese are the most likely components that reduce the bitter peptides
created by the cheese production (Homma et al. 2012). When developing a reduced fat
cheese, researchers could determine how much fat could be removed before the cheese
became bitter to consumers after determining the average hydrophobicity of the
peptides (of course, many other sensory characteristics need to be addressed, but this is
part of the equation). Additionally, this research could lend some learning to
67
development of coffee beverages that have a lot of inherently bitter compounds from
the bean. Those that dislike that bitterness often seek out beverages that have added
milk ingredients in addition to large amounts of sugar and flavor to reduce the
perception of bitter. Those consumers who do not wish to drink sweetened coffee
beverages but desire a reduction of bitterness could benefit from a product that uses an
optimized fat amount. Many hydrophobic compounds are found in coffee above taste
threshold such as quinoline and various pyrazines (McCamey et al. 1990). A small
amount of fat, such as an overall 2% fat content, would accomplish a reduction in
perception of bitter at the same level, or perhaps better, reduction caused by a
combination of milk, sugar, and flavor used currently.
Designing a novel food product or drug delivery system that uses fat to suppress the
bitterness of a particular compound should be approached thoughtfully. Only
hydrophobic compounds with highly positive Log Kow values are likely to be a successful
application of this work. Additionally, while I was able to clarify that aqueous
concentration was indicative of taste—at least in lipid emulsions—the bitterness
reduction caused by the increase in protein and/or fat is not nearly as dramatic as other
bitterness suppression techniques such as bitter blockers (Maehashi et al. 2008;
Gaudette and Pickering 2012). These techniques as well as small amounts of sweeteners
and flavors could be combined with the lipid approach proposed here for the most
effective bitter taste suppression. This research should therefore be treated as a
stepping stone to more fully understanding how bitterants are perceived and what
ingredient interactions alter that. From there, the ingredients can be optimized for
greater suppression.
6.2 Limitations and Recommendations for Future Work
There are limitations to this work, most notably protein binding was measured in the
sample before it was taken into the mouth where the interactions may change
completely. This was highlighted by the shifted dose response curve (Figure 4.3) that
was not fully explained by variables I tested. There are many components in saliva that
have demonstrated effects on taste perception (Harrison 1998; Melis et al. 2013) as well
as inducing depletion flocculation in emulsions (Vingerhoeds et al. 2005).
Preliminary results reported here show, salivary mucins do interact with bitterants and
could potentially disrupt binding inherently present in the food. Therefore, analyzing the
complex after contact with saliva would be a useful study. A future recommendation
would be to ask panelists to rate the bitterness of protein solutions, as was done in this
work, but then ask them to expectorate the samples into a cup for analysis. First, the
aqueous concentration should be measured to determine using the methods described
here. Next, the various proteins in the solution could be separated using size exclusion
chromatography. Each protein fraction could then be hydrolyzed, and the amount of
quinine could be determined. For example, if salivary proteins competitively bind
quinine, it would be expected that after being in the mouth, the amount of quinine
bound to WPI would be less than in the original solution. Quinine could be determined
68
to be bound to other salivary protein fractions. Additionally, a purified WPI fraction
could be re-investigated using NMR to determine whether the quinine-WPI complex
remained intact, and if the binding potential changed. Protease is a common salivary
component studied in this literature, and it would be valuable to determine if the
protein itself was affected and if it changed the binding. If salivary protease significantly
hydrolyzed the whey protein, that would also explain why a presumably protein-bound
bitterant can be tasted.
A similar approach could be taken to better understand the behavior of the emulsion
system. Being able to model exactly where the quinine is in the system would allow
researchers to better understand how changes to the system are affecting the taste. In
order to do this, the lipid fraction could be collected and analyzed using normal phase
HPLC as well as the aqueous as was done in this work. Then, the remaining quinine not
detected in the two phases would be bound to the protein. Additionally, the above
panelist experiment could be repeated with the emulsion samples followed by particle
size analysis (to ensure emulsion stability) and quantification of any changes in the
phase location of quinine. NMR could also be used in the collected emulsions following
filtration and separation as was outlined below to determine whether or not the saliva
changed the complex or conformation of WPI. The aqueous concentration was not
completely indicative of the bitterness rating of quinine in the emulsion samples, so
effects by saliva in the mouth are possible.
A second limitation is that the exact binding of quinine by WPI was not determined,
which makes the mechanism of bitterness suppression (or the lack thereof) hard to
explain. Because protein binding played a role in both the protein experiments and the
emulsion experiments, the binding specifics would be useful across this work.
One way to address this is to investigate the binding of quinine and WPI via nuclear
magnetic resonance (NMR). NMR has the ability to detect specific small molecule ligandprotein binding locations by mapping which hydrogen atoms are interacting between
the two (Williamson 2013). This would better classify how far into the interior of WPI
that quinine binds which would give a better indication of whether the protein-binding
truly shields the quinine from the bitter taste receptor. Additionally, this NMR binding
data could be used in conjunction with molecular modeling of taste receptor
interaction. An accurate binding location would allow for the development of a model
for the quinine-WPI complex. Then, the protein complex could be used in a theoretical
binding assay with a previously-sequenced bitter receptor such as the PTC receptor that
quinine interacts strongly with (Floriano et al. 2006). Controlled complex-receptor
modeling will provide better clarity on whether or not the protein bound-quinine is
accessible to the receptor. Finally, if any salivary components from the previous spit cup
test are implicated in altering protein binding, these could be included in the model for
potential taste activity.
These experiments would give more of an analytical measure to elucidate how the
environment of the mouth affects the physical structure of protein complexes and oil-inwater emulsions. Any changes in the structure could then be compared to the bitterness
69
ratings to determine whether or not they were responsible for deviations from the
aqueous concentration model proposed here.
70
6.3 References
Bennett SM, Zhou L, Hayes JE (2012) Using milk fat to reduce the irritation and bitter
taste of ibuprofen. Chemosens Percept 5:231–236. doi: 10.1007/s12078-012-91286
Drewnowski a, Gomez-Carneros C (2000) Bitter taste, phyonutrients, and the consumer:
a review. Am J Clin Nutr 72:1424–1435.
Floriano WB, Hall S, Vaidehi N, et al (2006) Modeling the human PTC bitter-taste
receptor interactions with bitter tastants. J Mol Model 12:931–941. doi:
10.1007/s00894-006-0102-6
Gaudette NJ, Pickering GJ (2012) The efficacy of bitter blockers on health-relevant
bitterants. J Funct Foods 4:177–184. doi: 10.1016/j.jff.2011.10.003
Harrison M (1998) Effect of breathing and saliva flow on flavor release from liquid foods.
J Agric Food Chem 46:2727–2735.
Homma R, Yamashita H, Funaki J, et al (2012) Identification of bitterness-masking
compounds from cheese. J Agric Food Chem 60:4492–4499. doi:
10.1021/jf300563n
Keast RSJ (2008) Modification of the bitterness of caffeine. Food Qual Prefer 19:465–
472. doi: 10.1016/j.foodqual.2008.02.002
Kontopidis G, Holt C, Sawyer L (2004) Invited Review: β-Lactoglobulin: Binding
Properties, Structure, and Function. J Dairy Sci 87:785–796. doi:
10.3168/jds.S0022-0302(04)73222-1
Maehashi K, Matano M, Nonaka M, et al (2008) Riboflavin-binding protein is a novel
bitter inhibitor. Chem Senses 33:57–63. doi: 10.1093/chemse/bjm062
Mattes RD (2007) Effects of linoleic acid on sweet, sour, salty, and bitter taste
thresholds and intensity ratings of adults. Am J Physiol Gastrointest Liver Physiol
292:G1243–G1248. doi: 10.1152/ajpgi.00510.2006
McCamey DA, Thorpe TM, McCarthy JP (1990) Coffee Bitterness. Dev Food Sci 25:169–
182.
Melis M, Aragoni MC, Arca M, et al (2013) Marked increase in PROP taste
responsiveness following oral supplementation with selected salivary proteins or
their related free amino acids. PLoS One 8:e59810. doi:
10.1371/journal.pone.0059810
Mennella JA, Beauchamp GK (2008) Optimizing oral medications for children. Clin Ther
30:2120–2132. doi: 10.1016/j.clinthera.2008.11.018
Metcalf KL, Vickers ZM (2001) Taste intensities of oil-in-water emulsions with varying fat
content. J Sens Stud 17:379–390.
Negri R, Di Feola M, Di Domenico S, et al (2012) Taste Perception and Food Choices. J
Pediatr Gastroenterol Nutr 54:624–629. doi: 10.1097/MPG.0b013e3182473308
71
Shahiwala A (2011) Formulation approaches in enhancement of patient compliance to
oral drug therapy. Expert Opin Drug Deliv 8:1521–1529. doi:
10.1517/17425247.2011.628311
Steiner JE, Glaser D, Hawilo ME, Berridge KC (2001) Comparative expression of hedonic
impact: Affective reactions to taste by human infants and other primates. Neurosci
Biobehav Rev 25:53–74. doi: 10.1016/S0149-7634(00)00051-8
Thurgood JE, Martini S (2010) Effects of three emulsion compositions on taste
thresholds and intensity ratings of five taste compounds. J Sens Stud 25:861–875.
doi: 10.1111/j.1745-459X.2010.00311.x
Vingerhoeds MH, Blijdenstein TBJ, Zoet FD, van Aken G a. (2005) Emulsion flocculation
induced by saliva and mucin. Food Hydrocoll 19:915–922. doi:
10.1016/j.foodhyd.2004.12.005
Williamson MP (2013) Using chemical shift perturbation to characterise ligand binding.
Prog Nucl Magn Reson Spectrosc 73:1–16. doi: 10.1016/j.pnmrs.2013.02.001
Wu SY, Pérez MD, Puyol P, Sawyer L (1999) β-Lactoglobulin binds palmitate within its
central cavity. J Biol Chem 274:170–174. doi: 10.1074/jbc.274.1.170
72
APPENDIX
Chapter 1 Copyright Permissions
License Agreement for Figure 1.1, from Mennella et al. 2013.
This image was published in Clinical Therapeutics, Volume 35, Mennella, JA,
Spector, AC, Reed, DR, Coldwell, SE, The Bad Taste of Medicines: Overview of
Basic Research on Bitter Taste, 1225-1246, Copyright Elsevier (2013)
This is a License Agreement between Kelsey N Tenney ("You") and Elsevier
("Elsevier") provided by Copyright Clearance Center ("CCC"). The license
consists of your order details, the terms and conditions provided by Elsevier, and
the payment terms and conditions.
All payments must be made in full to CCC. For payment instructions, please
see information listed at the bottom of this form.
Supplier
Elsevier Limited
The Boulevard,Langford Lane
Kidlington,Oxford,OX5 1GB,UK
Registered Company
Number
1982084
Customer name
Kelsey N Tenney
Customer address
720 Stratford Dr. Apt 9
STATE COLLEGE, PA 16801
License number
3821901102320
License date
Mar 04, 2016
Licensed content publisher
Elsevier
Licensed content publication Clinical Therapeutics
Licensed content title
The Bad Taste of Medicines: Overview of Basic Research
on Bitter Taste
Licensed content author
Julie A. Mennella,Alan C. Spector,Danielle R. Reed,Susan
E. Coldwell
Licensed content date
August 2013
Licensed content volume
number
35
Licensed content issue
number
8
Number of pages
22
Start Page
1225
End Page
1246
Type of Use
reuse in a thesis/dissertation
73
Portion
figures/tables/illustrations
Number of
figures/tables/illustrations
1
Format
both print and electronic
Are you the author of this
Elsevier article?
No
Will you be translating?
No
Original figure numbers
figure 1
Title of your
thesis/dissertation
Bitter: Understanding how protein, lipids, and oil-inwater emulsions influence perception
Expected completion date
Apr 2016
Estimated size (number of
pages)
90
Elsevier VAT number
GB 494 6272 12
Permissions price
0.00 USD
VAT/Local Sales Tax
0.00 USD / 0.00 GBP
Total
0.00 USD
Terms and Conditions
INTRODUCTION
1. The publisher for this copyrighted material is Elsevier. By clicking "accept" in
connection with completing this licensing transaction, you agree that the
following terms and conditions apply to this transaction (along with the Billing
and Payment terms and conditions established by Copyright Clearance Center,
Inc. ("CCC"), at the time that you opened your Rightslink account and that are
available at any time at http://myaccount.copyright.com).
GENERAL TERMS
2. Elsevier hereby grants you permission to reproduce the aforementioned
material subject to the terms and conditions indicated.
3. Acknowledgement: If any part of the material to be used (for example, figures)
has appeared in our publication with credit or acknowledgement to another
source, permission must also be sought from that source. If such permission is
not obtained then that material may not be included in your publication/copies.
Suitable acknowledgement to the source must be made, either as a footnote or in
a reference list at the end of your publication, as follows:
"Reprinted from Publication title, Vol /edition number, Author(s), Title of article
/ title of chapter, Pages No., Copyright (Year), with permission from Elsevier
[OR APPLICABLE SOCIETY COPYRIGHT OWNER]." Also Lancet special
credit - "Reprinted from The Lancet, Vol. number, Author(s), Title of article,
Pages No., Copyright (Year), with permission from Elsevier."
4. Reproduction of this material is confined to the purpose and/or media for
which permission is hereby given.
5. Altering/Modifying Material: Not Permitted. However figures and illustrations
may be altered/adapted minimally to serve your work. Any other abbreviations,
74
additions, deletions and/or any other alterations shall be made only with prior
written authorization of Elsevier Ltd. (Please contact Elsevier at
[email protected])
6. If the permission fee for the requested use of our material is waived in this
instance, please be advised that your future requests for Elsevier materials may
attract a fee.
7. Reservation of Rights: Publisher reserves all rights not specifically granted in
the combination of (i) the license details provided by you and accepted in the
course of this licensing transaction, (ii) these terms and conditions and (iii)
CCC's Billing and Payment terms and conditions.
8. License Contingent Upon Payment: While you may exercise the rights
licensed immediately upon issuance of the license at the end of the licensing
process for the transaction, provided that you have disclosed complete and
accurate details of your proposed use, no license is finally effective unless and
until full payment is received from you (either by publisher or by CCC) as
provided in CCC's Billing and Payment terms and conditions. If full payment is
not received on a timely basis, then any license preliminarily granted shall be
deemed automatically revoked and shall be void as if never granted. Further, in
the event that you breach any of these terms and conditions or any of CCC's
Billing and Payment terms and conditions, the license is automatically revoked
and shall be void as if never granted. Use of materials as described in a revoked
license, as well as any use of the materials beyond the scope of an unrevoked
license, may constitute copyright infringement and publisher reserves the right to
take any and all action to protect its copyright in the materials.
9. Warranties: Publisher makes no representations or warranties with respect to
the licensed material.
10. Indemnity: You hereby indemnify and agree to hold harmless publisher and
CCC, and their respective officers, directors, employees and agents, from and
against any and all claims arising out of your use of the licensed material other
than as specifically authorized pursuant to this license.
11. No Transfer of License: This license is personal to you and may not be
sublicensed, assigned, or transferred by you to any other person without
publisher's written permission.
12. No Amendment Except in Writing: This license may not be amended except
in a writing signed by both parties (or, in the case of publisher, by CCC on
publisher's behalf).
13. Objection to Contrary Terms: Publisher hereby objects to any terms
contained in any purchase order, acknowledgment, check endorsement or other
writing prepared by you, which terms are inconsistent with these terms and
conditions or CCC's Billing and Payment terms and conditions. These terms and
conditions, together with CCC's Billing and Payment terms and conditions
(which are incorporated herein), comprise the entire agreement between you and
publisher (and CCC) concerning this licensing transaction. In the event of any
conflict between your obligations established by these terms and conditions and
those established by CCC's Billing and Payment terms and conditions, these
terms and conditions shall control.
75
14. Revocation: Elsevier or Copyright Clearance Center may deny the
permissions described in this License at their sole discretion, for any reason or no
reason, with a full refund payable to you. Notice of such denial will be made
using the contact information provided by you. Failure to receive such notice
will not alter or invalidate the denial. In no event will Elsevier or Copyright
Clearance Center be responsible or liable for any costs, expenses or damage
incurred by you as a result of a denial of your permission request, other than a
refund of the amount(s) paid by you to Elsevier and/or Copyright Clearance
Center for denied permissions.
LIMITED LICENSE
The following terms and conditions apply only to specific license types:
15. Translation: This permission is granted for non-exclusive world English
rights only unless your license was granted for translation rights. If you licensed
translation rights you may only translate this content into the languages you
requested. A professional translator must perform all translations and reproduce
the content word for word preserving the integrity of the article.
16. Posting licensed content on any Website: The following terms and
conditions apply as follows: Licensing material from an Elsevier journal: All
content posted to the web site must maintain the copyright information line on
the bottom of each image; A hyper-text must be included to the Homepage of the
journal from which you are licensing at
http://www.sciencedirect.com/science/journal/xxxxx or the Elsevier homepage
for books at http://www.elsevier.com; Central Storage: This license does not
include permission for a scanned version of the material to be stored in a central
repository such as that provided by Heron/XanEdu.
Licensing material from an Elsevier book: A hyper-text link must be included to
the Elsevier homepage at http://www.elsevier.com . All content posted to the web
site must maintain the copyright information line on the bottom of each image.
Posting licensed content on Electronic reserve: In addition to the above the
following clauses are applicable: The web site must be password-protected and
made available only to bona fide students registered on a relevant course. This
permission is granted for 1 year only. You may obtain a new license for future
website posting.
17. For journal authors: the following clauses are applicable in addition to the
above:
Preprints:
A preprint is an author's own write-up of research results and analysis, it has not
been peer-reviewed, nor has it had any other value added to it by a publisher
(such as formatting, copyright, technical enhancement etc.).
Authors can share their preprints anywhere at any time. Preprints should not be
added to or enhanced in any way in order to appear more like, or to substitute for,
the final versions of articles however authors can update their preprints on arXiv
or RePEc with their Accepted Author Manuscript (see below).
If accepted for publication, we encourage authors to link from the preprint to
their formal publication via its DOI. Millions of researchers have access to the
76
formal publications on ScienceDirect, and so links will help users to find, access,
cite and use the best available version. Please note that Cell Press, The Lancet
and some society-owned have different preprint policies. Information on these
policies is available on the journal homepage.
Accepted Author Manuscripts: An accepted author manuscript is the
manuscript of an article that has been accepted for publication and which
typically includes author-incorporated changes suggested during submission,
peer review and editor-author communications.
Authors can share their accepted author manuscript:
 
immediately
o via their non-commercial person homepage or blog
o by updating a preprint in arXiv or RePEc with the accepted
manuscript
o via their research institute or institutional repository for
internal institutional uses or as part of an invitation-only
research collaboration work-group
o directly by providing copies to their students or to research
collaborators for their personal use
o for private scholarly sharing as part of an invitation-only
work group on commercial sites with which Elsevier has
an agreement
 
after the embargo period
o via non-commercial hosting platforms such as their
institutional repository
o via commercial sites with which Elsevier has an agreement
In all cases accepted manuscripts should:
 
link to the formal publication via its DOI
 
bear a CC-BY-NC-ND license - this is easy to do
 
if aggregated with other manuscripts, for example in a repository or
other site, be shared in alignment with our hosting policy not be added to
or enhanced in any way to appear more like, or to substitute for, the
published journal article.
Published journal article (JPA): A published journal article (PJA) is the
definitive final record of published research that appears or will appear in the
journal and embodies all value-adding publishing activities including peer review
co-ordination, copy-editing, formatting, (if relevant) pagination and online
enrichment.
Policies for sharing publishing journal articles differ for subscription and gold
open access articles:
Subscription Articles: If you are an author, please share a link to your article
rather than the full-text. Millions of researchers have access to the formal
publications on ScienceDirect, and so links will help your users to find, access,
cite, and use the best available version.
Theses and dissertations which contain embedded PJAs as part of the formal
submission can be posted publicly by the awarding institution with DOI links
back to the formal publications on ScienceDirect.
77
If you are affiliated with a library that subscribes to ScienceDirect you have
additional private sharing rights for others' research accessed under that
agreement. This includes use for classroom teaching and internal training at the
institution (including use in course packs and courseware programs), and
inclusion of the article for grant funding purposes.
Gold Open Access Articles: May be shared according to the author-selected
end-user license and should contain a CrossMark logo, the end user license, and a
DOI link to the formal publication on ScienceDirect.
Please refer to Elsevier's posting policy for further information.
18. For book authors the following clauses are applicable in addition to the
above: Authors are permitted to place a brief summary of their work online
only. You are not allowed to download and post the published electronic version
of your chapter, nor may you scan the printed edition to create an electronic
version. Posting to a repository: Authors are permitted to post a summary of
their chapter only in their institution's repository.
19. Thesis/Dissertation: If your license is for use in a thesis/dissertation your
thesis may be submitted to your institution in either print or electronic form.
Should your thesis be published commercially, please reapply for permission.
These requirements include permission for the Library and Archives of Canada to
supply single copies, on demand, of the complete thesis and include permission
for Proquest/UMI to supply single copies, on demand, of the complete thesis.
Should your thesis be published commercially, please reapply for permission.
Theses and dissertations which contain embedded PJAs as part of the formal
submission can be posted publicly by the awarding institution with DOI links
back to the formal publications on ScienceDirect.
Elsevier Open Access Terms and Conditions
You can publish open access with Elsevier in hundreds of open access journals or
in nearly 2000 established subscription journals that support open access
publishing. Permitted third party re-use of these open access articles is defined by
the author's choice of Creative Commons user license. See our open access
license policy for more information.
Terms & Conditions applicable to all Open Access articles published with
Elsevier:
Any reuse of the article must not represent the author as endorsing the adaptation
of the article nor should the article be modified in such a way as to damage the
author's honour or reputation. If any changes have been made, such changes must
be clearly indicated.
The author(s) must be appropriately credited and we ask that you include the end
user license and a DOI link to the formal publication on ScienceDirect.
If any part of the material to be used (for example, figures) has appeared in our
publication with credit or acknowledgement to another source it is the
responsibility of the user to ensure their reuse complies with the terms and
conditions determined by the rights holder.
Additional Terms & Conditions applicable to each Creative Commons user
license:
78
CC BY: The CC-BY license allows users to copy, to create extracts, abstracts
and new works from the Article, to alter and revise the Article and to make
commercial use of the Article (including reuse and/or resale of the Article by
commercial entities), provided the user gives appropriate credit (with a link to the
formal publication through the relevant DOI), provides a link to the license,
indicates if changes were made and the licensor is not represented as endorsing
the use made of the work. The full details of the license are available at
http://creativecommons.org/licenses/by/4.0.
CC BY NC SA: The CC BY-NC-SA license allows users to copy, to create
extracts, abstracts and new works from the Article, to alter and revise the Article,
provided this is not done for commercial purposes, and that the user gives
appropriate credit (with a link to the formal publication through the relevant
DOI), provides a link to the license, indicates if changes were made and the
licensor is not represented as endorsing the use made of the work. Further, any
new works must be made available on the same conditions. The full details of the
license are available at http://creativecommons.org/licenses/by-nc-sa/4.0.
CC BY NC ND: The CC BY-NC-ND license allows users to copy and distribute
the Article, provided this is not done for commercial purposes and further does
not permit distribution of the Article if it is changed or edited in any way, and
provided the user gives appropriate credit (with a link to the formal publication
through the relevant DOI), provides a link to the license, and that the licensor is
not represented as endorsing the use made of the work. The full details of the
license are available at http://creativecommons.org/licenses/by-nc-nd/4.0. Any
commercial reuse of Open Access articles published with a CC BY NC SA or CC
BY NC ND license requires permission from Elsevier and will be subject to a
fee.
Commercial reuse includes:
 
Associating advertising with the full text of the Article
 
Charging fees for document delivery or access
 
Article aggregation
 
Systematic distribution via e-mail lists or share buttons
Posting or linking by commercial companies for use by customers of those
companies.
20. Other Conditions:
v1.8
Questions? [email protected] or +1-855-239-3415 (toll free in
the US) or +1-978-646-2777.
79
License Agreement for Figure 1.2, from Matsuo 2000.
This image was published in Critical Reviews in Oral Biology & Medicine, Volume
11, Matsuo R, Role of Saliva in the Maintenance of Taste Sensitivity, 216-229,
Copyright SAGE Journals (2000).
Thank you for your email. Please find your intended use below and read through
the instructions.
FOR DISSERTATION AND THESIS USE
If you are requesting to reuse up to one full article in your unpublished
dissertation or thesis than please accept this automatic response as written
permission providing full reference to the original author and SAGE publication
is included. Where practicable please notify the author of your intention to
reuse the material and hyperlink to the article on SAGE journals.
License Agreement for Figure 1.3, from Coupland and Hayes 2014.
This image was published in Pharmaceutical Research, Volume 31, Coupland JN,
Hayes JE, Physical Approaches to Masking Bitter Taste: Lessons from Food and
Pharmaceuticals, 2921-2939, Copyright Springer (2014).
This is a License Agreement between Kelsey N Tenney ("You") and Springer
("Springer") provided by Copyright Clearance Center ("CCC"). The license
consists of your order details, the terms and conditions provided by Springer, and
the payment terms and conditions.
All payments must be made in full to CCC. For payment instructions, please
see information listed at the bottom of this form.
License Number
3816540886673
License date
Feb 26, 2016
Licensed content publisher
Springer
Licensed content publication Pharmaceutical Research
Licensed content title
Physical Approaches to Masking Bitter Taste: Lessons
from Food and Pharmaceuticals
Licensed content author
John N. Coupland
Licensed content date
Jan 1, 2014
Volume number
31
80
Issue number
11
Type of Use
Thesis/Dissertation
Portion
Figures/tables/illustrations
Number of
figures/tables/illustrations
1
Author of this Springer
article
No
Order reference number
None
Original figure numbers
Figure 3
Title of your thesis /
dissertation
Bitter: Understanding how protein, lipids, and oil-inwater emulsions influence perception
Expected completion date
Apr 2016
Estimated size(pages)
90
Total
0.00 USD
Terms and Conditions
Introduction
The publisher for this copyrighted material is Springer. By clicking "accept" in
connection with completing this licensing transaction, you agree that the
following terms and conditions apply to this transaction (along with the Billing
and Payment terms and conditions established by Copyright Clearance Center,
Inc. ("CCC"), at the time that you opened your Rightslink account and that are
available at any time at http://myaccount.copyright.com).
Limited License
With reference to your request to reuse material on which Springer controls the
copyright, permission is granted for the use indicated in your enquiry under the
following conditions:
- Licenses are for one-time use only with a maximum distribution equal to the
number stated in your request.
- Springer material represents original material which does not carry references to
other sources. If the material in question appears with a credit to another source,
this permission is not valid and authorization has to be obtained from the original
copyright holder.
- This permission
• is non-exclusive
• is only valid if no personal rights, trademarks, or competitive products are
infringed.
• explicitly excludes the right for derivatives.
- Springer does not supply original artwork or content.
- According to the format which you have selected, the following conditions
apply accordingly:
• Print and Electronic: This License include use in electronic form provided it
is password protected, on intranet, or CD-Rom/DVD or E-book/E-journal. It may
81
not be republished in electronic open access.
• Print: This License excludes use in electronic form.
• Electronic: This License only pertains to use in electronic form provided it is
password protected, on intranet, or CD-Rom/DVD or E-book/E-journal. It may
not be republished in electronic open access.
For any electronic use not mentioned, please contact Springer at
[email protected].
- Although Springer controls the copyright to the material and is entitled to
negotiate on rights, this license is only valid subject to courtesy information to
the author (address is given in the article/chapter).
- If you are an STM Signatory or your work will be published by an STM
Signatory and you are requesting to reuse figures/tables/illustrations or single text
extracts, permission is granted according to STM Permissions Guidelines:
http://www.stm-assoc.org/permissions-guidelines/
For any electronic use not mentioned in the Guidelines, please contact Springer at
[email protected]. If you request to reuse more content than
stipulated in the STM Permissions Guidelines, you will be charged a permission
fee for the excess content.
Permission is valid upon payment of the fee as indicated in the licensing process.
If permission is granted free of charge on this occasion, that does not prejudice
any rights we might have to charge for reproduction of our copyrighted material
in the future.
-If your request is for reuse in a Thesis, permission is granted free of charge
under the following conditions:
This license is valid for one-time use only for the purpose of defending your
thesis and with a maximum of 100 extra copies in paper. If the thesis is going to
be published, permission needs to be reobtained.
- includes use in an electronic form, provided it is an author-created version of
the thesis on his/her own website and his/her university’s repository, including
UMI (according to the definition on the Sherpa website:
http://www.sherpa.ac.uk/romeo/);
- is subject to courtesy information to the co-author or corresponding author.
Geographic Rights: Scope
Licenses may be exercised anywhere in the world.
Altering/Modifying Material: Not Permitted
Figures, tables, and illustrations may be altered minimally to serve your work.
You may not alter or modify text in any manner. Abbreviations, additions,
deletions and/or any other alterations shall be made only with prior written
authorization of the author(s).
Reservation of Rights
Springer reserves all rights not specifically granted in the combination of (i) the
license details provided by you and accepted in the course of this licensing
transaction and (ii) these terms and conditions and (iii) CCC's Billing and
Payment terms and conditions.
License Contingent on Payment
While you may exercise the rights licensed immediately upon issuance of the
82
license at the end of the licensing process for the transaction, provided that you
have disclosed complete and accurate details of your proposed use, no license is
finally effective unless and until full payment is received from you (either by
Springer or by CCC) as provided in CCC's Billing and Payment terms and
conditions. If full payment is not received by the date due, then any license
preliminarily granted shall be deemed automatically revoked and shall be void as
if never granted. Further, in the event that you breach any of these terms and
conditions or any of CCC's Billing and Payment terms and conditions, the license
is automatically revoked and shall be void as if never granted. Use of materials as
described in a revoked license, as well as any use of the materials beyond the
scope of an unrevoked license, may constitute copyright infringement and
Springer reserves the right to take any and all action to protect its copyright in the
materials.
Copyright Notice: Disclaimer
You must include the following copyright and permission notice in connection
with any reproduction of the licensed material:
"Springer book/journal title, chapter/article title, volume, year of publication,
page, name(s) of author(s), (original copyright notice as given in the publication
in which the material was originally published) "With permission of Springer"
In case of use of a graph or illustration, the caption of the graph or illustration
must be included, as it is indicated in the original publication.
Warranties: None
Springer makes no representations or warranties with respect to the licensed
material and adopts on its own behalf the limitations and disclaimers established
by CCC on its behalf in its Billing and Payment terms and conditions for this
licensing transaction.
Indemnity
You hereby indemnify and agree to hold harmless Springer and CCC, and their
respective officers, directors, employees and agents, from and against any and all
claims arising out of your use of the licensed material other than as specifically
authorized pursuant to this license.
No Transfer of License
This license is personal to you and may not be sublicensed, assigned, or
transferred by you without Springer's written permission.
No Amendment Except in Writing
This license may not be amended except in a writing signed by both parties (or,
in the case of Springer, by CCC on Springer's behalf).
Objection to Contrary Terms
Springer hereby objects to any terms contained in any purchase order,
acknowledgment, check endorsement or other writing prepared by you, which
terms are inconsistent with these terms and conditions or CCC's Billing and
Payment terms and conditions. These terms and conditions, together with CCC's
Billing and Payment terms and conditions (which are incorporated herein),
comprise the entire agreement between you and Springer (and CCC) concerning
this licensing transaction. In the event of any conflict between your obligations
83
established by these terms and conditions and those established by CCC's Billing
and Payment terms and conditions, these terms and conditions shall control.
Jurisdiction
All disputes that may arise in connection with this present License, or the breach
thereof, shall be settled exclusively by arbitration, to be held in the Federal
Republic of Germany, in accordance with German law.
Other conditions:
V 12AUG2015
Questions? [email protected] or +1-855-239-3415 (toll free in
the US) or +1-978-646-2777.
84