Professional Documents
Culture Documents
DOI 10.1007/s00421-014-2987-6
ORIGINAL ARTICLE
Received: 15 April 2014 / Accepted: 20 August 2014 / Published online: 19 September 2014
© Springer-Verlag Berlin Heidelberg 2014
13
130 Eur J Appl Physiol (2015) 115:129–138
13
Eur J Appl Physiol (2015) 115:129–138 131
and six solutions of a single NaCl concentration (Sigma- One-way repeated measures ANOVA was used to compare
Aldrich, St. Loius, MO, USA). NaCl was presented in a the responses between concentrations followed by Fisher’s
standardized concentration of 0.1 M to serve as a dishabitu- LSD post hoc tests, with differences accepted at p < 0.05
ator (Epstein et al. 1992). All solutions were prepared with (Statistical Package for the Social Sciences Version 17.0 for
doubly deionized water (Direct-Q 5 Ultrapure Water Sys- Windows software, Chicago, IL, USA). Participants were
tem, Millipore, MA, USA). then divided into groups based on their PROP tasting sta-
On the day of testing, all solutions were kept at room tus, with the ANS and subjective responses compared using
temperature (24 °C) and placed on a table in front of the two-way mixed-model ANOVA with Fisher’s LSD post hoc
participant in identical plastic cups. At the onset of the test, tests. The ANS and subjective responses to mouth rinsing
participants first tasted one of the NaCl solutions, then alter- with a 10 mM solution were compared with the responses
nated quinine and NaCl solutions. Quinine solutions were of the combined mouth rinsing and ingestion of a 2 mM
presented in order of ascending concentration to prevent the solution using Students t-tests. All values, unless otherwise
expected nausea associated with tasting the highest concen- stated, are expressed as mean ± SEM.
tration of quinine from invalidating the assessment of sub-
sequent solutions. Each solution was rinsed in the mouth Main study: effect of quinine mouth rinsing on sprint
for 10 s before being expelled. Immediately following qui- performance
nine expulsion, participants gave subjective ratings of taste
intensity (INT) and unpleasantness (UNP) using a general Twelve trained male cyclists (mean ± SD age 27 ± 3 years,
labeled magnitude scale (gLMS; Green et al. 1996), with height 180 ± 8 cm, mass 75.64 ± 11.51 kg, V̇ O2peak
adjectives altered to specify either intensity or unpleasant- 61.30 ± 6.91 mL/kg/min) were recruited for this study.
ness. Participants also gave ratings of nausea (NAS) using a Participants were fully informed of the testing procedures
100 mm visual analog scale (VAS) bounded by the descrip- before their written consent was obtained. However, to min-
tors “no nausea” and “extreme nausea” (Muth et al. 1996). imize the possibility of a placebo effect, they were deceived
ANS responses to quinine were continuously assessed about the true purpose of the study, and were instead
indirectly by measuring skin conductance (SC) amplitude informed that the aim of the study was to determine the
and instantaneous heart rate (IHR) amplitude, as described effect of different taste solutions on the metabolic responses
in Gam et al. (2014). In addition to the amplitude of the SC to maximal exercise. After completing all trials, participants
response, the ohmic perturbation duration (OPD) index was were personally debriefed as to the true aim of the study.
also analyzed, which has been shown to reflect the emo- The procedures were approved by the Human Research Eth-
tional load of the stimulus (Vernet-Maury et al. 1995). The ics Committee of The University of Western Australia.
next solution (either quinine or NaCl) was administered
once skin conductance and heart rate returned to base- Experimental design
line levels and were no longer fluctuating (approximately
2–3 min). In the rest period between solutions, participants Each participant visited the laboratory on five separate occa-
rinsed their mouth with water ad libitum. Preliminary work sions, each separated by 7 days and conducted at the same
was performed to ensure that the rest period between con- time of the day for each participant. The initial visit involved
secutive quinine exposures at a given concentration did not the assessment of PROP tasting status and V̇ O2peak prior to
result in habituation of taste or ANS responses. familiarization with the mouth rinsing procedure and sprint
protocol to be used in the subsequent experimental trials.
Session 3: comparison of the effect of ingesting a 2 mM For the following four experimental trials, participants
quinine solution and a 10 mM quinine mouth rinse on sub‑ completed a 30-s maximal cycling sprint immediately after
jective and ANS responses To compare the subjective and rinsing their mouth and expelling either plain water (WAT),
ANS responses between the mouth rinsing protocol adopted a sweet glucose solution (GLU), a bitter quinine solution
in this study and the mouth rinsing plus ingestion protocol (QUI), or a no-rinse control (CON) administered in a rand-
described in Gam et al. (2014), these two protocols were omized, counterbalanced order. The glucose solution acted
compared with ANS and subjective responses assessed as as an additional placebo solution given that mouth rinsing
described above. with such a solution does not affect sprint performance
(Chong et al. 2011).
Data and statistical analysis
Familiarization session
Psychophysical curves were graphed for the mean
ANS (SC, IHR, OPD) and subjective (INT, UNP, NAS) Individual differences in bitter taste perception were first
responses across the range of solution concentrations. taken into account, with participants’ PROP tasting status
13
132 Eur J Appl Physiol (2015) 115:129–138
determined as described previously. Following the PROP Before the commencement of the mouth rinse pro-
tasting test, participants completed aV̇ O2peak test on an air- tocol and at 0 and 7 min post-sprint, heart rate (HR) was
braked cycle ergometer (Evolution Pty. Ltd., Adelaide, Aus- recorded and each participant provided subjective ratings
tralia) using the protocol described in Gam et al. (2014). of perceived exertion (RPE; Borg 1982) and nausea. Nau-
Following the V̇ O2peak test, participants completed the sea ratings were made using a 100 mm visual analog scale
sprint cycling protocol to be used in the subsequent experi- anchored with the descriptors “No Nausea” and “Extreme
mental trials preceded by a plain water rinse to become Nausea” (Muth et al. 1996). At the same timepoints, a cap-
accustomed to the experimental procedures. Before leaving illary blood sample (125 µL) was taken from the fingertip
the laboratory, participants were given a food and physical (Clinitubes, Radiometer, Copenhagen) and analyzed imme-
activity diary, and were asked to record all food and drink diately for blood lactate and glucose using a blood gas ana-
intake and physical activity in the 24 h prior to each experi- lyzer (ABL™ 725, Radiometer, Copenhagen).
mental trial. Energy intake (kilojoules) was quantified from
the food diaries using commercially available software Statistical analyses
(Foodworks; Xyris Software, Kenmore Hills, QLD, Aus-
tralia). A copy of the diary from the first trial was returned The effects of mouth rinse treatment on each performance
to each participant, and they were asked to replicate the variable were compared using two-way repeated measures
same diet and activity pattern in the 24 h prior to each trial ANOVA followed by Fisher’s LSD post hoc tests, with dif-
(Jeacocke and Burke 2010). Compliance was confirmed ferences accepted at p < 0.05 (Statistical Package for the
upon arrival to the laboratory for each experimental trial Social Sciences Version 17.0 for Windows software, Chi-
by inspection of food diaries from the previous 24 h. Par- cago, IL, USA). Cohen’s effect size (ES) statistics were
ticipants were also instructed to fast overnight prior to each also used to highlight any trends. All values, unless other-
trial and to avoid strenuous exercise, alcohol and caffeine wise stated, are expressed as mean ± SEM.
in the 24 h preceding each testing session.
On arriving at the lab, participants had their body mass meas- Preliminary study: dose response relationship
ured and were fitted with a heart rate monitor (Garmin Ltd., between quinine concentration and bitter taste intensity
Kansas, USA). They then performed a 4-min light cycling and ANS responses
warm up at 40 % of V̇ O2peak followed by a 2–3-s practice
start to become accustomed to the amount of force required Psychophysical curves were constructed for taste intensity,
to start the pedals moving (Exertech EX-10 front access unpleasantness and nausea. As the concentration of quinine
cycle ergometer, Repco Cycle Company, Huntingdale, VIC, increased, there were significant increases (p < 0.05) in the
USA). Seat height was standardized for each participant. perception of taste intensity and unpleasantness (Fig. 1).
After 10 min of rest, participants were given either 25 mL The curve for taste intensity plateaued at 6 mM, indicat-
of plain water (WAT), 25 mL of a 7.1 % glucose solution ing that an increase in concentration above 6 mM no longer
(GLU; Glucodin, Boots Healthcare, NSW, Australia), 25 mL caused a subsequent increase in the perception of inten-
of a 10 mM quinine solution (QUI; Sigma-Aldrich, St. sity. The unpleasantness curve did not plateau, with sig-
Louis, MO, USA), or they were not given any mouth rinse at nificant differences between each successive concentration
all (CON). Participants were instructed to rinse their mouth (p < 0.05, Fig. 1). There were no significant differences
for 10 s and then expel the solution into a plastic cup. (p > 0.05) between concentrations for nausea (Fig. 1). As
Immediately after rinsing and expelling the solution, the concentration of quinine increased, there were signifi-
participants performed a 30-s maximal sprint effort, ini- cant increases in ANS responses (p < 0.05) up to a maxi-
tiated in a standing position with the preferred foot start- mum at 4, 6, and 6 mM for SC amplitude, OPD, and HR
ing at the “2 o’clock” position. The cycle ergometer was amplitude, respectively (Fig. 2).
interfaced with a customized program (Cyclemax, School
of Sport Science, Exercise and Health, The University of Effect of PROP tasting status on the dose response
Western Australia) to allow for the measurement of mean relationship between quinine concentration and bitter taste
power output over 0–30 s (Pmean), peak power output intensity and ANS responses
(Ppeak), mean power output over 0–10 s (P0–10), 10-20 s
(P10–20), and 20–30 s (P20–30) of the sprint. Participants Of the 18 participants, 5 were nontasters, 5 were medium
were instructed to cycle in an “all-out” manner for 30 s tasters, and 8 were supertasters. When participants were
without pacing themselves. grouped accordingly, there were no significant differences
13
Eur J Appl Physiol (2015) 115:129–138 133
Intensity (gLMS)
80
as well as in PROP supertasters
(triangles), medium tast- 60
ers (squares) and non-tasters # #
(diamonds). Hash symbols a 40 # #
significant difference from
preceding concentration level 20
(p < 0.05). Values are expressed
as mean ± SEM (n = 18) 0
0 2 4 6 8 10 0 2 4 6 8 10
b 100
Unpleasantness
80
60 # #
(gLMS)
# #
# #
40
# #
20
0
0 2 4 6 8 10 0 2 4 6 8 10
c 100
80
Nausea (mm)
60
40
20
0
0 2 4 6 8 10 0 2 4 6 8 10
Quinine Concentration (mM) Quinine Concentration (mM)
(p > 0.05) between groups for any of the ANS variables Main study: effect of quinine mouth rinse on sprint
examined here, or for subjective ratings of taste inten- performance
sity, unpleasantness, and nausea level. This indicates that
sensitivity to PROP may not be related to the ANS and Nutritional intake and environmental conditions
perceptual responses to quinine, which is in agreement
with other studies (Delwiche et al. 2001; Keast and Roper There was no significant difference in either total energy
2007) who have found no correlation between PROP sen- intake (CON 10,568 ± 224; QUI 9,976 ± 209; WAT
sitivity and sensitivity to other bitter compounds, includ- 10,723 ± 252; GLU 10,299 ± 229 kJ; p = 0.391) or car-
ing quinine. bohydrate intake (CON 324 ± 23; QUI 319 ± 20; WAT
340 ± 27; GLU 331 ± 24 g; p = 0.480) for the 24 h prior
Comparison between mouth rinsing with a 10 mM quinine to each experimental trial. Laboratory temperature and rel-
solution and ingesting a 2 mM quinine solution ative humidity were similar between trials (p > 0.05).
13
134 Eur J Appl Physiol (2015) 115:129–138
Amplitude (µSiemens)
well as in PROP supertasters #
∆ Skin Conductance
5
(triangles), medium tasters
(squares) and non-tasters (dia‑ 4
monds). Hash symbols indicates
a significant difference from 3
preceding concentration level
(p < 0.05). Values are expressed 2
as mean ± SEM (n = 18)
1
0
0 2 4 6 8 10 0 2 4 6 8 10
b 70
# #
60 #
#
50
∆ OPD (s)
40
30
20
10
0
0 2 4 6 8 10 0 2 4 6 8 10
c
25
∆ Heart Rate Amplitude
20
#
# #
#
15
(bpm)
10
0
0 2 4 6 8 10 0 2 4 6 8 10
Quinine Concentration (mM) Quinine Concentration (mM)
Fig. 3 Comparison of a sub- a b
jective and b ANS responses 60 70
between mouth rinsing then
ingesting a 2 mM quinine 50 60
solution (filled squares) and
mouth rinsing a 10 mM quinine 50
40
solution (open squares). Values 40
are expressed as mean ± SEM 30
(n = 18) 30
20
20
10 10
0 0
Intensity Hedonic Nausea SC amplitude OPD (s) HR amplitude
(gLMS) (gLMS) (mm) (µsiemens) (bpm)
13
Eur J Appl Physiol (2015) 115:129–138 135
Table 1 Mean power (Pmean), peak power (Ppeak), mean power over 10, 20 and 30 s (P0–10, P 10–20 and P20–30) and during a 30-s sprint effort
Performance Treatment Result Mean relative change p Mean relative change p
variable from QUI (%); ES from CON (%); ES
and P20–30 between treatments (p > 0.05; Table 1). Only described them with the appropriate descriptor (e.g. bitter,
small/trivial effect sizes were observed for the performance sweet).
variables in the quinine condition compared to the other
three conditions (Table 1).
Discussion
Heart rate and subjective ratings
We recently reported that combining mouth rinsing and the
Heart rate and RPE were similar at all time points between ingestion of a bitter tasting 2 mM quinine solution signifi-
treatments (p > 0.05; Fig. 4). Within each trial, both heart cantly improved performance of a 30-s cycling sprint (Gam
rate and RPE significantly increased in response to the 30-s et al. 2014). We also showed that the strong bitterness of
sprint (p < 0.05), before decreasing during the 7-min post- such a solution produced an unpleasant taste and evoked
sprint period. Nausea ratings were similar between trials at nausea at higher concentrations, thus limiting the concen-
all time points (p > 0.05; Fig. 4). tration of quinine that could be tested (Gam et al. 2014).
Given the possibility that mouth rinsing with quinine with-
Blood variables out ingesting it may not produce nausea, a mouth rinse only
protocol may be a more practical approach to administer
The blood lactate response to the 30-s sprint was similar quinine for improving exercise performance. The primary
between trials (p > 0.05; Fig. 4), and this variable changed purpose of this study was to determine whether mouth rins-
significantly within trials in response to exercise (p < 0.05; ing with a strong quinine concentration improves maximal
Fig. 4). There were no differences in blood glucose concen- cycling sprint performance. To meet our objective, a pre-
tration between trials or over time (p > 0.05, Fig. 4). liminary study had to be performed to identify the con-
centration of quinine in the mouth rinse that optimizes
PROP tasting taste intensity and ANS responses while limiting feelings
of nausea. Here we show that mouth rinsing with a bitter
Of the 12 participants, 2 were PROP nontasters, 4 were tasting quinine solution alone does not affect performance
medium tasters, and 6 were supertasters. All participants (mean and peak power) of a maximal 30-s cycling sprint
were able to taste both the QUI and GLU solutions and and has no effect on heart rate, RPE, blood variables, or
13
136 Eur J Appl Physiol (2015) 115:129–138
5
(mmol/L)
RPE
10
3
2 5
1
0 0
QUI CON WAT GLU QUI CON WAT GLU
e
100
80
Nausea (mm)
60
40
* * * *
* * *
20 *
0
QUI CON WAT GLU
subjective ratings of nausea compared to placebo and con- of precautions were taken to avoid any placebo effect. Par-
trol treatments. ticipants were deceived about the true purpose of the study,
The small effect sizes and lack of statistical significance being told that this study was concerned with examining
observed in this study suggest that quinine mouth rinsing the effect of different taste solutions on the responses of
has no effect on cycling sprint performance. Importantly, several blood variables (i.e., blood glucose and blood lac-
there was no order effect of trial administration for any of tate) to a maximal sprint effort. To further minimize the
the performance variables measured (p > 0.05). In addi- possibility of a placebo effect, two placebo treatments were
tion, the coefficients of variation between the familiariza- administered, namely a plain water and a sweet glucose
tion sprint (which included a water rinse) and the water solution, with both known not to enhance maximal sprint
rinse experimental sprint were 2.5 and 2.8 % for mean and performance (Chong et al. 2011). Overall, the success of
peak power output, respectively, suggesting limited intra- the aforementioned precautions at preventing a placebo
individual variability of our participants between testing effect is suggested by the absence of significant differences
sessions. in any of the performance measures between the quinine
Since our participants could taste the difference mouth rinse treatment and the water and glucose no-rinse
between the mouth rinse solutions tested here, a number conditions.
13
Eur J Appl Physiol (2015) 115:129–138 137
13
138 Eur J Appl Physiol (2015) 115:129–138
Pottier A, Bouckaert J, Gilis W, Roels T, Derave W (2010) Mouth Vernet-Maury E, Robin O, Dittmar A (1995) The ohmic perturbation
rinse but not ingestion of a carbohydrate solution improves 1-h duration, an original temporal index to quantify electrodermal
cycle time trial performance. Scand J Med Sci Spor 20:105–111 responses. Behav Brain Res 67:103–107
Rollo I, Williams C, Gant N, Nute M (2008) The influence of car- Wicks D, Wright J, Rayment P, Spiller R (2005) Impact of bitter taste
bohydrate mouth rinse on self-selected speeds during a 30-min on gastric motility. Eur J Gastroen Hepat 17:961–965
treadmill run. Int J Sport Nutr Exerc Metab 18:585–600
Tepper BJ, Christensen CM, Cao J (2001) Development of brief
methods to classify individuals by PROP taster status. Physiol
and Behav 73:571–577
13