Primate Sense of Taste: Behavioral and Single Chorda Tympani and
Glossopharyngeal Nerve Fiber Recordings in the Rhesus Monkey,
Macaca mulatta
GÖRAN HELLEKANT, 1 VICKTORIA DANILOVA, 1 AND YUZO NINOMIYA 2
1
The University of Wisconsin and Wisconsin Regional Primate Center, Madison, Wisconsin 53706; and
2
School of Dentistry, Asahi University, Gifu, Japan
INTRODUCTION
It is evident that our growing knowledge of taste differences between species necessitates a more cautious choice
of animal model in our attempts to understand the human
sense of taste. Many of these species differences reflect phylogenetic relationships.
In primates, influence of the phylogeny has been repeatedly documented with regard to the taste of sweeteners and
the effects of sweet taste modifiers (cf. Glaser et al. 1978;
Hellekant 1975; Hellekant and van der Wel 1989). In nonprimates, species differences affect not only the taste of
sweeteners, but also the taste of other compounds. It is evi-
dent, for example, that the threshold for a rejection of denatonium benzoate is higher in rodents than in humans (Whitney et al. 1990). Tannins are integral parts of the food for
several lemuriform primates, whereas their aversive taste to
humans prevents their consumption (Hellekant et al. 1993).
In salty taste, the varying species effect of amiloride on the
response to NaCl serves as an example (e.g., Ninomiya and
Funakoshi 1988; Tonosaki and Funakoshi 1989).
The effects of phylogenetic differences are not limited to
the taste of compounds representing the four taste qualities.
They also affect other compounds. Ethanol may serve as
one example; recordings from the human taste nerve, the
chorda tympani proper (CT), give a response that is quite
different from that in the CT of cats, dogs, and rats (Diamant
et al. 1965; Hellekant 1965a,b).
From this follows that the closer the phylogenetic relationship is between the animal model and humans the more applicable are the data acquired. For these reasons nonhuman primates
such as the rhesus monkey (Macaca mulatta) can be assumed
to offer a better model than, for example, dogs, but knowledge
is still needed of the similarities and dissimilarities of the sense
of taste of M. mulatta with the human sense of taste.
The response in CT taste fibers of macaques has been
recorded earlier. The most notable studies are by Sato et al.
(1975) and Brouwer et al. (1983), although the early study
by Gordon et al. (1959) should not be forgotten. However,
only one study of the other taste nerve of the tongue, the
glossopharyngeal nerve (NG), has been published (Farbman
and Hellekant 1989).
In this study we present recordings of the taste response
from both the CT and NG of rhesus monkeys and relate
them with behavioral observations. The closeness between
chimpanzee and human is then used to assess the usefulness
of the rhesus as the animal model for humans by comparing
the sense of taste of rhesus and chimpanzee. We thus attempt
to assess the validity of the rhesus model for the human
sense of taste by comparing the data here with data from
our earlier studies in chimpanzee (Hellekant and Ninomiya
1991, 1994a,b; Hellekant et al. 1985, 1996, 1997; Ninomiya
and Hellekant 1991).
METHODS
Electrophysiology
Twenty rhesus monkeys (M. mulatta) of both sexes
were used for electrophysiological recordings. Their weights
ranged from 3.4 to 7.6 kg.
ANIMALS.
0022-3077/97 $5.00 Copyright q 1997 The American Physiological Society
978
/ 9k0c$$fe23
J176-6
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
Hellekant, Göran, Vicktoria Danilova, and Yuzo Ninomiya.
Primate sense of taste: behavioral and single chorda tympani and
glossopharyngeal nerve fiber recordings in the rhesus monkey, Macaca mulatta. J. Neurophysiol. 77: 978–993, 1997. The responses
of 51 chorda tympani proper (CT) and 33 glossopharyngeal (NG)
neural taste units from the rhesus monkey (Macaca mulatta) were
recorded during stimulation of either the anterior (CT) or posterior
(NG) part of the tongue with 26 stimuli that taste salty, umami,
sour, bitter, and sweet to humans. In the CT, hierarchical cluster
analysis separated four major clusters. The N and S clusters were
most populous, followed by the H cluster and a small Q cluster.
NaCl, monosodium glutamate (MSG), and MSG with guanosine
5 *-monophosphate were the best stimuli in the N cluster. Amiloride
suppressed responses to NaCl. KCl did not stimulate fibers from
this cluster. S cluster fibers were characterized by strong responses
to all sweeteners. The H cluster responded best to acids but also
to some of the sweeteners such as xylitol, fructose, and sucrose.
Q fibers responded well to quinine hydrochloride (QHCl) and
caffeine, but not to denatonium benzoate. In the NG, hierarchical
cluster analysis separated three major clusters. Q fibers formed the
largest cluster. QHCl, caffeine, and sucrose octa-acetate but not
denatonium benzoate elicited very strong responses in these fibers.
S fibers formed a second cluster. Although most of the sweeteners
stimulated the S fibers, their responses were not so pronounced as
in CT S fibers. The small M cluster was formed by fibers that
responded best to MSG. They also responded to NaCl and acids.
Two bottle preference tests showed a positive relationship between
a sweetener’s ability to stimulate the taste fibers and the animals’
consumption. Thus the most-liked sweeteners stimulated the S cluster fibers of CT best, whereas less-liked sweeteners such as Dphenylalanine elicited a response in Q fibers and sodium cyclamate
stimulated N fibers. The results show that both CT and NG taste
fibers of M. mulatta group according to the human concepts of
taste qualities.
CHORDA TYMPANI AND GLOSSOPHARYNGEAL TASTE FIBERS IN RHESUS
Table 1 presents the compounds and their concentrations for both series of experiments. The compounds were chosen
to represent the human taste qualities. Their concentrations were
STIMULI.
TABLE
1.
influenced by our results from chimpanzees, especially with regard
to the sweeteners, but we also tried to maintain a continuity with
other earlier studies. Generally we attempted to use the same array
of stimuli for both nerves. However, there were a few exceptions,
as shown in Table 1. The structure of super-aspartame, which
incorporates aspartame and suosan into a single molecule, has been
given in Hellekant and Walters (1991). Structural formulas for
three guanidine analogues are shown in Fig. 1.
All compounds except quinine hydrochloride (QHCl), which
for solubility reasons was dissolved in distilled water, were dissolved in artificial saliva (Table 2).
STIMULATION.
The tastants were delivered to the tongue with
an open flow system (Taste-O-Matic) controlled by a computer
(Hellekant and Roberts 1995). It delivered the solutions at given
intervals, over a preset time, under conditions of constant flow and
temperature (337C). Two periods of taste stimulation were used,
5 and 8 s. During the CT experiments the flow was directed over
the fungiform papillae. To ensure that the foliate and vallate taste
buds were stimulated in the NG experiments the tongue was
stretched. Between stimulations the tongue was rinsed for 55–52
s with artificial saliva.
RECORDING TECHNIQUE.
Single-nerve impulses from the CT or
NG were recorded with a PAR 113 amplifier, monitored over a
loudspeaker and an oscilloscope, and fed into a recorder (Gould
ES 1000). An impulse amplitude analyzer with a window with
adjustable upper and lower levels was used to discriminate between
nerve impulses. These levels could be set so that when a nerve
impulse exceeded the lower but not the upper level it triggered a
List of solutions used in electrophysiological and behavioral experiments
CT Series
NG Series
NaCl (0.07 M)
NaCl (0.07 M) / Amiloride (0.5 mM)
NaCl (0.07 M) / Novobiocin (1 mM)
KCl (0.1 M)
KCl (0.1 M) / Amiloride (0.5 mM)
MSG (70 mM)
MSG (70 mM) / GMP (0.3 mM)
Citric acid (40 mM)
Aspartic acid (50 mM)
HCl (10 mM)
Quinine HCl (5 mM)
Caffeine (0.1 M)
Denatonium benzoate (0.01 mM)
Sucrose (0.3 M)
Fructose (0.3 M)
Aspartame (5 mM)
Alitame (0.3 mM)
Super-aspartame (0.1 mM)
Acesulfame-K (3.5 mM)
Saccharin-Na (1.6 mM)
Cyclamate-Na (10 mM)
Stevioside (0.87 mM)
Xylitol (0.8 M)
D-Tryptophan (0.03 M)
D-Phenyalanine (0.1 M)
D-Glucose (0.3 M)
L-Glucose (0.3 M)
Galactose (0.3 M)
Dulcin (1.6 mM)
SC-45647 (0.04 mM)
NC-00174 (0.03 mM)
NC-00351 (0.03 mM)
Brazzein (0.3%)
Monellin (0.1%)
TBP Test
NaCl (0.1 M)
KCl (0.1 M)
MSG (70 mM)
Citric acid (40 mM)
Aspartic acid (50 mM)
HCl (10 mM)
Quinine HCl (5 mM)
Caffeine (0.1 M)
Denatonium benzoate (0.1 mM)
Sucrose octaacetate (SOA) (1 mM)
Sucrose (0.3 M)
Fructose (0.3 M)
Aspartame (5 mM)
Alitame (0.3 mM)
Acesulfame-K (3.5 mM)
Cyclamate-Na (10 mM)
Stevioside (0.87 mM)
Xylitol (0.8 M)
D-Phenyalanine
(0.1 M)
Sucrose (0.3 M)
Aspartame (5 mM)
Alitame (0.3 mM)
Super-aspartame (0.1 mM)
Acesulfame-K (3.5 mM)
Saccharin-Na (1.6 mM)
Cyclamate-Na (10 mM)
Stevioside (0.87 mM)
Xylitol (0.8 M)
D-Tryptophan (0.03 M)
D-Phenyalanine (0.1 M)
NC-00174 (0.03 mM)
NC-00351 (0.03 mM)
Brazzein (0.3%)
CT, chorda tympani proper; NG, glossopharyngeal nerve; TBP, 2-bottle preference; MSG, monosodium glutamate; GMP, guanosine 5*-monophosphate.
/ 9k0c$$fe23
J176-6
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
SURGERY.
Anesthesia was initiated with ketamine (50 mg im).
The monkey was then intubated and the anesthesia maintained
with halothane (0.7–1.0%). Fluid losses were replaced with 5%
dextrose and lactated Ringer solution. Body temperature, heart rate,
and respiratory rate were continuously monitored.
The method of dissecting the right CT was recently described
(Hellekant and Roberts 1995). Briefly, an incision was made along
the mandibular angle between the rostral lobes of the parotid gland
and the mandibular bone. Then the tissue attached to the mandibular angle was dissected through and the caudomedial side of the
pterygoid muscle was followed down to its origin at the pterygoid
plate of the skull to the CT.
The method of dissecting the right lingual branch of the NG
was also recently described (Hellekant and Roberts 1995). The
incision was made ventral to the mandibular angle, ventral to the
rostral lobes of the parotid gland to reach the nerve dorsal and
medial to the much larger hypoglossal nerve. The NG distinguishes
itself with a very typical meandering course caudal to the hyoid
bone. For single-fiber recording it was not necessary to cut either
the CT or NG. Instead, single fibers were dissected off fine strands
of the nerves. The method left a major part of the nerve trunk
intact.
At the end of the experiment, deeper structures were sutured
with absorbable suture as needed. The skin was closed with nylon
sutures that were removed after 7–10 days.
979
980
G. HELLEKANT, V. DANILOVA, AND Y. NINOMIYA
Two-bottle preference tests were
carried through with 16 animals (10 males and 6 females) housed
individually. The animals had uninterrupted water access. The experiments were preceded by a training period. Different animals
were used for the electrophysiology.
Table 1, third column, lists the compounds used in the behavioral
experiments. They were dissolved in distilled water. One bottle
contained 30 ml of the test solution and the other 30 ml of distilled
water. The bottles were left on the cages until one bottle was
emptied or for at most 15 min. Each solution was tested twice with
changed position of the bottles. Only one sweetener was tested
each day.
The preference ratio was calculated as consumption of sweetener
divided by the total volume consumed. The results were averaged.
A preference ratio ú0.5 was chosen as the criterion for preference.
A two-tailed t-test was used to compare mean preference ratios
with 0.5. To compare preferences between all sweeteners, the results were first assessed with the use of analysis of variance. Then
Student-Newman-Keuls test was used to make all pairwise comparisons. P õ 0.01 was chosen as significance level.
BEHAVIORAL EXPERIMENTS.
Single-fiber response in the CT
FIG . 1. Structures of 3 guanidine derivatives, SC-45647, NC-00174, and
NC-00351, which are sweet to humans.
pulse, which was fed into an IBM computer via a DAS-Keithley
interface and processed by a program. The program stored the
interval between pulses together with information on the stimulus.
At the same time it controlled the Taste-O-Matic. The identity of
the stimulus, its order, and the level of nerve activity before, during,
and after each stimulation were continuously presented on the computer screen and printed out during the experiment.
DATA ANALYSES.
The spontaneous activity before each stimulation was deducted from the responses. Here we define spontaneous
activity as the impulse activity preceding the stimulation during
rinsing of the tongue with artificial saliva for 5 s. A fiber was
considered to be responsive to a stimulus if the nerve impulse rate
during the first 5 s of stimulation was ú2 times the SD of the
spontaneous activity of the fiber.
To facilitate comparisons with data from other species and studies, the responses of CT and NG fibers to the four basic stimuli
(NaCl, citric acid, QHCl, and sucrose) were used to categorize
each fiber by its best stimulus (Frank 1973) and to calculate the
breadth of responsiveness (H) for each fiber. H was calculated
according to the formula H Å 0 K pi log pi, where K is a scaling
constant (1.6) and pi is the proportional response to each of the
four basic stimuli (Smith and Travers 1979).
Hierarchical cluster analysis and multidimensional scaling analyses were performed with the statistical package SYSTAT for
Macintosh, version 5.2. Intercluster similarity was measured with
the use of the Pearson correlation coefficient and cluster analysis
proceeded according to the average linkage method.
/ 9k0c$$fe23
J176-6
Figure 2 shows an example of the nerve impulses recorded
from single CT taste fiber (RH95D19H). Here we present
only a part of the recordings from this unit. However, it is
evident that NaCl, the acids, and to some extent sodium
cyclamate gave a response. Some of these features will be
discussed later.
Figure 3 gives an overview of the responses in all 51 CT
fibers with the use of a topographical method in which the area
of the circles represents impulse activity over the first 5 s of
stimulation. The legend to Fig. 3 relates the area of the circles
with the nerve responses. Open circles indicate a suppression of
the nerve activity during a stimulation. The stimuli are arranged
along the X-axis in the order of salt, umami compounds, acids,
bitter compounds, and sweet compounds. The fibers are ordered
along the Y-axis in groups, starting with the fibers responding
best to NaCl, followed by the fibers mainly stimulated by citric
acid, QHCl, and finally sucrose.
The average spontaneous activity was 16.4 { 1.5 (SE)
impulses measured over 5 s before stimulations.
Hierarchical cluster analysis
The responses of 47 of these CT fibers to 25 stimuli were
included in the hierarchical cluster analysis. Because four
fibers were tested with only five stimuli, we did not include
them in the analysis. The result is represented as a dendroTABLE
2.
Composition of artificial saliva
Salt
Concentration, mM
Amount
NaCl
KCl
NaHCO3
KHCO3
HCl
CaCl2
MgCl2
K2HPO4
KH2PO4
4
10
6
6
3.6
0.5
0.5
0.24
0.24
230 mg/l
746 mg/l
504 mg/l
588 mg/l
6 ml
55.5 mg/l
47.5 mg/l
42 mg/l
33.3 mg/l
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
RESULTS
CHORDA TYMPANI AND GLOSSOPHARYNGEAL TASTE FIBERS IN RHESUS
981
FIG . 2. Recordings from the chorda tympani
proper (CT) NaCl-best filament RH95D19H
during stimulation with NaCl, NaCl with amiloride, KCl, citric acid, aspartic acid, sodium cyclamate, sucrose, saccharin, potassium acesulfam (ace-K), aspartame (APM), alitame, and
quinine hydrochloride (QHCl). Onset and offset of stimuli are shown as changes in bar code.
Average response profiles
Figure 5 shows the average response profiles of these four
clusters. The stimuli are listed along the X-axis, whereas the
average impulse activity measured over 5 s is plotted along
the Y-axis. The vertical lines illustrate the SE of these averages. In the following we present each cluster.
N cluster
As shown in Fig. 4, this cluster included 19 units and was
the largest one. NaCl elicited the strongest response in 15 of
these fibers. There were four monosodium glutamate (MSG)/
acid-best fibers, judged by the stimulus that elicited the largest
responses. Fibers RH92M12B and RH92M12C were not subjected to cluster analysis, but we included them among the NaClbest fibers in Figs. 3 and 5A because among the four basic
stimuli, NaCl elicited the largest response in these fibers.
Figure 5 A shows that the N cluster was characterized
by strong responses to NaCl and MSG alone or mixed
with guanosine 5 *-monophosphate. Amiloride suppressed the response to NaCl ( Figs. 2, 3, and 5 A ) . Novobiocin, which has been reported to enhance the salt
response in some species ( cf. Feigin et al. 1994 ) , had
no effect here. Interestingly, KCl elicited no response in
these fibers, whereas citric and aspartic acid evoked a
moderate response ( Figs. 2, 3, and 5 A ) . These fibers did
not respond to bitter compounds, and of the sweeteners
only sodium cyclamate and xylitol elicited some response in these fibers ( Figs. 2, 3, and 5 A ) .
/ 9k0c$$fe23
J176-6
H cluster
Figure 5B shows the average response profile of eight units.
Citric and aspartic acids elicited the best responses in these fibers.
Generally the sweeteners did not stimulate fibers of this cluster.
The exceptions were the carbohydrates and xylitol. Xylitol
evoked Ç70–80% of the response to citric acid. In fiber
RH92U23D, xylitol elicited the strongest response among all
stimuli. As can be seen in Fig. 5B, the fibers of this cluster did
not respond to salty and bitter compounds.
Q cluster
Figure 5C presents the average response profile for the Q
cluster. This cluster consisted of four QHCl-best fibers. Fiber
RH92U24K was not subject to cluster analysis, but we included it among the QHCl-best fibers in Figs. 3 and 5C
because QHCl elicited the largest response in this fiber. Figure 5C shows that these fibers responded well to QHCl and
caffeine but not to 0.01 mM denatonium benzoate. Considering the fact that amiloride tastes bitter to humans, it is interesting that NaCl with amiloride and KCl with amiloride
elicited a response in these Q fibers. They also responded
to saccharin, xylitol, D-phenylalanine, D-tryptophane, and
carbohydrates.
S cluster
The S fiber cluster was the second largest and consisted
of 16 fibers. One fiber (RH92M12D) was not subject to
cluster analysis, but we included it in Fig. 5D because among
the four basic stimuli sucrose elicited the best response in
this fiber. Figure 5D shows that these fibers responded to
every sweet stimulus. These included 0.3 M D- and L-glucose; 0.3 M galactose; 1.6 mM dulcin; three guanidine deriv-
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
gram in Fig. 4. The identity number of the fiber and response
category on the basis of its best response to the four basic
solutions are listed on the left side. The analysis distinguished four major clusters: N, H, S, and Q clusters.
982
G. HELLEKANT, V. DANILOVA, AND Y. NINOMIYA
atives (0.04 mM SC-45647, 0.03 mM NC-00174, and 0.03
mM NC-00351); brazzein; and monellin. The S fibers quite
often responded to aspartic acid and, to a lesser extent, MSG
and citric acid, but not to salts and bitter compounds
(Fig. 5D).
Multidimensional scaling
On the basis of a correlation matrix of the stimuli, we
performed multidimensional scaling. The spatial representation of the similarities among 25 stimuli is shown in
/ 9k0c$$fe23
J176-6
Fig. 6. The stress value is 0.047. All sweeteners formed
a tight group separate from the other stimuli. Another
compact group included NaCl alone or mixed with novobiocin and the umami compounds. Then caffeine and
QHCl stayed very close but denatonium benzoate was
positioned separately. KCl and mixtures of salts with
amiloride were located most closely to the bitter compounds. It is interesting and corroborates Fig. 5C that
the addition of amiloride moved NaCl into the bitter
group. Less evident, perhaps, is the grouping of cyclamate among salty compounds, but, as shown in Fig. 5 A,
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
FIG . 3. Overview of the response profiles of 51 CT
single fibers with the use of a topographical method. Area
of circles: impulse activity over the 1st 5 s of stimulation.
Open circles: inhibition. Absence of a mark: data are
missing. The stimuli were arranged along the X-axis in
order of salt, sour, bitter, and sweet. The fibers were
arranged along the Y-axis in groups: NaCl-, acid-, QHCl-,
and sucrose-best fibers. MSG, monosodium glutamate;
GMP, guanosine 5 *-monophosphate.
CHORDA TYMPANI AND GLOSSOPHARYNGEAL TASTE FIBERS IN RHESUS
it gave a response in N cluster fibers too. Citric and
aspartic acid were located close to each other, with HCl
somewhat distant.
Single-fiber response in the NG
Figure 7 shows as an example nerve impulses recorded
from the single NG fiber RH94Y23D. Here we present only
a part of the recordings from this unit. As can be seen from
the responses shown, it responded to stimulation with QHCl
and caffeine, and displayed some response to KCl, sucrose,
and D-phenylalanine. It was characterized as a QHCl-best
fiber.
983
Figure 8 gives an overview of the responses in all 33 NG
fibers. The results were plotted in the same way as in Fig.
3. Figure 8 shows that the fibers responding to bitter compounds constituted the largest group.
The average spontaneous activity was 8.6 { 1.3 (SE)
impulses measured over 5 s before stimulations.
Hierarchical cluster analysis
The responses of all NG fibers to 20 stimuli were subjected to hierarchical cluster analysis. The result is represented as a dendrogram in Fig. 9. The analysis resulted in
three major clusters: M, Q, and S clusters. The identity num-
/ 9k0c$$fe23
J176-6
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
FIG . 4. Hierarchical cluster analysis of the response profiles
for 47 CT fibers. Intercluster similarity was measured with the
use of the Pearson correlation coefficient and cluster analysis proceeded according to the average linkage method. Fiber number
and response category on the basis of its response to the 4 basic
solutions are listed on the left. N, H, Q, and S: NaCl-, citric
acid-, QHCl-, and sucrose-best fibers, respectively.
984
G. HELLEKANT, V. DANILOVA, AND Y. NINOMIYA
ber of the fiber and response category on the basis of its
best response to the four basic solutions are listed on the
left side of the dendrogram. Figure 10 shows the average
response profiles of these three clusters. In the following we
present each cluster of fibers.
citric acid, and aspartic acid elicited a substantial response
in these fibers, whereas the bitter compounds and the sweeteners did not stimulate. The M cluster included three NaClbest and two citric acid-best fibers according to their responses to the four basic stimuli.
M cluster
Q cluster
The average response profile of the M cluster in Fig. 10A
shows that MSG elicited the largest response, and NaCl,
The cluster analysis in Fig. 9 confirmed that the fibers
mainly responding to bitter compounds constituted the
/ 9k0c$$fe23
J176-6
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
FIG . 5. Average response profiles of 4
major clusters of rhesus CT fibers. Error
bars: SE. Hatched columns: salts. Dark
gray columns: umami compounds. Open
columns: acids. Light gray columns: bitter
compounds. Black columns: sweeteners.
CHORDA TYMPANI AND GLOSSOPHARYNGEAL TASTE FIBERS IN RHESUS
985
FIG . 6. Distribution of 23 stimuli in a 3-dimensional space
resulting from multidimensional scaling. The distribution was
calculated with the use of Pearson correlation coefficient between
stimuli across 47 CT fibers. Kruskal stress value: 0.048.
rates the effects of NaCl with amiloride on the CT. The Q
cluster did not respond to NaCl, acids, or sweeteners, but
sucrose, fructose, xylitol, and D-phenylalanine elicited some
response.
S cluster
S fibers constituted the second largest, but considerably
smaller, group of eight fibers in the NG (Fig. 9). They
FIG . 7. Recordings from glossopharyngeal nerve (NG) QHCl-best filament RH94Y23D during stimulation with QHCl,
caffeine, KCl, NaCl, sodium cyclamate, citric acid, sucrose, aspartame, potassium acesulfam, alitame, stevioside, and Dphenylalanine. Onset and offset of stimuli are shown as changes in bar code.
/ 9k0c$$fe23
J176-6
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
largest group in the NG (20 fibers of a total of 33 single
fibers). We added sucrose octa-acetate (SOA), a compound
bitter to humans, to our array of stimuli for the NG recordings. The data in Fig. 10B indicate that SOA stimulated
these fibers as well as QHCl and caffeine did. The lack of
a response to denatonium benzoate corroborates our findings
in the CT. The finding that NaCl with amiloride elicited a
larger response in the Q cluster than in the M cluster corrobo-
986
G. HELLEKANT, V. DANILOVA, AND Y. NINOMIYA
did not respond to stimulation with salts, acids, and bitter
solutions. Despite the fact that sweeteners were their best
stimuli, their response to these was small and inconsistent
(Fig. 10C). Further, aspartame and sodium cyclamate elicited small responses, whereas some of the other sweeteners,
such as stevioside, did not elicit any response in these fibers.
Multidimensional scaling
Figure 11 shows the results of multidimensional scaling
for 18 stimuli. The separation of the tastants in space is less
evident than for the CT data. Dimension 2 separated a group
consisting of citric and aspartic acid, NaCl, sodium cyclamate, and MSG. Dimension 1 suggested a group of sweeteners in the background of the left side and a group of bitter
compounds in the foreground. Because SOA and denatonium
benzoate were not used in all fibers, they were not included
in the analysis.
Comparison of CT and NG fibers
Figure 12 presents the distribution of the different types
of single taste fibers in the CT and NG according to their
/ 9k0c$$fe23
J176-6
best response to one of the four basic stimuli. In the CT
the two largest groups were the NaCl- and sucrose-best
fibers. In the NG there were few NaCl-best fibers. However, the response to NaCl in the NaCl-best fibers of the
CT ( 124.4 { 17.3 impulses per 5 s ) did not differ ( tailed
t-test, P ú 0.1 ) from the response in the NaCl-best fibers
of the NG ( 97.1 { 85.5 impulses per 5 s ) .
The percentage sucrose-best fibers was about the same
in the NG as in the CT. But the response to sucrose in
the NG sucrose-best fibers ( 56.5 { 11.8 impulses per 5
s ) was significantly smaller ( 2-tailed t-test, P õ 0.01 )
than in the CT ( 119.4 { 13.8 impulses per 5 s ) . This
difference in responses is indicated by the asterisk in
Fig. 12.
In the NG the QHCl-best fibers were more populous
than in the CT, but their responses to QHCl did not differ
significantly ( NG: 116.5 { 15.8 impulses per 5 s; CT:
173.8 { 52.5 impulses per 5 s ) .
Although the percentage of citric-acid-best fibers was
higher in CT than in NG, their responses to citric acid
did not differ significantly ( CT: 98.1 { 14.5 impulses
per 5 s; NG: 62.4 { 34.1 impulses per 5 s ) .
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
FIG . 8. Overview of the response profiles of 33
NG single fibers with the use of a topographical
method. Area of circles: impulse activity over the
1st 5 s of stimulation. Open circles: inhibition. Absence of mark: data are missing.
CHORDA TYMPANI AND GLOSSOPHARYNGEAL TASTE FIBERS IN RHESUS
987
Breadth of tuning
Table 3 shows the breadth of tuning (H) for each group
of fibers in both nerves. The H value varied between 0.37
and 0.81, with the lowest for the QHCl-best fibers in both
CT and NG fibers. This means that the QHCl-best fibers
were the most specific in both nerves.
Behavioral experiments
Figure 13 presents the results of the two-bottle preference experiments with the sweeteners listed in Table 1.
For all solutions, which are sweet to humans, the mean
preference ratios were significantly larger than the
chance level ( 0.5 ) . Summarizing pairwise comparisons,
the preference ratios for cyclamate and D-phenylalanine
were significantly lower than for other sweeteners. The
/ 9k0c$$fe23
J176-6
preference ratio for NC-00351 was lower than ratios for
super-aspartame and sucrose.
DISCUSSION
In the CT the hierarchical cluster analysis separated four
major clusters. The N and S clusters were most populous,
followed by the H cluster and a small Q cluster. In the N
cluster, NaCl and MSG were the best stimuli. Amiloride
always suppressed the response to NaCl, whereas KCl
did not stimulate fibers from this cluster. All sweeteners
stimulated the S cluster fibers. The H cluster fibers were
characterized by strong responses to acids. Xylitol and
carbohydrates elicited a response in some of these fibers.
Q fibers responded well to QHCl and caffeine, but not to
denatonium benzoate.
In the NG the analysis separated three major clusters.
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
FIG . 9. Hierarchical cluster analysis of the response profiles for
33 NG single fibers. Intercluster similarity was measured with the
use of the Pearson correlation coefficient and cluster analysis proceeded according to the average linkage method. Fiber number and
response category on the basis of its response to the 4 standard
solutions are listed on the left.
988
G. HELLEKANT, V. DANILOVA, AND Y. NINOMIYA
The Q cluster was the largest. QHCl, caffeine, and SOA
but not denatonium benzoate elicited very strong responses in these fibers. Sucrose-best fibers formed a second S cluster. Although most of the sweeteners stimulated
these fibers, the responses were not so pronounced as in
CT S cluster. The smallest cluster was formed by fibers
that responded best to MSG. They also responded to NaCl
and acids.
We first discuss the taste fibers of the NG, then compare
CT results with data from other studies on CT and cortex,
and finally examine the similarity between the sense of
/ 9k0c$$fe23
J176-6
taste in rhesus monkey with that of human and chimpanzee, in an attempt to decide to what extent the rhesus
monkey can serve as the animal model in taste.
Taste fibers in the NG
There is to our knowledge no single-fiber study published from the NG of a macaque, although there is an
earlier study, which presented summated recordings
( Farbman and Hellekant 1989 ) . This prevents direct comparisons of fiber distribution, etc. However, the parallel
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
FIG . 10. Average response profiles of 3 major clusters of rhesus NG fibers: M fiber cluster, Q fiber cluster,
and S fiber cluster. Error bars: SE. Different patterns of
columns indicate different taste qualities, as in Fig. 4.
CHORDA TYMPANI AND GLOSSOPHARYNGEAL TASTE FIBERS IN RHESUS
989
FIG . 11. Distribution of 18 stimuli in a 3-dimensional
space resulting from multidimensional scaling. The distribution was calculated with the use of Pearson correlation coefficient between stimuli across 33 NG fibers. Kruskal stress
value: 0.067.
Taste fibers in the CT
In contrast, there are some studies from the CT nerve of
macaques. Zotterman, who was the first to publish taste fiber
recordings (Zotterman 1935), was also one of the authors
of the first study of the CT of M. mulatta (Gordon et al.
1959). That study concluded that ‘‘certain gustatory fibers
responded very specifically to one class of substances only
e.g. to salt, acid, or to QHCl.’’ Also noted was the link
between compounds within the same taste quality: ‘‘fibers
responding to sucrose almost always responded to saccharin
as well.’’
After miraculin, acids taste sweet to humans (e.g., Kuri-
hara et al. 1969) and the summated CT response to acids is
enhanced in macaques (Diamant et al. 1972; Hellekant 1977;
Hellekant et al. 1974, 1976). In 1983, we published a singlefiber study of the CT of M. mulatta (Brouwer et al. 1983)
based on these earlier findings. The study showed that after
miraculin, sucrose-best fibers responded to acids as if the
acids had become sweeteners.
In this context it is interesting that one puzzling observation in the study by Brouwer et al. (1983) may be explained
by observations here. We used sucrose and citric acid to
distinguish and classify the sucrose-best and acid-best fiber
types and found an overlap between these stimuli. This corroborates Fig. 5B here, which shows that sucrose stimulates
to some extent fibers in our H cluster. Judged by the responses to fructose, D-glucose, and L-glucose, probably carbohydrates, in general, stimulate these H fibers. Consequently carbohydrates are less suited for distinguishing
between acid-best and sweet-best fibers. One of the highpotency sweeteners, e.g., aspartame, would have been a better choice than sucrose because, as shown in Fig. 5B, aspartame did not stimulate the H cluster. This corroborates a
recent study in chimpanzee that shows that high-potency
FIG . 12. Distribution of 4 different types of single taste fibers according their best response to 1 of the 4 basic stimuli in
CT and NG. Asterisk: statistically significant difference between
responses to sucrose in CT and NG sucrose-best fibers.
/ 9k0c$$fe23
J176-6
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
is evident between our finding of a large group of QHClbest fibers in the NG and the reports of human taste sensitivity to bitterness, prevalently from the back of the tongue
( Hänig 1901; cf. Pfaffmann et al. 1971 ) , especially if we
compare it with the many sweet-best and salt-best fibers
found in the CT of the rhesus monkey and the sensitivity
to salt and sweet reported as being most prominent on the
front of the human tongue.
990
TABLE
G. HELLEKANT, V. DANILOVA, AND Y. NINOMIYA
3.
Breadth of tuning of CT and NG fibers
CT Fibers
N
NaCl-best fibers
Citric-acid-best fibers
QHCl-best fibers
Sucrose-best fibers
All fibers
17
13
5
16
51
NG Fibers
H
0.59
0.57
0.38
0.51
0.54
{
{
{
{
{
N
0.05
0.04
0.13
0.05
0.03
2
2
21
8
33
H
0.64
0.81
0.46
0.47
0.49
{
{
{
{
{
0.12
0.13
0.03
0.09
0.03
H values are means { SE. N, number of fibers. QHCl, quinine hydrochloride; for other abbreviations, see Table 1.
Cortical studies
Similar ranges of taste stimuli have been employed in
macaques earlier. There are, as a matter of fact, more recordings from cortical neurons than from peripheral neurons
in macaques, although from a different species, M. fascicularis. Many studies originate from the laboratories of Rolls
and Scott, who in a number of studies have presented data
from the primary taste area in awake animals (Baylis and
Rolls 1991; Plata-Salaman et al. 1992, 1993, 1995; Scott et
al. 1991, 1994). It is therefore of interest to relate taste
responses in the CT and NG with some of these findings
from the cortical taste area.
/ 9k0c$$fe23
J176-6
Validity of the macaque model
How good a model is M. mulatta for the study of the
human sense of taste? One answer to this question can
be obtained by comparing the results here with human
psychophysical observations. The parallel is evident between the human taste sensitivity to bitter taste, reported
as being predominantly from the back of the tongue, and
the large number of QHCl-best fibers in the rhesus NG
mediating taste from the back of the tongue. However,
this feature is also found in other species ( Frank 1991;
Hanamori et al. 1988 ) , although this seems not to be the
case in all species ( e.g., Hård af Segerstad and Hellekant
1989 ) . Similarly, the larger number of salt-best fibers
and the more intense responses to sweet in the CT are in
agreement with the higher sensitivity to sweet and salt
stimuli from the CT tongue area in humans ( Collings
1974; Hänig 1901 ) .
It is well known that rhesus monkeys have a sweet tooth.
We used this in the two-bottle preference tests. As shown
in Fig. 13, all the tastants, which have been described as
being sweet to humans, were liked. The low liking for Dphenylalanine and sodium cyclamate parallels observations
in humans, who rate them as less attractive than, e.g., sucrose
(Schiffman et al. 1979). For evident reasons it is impossible
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
sweeteners stimulate sucrose-best fibers more selectively
than sucrose and other carbohydrates (Hellekant et al. 1997).
A study in M. fascicularis by Sato et al. (1975), later
reanalyzed in Sato et al. 1994, is of particular interest here.
Sato et al. concluded that the CT fibers could be classified
into four categories, depending on their best sensitivity to
one of the four basic stimuli. Sucrose-best, NaCl-best, and
QHCl-best fibers were rather specifically sensitive to sucrose, NaCl, and QHCl, respectively, whereas acid-best fibers also responded relatively well to a series of salts in
addition to 0.01 M HCl. Interestingly, all NaCl-best fibers
in the study by Sato et al. and in this one show the same
lack of a response to KCl.
In agreement with our CT observations, the bitter sensitive
group in the study by Sato et al. was the smallest, Ç16%
of the total fibers. Saccharin produced a good response in
sucrose-best fibers, but the bitter-sensitive fibers also responded to saccharin, more to 0.3 M than to 0.01 M. Considering the increased bitter taste with increasing concentration
of saccharin, this is interesting and fits the conclusion below.
Because there was some discrepancy in breadth of tuning
between the data of Sato et al. (1994) and ours, we recalculated breadth of tuning in our study with the use of HCl
instead of citric acid. As shown in Table 4, we arrive at
virtually the same average. Only the order between fiber
groups is somewhat different.
Further, Sato et al. found that macaque monkey taste fibers
are more narrowly tuned to one of the four basic taste stimuli
than CT fibers of rats, hamsters, and squirrel monkeys (cf.
Sato et al. 1994), which is in agreement with this study.
They concluded, in agreement with our conclusions here,
‘‘that in the monkey CT each of the four classes of fibers
predominantly contributes to mediate gustatory information
for sweet, salty, sour and bitter stimuli.’’
The cortical results gave a large, well-delineated S cluster
and N cluster (Plata-Salaman et al. 1993; Scott et al. 1994)
and less delineated H and Q clusters (Plata-Salaman et al.
1993, 1995). This agrees better with our findings in CT
fibers than with the large Q cluster and low-level response
to sweeteners in the NG. The most likely explanation is that
when working with an awake monkey the tastants will
mainly stimulate the anterior tongue. Consequently the results of, for example, cluster analyses of the responses in
cortical neurons conform better with the results from the CT
than from the NG.
Cortical neurons seem to be less specific. Thus the highest
mean value for breadth of tuning of CT and NG fibers is
0.54, which is lower than the lowest mean value obtained
for cortical taste neurons, 0.59 (Plata-Salaman et al. 1993).
This suggests that cortical taste neurons are more generalist
than the peripheral ones. Considering their integrative role
in taste, this is not unexpected. The important point is that
all the way through the taste system a relationship between
human taste qualities and neuron identity is maintained in
macaques.
Baylis and Rolls (1991) identified an umami group of cortical
neurons. Here we obtained as good a response to MSG as to
NaCl in the N fibers of the CT (Fig. 5). As a matter of fact,
in four fibers MSG elicited the largest response. Our cluster
analysis of the fibers in the NG separated a group that predominantly responded to MSG. Two conclusions can be drawn. One
is that MSG is a powerful taste stimulus to macaques. This
agrees with and supports the results of Baylis and Rolls, who
concluded that ‘‘glutamate, which produces umami taste in humans, is approximately as well represented as are the tastes
produced by: glucose, NaCl, HCl and QHCl.’’ On the other
hand, our cluster analysis did not indicate that MSG was handled
as a separate taste quality, which seems to be the conclusion
reached by Baylis and Rolls (1991).
CHORDA TYMPANI AND GLOSSOPHARYNGEAL TASTE FIBERS IN RHESUS
TABLE
4.
991
Breadth of tuning of CT single fibers
Present study
Sato et al. (1994)
NaCl-Best
Fibers
HCl-Best
Fibers
QHCl-Best
Fibers
Sucrose-Best
Fibers
Mean
0.23
0.35
0.45
0.63
0.32
0.37
0.42
0.28
0.36
0.38
For abbreviations, see Tables 1 and 3.
Comparison with chimpanzee
As mentioned earlier, we are publishing a series of studies in chimpanzee. These show a strong similarity between
FIG . 13. Results of 2-bottle preference tests with the array of sweeteners
used in electrophysiological experiments. Data were averaged for 16 monkeys. Error bars: SE. Horizontal line: chance level (0.5).
/ 9k0c$$fe23
J176-6
the sense of taste in chimpanzees and humans ( Hellekant
and Ninomiya 1991, 1994a,b; Hellekant et al. 1985, 1996,
1997; Ninomiya and Hellekant 1991 ) . This is not unexpected, considering that the chimpanzee is phylogenetically closer to humans than any other species ( cf. Begun
1992; Miyamoto et al. 1987; Pilbeam 1984 ) and shows
strong similarities in food choices ( cf. Nowak 1991 ) .
Thus it seems that a comparison between the characteristics of the CT nerve of chimpanzee and rhesus monkey
could yield data applicable to the question on how good
a model the rhesus monkey is.
It seems to us that the most important difference is in the
way gymnemic acids affect sweet taste. In humans (Bartoshuk et al. 1969) and chimpanzees (Hellekant et al. 1985,
1996, 1997), gymnemic acids abolish sweet taste in general,
whereas they have no effects on the taste of sweeteners in
macaques (Hellekant et al. 1974). Here we observe five
considerably smaller differences between chimpanzee and
rhesus monkey.
1) Acids never stimulated S fibers in the chimpanzee;
here they elicited some response in S fibers.
2) The sensitivity to denatonium benzoate is higher in
the chimpanzee CT, where 0.001 mM elicited a response,
compared with the 0.01 and 0.1 mM used here with almost
no effect. The monkey seems to share less sensitivity to
denatonium benzoate with some other species, e.g., mice
(Whitney et al. 1990). However, when a small response
occurred it was only observed in Q fiber.
3) Here, we found one group of NaCl-best fibers that did
not respond to KCl and whose response to NaCl was generally suppressed by amiloride. In chimpanzee we identified
three fiber groups within the N cluster: fibers responding
best to NaCl but not to KCl, fibers responding best to both
NaCl and KCl, and fibers responding best to MSG (Hellekant et al. 1997). Only the first group shared features with
the NaCl-best group in rhesus.
4) The effects of umami compounds (MSG, guanosine
5 *-monophosphate) were different. In the chimpanzee these
compounds stimulated, in addition to fibers of the M cluster,
some sucrose-best fibers. Here they stimulated mostly fibers
of the N cluster. This may indicate that to the rhesus monkey
the taste of MSG does not carry any sweet qualities.
5) Finally, the CT fibers of rhesus monkey were more
broadly tuned than the CT fibers of chimpanzee (mean value
of entropy for all fibers: 0.53 vs. 0.3). However, these differences are only minor, not major as when a compound with
an intense sweet taste to humans has no taste at all to an
animal species (e.g., Glaser et al. 1978)
To summarize, the taste fibers of M. mulatta cluster in
both CT and NG according to the human concepts of taste
qualities. The finding of a strong representation of QHCl-
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
to know how, e.g., aspartame tastes to the monkey. It may
also seem that the term ‘‘like’’ is anthropomorphic, but it
best describes the animals’ reactions when they eagerly drink
and empty the bottle with 50 ml sweetener within 1 min.
Thus the results suggested a similar taste of the sweeteners
used here in both monkey and humans.
Further, the human response to acids and QHCl, especially
in children (Steiner 1994), parallels that of rhesus monkeys.
Our earlier two-bottle preference study (Brouwer et al. 1983)
shows that acids are rejected by rhesus monkeys. As in humans,
the rejection can be overcome by addition of sucrose or by
application of miraculin (Brouwer et al. 1983). QHCl is strongly
disliked by rhesus monkeys. The rejection can also be overcome
with sweeteners (Hellekant 1980). The parallel with human
experiences, especially with children, is evident.
With regard to the possibility of direct comparisons between this study and human taste nerve recordings, there are
few recordings from human taste nerves and none from single taste fibers (Diamant et al. 1963, 1972; Oakley 1985;
Zotterman 1971). This limits comparisons between nerve
responses in humans and macaques.
992
G. HELLEKANT, V. DANILOVA, AND Y. NINOMIYA
This study was supported by National Institute on Alcohol Abuse and
Alcoholism Grant AA-09391.
Address for reprint requests: G. Hellekant, Dept. of Animal Health and
Biomedical Sciences, 1655 Linden Dr., Madison, WI 53706.
Received 5 March 1996; accepted in final form 28 October 1996.
REFERENCES
BARTOSHUK, L. M., DATEO, G. P., VANDERBELT, D. J., BUTTRICK, R. L.,
AND LONG, L., JR . Effects of Gymnema sylvestre and Synsepalum dulcificum on taste in man. In: Olfaction and Taste, edited by C. Pfaffmann.
New York: Rockefeller Univ. Press, 1969, vol. III, p. 436–449.
BAYLIS, L. L. AND ROLLS, E. T. Responses of neurons in the primate taste
cortex to glutamate. Physiol. Behav. 49: 973–979, 1991.
BEGUN, D. R. Miocene fossil hominids and the chimp-human clade. Science
Wash. DC 257: 1929–1933, 1992.
BROUWER, J. N., GLASER, D., HARD AF SEGERSTAD, C., HELLEK ANT, G.,
NINOMIYA, Y., AND VAN DER WEL, H. The sweetness-inducing effect of
miraculin; behavioral and neurophysiological experiments in the rhesus
monkey (Macaca mulatta). J. Physiol. Lond. 337: 221–240, 1983.
COLLINGS, V. D. Human taste response as a function of locus of stimulation on the tongue and soft palate. Percept. Psychophys. 1: 169 – 174,
1974.
DIAMANT, H., FUNAKOSHI, M., STRÖ M, L., AND ZOTTERMAN, Y. Electrophysiological studies on human taste nerves. In: Olfaction and Taste,
edited by Y. Zotterman. Oxford, UK: Pergamon, 1963, vol. I, p. 193–
203.
DIAMANT, H., HELLEK ANT, G., AND ZOTTERMAN, Y. The effect of miraculin
on the taste buds of man, monkey and rat. In: Olfaction and Taste, edited
by D. Schneider. Stuttgart, Germany: Wissenschaftliche, 1972, vol. IV,
p. 241–244.
DIAMANT, H., OAKLEY, B., STRÖ M, L., WELLS, C., AND ZOTTERMAN, Y. A
comparison of neural and psychophysical responses to taste stimuli in
man. Acta Physiol. Scand. 64: 67–74, 1965.
FARBMAN, A. I. AND HELLEK ANT, G. Evidence for a novel mechanism of
binding and release of stimuli in the primate taste bud. J. Neurosci. 9:
3522–3528, 1989.
FEIGIN, A. M., BRAND, J. G., NINOMIYA, Y., BRYANT, B. P., BEZRUKOV,
S. M., MOORE, P. A., VODYANOY, I., AND TEETER, J. H. Effect of novobiocin on cation channel formation and enhancement of salt taste. In: Olfaction and Taste, edited by K. Kurihara, N. Suzuki, and H. Ogawa. Tokyo:
Springer-Verlag, 1994, vol. XI, p. 104–107.
FRANK, M. E. An analysis of hamster afferent taste nerve response functions. J. Gen. Physiol. 61: 588–618, 1973.
FRANK, M. E. Taste-responsive neurons of the glossopharyngeal nerve of
the rat. J. Neurophysiol. 65: 1452–1463, 1991.
GLASER, D., HELLEK ANT, G., BROUWER, J. N., AND VAN DER WEL, H. The
taste responses in primates to the proteins thaumatin and monellin and
their phylogenetic implications. Folia Primatol. 29: 56–63, 1978.
GORDON, G., KITCHELL, R., STRÖ M, L., AND ZOTTERMAN, Y. The response
/ 9k0c$$fe23
J176-6
pattern of taste fibers in the chorda tympani of the monkey. Acta Physiol.
Scand. 46: 119–132, 1959.
HANAMORI, T., MILLER, I. J., AND SMITH, D. V. Gustatory responsiveness
of fibers in the hamster glossopharyngeal nerve. J. Neurophysiol. 60:
478–498, 1988.
HÄ NIG, D. P. Zur Psychophysik des Geschmacksinnes. Philos. Stud. 17:
576–623, 1901.
HÅRD AF SEGERSTAD, C. H. AND HELLEK ANT, G. The sweet taste in the
calf. II. Glossopharyngeal nerve responses to taste stimulation of the
tongue. Physiol. Behav. 45: 1043–1047, 1989.
HELLEK ANT, G. Electrophysiological investigation of the gustatory effect
of ethyl alcohol. I. The summated response of the chorda tympani in the
cat, dog and rat. Acta Physiol. Scand. 64: 392–397, 1965a.
HELLEK ANT, G. Electrophysiological investigation of the gustatory effect
of ethyl alcohol. II. A single fibre analysis in the cat. Acta Physiol. Scand.
64: 398–406, 1965b.
HELLEK ANT, G. Different types of sweet receptors in mammals. In: Olfaction and Taste, edited by D. A. Denton and J. P. Coghlan. New York:
Academic, 1975, vol. V, p. 15–21.
HELLEK ANT, G. Effects of miraculin and gymnemic acid in the rhesus
monkey (Macaca mulatta). In: Food Intake and Chemical Senses, edited
by Y. Katsuki, M. Sato, S. F. Takagi, and Y. Oomura. Tokyo: Univ. of
Tokyo Press, 1977, p. 201–210.
HELLEK ANT, G. Preference for sweet taste and the taste of sweeteners in
animals. J. Anim. Physiol. Anim. Nutr., Suppl. 11: 43–52, 1980.
HELLEK ANT, G., GLASER, D., BROUWER, J. N., AND VAN DER WEL, H. Gustatory effects of miraculin, monellin and thaumatin in the Saguinus midas
tamarin monkey studied with electrophysiological and behavioral techniques. Acta Physiol. Scand. 97: 241–250, 1976.
HELLEK ANT, G., HAGSTROM, E. C., KASAHARA, Y., AND ZOTTERMAN, Y.
On the gustatory effects of miraculin and gymnemic acid in monkey.
Chem. Sens. Flavor 1: 137–145, 1974.
HELLEK ANT, G., HÅRD AF SEGERSTAD, C., ROBERTS, T., VAN DER WEL, H.,
BROUWER, J. N., GLASER, D., HAYES, R. J., AND EICHBERG, J. W. Effects
of gymnemic acid on the chorda tympani proper nerve responses to sweet,
sour, salty and bitter taste stimuli in chimpanzee. Acta Physiol. Scand.
124: 339–408, 1985.
HELLEK ANT, G., HLADIK, C. M., DENNYS, V., SIMMEN, B., ROBERTS, T. W.,
GLASER, D., DUBOIS, G., AND WALTERS, D. E. On the sense of taste in
two malagasy primates (Microcebus murinus and Elemur mongoz).
Chem. Senses 18: 307–320, 1993.
HELLEK ANT, G. AND NINOMIYA, Y. On the taste of umami. Physiol. Behav.
49: 927–934, 1991.
HELLEK ANT, G. AND NINOMIYA, Y. Umami taste in chimpanzee (Pan troglodytes) and rhesus monkey (M. mulatta). In: Olfaction and Taste, edited
by K. Kurihara, N. Suzuki, and H. Ogawa. Tokyo: Springer-Verlag,
1994a, vol. XI, p. 365–368.
HELLEK ANT, G. AND NINOMIYA, Y. Bitter taste in single chorda tympani
taste fibers from chimpanzee. Physiol. Behav. 56: 1185–1188, 1994b.
HELLEK ANT, G., NINOMIYA, Y., AND DANILOVA, V. Taste in chimpanzee.
II. Single chorda tympani fibers. Physiol. Behav. In press.
HELLEK ANT, G., NINOMIYA, Y., DUB OIS, G. E., D ANILOVA, V., AND
ROBERTS, T. W. Taste in chimpanzee. I. The response to sweeteners
and the effect of gymnemic acid. Physiol. Behav. 60: 469 – 479,
1996.
HELLEK ANT, G. AND ROBERTS, T. W. Whole nerve and single fiber recordings in non-human primates. In: Experimental Cell Biology of Taste
and Smell, edited by A. L. Spielman and J. G. Brand. Boca Raton, FL:
CRC, 1995, p. 277–290.
HELLEK ANT, G. AND VAN DER WEL, H. Taste modifiers and sweet proteins.
In: Neural Mechanisms of Taste, edited by R. Cagan. Boca Raton, FL:
CRC, 1989, p. 85–96.
HELLEK ANT, G. AND WALTERS, D. E. On the sweetness of five high-potency
sweeteners in rats. In: Sweet-Taste Chemoreception, edited by M. Mathlouthi, J. A. Kanters, and G. G. Birch. London: Elsevier, 1991, p. 373–
386.
KURIHARA, K., KURIHARA, Y., AND BEIDLER, L. M. Isolation and mechanism
of taste modifiers. Taste-modifying protein and gymnemic acids. In: Olfaction and Taste, edited by C. Pfaffmann. New York: Rockefeller Univ.
Press, 1969, vol. III, p. 450–469.
MIYAMOTO, M. M., SLIGHTOM, J. L., AND GOODMAN, M. Phylogenetic relations of humans and African apes from DNA sequences in the psi etaglobin region. Science Wash. DC 238: 369–373, 1987.
NINOMIYA, Y. AND FUNAKOSHI, M. Amiloride inhibition of responses of rat
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
best fibers from the back of the tongue, and of NaCl- and
sucrose-best fibers from the front of the tongue, although
not unique among mammals, corresponds with the human
distribution of mostly bitter taste on the back and sweet /
salty from the front. The fiber pattern of the CT of the
rhesus monkey mirrors that of the chimpanzee and its
behavioral response to tastants parallels the psychophysical and electrophysiological observations in humans well
enough to support the conclusion that the rhesus monkey,
from most points of view, can serve as a relevant animal
model of the human sense of taste. We conclude that the
sense of taste in macaques is quite similar to that of humans and different from that of commonly used nonprimate species. This conclusion forms the basis for and may
in some cases necessitate the use of higher primates as
models for the human taste of both ‘‘standard’’ tastants,
such as bitter or sweet compounds, and compounds with
an enigmatic taste, such as ethanol.
CHORDA TYMPANI AND GLOSSOPHARYNGEAL TASTE FIBERS IN RHESUS
/ 9k0c$$fe23
J176-6
chorda tympani fibers in the cynomolgus monkey. Chem. Senses 19:
381–400, 1994.
SCHIFFMAN, S. S., REILLY, D. A., AND CLARK, T. B. Qualitative differences
among sweeteners. Physiol. Behav. 23: 1–9, 1979.
SCOTT, T. R., PLATA-SALAMAN, C. R., SMITH, V. L., AND GIZA, B. K. Gustatory neural coding in the monkey cortex: stimulus intensity. J. Neurophysiol. 65: 76–86, 1991.
SCOTT, T. R., PLATA-SALAMAN, C. R., AND SMITH SWINTOSKY, V. L. Gustatory neural coding in the monkey cortex: the quality of saltiness. J.
Neurophysiol. 71: 1692–1701, 1994.
SMITH, D. V. AND TRAVERS, J. B. A metric for the breadth of tuning of
gustatory neurons. Chem. Sens. Flavor 4: 215–229, 1979.
STEINER, J. E. Behaviour manifestations indicate of hedonics and intensity
in chemosensory experience. In: Olfaction and Taste, edited by K. Kurihara, N. Suzuki, and H. Ogawa. Tokyo: Springer-Verlag, 1994, vol. XI,
p. 284–287.
TONOSAKI, K. AND F UNAKOSHI, M. Amiloride does not block taste transduction in the mouse. Comp. Biochem. Physiol. 94: 659 – 661, 1989.
WHITNEY, G., MAGGIO, J. C., AND HARDER, D. B. Manifestation of the
major gene influencing sucrose octaacetate (SOA) tasting among mice:
classic taste qualities. Chem. Senses 15: 243–252, 1990.
ZOTTERMAN, Y. Action potentials in the glossopharyngeal nerve and in the
chorda tympani. Scand. Arch. Physiol. 72: 73–77, 1935.
ZOTTERMAN, Y. The recording of the neural response from human taste
nerves. In: Handbook of Sensory Physiology: Chemical Senses, Taste,
edited by L. M. Beidler. Berlin: Springer-Verlag, 1971, vol. 2, p. 102–
115.
09-04-97 20:30:55
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 14, 2017
single chorda tympani fibers to chemical and electrical tongue stimulations. Brain Res. 451: 319–325, 1988.
NINOMIYA, Y. AND HELLEK ANT, G. Specific taste sensitivity of single chorda
tympani fibers in chimpanzee. Proc. Jpn. Symp. Taste Smell 25: 313–
316, 1991.
NOWAK, R. Walker’s Mammals of the World. Baltimore, MD: Johns Hopkins Univ. Press, 1991.
OAKLEY, B. Taste responses of human chorda tympani nerve. Chem. Senses
10: 469–481, 1985.
PFAFFMANN, C., BARTOSHUK, L. M., AND MC BURENY, D. H. Taste psychophysics. In: Handbook of Sensory Physiology: Chemical Senses, Taste,
edited by L. M. Beidler. Berlin: Springer-Verlag, 1971, vol. 2, p. 75–
101.
PILBEAM, D. The descent of hominoids and hominids. Sci. Am. 250: 84–
96, 1984.
PLATA-SALAMAN, C. R., SCOTT, T. R., AND SMITH SWINTOSKY, V. L. Gustatory neural coding in the monkey cortex: L-amino acids. J. Neurophysiol.
67: 1552–1561, 1992.
PLATA-SALAMAN, C. R., SCOTT, T. R., AND SMITH-SWINTOSKY, V. L. Gustatory neural coding in the monkey cortex: the quality of sweetness. J.
Neurophysiol. 69: 482–493, 1993.
PLATA-SALAMAN, C. R., SCOTT, T. R., AND SMITH-SWINTOSKY, V. L. Gustatory neural coding in the monkey cortex: acid stimuli. J. Neurophysiol.
74: 556–64, 1995.
SATO, M., OGAWA, H., AND YAMASHITA, S. Response properties of macaque
monkey chorda tympani fibers. J. Gen. Physiol. 66: 781–810, 1975.
SATO, M., OGAWA, H., AND YAMASHITA, S. Gustatory responsiveness of
993