Neuronal responses of the nucleus tractus solitarius to oral stimulation with umami substances

Physiology& Behavior,Vol. 49, pp. 935-941. ©Pergamon Press pie, 1991. Printed in the U.S.A.

0031-9384/91 $3.00 + .00

Neuronal Responses of the Nucleus Tractus Solitarius to Oral Stimulation With Umami Substances AKIRA ADACHI AND MITSUNORI AOYAMA*

Department of Physiology, *Department of Operative Dentistry Okayama University Dental School, Shikata-cho, Okayama, Japan 700

ADACHI, A. AND M. AOYAMA. Neuronal responses of the nucleus tractus solitarius to oral stimulation with umami substances. PHYSIOL BEHAV 49(5) 935-941, 1991.--In order to investigate coding mechanisms of special taste modality (umami), responses of neurons within the nucleus tractus solitarius (NTS) to oral stimulation with monosodium glutamate, disodium 5'-inosinate (IMP) or their mixture were recorded in the conventional electrophysiological method. Results obtained were as follows: Neither MSGbest nor IMP-best neuron was recorded within the NTS as in the primary taste afferents. Some of the sucrose-best neurons, NaC1best neurons and HCl-best neurons responded to oral stimulation with MSG or IMP. A remarkable synergistic effect was observed in all of the sucrose-best neurons and in some of the NaCl-best neurons but not in all of the HCl-best neurons, when the mixed solution of MSG and IMP was applied into the oral cavity. As to the sucrose-best neurons, potency of the synergism was positively correlated with the responsiveness to sucrose. No correlation was recognized between them in the case of NaCl-best neurons. These results suggest a view that the sucrose-best neurons and the NaCl-best neurons which show the synergism may participate in coding umami taste. Nucleus tractus solitarius

Umami taste

Synergism

MONOSODIUM glutamate (MSG) and 5'-nucleotides have been used as food additives to enhance the taste quality of food and their special taste-increasing palatability was designated in Japanese as umami. A large number of neurophysiological studies have been carried out to solve a transduction and coding mechanisms of this taste modality, umami (4, 7, 10-12, 17, 18). MSGspecific primary afferent was found in the cat chorda tympani (4) and in the mouse glossopharyngeal nerve (7). However, these findings seem to be an exceptional case. In general, no MSGspecific fiber has been identified either in the chorda tympani or the glossopharyngeal nerve of various experimental animals. Otherwise, a multidimensional scaling analysis applied on the spatial representation of the similarities of various taste stimuli elucidated that umami is independent from the classical four primary tastes, saltiness, sweetness, sourness and bitterness (16). This discrepancy indicates a necessity for further neurophysiological analysis of umami taste. Among these studies, it is worthy of notice that a mixture of MSG and IMP causes synergism which potentiates only umami modality but not others. We anticipated that synergism observed in neuronal responses seems to provide a promising clue for the explanation of these poorly understood sensory phenomena. In addition, responses of the secondary afferent neurons to oral stimulation with umami taste solution have not been studied in much detail. The purpose of this study is to delineate responses of neurons within the nucleus tractus solitarius to oral stimulation with MSG and/or disodium 5'-inosinate (IMP), emphasizing synergism of

the mixture of MSG and IMP, in order to obtain further information of the transduction and coding mechanisms of umami taste. METHOD Male Sprague-Dawley rats (Charles River of Japarl) weighing 230--400 g were used. They were anesthetized with a single intraperitoneal injection of urethane-chloralose (urethane, 0.8 g/kg; chloralose, 65 mg/kg body wt.). Anesthesia was maintained by subsequent IP administrations of the drugs. The trachea was cannulated. An oral catheter (PE-50 polyethylene tubing with a flared end) for purpose of oral stimulation with sapid solution was inserted into the oral cavity through the bucca. The animal was mounted on a stereotaxic apparatus and the medulla oblongata was exposed by removing a part of the occipital bone and aspirating a caudal portion of the cerebellum. Care was taken not to break the auris media by inserting a pair of ear bars into the auditory meatus. Glass pipette electrodes filled with 0.5 M sodium acetate and 2% pontamine sky blue and conventional electrophysiological equipments were used for recording unitary discharges. Recording sites were determined afterwards histologically. Sapid solutions employed in this experiment were as follows: 0.1 M and 0.01 M NaC1, 0.5 M sucrose, 0.01 M HC1, 0.01 M quinine-HC1, 0.01 M MSG, and 0.001 M IMP. In order to examine synergism, mixture of 0.01 M MSG and 0.001 M IMP was used. A mixture of 0.1 M NaC1, 0.01 M MSG and 0.001 M IMP was also em-

935

936

A D A C H I AND A O Y A M A

0.1M

Impulses/sec

NaCI

i!r!Duises/sec 0.1M

IJ ~ ] i "U

' I'

0.5M

[IIH[ l ilL _;

[ii.

;

IIIrl'

r~

1-'1

l

I~I

r

l ,~

[I

r~

. i .I

11 "

......

f-

i.;

I

l

NaCI

r '

4[i

~-

20

[-

4 c~

f

o 20

Suc f

0.5M

Suc

10

0

O.01M

MSG O.01M

MSG

F "i

""

0.001M

I~1~1

V]II!"I I

q-I"FL

]

1t1 " 111! I I 3

II

IMP

O.O01M

J q

"

0,01M

L I

lib Fill

MSG+0.001M

i ~'

~; ql

t -- I

-

LL L_L IT I ~

irk ilL Ill l~q "

l

IMP

r

Jl,,

~r

0

i,~llql'IPl'~FIP;lINilqH~!~ ! lq,,IrIll

II'~F~[iTII#]II~I~111 ~ ql~r!

,rl,[,],]] }II~,'i;i[

I

0 L

.,.t,a ....

IMP

~II ii

i~lllll llll~lllllllliilll il|il{lll IiII~ ltlillillitillllllll~lllI|I

Ir!

IIIIII!IIIIIIII~ IIiI~11~11lllllll~fll ~III II~,~ I~ iII~II'~11111l~illIIIIII~II

11

5

j=

O.01M

MSG÷O.OOIM

IMP

v 20

0

5sec

30sec

FIG. 1. Responses of a NTS neuron to oral stimulation with various sapid solutions. Right traces are oscillograph records of unit discharges and left graphs are pen recordings of discharge rate vs. time by means of a pulse rate meter. Arrows indicate time when the taste stimuli were applied. Note 0.5 M sucrose (Suc) or the mixture of 0.01 M MSG and 0.001 M IMP elicits a marked increase in discharge rate. The synergistic ratio: 1.60

ployed. These solutions were infused into the oral cavity via the oral catheter by 2 ml at a room temperature. After intraoral stimulation for 30 s with each solution, the oral cavity was rinsed with distilled water via the same catheter until a discharge rate

5sec

FIG. 2. Responses of another neuron are illustrated in a similar fashion in Fig. 1. Note a marked response to NaC1, MSG or the mixture. No synergism is recognized. The synergistic ratio: 0.96

returned to its spontaneous level. The magnitude o f neuronal response was expressed by number o f impulses which were obtained by subtracting number o f impulses appeared during 5 s immediately before stimulation from number o f impulses elicited during the initial 5 s after stimulation. Synergistic ratio was calculated by a formula: R ( M S G +

R(MSG+IMP)

[~

R(MSG) R(IMP)

~ 400 _E

30[

20(

10C

!~!ii

iliii a

,~ i~iiiiil

t

i i!!il c

Non synergistic

J

type neurons

30sec

k

f

n

o

p

q

r

s

Synergistic type neurons

FIG. 3. Comparison between R(MSG + IMP) and arithmetic sum of R(MSG) and R(IMP) on 19 neurons examined in this experiment. Difference of the synergistic type neurons from the nonsynergistic ones are clearly presented. See text for more detail.

NTS NEURON AND UMAMI TASTE

937

TABLE 1 NUMERICAL D A T A OF SYNERGISTIC RATIOS C A L C U L A T E D O N EACH NEURON PRESENTED IN FIG. 3

Nonsynergistic Type Neurons

Synergistic ratio Mean ---SD

a

b

1.07

0.96

d

e

f

g

h

i

j

k

1

0.92 1.09 0.87 0.87 1.18 1.26 0.91 0.78 0.89 0.66

0.96 _+0.17" Synergistic Type Neurons n o p q r

m Synergistic ratio Mean -+SD

c

1.47 1.94 1.65 ---0.23*

s

1.47 2.02 1.53 1.54 1.60

*p<0.001.

IMP)/R(MSG) + R(IMP) where R(MSG + IMP) is the magnitude of response to the mixture of MSG and IMP; R(MSG) or R(IMP) is the magnitude to single solution of MSG or IMP respectively. Because the synergistic ratios of nonsynergistic responses were not always exactly 1.00 but some of them were less than 1.00, the synergistic neuron was evaluated by the criterion that the synergistic ratio was more than 1.32 which was derived from the statistical analysis.

Figure 2 shows a sample record of responses that no synergism was observed. Depending on the same criteria in Fig. 1, this neuron is one of the NaCl-best neurons (N-best). No synergistic effects were observed, because the synergistic ratio was 0.96 calculated by the formula described in the Method section, being significantly less than the threshold value 1.32. The umami taste substances, MSG, IMP and their mixture employed in this experiment elicited the responses of the NTS neurons only to the S-best, the N-best and the H-best neurons more or less, if not all. The magnitudes of responses of 19 neurons (from a to s) responded to these urnami stimuli are presented in Fig. 3. The magnitude of response to IMP single solution, R(IMP), indicated by a cross-ruled bar is added on the top of shaded bar which presents the magnitude of response to MSG single solution, R(MSG), in order to show comparison of addition of R(IMP) and R(MSG) with the magnitude of response to the mixture, R ( M S G + I M P ) , indicated by clear bar. The responses of nonsynergistic type neurons are shown from a to 1 and these of synergistic type neurons are also shown from m to s. This bar graph (Fig. 3) clearly demonstrates that there exist two different types of neurons within the NTS; one is nonsynergistic and the other is synergistic. It is also evident that neither R(MSG) nor R(IMP) relates to the degree of synergism. Numerical data of Fig. 3 were listed on Table 1.

RESULTS

A total of 56 neurons were explored. Among them, 19 were possible to complete the examination for responsiveness to all test solutions. All other neurons that lost neural activity during the examination were eliminated from the data. A typical response of a NTS neuron which showed the synergism is illustrated in Fig. 1. This neuron displays the most marked response to 0.5 M sucrose among the other primary taste modalities. Less response was elicited by either HC1 or quinine-HC1 (these responses are not shown in Fig. 1). Therefore, this can be identified as the sucrose-best neuron (S-best) depending upon the criteria according to a labelled-line notion. It is noted that the mixture of MSG and IMP applied into the oral cavity elicits a marked response, so the synergistic ratio was 1.60.

TABLE 2 N U M E R I C A L DATA O F R(NaC1 + M S G + IMP)/R(NaCI) + R(MSG + IMP) C A L C U L A T E D ON EACH NEURON PRESENT IN FIG. 7

Nonsynergistic Type Neurons a

R(NaCI+MSG+IMP)

0.70

R(NaCI)+ R(MSG+ IMP)

Mean --+SD

b

c

d

e

f

h

Synergistic type neurons R(NaO + MSG+ IMP) R(NaCl)+ R(MSG+ IMP)

Mean - SD

0.40

j

k

1

0.49 0.59 0.68 0.58 0.70 0.43 0.86 0.29 0.55 1.45

0.67 _ 0.30 m

i

n

o

p

q

r

s

0.71 0.72 0.74 0.46 0.64 0.47

0.59 ___0.14

Note most of these neurons except neuron 1 are less than 1.00.

938

ADACHt AND AOYAMA

j m

t

0~ 400

o ,,

NaCI-best

;

(n 7)

HCI best ,n 4)

400[ua

r

I I

~,

300!

3001

-

!

o

i Suc-best (n=4)

200

200i

o

Q

J

u~ ,,',',/

,

~

100

.

100 _ ,

I00

' "i', '

p,

b ,?'

0

43

0

S

N

H

Q

S

MSG

N

H

Q

S

MSG

N

H

Q

MSG

FIG. 4. Neurons examined are classified into 3 groups according to the labelled-line coding notion. The magnitude of response to each sapid solution (S; 0.5 M sucrose, N; 0.1 M NaCI, H; 0.01 M HCI, Q; 0.01 M quinine-HC1) is plotted. Filled circles indicate the responses of neurons which induced the synergism, while open circles indicate those of the nonsynergistic ones. No Q-best neuron was recorded in this experiment.

o r 0.64 (0.001
2

,A, c

Suc best

2 • "~

A NaCl best I-1 HCI best

[]

• O Not classified

1

0 .... 0

100 RISuc)

Impulses/5sec

o 3 -

r =-0.18 (0.4 < P < 0 . 5 )

2 u~ o~ co

2 -

•

~-

z~ I

OO

0

D

[]

a.

[]

I

I

I

I

100

200

300

400 R(NaCI)

Impulses/5sec

FIG. 5. Correlation between the synergistic ratios and the magnitudes of responses to sucrose [R(Suc)] is presented in the upper graph. Note that a good correlation exists between them. In the lower graph, the synergistic ratios are plotted against the magnitudes of responses to NaC1 [R(NaC1)]. No correlation can be found in this case. All filled marks indicate neurons which induced the synergism and open ones indicate nonsynergistic neurons. Circles indicate not classified neurons because all tests were not completed,

NTS NEURON AND UMAMI TASTE

939

Impulses/sec

0.1M NaCl

20

nhllll II !IImLL2 ]lllllll ~"a'='|ll lilllll dlallliiMI llil Illili

IIIIlll /lllllllll~llllllklhllllllllil/lll

Llilll

10 0

O.01M MSG+O.OO1M IMP

3°I

20 id~ll'i~lJlll ' hr ~llh~ill iHlllJIql~lltlhll

II ~1,11 fldllltl I~1 Illll. alklllflJ

II1~.1

~ [ I n l l ~ I]]FI

II u l u i n

I~111|' I

I, lHrlnlmll iii q l l l l l l l l l ~ l l t l l l l l l

IIIll[ll]

Ill n II[I I"11~111~[ p

10 o

0.1M NaCI+O.O1M MSG+O.OO1M IMP

40 r 3O

T

2o

! I!~li! ,lll!~l= j =[l:[ IIIiilllll!tl~ MI qllhlj,lqlh I i~!,IILIIH!~III rP~hlH!III I~l~l~ IH ~ll,lllLJrllll I~1 i~qllr~[l~lil~ld

I~lllllll[ipll~ill,llilqlllllllllii I

f~ljqFiiiiiiIil[p[qlllllllllrllq~n~llqMiiillll~tiiFllll p

10

o

L

0.1M NaCI

~

J.LL~ i~!l! I Ill

Ib'll IIIllllltlll~|llidl Iltll

Ill|lllllll~lltltlhlllllllill~llll'

liill

I~ CTIIIIJIIlli|~IPIII~IITII~I~'IPlil

lilpl[llllll~ll'llllllTllllml

~P I qlll~Ullll I

Illlllll

5sec

2o 10 0

30sec

FIG. 6. Comparisons of R(NaC1+ MSG + IMP) with arithmetic sum of (NaC1) and R(MSG + IMP). All 18 neurons show the former is less than the latter, except the neurons I.

Neurons examined in this experiment were classified into three types according to the labelled-line notion. The magnitudes of responses of neurons which were most sensitive to the sweet sucrose (S-best), the salty NaC1 (N-best) or the sour HC1 (H-best) are depicted in Fig. 4. The left graph shows response patterns of the S-best neurons, the middle one shows those of the N-best and the fight hand one shows those of the H-best. These graphs clearly represent that all S-best neurons show the synergism when MSG is mixed with IMP, but no H-best ones do it. Only 2 N-best neurons among 7 show the synergism, even though all this type of neurons responded to MSG. The response to MSG of nonsynergistic N-best as well as H-best neurons could be induced by sodium ion of MSG. In Fig. 5, the magnitude of response to sucrose R(Suc) or the magnitude of response to NaC1 R(NaC1), were plotted against the synergistic ratios. The synergistic ratios are positively correlated to R(Suc) as presented in the upper graph of Fig. 5. No correlation between the synergistic ratios and R(NaC1) was recognized (the lower graph). This evidence suggests a possibility that some of the S-best neurons may predominantly code umami taste modality. Because MSG elicited the marked response of the N-best neuron probably due to sodium ion, effects of the mixture of MSG and IMP that induces the synergism of responses to NaC1 were examined. The most typical effect induced by blending the mixture of MSG and IMP in NaC1 solution is illustrated in Fig. 6. It is noted that R(NaC1) is almost completely suppressed by blendhag with the mixture of MSG and IMP even though no responses are elicited by the mixture (see the 2nd trace from the top). It is certain that the suppression was not caused by an accidental decrease in neuronal activity, because the same R(NaC1) was repeatedly elicited as shown in the bottom trace in this figure. The suppressive effect on R(NaC1) due to blend the mixture was mostly observed ha the neurons responded to the umami stim-

uli as presented in Fig. 7. As in Fig. 3, R(MSG + IMP), clear bar, is added on the top of dotted bar presenting R(NaC1) for purpose of comparison with R(NaC1 + MSG + IMP) indicated by hatched bar. Apparently, this graph shows that R(NaCI) of all neurons except one (neuron 1) are suppressed by the blend. Numerical data of Fig. 7 were presented on Table 2. DISCUSSION

Histological examination of the recording sites revealed that these were localized within the most rostral portion of the NTS to where the gustatory signals from the anterior tongue (chorda tympani) and from the nasoincisor ducts (superficial greater petrosal nerve) project (3). The nasoincisor ducts, as well as the anterior tongue, are least sensitive to quinine-HC1 (6, 8, 9). It was because of this that we failed to record a response to oral stimulation with quinine-HCl. The time courses of the NTS responses to oral stimulation with sucrose observed in the present experiment are quite similar to those of stimulation of the nasoincisor ducts with sucrose (14). This implies that the sucrose responses in this experiment were derived from nasoincisor ducts stimulation. It is interesting that the taste receptors located at the nasoincisor ducts also respond to either MSG or IMP. In addition, it is noted that the synergism primarily observed in the chorda tympani response (1, 2, 4, 1012, 18) was also recognized in the superficial greater petrosal nerve response in the rat as well. It is well known that the mixture of MSG and 5'-nucleotides induces the synergism at the taste receptor sites (13). Psychophysical test also revealed the synergism that the unique umami modality is greatly enhanced but not others when a small amount of IMP is mixed with MSG (5,15). Combining the above evidences, it is conceivable that a group of neurons inducing the synergism can be closely related to code umami taste. We observed that all S-best neurons within the NTS induced the syner-

940

ADACHI AND AOYAMA

8ooj

R(NaCI+MSG+IMP} R(NaCb R(MSGqMP)

q

600

E~_]

400 1

a

b

d

e

f

h

i

I

Non synergistic type neurons

k

m

n

q

p

o

r

s

Synergistic type neurons

FIG. 7. Suppression of R(NaC1) by blending the mixture of MSG and IMP is illustrated in the similar fashion in Fig. 1.

gism but only two among 7 N-best neurons did as far as examined. It is possible that nonsynergistic S-best neurons may be found in the future. However, the S-best neurons exert the synergism more predominantly than the N-best ones. Thus it is likely that the most S-best neurons and the selected N-best neurons which induce the synergism extensively related to the coding mechanisms of umami modality within the furst order taste relay, the NTS. It is worth noting that the responses to NaC1 were suppressed by blending the mixture of MSG and IMP. This kind of modulation has not been observed in the primary afferent, the chorda tympani response. An additive effect is in a common phenomenon recognized as the peripheral gustatory nerve responses (4, ! 8). However, it is obscure whether the mixture suppresses the response to NaC1 or NaC1 does the response to the rmxture. The

evidence presented in Fig. 7 might support the former possibility, because even though this neuron is insensitive to oral stimulation with the mixture, the response to NaC1 is completely suppressed. Therefore, it is suggested that some of the N-best neurons not responding to the mixture are postsynaptically modulated by a primary afferent responding to the mixture, probably via some intemeurons within the NTS. This modulation or interaction occurring within the relay nucleus may play a role to perception of umami taste modality. ACKNOWLEDGEMENTS This work was supported by a Grant-in-Aid for General Scientific Research 62480378 and 02670831 from the Ministry of Education, Science and Culture in Japan and by a grant from Umami Manufacturers Association of Japan.

REFERENCES 1. Adachi, A. Study on the taste mechanisms of Na-glutamate and Nainosinate. J. Physiol. Soc. Jpn. 24:607--613; 1962 (in Japanese with English summary). 2. Adachi, A. Neurophysiological study on taste effectiveness of seasoning. J. Physiol. Soc. Jpn. 26: 347-355; 1964 (in Japanese with English summary). 3. Hamilton, R. B.; Norgren, R. Central projections of gustatory nerves in the rat. J. Comp. Neurol. 222:560-577; 1984. 4. Kawamura, Y.; Adachi, A.; Ohara, M.; Ikeda, S. Neurophysiological studies on taste effectiveness of flavor enhancing substances. Amino Acid Nucl. Acid 10:168-178; 1964 (in Japanese with English summary). 5. Kuninaka, A.; Kibi, M.; Sakaguchi, K. History and development of flavor nucleotides. Food Technol. 18:287-293; 1964. 6. Nejad, M. S. The neural activity of the greater superficial petrosal nerve of the rat in response to chemical stimulation of the palate. Chem. Senses 11:283-293; 1986. 7. Ninomiya, Y,; Funakoshi, M. Qualitative discrimination among "Umami" and the four basic taste substances in mice. In: Kawamura, Y.; Kare, M. R., eds. New York: Dekker; 1987:365-385. 8. Oaldey, B. Altered taste responses from cross-regenerated taste nerves in the rat. In: Hayashi, T., ed. Olfaction and taste II. Oxford:

Pergamon Press; 1967:535-547. 9. Pfaffmann, C.; Fisher, G. L.; Frank, M. The sensory and behavioral factors in taste preference. In: Hayashi, T., ed. Olfaetion and taste II. Oxford: Pergamon Press; 1967:361-381. 10. Sato, M.; Akaike, N.; Yamashita, S. 5'-ribonucleotides as gustatory stimuli in rats. Electrophysiological studies. Jpn. J. Physiol. 15:5370; 1965. 11. Sato, M.; Yamashita, S.; Ogawa, H. Patterns of impulses produced by MSG and 5'-ribonucleotides in taste units of the rat. In: Hayashi, T,, ed. Olfaction and taste II. Oxford: Pergamon Press; 1967:399410. 12. Sato, M.; Yamashita, S.; Ogawa, H. Potentiation of gustatory response to monosodium glutamate in rat chorda tympani fibers by addition of 5'-ribonucleotides Jpn. J. Physiol. 20:444-464; 1970. 13. Torii, K.; Cagan, R, H. Biochemical studies of taste sensation. IX. Enhancement of L-[3H] glutamate binding to bovine taste papillae by 5'-ribonucleotides. Biochem. Biophys. 627:313-323; 1980. 14. Travers, S. P.; Norgren, R. The time course of solitary nucleus gustatory responses: influence of stimulus and site of application. Chem. Senses 14:55-74; 1989. 15. Yamaguchi, S.; Yoshikawa, T.; Ikeda, S.; Ninomiya, Y. Measurement of relative taste intensity of some L-a-amino acids and 5'-nu-

NTS NEURON AND UMAMI TASTE

cleotides. J. Food Sci. 36:846-849; 1971. 16. Yamaguchi, S. Fundamental properties of Umarni in human taste sensation. In: Kawamura, Y.; Kare, M. R., eds. Umami: A basic taste. New York: Dekker; 1987:41-73. 17. Yamamoto, T.; Matsuo, R.; Kiyomitsu, Y.; Kitamura, R. Taste effects of 'umami' substances in hamsters as studied by electrophysio-

941

logical and conditioned taste aversion techniques. Brain Res. 451: 147-162; 1988. 18. Yoshii, K.; Yokouchi, C.; Kurihara, K. Synergistic effects of 5'-nucleotides on rat taste responses to various amino acids. Brain Res. 367:45-51; 1986.