71

Brain Research, 583 (1992) 71-80 © 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00

BRES 17852

Effects of inorganic constituents of saliva on taste responses of the rat chorda tympani nerve Ryuji Matsuo a

a

and Takashi Yamamoto

b

Departmentof Oral Physiology, Faculty of Dentistry, and b Departmentof Behavioral Physiology, Facultyof Human Sciences, Osaka University, Suita, Osaka (Japan) (Accepted 4 February 1992)

Key words: Rat; Saliva; Taste; Chorda tympani; Electrolytes; pH

The effects of saliva on the taste responses of the chorda tympani nerve to the 4 standard chemical stimuli (sucrose, NaC!, HCI, and quinine hydrochloride) and water were investigated in anesthetized rats. When the tongue was adapted to pilocarpine-stimulated whole saliva (pH 8.7), the magnitude of neural response to sucrose was about 2 times that obtained when the tongue was adapted to distilled water. Under saliva-adapted conditions, the magnitude of responses to other taste stimuli was reduced by 10-30%, and the water response appeared. These changes were dependent on the concentration of electrolytes (Na +, K+, C!-, and HCO;) and on the pH of the saliva. When the tongue was adapted to 10-30 mM NaHC0 3 (pH 8.4-8.6), taste and water responses were similar to those under saliva-adapted conditions. Single fiber analyses revealed that the enhancement of the sucrose response after adaptation to NaHC0 3 was produced by an increased overall activity of sucrose-responsive fibers. The correlation coefficients of the magnitude of the taste responses between the 4 taste stimuli remained unchanged, but the water response showed a high correlation to HCI and quinine hydrochloride responses after adaptation. Possible mechanisms for the effects of saliva on taste and water responses were discussed.

INTRODUCfION

The outer surface of mammalian taste receptor cells is normally covered with saliva, suggesting that the salivary environment plays an important role in the initial events of gustatiorrv". Neural taste responses obtained in recent chronic and semi-chronic animal experiments 17,2i ,22,39 distinctly differ from those obtained in acute experiments. One of the candidate parameters responsible for this difference is alteration of the salivary environment. In our previous study 17, we showed that the sweet taste response of the rat chorda tympani nerve was 2-4 times larger, while the NaCI response was slightly smaller, when animals were awake (saliva-adapted condition) than when they were anesthetized (water-adapted condition). We concluded that these effects were mainly due to the inorganic components of saliva. Norgren and co-workers 2i ,22 , in their studies on the neural responses of conscious rats, have also suggested that saliva affects the taste and water responses of the neurons in the nucleus of the solitary tract and parabrachial nucleus. Further, Rehn-

berg et al.28 , using anesthetized hamsters, have demonstrated that salivary ions are able to alter taste responses of the chorda tympani nerve in these animals. The aim of this study was to clarify which constituents of saliva affect taste responses, and how the initial process of gustation is modified by salivary constituents. For this purpose we used anesthetized rats in which the taste responses of the chorda tympani nerve to each of the 4 standard chemical stimuli and water were compared under saliva-adapted and wateradapted conditions. Further, the effects of salivary ionic constituents and pH were analyzed in multi- and single-unit taste nerve responses on the basis of crossadaptation. A preliminary account of this study has previously been reported in abstract form". MATERIALS AND METHODS

Recording from the chorda tympani nerve Male Wistar rats weighing 250-360 g were deeply anesthetized with intraperitoneal injections of a mixture of urethane (700 mgykg)

Correspondence: R. Matsuo, Department of Oral Physiology, Faculty of Dentistry, Osaka University, 1-8 Yamadaoka, Suita, Osaka 565, Japan.

72 and a-chloralose (70 mgykg), The trachea was cannulated, and the hypoglossal nerve was cut bilaterally to prevent possible tongue movement. The left chorda tympani nerve was dissected free from the point where it joined the lingual nerve to its exit from the bulla, where it was cut. Por recording multi-unit discharges from the whole nerve, the nerve was desheathed and lifted onto a silver wire recording electrode (100 ,urn in diameter) with a chloridized Ag/AgC! indifferent electrode attached to nearby tissues. The potentials were amplified with a pre-amplifier (WPI, DAM-SA), displayed on an oscilloscope (Tektronix 5110), and stored on analog tape. Por measuring the magnitude of response discharges, the stored data were fed into a digitally controlled summator ", and the output of the summator was supplied to a mini-computer system (San-ei Co., Signal processor ?T18). Por each test stimulus, we measured the spontaneous discharges for 30 s just before stimulation, and the discharges for 30 s after stimulation commenced. The magnitude of each response was then obtained by subtracting the spontaneous activity from the activity during stimulation. Por recording functional single fiber discharges, the nerve was dissected into small fiber bundles. The recording equipment was identical to that used for the whole nerve recordings, except for the sumrnator, The stability of the recording was judged by the amplitude and wave form of action potentials displayed on an oscilloscope. The stored data of functional single units were later photographed, and discharge rates were counted with a rate/interval monitor through a spike amplitude discriminator (Dia Medical System Co., Spike counter DSE-325P).

saliva, dialysed saliva, or salt solutions); and final control (the tongue was adapted to water again). Each test stimulus was applied to the anterior tongue for 30 s (0.5 mljs by gravity flow), followed by rinsing with distilled water (in control), and saliva, dialysed saliva, or adapting solutions (in the adaptation test) for 30 s. When the adaptation test was performed with saliva or dialysed saliva (described below), the tongue was coated with saliva or dialysed saliva for 20 min. Taste stimuli dissolved in distilled water were then applied. The final control was obtained after a lO-min water rinse. These adaptation and rinsing times were based on those in our previous study 17. When the adaptation test was performed with salt solutions, the tongue was adapted to a salt solution for 5 min. The final control was obtained after a 5-min water rinse. These adaptation and rinsing times were based on those in our pilot study, which showed that the effects of adaptation became stable after 5-min adaptation and disappeared after a 5-min water rinse. In this session, a cross-adaptation test was performed to prevent contamination of the water response; test stimuli were made up with the same adapting solution, except for quinine hydrochloride (for reasons outlined below). During the recording session, the stability of each multi-unit preparation was judged from the control response to 0.1 M NaC!. A recording was considered to be stable when the magnitude of the 0.1 M NaCI response of the initial and final controls deviated by < 10%. Only data from stable recordings were used for data analysis. Toste stimuli and adapting solutions The taste stimuli were: 0.1 M NaC!, 0.5 M sucrose, 0.01 M HCl, and 0.02 M quinine hydrochloride. They were dissolved in distilled water or in either one of the following adapting solutions: 1-100 mM; NaCI, KCl, NaHC0 3 , and KHC0 3 , 1-5 mM; CaCl z, MgCl z , NaP, NaI, and NaH ZP04 , pH 3-11; HCl or NaOH. Quinine hydrochloride was not dissolved at pH > 7.4, or with > 30 mM NaHC0 3 and KHC0 3 . When the effects of adaptation to 30 and 100 mM

Application of taste stimuli The tongue was gently protruded with a hook, and rinsed with distilled water for 30 min to prevent the effects of saliva'". Then the following recording session was carried out: initial control (the tongue was adapted to water); adapting test (the tongue was adapted to

..

A

, \

-."'--.

...."'-.oJ
-

'Sucrose'

,

,

.~

'Sucrose'

.~

. ~-.;...v.:'}' "

• HGI

,f'.".... .~v':J."'Iy',(,,~,

~

'I/o.

:!'or............

....\\..: «.

'

\

.

.'\'v.,\J."~,,~,,.:, -

.-.10, t. I .,......",., ~-"'.,;~.-J-..,.#o;J--..-J"'.\.'~"

• Water '

- -AQuinine '

J otIt'~rA\:

r ;-

"--""-' .......J,).'V,;J.~r'-...~.-

• Quinine'

• Water ' 30 s

Fig. 1. Effects of saliva on whole nerve chorda tympani responses to 0.1 M NaC!, 0.5 M sucrose, 0.01 M HCI, 0.02 M quinine hydrochloride, and water. Recordings A, B, C, and D were obtained successively. A: initial control responses obtained after the tongue was rinsed with water for 30 min; B: response to pilocarpine-stimulated whole salvia; C: responses recorded after the tongue was adapted to saliva for 20 min; D: final control response after a lO-min water rinse. Each recording was drawn by a pen-recorder after processing with a digitally controlled summator, using a 0.5-s integrating interval. The upward and downward arrowheads indicate application of test stimuli, and rinses with water (A and D) and saliva (C), respectively.

73 bicarbonate salts were examined, quinine was dissolved in 10 mM bicarbonate salts. All the taste stimuli and the adapting and rinsing solutions were freshly prepared with distilled water and reagent grade chemicals, and were kept at room temperature (21-24°C).

z c:: '" c:: u0 0

c.

100

Saliva ad3P1

60

CIl

20

.~

Cii

Qj

II:

0

(II

e

150

Q. UI

(II

a:

100

(II

•

I

0*---6* 1* r.>: t I~· &:::::==::8 A-t~ • 1 1 aT T

50

(II

100

Q. UI (II

a:

1~

-1

-

~

·~~~l *!?* * *

-3

_ _~ _ _

-*

*

1ii

0 -3

-1

100

TT TT 1 tec:::::::::=::~~T~ 11 Tl_:l~TT* 1

T

*t~ *

50

-2

Quinine

•

.~

II:

-2

+~~=---=::e-.2________ * .trT*

(II

a;

~

Hel

0

w

0 -3

e

~"";--io~ll-· ~ =--==~*

50

0

UI

100

1

t

a

- NaHCOa adapt. o NaCladapt. • KHCOa adapt. A KCI adapt.

~%=---+-.~ ~Yl: 1: l~T:

1ii

a:

H

1* T* TL---,~'*

.~

a;

s

NaCI

a

0

N

Fig. 2. Effects of saliva and dialysed saliva on relative magnitude of multi-unit responses to 0.1 M NaCl (N), 0.5 M sucrose (S), 0.01 M HCI (H), 0.02 M quinine hydrochloride (Q), and water (W). To estimate response magnitude, the total amount of the digitally summated O.5-s response was measured for 30 s after the commencement of test stimulation, and the amount of spontaneous discharge before each stimulus was subtracted. The magnitude of the control response to NaCI, under distilled water adaptation conditions, was taken as standard (100%). Open columns, control responses obtained after water rinse; solid columns, responses after the tongue was adapted to whole saliva for 20 min; dotted columns, responses after adaptation to dialysed saliva. Each column, except for the control response to NaCI, shows mean±S.D. (n = 8). • "P < 0.01 (Wilcoxon signed ranksum test).

Sucrose

UI

adapt.

40

'"

Effects of saliva Taste responses to the 4 standard basic taste solutions and water were analyzed in the presence and

~

Olalysed saliv a

80

a> II:

RESULTS

200

ConUol

0

-: .t... : : -=t* i*

-2 Log Molar Cone. of Adapting Sol.

-1

50

*

~~T*

11~-T* lA* I

0

H

0

-3

-2

-1

Log Molar Cone. of Adapting Sol.

Fig. 3. Effects of adaptation to various concentrations of NaHC0 3 (solid circles), KHC0 3 (solid triangles), NaCl (open circles), and KCI (open triangles) on the magnitude of multi-unit responses to 0.5 M sucrose, 0.1 M NaCl, 0.01 M HCI, and 0.02 M quinine hydrochloride. The abscissa indicates the concentration of the adapting solutions. The ordinate indicates the relative magnitude of response, when the control response (after water adaptation) to each taste stimulus was taken as standard (100%). Each taste stimulus, except for quinine hydrochloride, contained the same concentration of the salt in the adapting solution. When the effect of > 30 mM NaHC0 3 and KHC0 3 on the quinine hydrochloride response was examined, the solutions of quinine hydrochloride contained 10 mM NaHC0 3 or KHC0 3 • Each plotted point shows mean relative response magnitude ± S.D. tn = 8). "P < 0.05, "P < 0.01 (Wilcoxon signed rank-sum test).

74 absence of saliva on the tongue surface. Fig. 1 shows sample records of the multi-unit activity of the chorda tympani nerve. Initial control responses were recorded 30 min after saliva was rinsed away from the tongue with distilled water (Fig. l A), When collected saliva was applied to the water-adapted tongue, a small phasic response was generated; this returned to the background level within a few min (Fig. l.B). Fig. lC shows taste and water responses recorded 20 min after the tongue was adapted to saliva. Under these conditions, sucrose and water responses increased, while NaCl, HCI, and quinine hydrochloride responses decreased. These salivary effects began appearing from 2 min after adaptation, were stable after 15-20 min adaptation, and lasted for several min after rinsing the tongue with water had commenced. As shown in Fig. ID, the effects of saliva disappeared after a lO-min water rinse. Fig. 2 shows the mean magnitudes of taste responses of the chorda tympani nerve obtained when the tongue was adapted to distilled water, whole saliva, or dialysed saliva. The control NaCl response under water adaptation was taken as standard (100%). After whole saliva adaptation, the mean magnitude of the sucrose response increased from 30% to 67% (P < 0.01, Wilcoxon signed rank-sum test), and a water response appeared. In contrast, NaCl, HCI, and quinine responses were reduced, .by 13%, 28%, and 22%, respectively (P < 0.01). When the tongue was adapted with the dialysed saliva which contained no inorganic constituents, the taste responses were essentially unchanged, suggesting that it is the inorganic constituents of saliva which are responsible for alterations of the magnitude of taste responses. Effects of inorganic constituents of saliva The principal ionic constituents of rat saliva are; Na+ (5-100 mEq/O, K+ (10-55 mEq/I), Cl- (10-100 mEq/O, and HC0 3 (5-50 mEq/O. In addition, the saliva also contains small amounts of Ca2+, Mg 2+, F-, 1-, and H2POi « 5 mEq/031. Therefore, adaptation of the tongue was performed with 1-100 mM NaCl, KCI, NaHC0 3, and KHC0 3, and with 1-5 mM CaCI2, MgCl2, NaF, NaI, and NaH 2P04 • Fig. 3 shows the effects of 5 min adaptation to NaHC0 3, KHC0 3, NaCl, and KCI on the magnitude of multi-unit taste responses, when the magnitude of responses under water adaptation to each of the 4 taste stimuli was taken as standard (100%). The magnitude of the sucrose response was significantly increased by adaptation to NaHC0 3 (at all concentrations, P < 0.01, Wilcoxon signed rank-sum test), KHC0 3 (at all concentrations, P < 0.01), and NaCl (at 10-30 mM, P < 0.05). This implies that HC0 3 and Na + increase the

..... 11-

~

0

..o zIII ~e t:" ot: 0 c. III o CD

a:

Water

50 0

~:::=--8

-====

CD

III

a:

-100

T

.-------1:,

l>---l> 0 _________ J.

1

2~~

• NaHC03 adapt. o NaCI adapt. ... KHC03 adapt. l> KCI adapt.

-50

> ;::

'i

... ...T --------==ii :::::::===.___________.1 ~ •

0

~

i

-3

-2

-1

Log Molar Cone. of Adapting Sol.

Fig. 4. Effects of adaptation to various concentrations of NaHC0 3 (solid circles), KHC0 3 (solid triangles), NaCl (open circles), and KCl (open triangles) on the magnitude of multi-unit responses to distilled water. The magnitude of response was expressed as a percentage of the control 0.1 M NaCl response. Each plotted point shows mean relative response magnitude ± S.D. (n = 8).

sucrose response; the maximal response was obtained by 10-30 mM NaHC0 3. The NaCl response was decreased by adaptation to NaCI (at 30 mM, P < 0.05; at 100 mM, P < 0.01), NaHC0 3 (at 30 mM, P < 0.05; at 100 mM, P < 0.01), and KCl (at 100 mM, P < 0.01), but not by adaptation to KHC0 3 • This implies that Na + and cr independently decrease the NaCl response. The HCl response was strongly decreased by NaHC0 3 (at > 10 mM, P < 0.01) and KHC0 3 (at > 10 mM, P < 0.01), and slightly decreased by NaCl (at 30 mM, P < 0.05; at 100 mM, P < 0.01) and KCI (at 100 mM, P < 0.01), suggesting that the depressant effect on HCl response is caused mainly by HC0 3. The quinine response was decreased by all adapting solutions; NaHC0 3 (at 100 mM, P < 0.01), KHC0 3 (at 100 mM, P < 0.01), NaCl (at 30 mM, P < 0.05; at 100 mM, P < 0.01), and KCI (at 30 mM, P < 0.05; at 100 mM, P < 0.01). All ions tested related to the decrease of quinine response. Fig. 4 shows the magnitude of taste nerve responses to distilled water when the tongue was adapted to various concentrations (1-100 mM) of NaCl, KCI, NaHC0 3 and KHC0 3; the control NaCl response was taken as standard (100%). After adaptation to 10-100 mM NaCl or KCI, the application of water reduced background neural activity. A water response was produced after adaptation to 3-100 mM NaHC0 3 or KHC0 3. This effect was more prominent in KHC0 3adapted than in NaHC0 3-adapted conditions at 10-100 mM, but the difference was not statistically significant. These effects of adaptation to salts disappeared within 5 min after the tongue was readapted with water. There were virtually no effects on taste and water responses with adapting solutions of 1-5 mM CaCl 2, MgCI2, NaF, NaI, and NaH 2P04 •

75

Effects of pH It is well known that bicarbonate contributes significantly to the buffer capacity of saliva. In this experiment, the pH of pilocarpine-stimulated whole saliva and dialysed saliva at room temperature were 8.1-9.2 (8.7 ± 0.2; mean ± S.D., n = 9) and 6.1-6.5 (6.3 ± 0.1, n = 7), respectively. The pH of the bicarbonate adapting solutions (NaHC0 3 and KHC0 3 ) were about 7.5 (I mM), 8.0 (3 mM), 8.4 (10 mM), 8.6 (30 mM), and 8.8 (l00 mM). The pH of the other adapting solutions were in the range of 6.1 to 6.5. To examine the effects of pH on taste nerve responses, the pH of the adapting, rinsing, and taste solutions were adjusted to levels ranging from 3 to 11 by adding HCI or NaOH. The responses to HCI and quinine hydrochloride were not tested in this session because of the acidity of 0.01 M HCI (pH 2.2) and the low solubility of 0.02 M quinine hydrochloride at pH above 7.4. As shown in Fig. 5, the magnitude of multiunit sucrose responses as compared with control responses at pH 6.2-6.5, was about 60% at pH 3-5, and about 150% at pH around 9.5. The NaCI response was stable at a pH range of 3 to 10. No obvious water response was observed at a pH range of 3 to 11. However, when KCI or NaCI was added to adapting solutions with a pH of more than about 8, a water response appeared; for example, about 10% of control NaCI response was produced by adaptation to 30 mM KCI or NaCl solution with a pH of 8.7-9.1.

--Control 0----- After adaptation

A i' .....

200

Sucrose

. . . .... .)-..

150

/

Gl

CIl

e

o

~.

Q

:

lr

. /-

100

Gl

lr

.

. .

~

i

"-~' -...

.~

",,'

~:.

50

3

B

4

5

6

7

8

9 10 11

NaCI

i'

';; 100 CIl

e o Q

CIl Gl

lr

50

Gl .~

1ii

a; lr

,