Taste responses of bullfrog to pungent stimuli

BRAIN RESEARCH ELSEVIER

Brain Research 637 (1994)68-72

Research Report

Taste responses of bullfrog to pungent stimuli Kiyonori Yoshii *, Takashi Matui Faculty of Pharmaceutical Sctences, Hokkatdo Unwerstty, Sapporo 060, Japan (Accepted 28 September 1993)

Abstract Taste responses of bullfrogs to various pungent compounds and taste substances were electrophysiologically recorded from glossopharyngeal nerves. The threshold concentrations were ~ 10 7 M for plperlne, ~ 10 -6 M for capsalcin and ~ 10 -4 M for allyl isothlocyanate. At any concentration examined, piperine was more potent than capsaicin. Both piperme and capsalcm elicited desensitizing responses, but the taste receptors recovered from the desensltizat~on w~thin 10 min after washing with deiomzed water. Cross-adaptation experiments revealed that capsaicin only partially desensitizes receptors for piperme, L-leucine, HCi or quinine. Perfusion of the lingual artery with a solution containing no added Ca decreased the responses to capsaicm. Such a solution has been shown to suppress the taste nerve responses by blocking synaptic transmissions between tastc cells and taste nerves [8]. These results suggest that the gustatory effects of capsalcin are different from its pharmacological effects on sensory neurons. It is likely that capsaicin and other pungent compounds, when they act as seasonings, stimulate taste cells rather than the free nerve endings of the sensory neurons.

Key words. Capsaicin; Piperme; Glossopharyngeal nerve; Electrophysiology; Cross-adaptation; Desensitization

1. Introduction Pungent compounds are the most appropriate word to denote 'pepper-flavored' compounds eliciting biting sensory effect [10]. In a broader sense, 'irritants" producing a simple unpleasant pricking pain could also be called pungent agents. Many pungent compounds involved in spices are familiar to us as seasonings. Although their structures have been elucidated [11], their stimulating mechanism as seasonings remain unknown since few electrophysiological experiments have focused on this oral, gustatory aspects of pungent stimuli. In contrast, pharmacological actions of the pungent components, especially those of capsaicin, a principal pungent component of hot peppers of the plant genus Capsicum, have been extensively investigated. An increasing number of studies attempt to construct the following hypothesis for the mechanism of action of capsaicin: capsaicin stimulates a putative receptor on a particular population of sensory neurons with unmyeli-

* Corresponding author. Present address: Department of Biochemical Engineering and Science, Kyushu Institute of Technology, lizuka, Fukuoka 820, Japan

0006-8993/94/$07.00 © 1994 Elsevier Soence B V. All rights reserved

SSDI 0 0 0 6 - 8 9 9 3 ( 9 3 ) E 1 3 6 7 - C

nated C-fiber-type processes, which are polymodal nociceptive detectors with free nerve endings peripherally [2,13]. Hence capsatcin and other pungent compounds added to foods as seasonings have been supposed to stimulate free nerve endings. However, the threshold concentration of capsaicin on human tongues for producing a simple, pleasantly warm sensation was lower than that for producing definite painful burning sensations [14]. It is also unlikely that many spices act only as irritants when added to dishes. Thus it is questionable whether the hypothesis above adequately accounts for the gustatory effects of capsaicin and various other pungent compounds. In the present study, we recorded bullfrog taste responses to several pungent stimuli. The results showed that the desensitizing effect and the order of effectiveness of the pungent stimuli on the bullfrog taste receptors were different from those on the sensory neurons [14,15]. The present results suggest that the pungent substances stimulate by a different mechanism from that involved in capsaicin stimulation of sensory neurons. The possibility that the pungent compounds stimulate taste cells directly is discussed.

K.Yoshn, T. Matut ~Brain Research 637 (1994) 68-72

2. Materials and methods 2.1. Recordings Neural responses of glossopharyngeal nerves of anesthetized adult bullfrogs, Rana catesbetana, were recorded as described m a previous paper [18]. Neural impulses were amplified and then s u m m a t e d with a time constant of 0 3 s. T h e height of the s u m m a t e d response was used as the m a g m t u d e of a neural response To calculate relatwe responses, the magnitude of the responses to test stimuli were dwided by the m e a n m a g m t u d e of the responses to control stimuh applied preceding and following to the apphcation of the test stlmuh. Antldromic neural impulses were also recorded from a single fungfform papdla with a suctton electrode according to Rapuzzl et al. [16]. The amphfied neural impulses were filtered (low frequency cut, 3 Hz and high frequency cut, 3 kHz) and stored m a data recorder We assessed the stabihty of the whole nerve responses by monitoring changes m the magnitude of the responses to control st~muh. In the suction electrode recordings, the n u m b e r and height of the ~mpulses were monitored. W h e n e v e r these factors of responses to the control stimuh were changed more than 20% of the preceding one, the preparation was &scarded

2 2 Perfuston of the hngual artery Perfus~on of the hngual artery w~th an artificial solution was carried out as described by N a g a h a m a et al. [8]. In brief, a polyethylene tube was cannulated into the lingual artery to perfuse the tongue wtth a saline solution (112 m M NaCI, 3.4 m M KC1, 3.6 m M MgSO4, 2.5 m M N a H C O 3, 10 U / m l s o d m m hepa~in, p H 7.2) containing either 1 m M CaCI 2 or no added Ca 2+ at a rate of 0.01 m l / s . The perfused solutions flowed out from the veto at the bottom of the tongue

69

sponses of a s~mllar magnitude to equahze the effect on the common steps though the concentrations may be different for each stimulus. All experiments were carried out at room temperature.

3. Results

3.1. Summated response pattern Fig. 1 shows the summated responses of the glossopharyngeal nerves to the pungent stimuli and other stimuli. The responses to piperine and capsaicin were phasic due to desensitization. The receptors completely recovered from the desensitizing effects in 10 min in the range of the concentrations examined (data not shown). On the other hand, allyl isothiocyanate at concentrations of 2 mM or lower elicited a phasic response followed by a tonic response. HC1 and quinine elicited phasic responses, and L-leucine elicited a phasic response followed by a tonic one.

3.2. Concentratton-response relationships Fig. 2 shows concentration-response relationships for the pungent stimuli. The threshold concentrations for piperine and capsaicin were ~ 10 - 7 M and ~ 1 0 - 6 M, respectively, indicating that the bullfrog is more sensitive to them than to salts [19] or amino acids [18]. The threshold concentration for allyl isothiocyanate

2.3 Sttmulattons Stimulating solutions were applied to bullfrog tongues for 10 to 60 s at intervals of 10 to 30 min. After the stimulation, the tongue was washed with deionized water, followed by application of the next stimulating solution. All chemicals used were analytical grade. Capsaicin and plperme were dissolved m ethanol and then diluted with deiomzed water. Ethanol concentrations in the sUmulating solutions of capsa~cm or p~perme were lower than 5 mM, whereas the threshold concentration of ethanol for producing neural responses is 10 m M [6], mdtcatmg that ethanol m the stimulating solutions &d not ehclt responses Other stimuli were dissolved in delomzed water Allyl isothlocyanate solution was freshly prepared.

1

10 ~M capsaicin

l

5 ~.M piperine

2 mM

Allyl isothiocyanate

2 4 Cross-adaptation Cross adaptation experiments were carried out as described m a previous paper [17]. Although the m e c h a n i s m of the cross-adaptation experiments ~s not completely understood, the basic ~dea for the cross-adaptation experiments ~s: a stimulus A would produce a response when receptor cells are completely adapted and desenslt~zed to stimulus B ff the receptor s~tes for stimulus A and stimulus B are different [1] Even if each stimulus has a different receptor, the stimulus B of higher concentrations would suppress the response to the stimulus A m some extent, since there may be common-steps/nonselective-paths m taste transduct~on [12]. Therefore concentrations at which each stimulus elicits a moderate response have been used in the crossadaptation experiments. In addition, each stimulus must ehclt re-

m

1 mM HCI

I

2 ~tM quinine

m

50 mM L-leucine

m

30 s

Fig. 1. S u m m a t e d responses of bullfrog taste nerves to pungent stlmuh and classical taste stimuli.

70

°20[

K Yo~hu, T Matut / Brain Research 63 7 (1994) 08- 72

3.3. Cross-adaptation experiments

1.s

I

-" .~0s ~C O.

8

10"

10"

10"

10"~4

10 .3

10-2

Concentration (M) Fig. 2 Relative magnitude of typical responses to plperme (closed orcles), capsalcln (open circles), and allyl tsothiocyanate (triangles) as a function of the log of stimulus concentration Peak height of the summated response was taken as the m a g m t u d e of the response Responses plotted were calculated relative to the response to 10 -5 M piperme Curves were drawn by eye.

was ~ 10 - 4 M. The responses to these pungent substances did not reach saturation levels in the range of concentrations examined. The mean responses (_+the standard deviations, n = 4) to 10 -5 M capsaicin and 10 -5 M allyl isothiocyanate were 0.53 _+ 0.13 and 0.00 + 0.00, respectively, where the responses to 10 -5 M piperine were taken as 1.0. Unpaired t-test showed that the differences among them were significant ( P < 0.01).

Fig. 3A shows cross-adapting effects ot piperine and capsaicm on the responses to each other. Capsaicln was applied after the response to piperine (peak a) declined by desensitization to piperine itself. Under these conditions, capsaacin elicited a substantial responses (peak b) though it was smaller than the control (peak c). Piperine, after adaptation with capsaxcin, also elicited a response that was substantial but smaller than the control (compare peak d with peak a). These results indicate that both p~perine and capsaicln only partially desensitize the receptor sites for each other. Fig. 3 also shows that plperine (B and C) and capsaicin (D and E) only partially desensitize the receptor sites for quinine and HCI. These results indicate that the receptor sites for these pungent compounds are not only independent from each other but different from those for quinine or HC1.

3. 4. Responses of a few glossopharyngeal heroes Fig. 4 shows antidromic neural impulses recorded with a suction electrode. The neural impulses elicited can be roughly classified into large and small spikes irrespective of the stimuli eliciting them. Capsaicin,

A

C

a

c

PQ

~p

~PH

Hp

d D

E

m

PC

Cp

30s

Fig 3 Slight cross desensitization of piperme and capsaicin on responses to each other or to quinine and HCI. A second stimulus was apphed after the response to the first stimulus was self-desensitized. Concentrations of stimuh were chosen to elicit responses of similar m a g m t u d e P 5 × 10 - 6 M piperlne; C, 10 -5 M capsalcm, Q, 2 × 10 - 6 M quinine, and H, 10 -3 M HCI

K Yoshn, T Matut/Bram Research 637 (1994) 68-72

71

3.5. Perfusion of lingual artery 10 ~tM capsaicm

5 #M piperme

..... l., J , tll,,;t .L

50 mM L-leucme

ls

Fig. 4 Neural responses to 10-5 M capsalcln, 5 x 10-6 M plperme, 50 mM L-leucme, and 10-5 M quinine recorded from single fungiform papillae with a suction electrode.

piperine and L-leucine elicited both types of spikes and quinine predominantly elicited the small one. The height of the neural impulse depends on parameters such as nerve diameters or conduction velocities of the nerves. Hence these results indicate that the characteristics of the taste nerves transmitting the responses to capsaicin and piperine are similar to the characteristics of the taste nerves transmitting the responses to Lleucine or quinine.

A

B

m

C

I

m

30s Fig 5 D e p e n d e n c e of responses to 10 -5 M capsaicin on interstitial Ca 2÷ concentrations (A) response when the lingual artery was perfused with sahne solution containing 1 m M CaCI 2 (a control solution) (B) response when the lingual artery was perfused with a no added Ca sahne solution, and (C) response during perfuslon with the control solution after the no added Ca saline solution

We investigated the dependence of the responses to capsaicin on interstitial Ca 2÷ concentrations by perfusing the lingual artery with a solution containing no added Ca 2÷. As Fig. 5 shows, the neural response to capsaicin is decreased in the absence of added Ca 2+ (B) and partially recovers in the presence of 1 mM Ca 2+ (C). The mean responses ( + s t a n d a r d deviation, n = 3) were 0.49 + 0.25 (no added Ca 2÷) and 0.70 + 0.26 (1 mM Ca 2÷, after perfusion with no added Ca 2+ solution). The initial, control responses of the tongues was taken as 1.0. Thus, perfusion with the no added Ca 2÷ solution led to a significant decrease in the responses ( P < 0.05, paired one-tail t-test). Although recovery of the response in the normal saline solution was poor, the increase in the normal saline solution was also significant ( P < 0.05, paired one-tail t-test) and there was no significant difference between the recovered response and the initial response.

4. Discussion Since the late N. Jancso demonstrated that sensory nerve endings subserving pain were stimulated and then desensitized by capsaicin [5], a growing number of studies have focused on the stimulating effect of capsaicin and its analogs on sensory neurons [2,14]. Based on these studies, pungent compounds, at least capsaicin, were assumed to stimulate the free nerve endings of the sensory neurons contained in taste nerve bundles rather than taste receptor cells. The present results, however, oppose this hypothesis and suggest that capsaicin stimulates the taste cells and thus elicits taste sensations. In free nerve endings of capsaicin-sensitive sensory neurons, capsaicin is more potent than piperine in inducing hypothermia [14]. Also capsaicin-desensitization, which lasts for hours or days, inhibits various types of piperine-induced pain such as hypothermia or eye-wiping [15]. The present experiments, in contrast, revealed the following results: (i) piperine was more potent in eliciting taste nerve responses than capsaicin at any concentration examined, (ii) the taste nerve responses to capsaicin were reproducible and completely recovered from the self-desensitizing effects in 10 min in the range of the concentrations examined, and (iii) crossadaptation experiments revealed that capsaicin did not desensitize receptors for piperine. These discrepancies suggest that the taste nerve responses of the bullfrog do not result from the activation of free nerve endings of capsaicin-sensitive sensory neurons contained in the taste nerve bundles.

72

K Yo~hu, T Matut / Bram Re~earcti 637 (1994) 68-72

The present experiments show that the perfusion with the no added Ca e+ solution decreases the responses to the pungent stimuh. Since a decrease in Ca concentration on voltage-gated Na channels enhances their excitability [3], the responses to the pungent stimuli might be increased, if free nerve endings transmit the responses. The neural responses of the bullfrog to various taste substances were reported to depend on Ca 2+ in the perfusing interstitial solutions, which suggests that Ca 2 + influx is essential for releasing a neurotransmitter m the synapses between taste cells and glossopharyngeal nerves of bullfrogs [8,9]. Thus, it is likely that pungent substances act as taste stimuli to stimulate taste cells in the bullfrog. The present results showed that the height of the impulses elicited by the pungent stimuh and the taste stimuli were similar to each other. Since the impulse height depends on the properties of nerve fibers whenever the recording conditions are the same, the results mdicate that both substances stimulate the nerves with similar properties. Although further experiments such as the measurement of the impulse height ehcited by free nerve endings are needed, the present results are consistent with our hypothesis. Substance P-like immunoreactive fibers were found to be consistently present in each fungiform papilla of bullfrogs [4]. However, they did not reach the free surface [7]. These findings indicate that the pungent stimuli do not contact the free nerve endings containing substance P and support our proposal that capsalcin stimulates the taste cells directly. Acknowledgment The authors wish to thank Dr K Kunhara (Dept of Pharmaceutmal Scmnces, Hokkaldo Umverslty) Dr K Toyoshlma (Dept of Oral Anatomy, Kyushu Dental College), Dr T Hanamon (Dept of Physiology, Mlyazakl Medical College), Dr T Mlyamoto (Dept. of Physmlogy, Nagasaki Umverslty School of Dentistry) for helpful advice

5. References [1] Beldler, L M., Taste receptor stimulation, J Gen Phvstol, 38 (1962) 107-151 [2] Buck, S.H and Burks, T F., The Neurophamaeology of cap-

~alcin review of some recent observation,, t'harmu¢ol Rel ~,~, (1086) 179 226 [3] Frankenbaeuser, B and Hodgkm, A.L, The action of calcium on the electrical properties of squid axoiis, J Phv~tol, 137 (1957) 218-244 [4] Hlrata, K and Kanaseki, T, Substance P-like lmmunoreactl,~e fibers in the frog taste organs, Evpertentta, 43 (1987) 386-389 [5] Jancso, N, Desensmzatum with capsatcln and related acylamldes as a tool for studying the function of pare receptors, In R K S Lim (Ed), Pharmacology of Patti Proceedings o] Tturd International Congres~ on Pharmacology, Pergamon, Oxford, 1968, pp 33-55 [6] Kashlwagura, T, Kamo, N, Kunhara, K and Kobatake, Y Responses of the frog gustatory receptors to various odorants, Comp Btochem Phystol, 56C (1977) 105-108 [7] Kuramoto, H , An lmmunohlstochemlcal study of cellular and nervous elements m the taste organ of the bullfrog, Rana ~atesbetana, Arch Hlstol Cytol, 51 (1988) 205-221, [8] Nagahama, S, Kobatake, Y and Kunhara, K. Effect of Ca -'+, cychc GMP, and cychc AMP added to artificial solution perfusmg lingual artery on frog gustatory nerve responses, J Gen Phystol, 80 (1982) 785-800. [9] Nagahama, S and Kunbara, K, Noreplnephrme as a possible transmitter involved in synaptlc transmission in lrog taste organs and Ca dependence of its release, J Gen Phystol, 85 (1985) 431-442 [10] Newman, A A , Natural and synthetic pepper-flavored substances (5) Pungency and structure relationships, Chem Product~ (Lond), 17 (1954) 14-18 [11] Newman, A A . , Natural and synthetic pepper-flavored substances (6) Collective list, Chem Products (Lond), 17 (1954) 102-106 [12] Smith, D V and Frank, M, Cross adaptation between salts in the chorda tympanl nerve of the rat, Phystol Behat', 8 (1972) 213-220 [13] Szallasl, A and Blumberg, P M., Reslmferatoxln and its analogs prowde novel insights into pharmacology of the vandlold (capsalcln) receptor, Life Sct, 47 (1990) 1399-1408 [14] Szolcs~nyL J , Capsatcln type pungent agents producing pyrexla In A S Mdton (Ed), Handbook of Eapertmental Pharmacology, Springer, Berlin, 1982, pp 437-478 [15] Szolcsdnyl, J , Capsalcm, Irritation, and desensitization In B G Green, J R. Mason, and M R Kare (Eds.), Chemical Seines, Marcel Dekker, Inc, New York and Basel, 1989, pp 141-168 [16] Rapuzzl, G and Casella, C, Innervatmn of the fungfform papillae m the frog tongue, J Neurophystol, 28 (1965) 154-165 [17] Yoshu, K. Kamo, N, Kunhara, K and Kobatake, Y, Gustatory responses ot eel palatine receptors to amino acids and carboxyhc acids, J Gen Phystol, 74 (1979) 301-317 [18] Yoshu, K, Kobatake, Y and Kunhara, K, Selective enhancement and suppressmn of frog gustatory responses to amino acxds, J Gen Phystol, 77 (198l) 373-385 [19] Yoshll, K, Klyomoto, Y and Kurlhara, K, Taste receptor mechanism of salts in frog and rat, Comp Btochem Physlol, 85A (1986) 501-507