- Email: [email protected]
Time-dependent expression of hypertonic effects on bullfrog taste nerve responses to salts and bitter substances
brain research 1556 (2014) 1–9
Available online at www.sciencedirect.com
www.elsevier.com/locate/brainres
Research Report
Time-dependent expression of hypertonic effects on bullfrog taste nerve responses to salts and bitter substances Kazunori Mashiyamaa, Yuhei Nozawab, Yoshitaka Ohtuboa, Takashi Kumazawac, Kiyonori Yoshiia,n a
Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology, Hibikino 2-4, Kitakyushu-shi 808-0196, Japan b Department of Life Science and Green Chemistry, Saitama Institute of Technology, Fukaya 369-0293, Japan c Graduate School of Engineering, Saitama Institute of Technology, Fukaya 369-0293, Japan
art i cle i nfo
ab st rac t
Article history:
We previously showed that the hypertonicity of taste stimulating solutions modified tonic
Accepted 4 February 2014
responses, the quasi-steady state component following the transient (phasic) component
Available online 8 February 2014
of each integrated taste nerve response. Here we show that the hypertonicity opens tight
Keywords:
junctions surrounding taste receptor cells in a time-dependent manner and modifies
Diffusion potentials
whole taste nerve responses in bullfrogs. We increased the tonicity of stimulating
Tight junctions
solutions with non-taste substances such as urea or ethylene glycol. The hypertonicity
Local circuit currents
enhanced phasic responses to NaCl40.2 M, and suppressed those to NaClo0.1 M, 1 mM
Glossopharyngeal nerves
CaCl2, and 1 mM bitter substances (quinine, denatonium and strychnine). The hypertonicity also enhanced the phasic responses to a variety of 0.5 M salts such as LiCl and KCl. The enhancing effect was increased by increasing the difference between the ionic mobilities of the cations and anions in the salt. A preincubation time 420 s in the presence of 1 M nontaste substances was needed to elicit both the enhancing and suppressing effects. Lucifer Yellow CH, a paracellular marker dye, diffused into bullfrog taste receptor organs in 30 s in the presence of hypertonicity. These results agreed with our proposed mechanism of hypertonic effects that considered the diffusion potential across open tight junctions. & 2014 Elsevier B.V. All rights reserved.
1.
Introduction
In scientific research, experimental conditions are normally changed one by one to examine the effect of each condition, but an exception occurs in investigating taste responses. The n
Corresponding author. Fax: þ81 93 695 6014. E-mail address: [email protected] (K. Yoshii). URL: http://www.brain.kyutech.ac.jp/ yoshii (K. Yoshii).
http://dx.doi.org/10.1016/j.brainres.2014.02.006 0006-8993 & 2014 Elsevier B.V. All rights reserved.
addition of taste substances to stimulating solutions simultaneously increases the concentration of taste substances and the tonicity of stimulating solutions. The increase of the tonicity is negligible whenever the concentration of taste substances is low (Caprio, 1975; Yoshii et al., 1979). However,
2
brain research 1556 (2014) 1–9
several taste substances such as salts and sugars are tested at high concentrations, which substantially increases the tonicity of the stimulating solutions (Kamo et al., 1978; Kumazawa and Kurihara, 1990; Ogawa et al., 1968; Yoshii et al., 1981, 1986). Each taste nerve response of bullfrogs consists of transient (phasic) and persistent (tonic) responses. Their generation mechanisms are complicated: Single taste receptor sites generate both responses (Kamo et al., 1980; Kashiwagura et al., 1980), but other taste receptor sites selectively generate either response (Yoshii et al., 1981). We previously showed that hypertonicity modified bullfrog tonic responses to NaCl and CaCl2, and proposed the following mechanism for the hypertonic effects (Beppu et al., 2012). Hypertonicity opens the tight junctions surrounding taste receptor cells, and produces a diffusion potential across the tight junctions that generates local circuit currents (Fig. 1). When the diffusion potential makes the inside of the tongue negative with respect to the outside, the local circuit current flows outwardly through the basolateral membrane of the taste receptor cells, which depolarizes the basolateral membrane, increases neurotransmitter release, and enhances taste nerve responses. If positive, the diffusion potential reverses the current and suppresses taste nerve responses (not shown). This mechanism predicts the following issues that we had not shown in the previous study (Beppu et al., 2012). The hypertonic effects modify not only tonic responses but also phasic responses to any taste substance in the same way, irrespective of their generation mechanisms. The hypertonicity opens tight junctions. Here, we have confirmed these predictions and shown the reliability of the proposed mechanism of
Fig. 1 – Proposed mechanism of hypertonic effects (Beppu et al., 2012). An equivalent electrical model of taste receptor cells (TRC), supporting cells (SC), and surrounding tight junctions (T). The opening of tight junctions under hypertonic solutions decreases the electric resistance of tight junctions (Rt), which increases local circuit currents generated by the diffusion potential (Et). The decrease in receptor membrane resistance (Rr) in response to salts also increases the local circuit current. An increased outward flow of the local circuit current depolarizes the basolateral membrane (Rb) of taste receptor cells, potentiates neurotransmitter release from the taste receptor cells, and enhances the taste nerve response. Inward currents reverse the effect (not shown). In the absence of stimulating salts, the receptor membrane resistance (Rr) of taste receptor cells remains high, which prevents an increase in local circuit currents.
hypertonic effects on taste responses. Part of the present results are published elsewhere in an abstract (Kumazawa et al., 2012; Mashiyama et al., 2007).
2.
Results
2.1.
Responses to NaCl
We investigated the hypertonic effects by increasing the tonicity of preincubating and stimulating solutions with 1 M urea, ethylenglycol, glycerol, and sorbitol. All of these chemicals elicited no responses, when they were applied alone in the 5 mM HEPES solution (Fig. 2a). Each integrated taste nerve response to 0.5 M NaCl consisted of a phasic and a following tonic response. Addition of one of the non-taste substances to stimulating solutions had no effects on the shape and magnitude of phasic responses, but it transformed the tonic responses into a bell shape and enhanced the magnitude as we described previously (Beppu et al., 2012). However, stimulation following a 30 s preincubation with 1 M urea significantly enhanced the magnitude of the phasic responses (Fig. 2a and b). The enhancing effect reached a maximum at 30 s preincubation, and decreased with increasing preincubation time. Also, a 30 s preincubation with 1 M ethylene glycol, glycerol, or sorbitol significantly enhanced the phasic responses, but no enhancing effect occurred without preincubation (Fig. 2c). In contrast to phasic responses, preincubation was not required to enhance tonic response magnitude (Fig. 2a). However, the stimulation itself seemed to act as a preincubation, since tonic responses were generated by exposing taste receptors to hypertonic stimulating solutions for a moment. Therefore, we investigated the tonic response magnitude every 10 s during stimulation. The tonic response magnitude reached a maximum 30 s after the onset of stimulation or preincubation, and then decreased (Fig. 2d), similar to the time-dependency of phasic response magnitude (Fig. 2b). In the previous study, we predicted an enhancing effect for NaCl4119 mM and a suppressing effect for NaClo119 mM (Beppu et al., 2012), because the diffusion potential calculated with Goldman's equation (see Section 3) changes it polarity and hence the direction of the local circuit current at 119 mM NaCl (see Fig. 8). However, we only found the enhancing effect, because the threshold NaCl concentration for tonic responses was 100 mM and NaClo119 mM elicited negligible tonic responses (Beppu et al., 2012). The present study showed that the threshold NaCl concentration for the phasic response waso10 mM (Fig. 3). We took advantage of phasic responses to investigate both the enhancing and the suppressing effect on them. As the proposed model had predicted, the phasic response to NaClr0.1 M was significantly suppressed and that to NaClZ0.2 M significantly enhanced in the presence of a 30 s preincubation with 1 M urea.
2.2.
Responses to other monovalent salts
A 30 s preincubation with 1 M urea significantly enhanced the magnitude of phasic responses to 0.5 M LiCl and NaBr as well as to NaCl, but did not enhance phasic responses to 0.5 M KCl
brain research 1556 (2014) 1–9
3
Fig. 2 – Preincubation effect on responses to NaCl. (a) Integrated taste nerve responses to 1 M non-taste substances (upper row), and to 0.5 M NaCl in the presence and absence of 1 M urea (middle row) or 1 M glycerol (lower row). Each row was obtained from the same frog. We adapted the tongue to 5 mM HEPES (pH 7.0), applied one of 1 M non-taste substances (open bars), and stimulated with 0.5 M NaCl (closed bars). Overlaps of bars show the application of the mixture of NaCl and one of the non-taste substances in this and following figures. All of these solutions were applied to the tongue through the same outlet. Note that the pair of preincubating and stimulating solutions was prepared with the same 1 M non-taste substances in this and following figures. (b) Relative magnitude of phasic responses to 0.5 M NaCl in the presence of 1 M urea as a function of preincubation time with 1 M urea. npo0.05, ANOVA followed by Scheffe's multiple comparison test. A significant difference appeared between the preincubation time of 0 and 30 s. (c) Relative magnitude of phasic responses to 0.5 M NaCl in the presence of 1 M non-taste substances as indicated without (open bars) and with a 30 s preincubation (closed bars). EG, ethylene glycol. npo0.05; two-tailed ttest. (d) Relative magnitude of tonic responses to 0.5 M NaCl measured every 10 s and plotted as a function of elapsed time after the onset of preincubation. Plotted data in (b) (c) and (d) are means and S.D. calculated relative to the magnitude of phasic responses to 0.5 M NaCl in the absence of non-taste substances and preincubation. Numerals by each data point are the number of bullfrogs examined in this and following figures.
or KBr (Fig. 4a and b). The ratio of enhancement, the magnitude of phasic responses in the presence of urea divided by that in the absence of urea, depended on the difference between the absolute value of the ionic mobility of ions contained in respective salts (Fig. 4c).
2.3.
Response to CaCl2
The addition of 1 M urea to the 1 mM CaCl2 stimulating solution did not suppressed the phasic response magnitude in the absence of preincubation with 1 M urea (two-tailed t-test, p40.05, n ¼3). The stimulation following a preincubation with 1 M urea suppressed both phasic and tonic response
magnitudes (Fig. 5a). A preincubation time Z 30 s significantly suppressed the phasic response magnitude (Fig. 5b), and the extent of suppression increased with increasing preincubation time. A 30 s preincubation in the presence of other 1 M non-taste substances significantly suppressed the phasic responses to 1 mM CaCl2, though no suppressing effect occurred in the absence of preincubation (Fig. 5c).
2.4.
Responses to bitter substances
Bitter substances, 1 mM quinine, denatonium, and strychnine, elicited only phasic responses (Fig. 6a). A 30 s preincubation in the presence of 1 M urea suppressed phasic
4
brain research 1556 (2014) 1–9
Fig. 3 – Enhancing and suppressing effects on responses to NaCl. (a) Integrated taste nerve responses to NaCl (closed bars) in the absence and presence of 1 M urea (open bars). (b) Relative magnitude of phasic responses to a series of NaCl solutions in the absence (open circles) and presence of 1 M urea (closed circles). Plotted data are means and S.D. calculated in the same way as shown in Fig. 2b–d. The hypertonic effects are statistically examined between 50 mM and 200 mM NaCl. npo0.05; nnpo0.01; two-tailed t-test.
responses to 1 mM quinine. Preincubation times Z20 s significantly suppressed the phasic responses (Fig. 6b), and the extent of suppression increased with increasing preincubation time. The 30 s preincubation in the presence of other non-taste substances significantly suppressed the phasic responses to 1 mM quinine, though no effects occurred on the phasic responses without preincubation (Fig. 6c). The 30 s preincubation with 1 M urea significantly suppressed phasic responses to the other bitter substances, though no effects occurred without preincubation (Fig. 6d).
2.5.
Fig. 4 – Responses to various monovalent salts. (a) Integrated taste nerve responses to 0.5 M salts (closed bars) in the absence (left traces of each pair) and presence of 1 M urea (open bars). (b) Relative magnitude of phasic responses to 0.5 M salts without (open bars) and with a 30 s preincubation with 1 M urea (closed bars). npo0.05; nnpo0.01; two-tailed ttest. (c) Relative magnitude of phasic responses to 0.5 M salts with a 30 s preincubation as a function of difference between the absolute value of the ionic mobility of respective cations and anions. Plotted data in (b) and (c) are means and S.D. calculated relative to the magnitude of phasic responses to respective 0.5 M salts in the absence of non-taste substances and preincubation.
Diffusion of Lucifer Yellow CH
We had assumed that hypertonicity opened tight junctions surrounding taste receptor cells (Beppu et al., 2012). In the present study, we examined this potential permeation by diffusing Lucifer Yellow CH into taste disks. Lucifer Yellow CH
has been used as a paracellular marker dye (Hashimoto et al., 2003; Krouwer et al., 2012; Segawa, 1994). The taste disks were labeled in 30 s when 0.1 mM Lucifer Yellow CH dissolved in a hypertonic solution with 1.4 M urea was applied to the tongue (Fig. 7a). We measured the fluorescent intensity of each pixel
brain research 1556 (2014) 1–9
5
subsequently decreased with increasing depth (Fig. 7b). In the absence of the hypertonicity, no diffusion occurred in the 30 s period. The other non-taste substances examined at 1.4 M, as well as urea, significantly increased the labeling ratio at 20 μm below the tongue surface in 30 s (Fig. 7c).
3.
Fig. 5 – Preincubation effect on responses to CaCl2. (a) Integrated taste nerve responses to 1 mM CaCl2 (closed bars) in the absence and presence of 1 M urea (open bars). (b) Relative magnitude of phasic responses to 1 mM CaCl2 in the presence of 1 M urea as a function of preincubation time with 1 M urea. nnpo0.01, ANOVA followed by Scheffe's multiple comparison test. We only show significant differences with respect to 0 s preincubation. (c) Relative magnitude of phasic responses to 1 mM CaCl2 in the presence of 1 M non-taste substances as indicated without (open bars) and with a 30 s preincubation (closed bars). n po0.05; nnpo0.01; two-tailed t-test. Plotted data in (b) and (c) are means and S.D. calculated relative to the magnitude of phasic responses to 1 mM CaCl2 in the absence of nontaste substances and preincubation.
along the centerline of the taste disk (a dotted line in Fig. 7b), averaged the pixel value over 80 μm to 0 μm, and divided each pixel value by the average. The ratio thus obtained increased with increasing distance from the surface of taste disks, peaked at 20 μm below the tongue surface, and
Discussion
In the previous study, we had shown that hypertonicity modulated the magnitude of the tonic response of bullfrog taste nerve responses to salts, and that the hypertonic effects depended on the diffusion potential across the tight junctions surrounding taste receptor cells (Fig. 1). In the present study, we showed that hypertonicity opened the tight junctions in a time-dependent manner, and modulated both phasic and tonic responses in a similar time-dependence. The present study showed significant lag times of 30 s for both the enhancing (Fig. 2b) and the suppressing effects (Fig. 5b and 6b). These results suggest that the same mechanism underlies both effects. Another present result showed that hypertonicity caused the diffusion of Lucifer Yellow CH into taste disks in 30 s (Fig. 7). The diffusion demonstrated the hypertonicity-dependent opening of tight junctions surrounding taste receptor cells. This is the first evidence of their opening in taste disks, though tight junctions open in the presence of hypertonicity in frog skin (Erlij and MartinezPalomo, 1972; Ussing, 1965, 1966). This opening confirmed our proposed mechanism for the hypertonic effects (Beppu et al., 2012), which we describe in the following paragraphs. The preincubation of tongues with hypertonic solutions enhanced the phasic responses to NaCl40.2 M and suppressed those to NaClo0.1 M (Fig. 3). Preincubation also suppressed the phasic responses to 1 mM CaCl2 and to 1 mM bitter substances (Figs. 5 and 6). We tested whether the mechanism we proposed explained these results. The mechanism claimed that these hypertonic effects depend on the direction of the local circuit current: local circuit current that flows outwardly through the basolateral membrane of taste receptor cells depolarizes the basolateral membrane, and inward currents hyperpolarize the basolateral membrane. It is likely that the depolarization enhances taste nerve responses and the hyperpolarization suppresses them. The direction of the local circuit current depends on the polarity of the diffusion potential that drives the local circuit current. Therefore, we investigated the polarity with Goldman's equation, I ¼ ∑Ii ¼
C Ci:o expððzi F=RTÞðφin φout ÞÞ FSðφin φout Þ ∑zi ui i;i : expððzi F=RTÞðφin φout ÞÞ 1 d
Each parameter is as follows: I, total currents (net currents); Ii, a current magnitude carried by the ion; F, Faraday constant; S, the cross-sectional area where ions diffuse; φin, the potential inside tight junctions; φout, the potential outside tight junctions; d, the distance of diffusion; zi, the valence of the ion; ui, ionic mobility of the ion; ci,i and ci,o, the concentrations of ions inside and outside tight junctions, respectively; R, the gas constant; T, the absolute temperature. Note that we take Δφ( ¼ φin φout) as the magnitude of a diffusion potential.
6
brain research 1556 (2014) 1–9
Fig. 6 – Preincubation effect on responses to bitter substances. (a) Integrated taste nerve responses to 1 mM quinine (closed bars) in the absence or presence of 1 M urea (open bars). (b) Relative magnitude of phasic responses to 1 mM quinine in the presence of 1 M urea as a function of preincubation time with 1 M urea. npo0.05; nnpo0.01, ANOVA followed by Scheffe's multiple comparison test. We only show significant differences with respect to 0 s preincubation. (c) Relative magnitude of phasic responses to 1 mM quinine in the presence of 1 M non-taste substances as indicated without (open bars) and with a 30 s preincubation (closed bars). npo0.05; nnpo0.01, two-tailed t-test. (d) Relative magnitude of phasic responses to 1 mM quinine, denatonium and strychnine in the presence of 1 M urea without (open bars) and with a 30 s preincubation (closed bars). nnpo0.01, two-tailed t-test. Plotted data in (b), (c), and (d) are means and S.D. calculated relative to the magnitude of phasic responses to 1 mM quinine (b) or respective 1 mM bitter substances ((c) and (d)) in the absence of non-taste substances and preincubation.
We substituted these parameters as follow; null for I because of the absence of any external current sources; 298 K for T; 115 mM NaCl, 2.0 mM KCl, and 1.6 mM CaCl2 (principal ions in the physiological saline of frogs) for ci,i; and the concentration of stimulating salts for ci,o. The ionic mobility is assumed to be equal to that in water at 298 K except that for quinine, which is 0.88 10 4 cm2/s/V in frog skin (Mudry et al., 2007). The concentration of stimulating solutions where the diffusion potential thus calculated changed the polarity was; 119 mM NaCl, 37 mM CaCl2, and 44 mM quinine (Fig. 8). Since the solubility of quinine is 52.6 mg/100 mL, 1.6 mM (Budavari, 1996), it is impossible to examine the enhancing effect on responses to quinine. These results clearly explained the hypertonic effects on the responses to NaCl, CaCl2 and quinine. We could not calculate the concentration of denatonium and strychnine, because the ionic mobility of these bitter substances was unknown. However, the concentration seems to be similar to that for quinine, because they are monovalent
cations and are similar to quinine in hydrophobicity and molecular mass. Therefore, we assume that the concentration is higher than 1 mM, and conclude that the mechanism is applicable to these bitter substances as well. The enhancing effect on NaCl decreased with increasing preincubation time, peaked at 30 s after the onset of preincubation, and decreased with preincubation times 430 s. The decrease seemed to result from the mixing of hypertonic solutions and body fluids in interstitial spaces, which decreased the diffusion potential in magnitude. Also the mixing increased the osmolarity of interstitial fluid, which removed water from taste receptor cells, increased their intracellular Kþ concentration, and hyperpolarized these taste receptor cells. Since rat taste receptor cells express aquaporins on the basolateral membrane of the cell (Watson et al., 2007), bullfrogs may express similar channels, which could account for the transcellular water movement proposed above. Although the mixing antagonized the suppression effect by decreasing the diffusion potential, the
brain research 1556 (2014) 1–9
7
Fig. 7 – Hypertonicity-dependent diffusion of Lucifer Yellow CH into taste disks. (a) A taste disk in a fungiform papilla (left), its fluorescent image (middle), and their overlay (right). Scale bars, 50 μm. Lucifer Yellow CH was applied to this taste disk in the presence of 1.4 M urea (see Section 4). (b) Relative pixel values along a dotted line (inset) as a function of distance from the tongue surface. Negative distances, outside tongues. We exposed taste disks to Lucifer Yellow CH in the absence (open circles) and presence of 1.4 M urea (closed circles), measured the fluorescent intensity of each pixel on the dotted line, averaged them over 80 μm to 0 μm, divided each pixel value by the average to obtain relative pixel values, and plotted the mean and S.D. of relative pixel values. Each pixel size was 0.66 μm 0.66 μm. (c) Relative pixel values at 20 μm apart from the tongue surface in the presence of respective non-taste substances as indicated. Control, relative pixel values in the absence of non-taste substances. nnpo0.01, ANOVA followed by Scheffe's multiple comparison test. We only show significant differences with respect to control.
8
brain research 1556 (2014) 1–9
the council of the Physiological Society of Japan, and were permitted by the Animal Institutional Review Board of Saitama Institute of Technology in accordance with the guidelines of the U.S. National Institutes of Health.
4.1.
Fig. 8 – Calculated diffusion potentials across tight junctions. Diffusion potentials were calculated with Goldman's equation and plotted as a function of the concentration of each taste substance in stimulating solutions. The diffusion potential was null in the presence of 119 mM NaCl, 37 mM CaCl2 or 44 mM quinine in respective stimulating solutions, and changes its polarity on either side. Broken line, the diffusion potential of quinine at concentrations higher than its solubility of 1.6 mM.
mixing-induced hyperpolarization accelerated the suppression. Therefore it is likely that taste nerve responses to any substance at any concentration are suppressed under steady state hypertonic conditions, though taste nerve responses are enhanced or suppressed under transient hypertonic conditions as the present results showed. We have shown the hypertonic effects in terms of diffusion potentials across tight junctions with an equivalent electrical circuit in this (Fig. 1) and a previous study (Beppu et al., 2012). In rodents, the role of diffusion potentials was assumed in term of transephthelial potentials in investigating anion effects on taste nerve responses to salts (Breza and Contreras, 2012; Rehnberg et al., 1993; Ye et al., 1991, 1993, 1994). Although the authors did not investigate associated hypertonic effects, they elucidated the role of transepithelial potentials in greater detail. Also one of these studies used an equivalent electrical circuit similar to ours (Ye et al., 1994). Hypertonicity enhanced the taste nerve responses of bullfrogs to NaCl4 100 mM. Chorda tympani nerve fibers of mammals generate substantial responses to sugar and non-sugar stimulants. Recordings from single non-sugar fibers may be useful in investigating the hypertonic effects in mammals. If a similar effect could enhance the salt taste in human beings, it would greatly improve the quality of life for people who require low Naþ diets.
4.
Experimental procedures
All experiments were carried out in Saitama Institute of Technology. All experimental protocols were conducted in compliance with the Guiding Principles for the Care and Use of Animals in the Field of Physiological Sciences approved by
Neural responses
The taste responses of glossopharyngeal nerves were recorded from anesthetized adult bullfrogs, Rana catesbeiana, as described previously (Beppu et al., 2012). In brief, we anesthetized bullfrogs of 200–300 g with intraperitoneal injection of urethane (0.2 g/100 g body weight), freed the glossopharyngeal nerve from surrounding tissues, placed it on a pair of Ag/AgCl electrodes, and covered the nerve with liquid paraffin. A small amount of urethane was added to maintain anesthetization. These animals were sacrificed by decapitation under anesthetization after the experiments. The neural activity of whole glossopharyngeal nerve thus recorded was amplified, band-pass filtered (150–3000 Hz), integrated with a time constant of 0.3 s, and displayed on a pen-recorder. The stimulus response magnitude was normalized with respect to the magnitude of pre-stimulus neural activity and expressed as relative response magnitude (Beppu et al., 2012). Whenever the magnitude of responses to the same control changed by more than 20% of the magnitude of the preceding one, we discarded the preparation.
4.2.
Stimulation
We adapted frog tongues to a HEPES solution (5 mM HEPES dissolved in deionized water, pH 7.0), applied preincubating solutions for different periods, and applied stimulating solutions. We added the same concentration of the same nontaste substances to the preincubating and stimulating solutions. The flow rate of all solutions was 100 mL/min. The stimulation interval was 410 min. Although water elicited substantial taste nerve responses in the absence of salts (Miyake et al., 1976; Zotterman, 1949), 5 mM HEPES completely suppressed taste nerve responses to them (Beppu et al., 2012). All stimulating and preincubating solutions were freshly prepared with the HEPES solution.
4.3.
Diffusion of Lucifer Yellow CH
We investigated the diffusion of Lucifer Yellow CH into taste disks, taste receptor organs of bullfrogs, in the presence and absence of hypertonicity. We washed the tongue of anesthetized bullfrogs for 30 min with a physiological saline (115 mM NaCl, 2.0 mM KCl, and 2.5 mM CaCl2, 2.0 mM MgCl2, 5 mM HEPES, pH 7.4), applied 0.1 mM Lucifer Yellow CH dissolved in the HEPES solution supplemented with or without 1.4 M urea, ethylene glycol, glycerol, or sorbitol for 30 s, and removed the tongue. The bullfrog was killed by decapitation under the anesthetization. We fixed the tongue with 15% formalin for 2–3 days, sectioned it at 60 μm (LinearSlicer Pro 7, Dosaka EM, Kyoto, Japan), and viewed the distribution of Lucifer Yellow CH in taste disks with a fluorescent microscope (BX60, Olympus, Tokyo) with a 10 objective. Images were recorded with a digital CCD camera and processed with Aquacosmos
brain research 1556 (2014) 1–9
software (ver. 2.5, Hamamatsu Photonics K.K., Hamamatsu, Japan).
Acknowledgments This work was partially supported by the “Nano Project” for Private Universities: 2011–2015, matching fund subsidy from Ministry of Education, Culture, Sports, Science and Technology.
r e f e r e n c e s
Beppu, N., Higure, Y., Mashiyama, K., Ohtubo, Y., Kumazawa, T., Yoshii, K., 2012. Hypertonicity augments bullfrog taste nerve responses to inorganic salts. Pflugers Arch. 463, 845–851. Breza, J.M., Contreras, R.J., 2012. Anion size modulates salt taste in rats. J. Neurophysiol. 107, 1632–1648. Budavari, S., 1996. The Merck Index. Merck Research Laboratories Division of Merck & Co., Inc., Whitehouse Station, N. J.. Caprio, J., 1975. High sensitivity of catfish taste receptors to amino acids. Comp. Biochem. Physiol. A: Mol. Integr. Physiol. 52, 247–251. Erlij, D., Martinez-Palomo, A., 1972. Opening of tight junctions in frog skin by hypertonic urea solutions. J. Membr. Biol 9, 229–240. Hashimoto, S., Murakami, M., Kanaseki, T., Kobayashi, S., Matsuki, M., Shimono, M., Segawa, A., 2003. Morphological and functional changes in cell junctions during secretory stimulation in the perfused rat submandibular gland. Eur. J. Morphol. 41, 35–39. Kamo, N., Kashiwagura, T., Kobatake, Y., Kurihara, K., 1978. Role of membrane-bound calcium in taste reception of the frog. J. Physiol. 282, 115–129. Kamo, N., Kashiwagura, T., Kurihara, K., Kobatake, Y., 1980. A theory of dynamic and steady responses in chemoreception. J. Theor. Biol. 83, 111–130. Kashiwagura, T., Kamo, N., Kurihara, K., Kobatake, Y., 1980. Interpretation by theoretical model of dynamic and steady components in frog gustatory response. Am. J. Physiol. 238, G445–G452. Krouwer, V.J., Hekking, L.H., Langelaar-Makkinje, M., ReganKlapisz, E., Post, J.A., 2012. Endothelial cell senescence is associated with disrupted cell–cell junctions and increased monolayer permeability. Vasc. Cell 4, 12. Kumazawa, T., Kurihara, K., 1990. Large enhancement of canine taste responses to sugars by salts. J. Gen. Physiol 95, 1007–1018. Kumazawa, T., Mashiyama, K., Ohtubo, Y., Yoshii, K., 2012. Bullfrog taste nerve responses under hyperosmotic conditions. In: Eighth Forum of Neuroscience. vol., ed. Barcelona, Spain, Abstract. no. 4173.
9
Mashiyama, K., Bunya, H., Higure, Y., Beppu, N., Yoshii, K., Kumazawa, T., 2007. Effects of osmotic pressure on the phasic component of bullfrog taste responses. Chem. Senses 32 (JASTS Abstract. J3). Miyake, M., Kamo, N., Kurihara, K., Kobatake, Y., 1976. Physicochemical studies of taste reception. III. Interpretation of the water response in taste reception. Biochim. Biophys. Acta 436, 843–855. Mudry, B., Carrupt, P.A., Guy, R.H., Delgado-Charro, M.B., 2007. Quantitative structure–permeation relationship for iontophoretic transport across the skin. J. Controlled Release 122, 165–172. Ogawa, H., Sato, M., Yamashita, S., 1968. Multiple sensitivity of chordat typani fibres of the rat and hamster to gustatory and thermal stimuli. J. Physiol 199, 223–240. Rehnberg, B.G., MacKinnon, B.I., Hettinger, T.P., Frank, M.E., 1993. Anion modulation of taste responses in sodium-sensitive neurons of the hamster chorda tympani nerve. J. Gen. Physiol. 101, 453–465. Segawa, A., 1994. Tight junctional permeability in living cells: dynamic changes directly visualized by confocal laser microscopy. J. Electron. Microsc. (Tokyo) 43, 290–298. Ussing, H.H., 1965. Relationship between osmotic reactions and active sodium transport in the frog skin epithelium. Acta Physiol. Scand 63, 141–155. Ussing, H.H., 1966. Anomalous transport of electrolytes and sucrose through the isolated frog skin induced by hypertonicity of the outside bathing solution. Ann. N. Y. Acad. Sci. 137, 543–555. Watson, K.J., Kim, I., Baquero, A.F., Burks, C.A., Liu, L., Gilbertson, T.A., 2007. Expression of aquaporin water channels in rat taste buds. Chem. Senses 32, 411–421. Ye, Q., Heck, G.L., DeSimone, J.A., 1991. The anion paradox in sodium taste reception: resolution by voltage-clamp studies. Science 254, 724–726. Ye, Q., Heck, G.L., DeSimone, J.A., 1993. Voltage dependence of the rat chorda tympani response to Naþ salts: implications for the functional organization of taste receptor cells. J. Neurophysiol. 70, 167–178. Ye, Q., Heck, G.L., DeSimone, J.A., 1994. Effects of voltage perturbation of the lingual receptive field on chorda tympani responses to Naþ and Kþ salts in the rat: implications for gustatory transduction. J. Gen. Physiol 104, 885–907. Yoshii, K., Kamo, N., Kurihara, K., Kobatake, Y., 1979. Gustatory responses of eel palatine receptors to amino acids and carboxylic acids. J. Gen. Physiol. 74, 301–317. Yoshii, K., Kobatake, Y., Kurihara, K., 1981. Selective enhancement and suppression of frog gustatory responses to amino acids. J. Gen. Physiol. 77, 373–385. Yoshii, K., Yokouchi, C., Kurihara, K., 1986. Synergistic effects of 50 -nucleotides on rat taste responses to various amino acids. Brain Res 367, 45–51. Zotterman, Y., 1949. The response of the frog’s taste fibres to the application of pure water. Acta Physiol. Scand 18, 181–189.
Available online at www.sciencedirect.com
www.elsevier.com/locate/brainres
Research Report
Time-dependent expression of hypertonic effects on bullfrog taste nerve responses to salts and bitter substances Kazunori Mashiyamaa, Yuhei Nozawab, Yoshitaka Ohtuboa, Takashi Kumazawac, Kiyonori Yoshiia,n a
Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology, Hibikino 2-4, Kitakyushu-shi 808-0196, Japan b Department of Life Science and Green Chemistry, Saitama Institute of Technology, Fukaya 369-0293, Japan c Graduate School of Engineering, Saitama Institute of Technology, Fukaya 369-0293, Japan
art i cle i nfo
ab st rac t
Article history:
We previously showed that the hypertonicity of taste stimulating solutions modified tonic
Accepted 4 February 2014
responses, the quasi-steady state component following the transient (phasic) component
Available online 8 February 2014
of each integrated taste nerve response. Here we show that the hypertonicity opens tight
Keywords:
junctions surrounding taste receptor cells in a time-dependent manner and modifies
Diffusion potentials
whole taste nerve responses in bullfrogs. We increased the tonicity of stimulating
Tight junctions
solutions with non-taste substances such as urea or ethylene glycol. The hypertonicity
Local circuit currents
enhanced phasic responses to NaCl40.2 M, and suppressed those to NaClo0.1 M, 1 mM
Glossopharyngeal nerves
CaCl2, and 1 mM bitter substances (quinine, denatonium and strychnine). The hypertonicity also enhanced the phasic responses to a variety of 0.5 M salts such as LiCl and KCl. The enhancing effect was increased by increasing the difference between the ionic mobilities of the cations and anions in the salt. A preincubation time 420 s in the presence of 1 M nontaste substances was needed to elicit both the enhancing and suppressing effects. Lucifer Yellow CH, a paracellular marker dye, diffused into bullfrog taste receptor organs in 30 s in the presence of hypertonicity. These results agreed with our proposed mechanism of hypertonic effects that considered the diffusion potential across open tight junctions. & 2014 Elsevier B.V. All rights reserved.
1.
Introduction
In scientific research, experimental conditions are normally changed one by one to examine the effect of each condition, but an exception occurs in investigating taste responses. The n
Corresponding author. Fax: þ81 93 695 6014. E-mail address: [email protected] (K. Yoshii). URL: http://www.brain.kyutech.ac.jp/ yoshii (K. Yoshii).
http://dx.doi.org/10.1016/j.brainres.2014.02.006 0006-8993 & 2014 Elsevier B.V. All rights reserved.
addition of taste substances to stimulating solutions simultaneously increases the concentration of taste substances and the tonicity of stimulating solutions. The increase of the tonicity is negligible whenever the concentration of taste substances is low (Caprio, 1975; Yoshii et al., 1979). However,
2
brain research 1556 (2014) 1–9
several taste substances such as salts and sugars are tested at high concentrations, which substantially increases the tonicity of the stimulating solutions (Kamo et al., 1978; Kumazawa and Kurihara, 1990; Ogawa et al., 1968; Yoshii et al., 1981, 1986). Each taste nerve response of bullfrogs consists of transient (phasic) and persistent (tonic) responses. Their generation mechanisms are complicated: Single taste receptor sites generate both responses (Kamo et al., 1980; Kashiwagura et al., 1980), but other taste receptor sites selectively generate either response (Yoshii et al., 1981). We previously showed that hypertonicity modified bullfrog tonic responses to NaCl and CaCl2, and proposed the following mechanism for the hypertonic effects (Beppu et al., 2012). Hypertonicity opens the tight junctions surrounding taste receptor cells, and produces a diffusion potential across the tight junctions that generates local circuit currents (Fig. 1). When the diffusion potential makes the inside of the tongue negative with respect to the outside, the local circuit current flows outwardly through the basolateral membrane of the taste receptor cells, which depolarizes the basolateral membrane, increases neurotransmitter release, and enhances taste nerve responses. If positive, the diffusion potential reverses the current and suppresses taste nerve responses (not shown). This mechanism predicts the following issues that we had not shown in the previous study (Beppu et al., 2012). The hypertonic effects modify not only tonic responses but also phasic responses to any taste substance in the same way, irrespective of their generation mechanisms. The hypertonicity opens tight junctions. Here, we have confirmed these predictions and shown the reliability of the proposed mechanism of
Fig. 1 – Proposed mechanism of hypertonic effects (Beppu et al., 2012). An equivalent electrical model of taste receptor cells (TRC), supporting cells (SC), and surrounding tight junctions (T). The opening of tight junctions under hypertonic solutions decreases the electric resistance of tight junctions (Rt), which increases local circuit currents generated by the diffusion potential (Et). The decrease in receptor membrane resistance (Rr) in response to salts also increases the local circuit current. An increased outward flow of the local circuit current depolarizes the basolateral membrane (Rb) of taste receptor cells, potentiates neurotransmitter release from the taste receptor cells, and enhances the taste nerve response. Inward currents reverse the effect (not shown). In the absence of stimulating salts, the receptor membrane resistance (Rr) of taste receptor cells remains high, which prevents an increase in local circuit currents.
hypertonic effects on taste responses. Part of the present results are published elsewhere in an abstract (Kumazawa et al., 2012; Mashiyama et al., 2007).
2.
Results
2.1.
Responses to NaCl
We investigated the hypertonic effects by increasing the tonicity of preincubating and stimulating solutions with 1 M urea, ethylenglycol, glycerol, and sorbitol. All of these chemicals elicited no responses, when they were applied alone in the 5 mM HEPES solution (Fig. 2a). Each integrated taste nerve response to 0.5 M NaCl consisted of a phasic and a following tonic response. Addition of one of the non-taste substances to stimulating solutions had no effects on the shape and magnitude of phasic responses, but it transformed the tonic responses into a bell shape and enhanced the magnitude as we described previously (Beppu et al., 2012). However, stimulation following a 30 s preincubation with 1 M urea significantly enhanced the magnitude of the phasic responses (Fig. 2a and b). The enhancing effect reached a maximum at 30 s preincubation, and decreased with increasing preincubation time. Also, a 30 s preincubation with 1 M ethylene glycol, glycerol, or sorbitol significantly enhanced the phasic responses, but no enhancing effect occurred without preincubation (Fig. 2c). In contrast to phasic responses, preincubation was not required to enhance tonic response magnitude (Fig. 2a). However, the stimulation itself seemed to act as a preincubation, since tonic responses were generated by exposing taste receptors to hypertonic stimulating solutions for a moment. Therefore, we investigated the tonic response magnitude every 10 s during stimulation. The tonic response magnitude reached a maximum 30 s after the onset of stimulation or preincubation, and then decreased (Fig. 2d), similar to the time-dependency of phasic response magnitude (Fig. 2b). In the previous study, we predicted an enhancing effect for NaCl4119 mM and a suppressing effect for NaClo119 mM (Beppu et al., 2012), because the diffusion potential calculated with Goldman's equation (see Section 3) changes it polarity and hence the direction of the local circuit current at 119 mM NaCl (see Fig. 8). However, we only found the enhancing effect, because the threshold NaCl concentration for tonic responses was 100 mM and NaClo119 mM elicited negligible tonic responses (Beppu et al., 2012). The present study showed that the threshold NaCl concentration for the phasic response waso10 mM (Fig. 3). We took advantage of phasic responses to investigate both the enhancing and the suppressing effect on them. As the proposed model had predicted, the phasic response to NaClr0.1 M was significantly suppressed and that to NaClZ0.2 M significantly enhanced in the presence of a 30 s preincubation with 1 M urea.
2.2.
Responses to other monovalent salts
A 30 s preincubation with 1 M urea significantly enhanced the magnitude of phasic responses to 0.5 M LiCl and NaBr as well as to NaCl, but did not enhance phasic responses to 0.5 M KCl
brain research 1556 (2014) 1–9
3
Fig. 2 – Preincubation effect on responses to NaCl. (a) Integrated taste nerve responses to 1 M non-taste substances (upper row), and to 0.5 M NaCl in the presence and absence of 1 M urea (middle row) or 1 M glycerol (lower row). Each row was obtained from the same frog. We adapted the tongue to 5 mM HEPES (pH 7.0), applied one of 1 M non-taste substances (open bars), and stimulated with 0.5 M NaCl (closed bars). Overlaps of bars show the application of the mixture of NaCl and one of the non-taste substances in this and following figures. All of these solutions were applied to the tongue through the same outlet. Note that the pair of preincubating and stimulating solutions was prepared with the same 1 M non-taste substances in this and following figures. (b) Relative magnitude of phasic responses to 0.5 M NaCl in the presence of 1 M urea as a function of preincubation time with 1 M urea. npo0.05, ANOVA followed by Scheffe's multiple comparison test. A significant difference appeared between the preincubation time of 0 and 30 s. (c) Relative magnitude of phasic responses to 0.5 M NaCl in the presence of 1 M non-taste substances as indicated without (open bars) and with a 30 s preincubation (closed bars). EG, ethylene glycol. npo0.05; two-tailed ttest. (d) Relative magnitude of tonic responses to 0.5 M NaCl measured every 10 s and plotted as a function of elapsed time after the onset of preincubation. Plotted data in (b) (c) and (d) are means and S.D. calculated relative to the magnitude of phasic responses to 0.5 M NaCl in the absence of non-taste substances and preincubation. Numerals by each data point are the number of bullfrogs examined in this and following figures.
or KBr (Fig. 4a and b). The ratio of enhancement, the magnitude of phasic responses in the presence of urea divided by that in the absence of urea, depended on the difference between the absolute value of the ionic mobility of ions contained in respective salts (Fig. 4c).
2.3.
Response to CaCl2
The addition of 1 M urea to the 1 mM CaCl2 stimulating solution did not suppressed the phasic response magnitude in the absence of preincubation with 1 M urea (two-tailed t-test, p40.05, n ¼3). The stimulation following a preincubation with 1 M urea suppressed both phasic and tonic response
magnitudes (Fig. 5a). A preincubation time Z 30 s significantly suppressed the phasic response magnitude (Fig. 5b), and the extent of suppression increased with increasing preincubation time. A 30 s preincubation in the presence of other 1 M non-taste substances significantly suppressed the phasic responses to 1 mM CaCl2, though no suppressing effect occurred in the absence of preincubation (Fig. 5c).
2.4.
Responses to bitter substances
Bitter substances, 1 mM quinine, denatonium, and strychnine, elicited only phasic responses (Fig. 6a). A 30 s preincubation in the presence of 1 M urea suppressed phasic
4
brain research 1556 (2014) 1–9
Fig. 3 – Enhancing and suppressing effects on responses to NaCl. (a) Integrated taste nerve responses to NaCl (closed bars) in the absence and presence of 1 M urea (open bars). (b) Relative magnitude of phasic responses to a series of NaCl solutions in the absence (open circles) and presence of 1 M urea (closed circles). Plotted data are means and S.D. calculated in the same way as shown in Fig. 2b–d. The hypertonic effects are statistically examined between 50 mM and 200 mM NaCl. npo0.05; nnpo0.01; two-tailed t-test.
responses to 1 mM quinine. Preincubation times Z20 s significantly suppressed the phasic responses (Fig. 6b), and the extent of suppression increased with increasing preincubation time. The 30 s preincubation in the presence of other non-taste substances significantly suppressed the phasic responses to 1 mM quinine, though no effects occurred on the phasic responses without preincubation (Fig. 6c). The 30 s preincubation with 1 M urea significantly suppressed phasic responses to the other bitter substances, though no effects occurred without preincubation (Fig. 6d).
2.5.
Fig. 4 – Responses to various monovalent salts. (a) Integrated taste nerve responses to 0.5 M salts (closed bars) in the absence (left traces of each pair) and presence of 1 M urea (open bars). (b) Relative magnitude of phasic responses to 0.5 M salts without (open bars) and with a 30 s preincubation with 1 M urea (closed bars). npo0.05; nnpo0.01; two-tailed ttest. (c) Relative magnitude of phasic responses to 0.5 M salts with a 30 s preincubation as a function of difference between the absolute value of the ionic mobility of respective cations and anions. Plotted data in (b) and (c) are means and S.D. calculated relative to the magnitude of phasic responses to respective 0.5 M salts in the absence of non-taste substances and preincubation.
Diffusion of Lucifer Yellow CH
We had assumed that hypertonicity opened tight junctions surrounding taste receptor cells (Beppu et al., 2012). In the present study, we examined this potential permeation by diffusing Lucifer Yellow CH into taste disks. Lucifer Yellow CH
has been used as a paracellular marker dye (Hashimoto et al., 2003; Krouwer et al., 2012; Segawa, 1994). The taste disks were labeled in 30 s when 0.1 mM Lucifer Yellow CH dissolved in a hypertonic solution with 1.4 M urea was applied to the tongue (Fig. 7a). We measured the fluorescent intensity of each pixel
brain research 1556 (2014) 1–9
5
subsequently decreased with increasing depth (Fig. 7b). In the absence of the hypertonicity, no diffusion occurred in the 30 s period. The other non-taste substances examined at 1.4 M, as well as urea, significantly increased the labeling ratio at 20 μm below the tongue surface in 30 s (Fig. 7c).
3.
Fig. 5 – Preincubation effect on responses to CaCl2. (a) Integrated taste nerve responses to 1 mM CaCl2 (closed bars) in the absence and presence of 1 M urea (open bars). (b) Relative magnitude of phasic responses to 1 mM CaCl2 in the presence of 1 M urea as a function of preincubation time with 1 M urea. nnpo0.01, ANOVA followed by Scheffe's multiple comparison test. We only show significant differences with respect to 0 s preincubation. (c) Relative magnitude of phasic responses to 1 mM CaCl2 in the presence of 1 M non-taste substances as indicated without (open bars) and with a 30 s preincubation (closed bars). n po0.05; nnpo0.01; two-tailed t-test. Plotted data in (b) and (c) are means and S.D. calculated relative to the magnitude of phasic responses to 1 mM CaCl2 in the absence of nontaste substances and preincubation.
along the centerline of the taste disk (a dotted line in Fig. 7b), averaged the pixel value over 80 μm to 0 μm, and divided each pixel value by the average. The ratio thus obtained increased with increasing distance from the surface of taste disks, peaked at 20 μm below the tongue surface, and
Discussion
In the previous study, we had shown that hypertonicity modulated the magnitude of the tonic response of bullfrog taste nerve responses to salts, and that the hypertonic effects depended on the diffusion potential across the tight junctions surrounding taste receptor cells (Fig. 1). In the present study, we showed that hypertonicity opened the tight junctions in a time-dependent manner, and modulated both phasic and tonic responses in a similar time-dependence. The present study showed significant lag times of 30 s for both the enhancing (Fig. 2b) and the suppressing effects (Fig. 5b and 6b). These results suggest that the same mechanism underlies both effects. Another present result showed that hypertonicity caused the diffusion of Lucifer Yellow CH into taste disks in 30 s (Fig. 7). The diffusion demonstrated the hypertonicity-dependent opening of tight junctions surrounding taste receptor cells. This is the first evidence of their opening in taste disks, though tight junctions open in the presence of hypertonicity in frog skin (Erlij and MartinezPalomo, 1972; Ussing, 1965, 1966). This opening confirmed our proposed mechanism for the hypertonic effects (Beppu et al., 2012), which we describe in the following paragraphs. The preincubation of tongues with hypertonic solutions enhanced the phasic responses to NaCl40.2 M and suppressed those to NaClo0.1 M (Fig. 3). Preincubation also suppressed the phasic responses to 1 mM CaCl2 and to 1 mM bitter substances (Figs. 5 and 6). We tested whether the mechanism we proposed explained these results. The mechanism claimed that these hypertonic effects depend on the direction of the local circuit current: local circuit current that flows outwardly through the basolateral membrane of taste receptor cells depolarizes the basolateral membrane, and inward currents hyperpolarize the basolateral membrane. It is likely that the depolarization enhances taste nerve responses and the hyperpolarization suppresses them. The direction of the local circuit current depends on the polarity of the diffusion potential that drives the local circuit current. Therefore, we investigated the polarity with Goldman's equation, I ¼ ∑Ii ¼
C Ci:o expððzi F=RTÞðφin φout ÞÞ FSðφin φout Þ ∑zi ui i;i : expððzi F=RTÞðφin φout ÞÞ 1 d
Each parameter is as follows: I, total currents (net currents); Ii, a current magnitude carried by the ion; F, Faraday constant; S, the cross-sectional area where ions diffuse; φin, the potential inside tight junctions; φout, the potential outside tight junctions; d, the distance of diffusion; zi, the valence of the ion; ui, ionic mobility of the ion; ci,i and ci,o, the concentrations of ions inside and outside tight junctions, respectively; R, the gas constant; T, the absolute temperature. Note that we take Δφ( ¼ φin φout) as the magnitude of a diffusion potential.
6
brain research 1556 (2014) 1–9
Fig. 6 – Preincubation effect on responses to bitter substances. (a) Integrated taste nerve responses to 1 mM quinine (closed bars) in the absence or presence of 1 M urea (open bars). (b) Relative magnitude of phasic responses to 1 mM quinine in the presence of 1 M urea as a function of preincubation time with 1 M urea. npo0.05; nnpo0.01, ANOVA followed by Scheffe's multiple comparison test. We only show significant differences with respect to 0 s preincubation. (c) Relative magnitude of phasic responses to 1 mM quinine in the presence of 1 M non-taste substances as indicated without (open bars) and with a 30 s preincubation (closed bars). npo0.05; nnpo0.01, two-tailed t-test. (d) Relative magnitude of phasic responses to 1 mM quinine, denatonium and strychnine in the presence of 1 M urea without (open bars) and with a 30 s preincubation (closed bars). nnpo0.01, two-tailed t-test. Plotted data in (b), (c), and (d) are means and S.D. calculated relative to the magnitude of phasic responses to 1 mM quinine (b) or respective 1 mM bitter substances ((c) and (d)) in the absence of non-taste substances and preincubation.
We substituted these parameters as follow; null for I because of the absence of any external current sources; 298 K for T; 115 mM NaCl, 2.0 mM KCl, and 1.6 mM CaCl2 (principal ions in the physiological saline of frogs) for ci,i; and the concentration of stimulating salts for ci,o. The ionic mobility is assumed to be equal to that in water at 298 K except that for quinine, which is 0.88 10 4 cm2/s/V in frog skin (Mudry et al., 2007). The concentration of stimulating solutions where the diffusion potential thus calculated changed the polarity was; 119 mM NaCl, 37 mM CaCl2, and 44 mM quinine (Fig. 8). Since the solubility of quinine is 52.6 mg/100 mL, 1.6 mM (Budavari, 1996), it is impossible to examine the enhancing effect on responses to quinine. These results clearly explained the hypertonic effects on the responses to NaCl, CaCl2 and quinine. We could not calculate the concentration of denatonium and strychnine, because the ionic mobility of these bitter substances was unknown. However, the concentration seems to be similar to that for quinine, because they are monovalent
cations and are similar to quinine in hydrophobicity and molecular mass. Therefore, we assume that the concentration is higher than 1 mM, and conclude that the mechanism is applicable to these bitter substances as well. The enhancing effect on NaCl decreased with increasing preincubation time, peaked at 30 s after the onset of preincubation, and decreased with preincubation times 430 s. The decrease seemed to result from the mixing of hypertonic solutions and body fluids in interstitial spaces, which decreased the diffusion potential in magnitude. Also the mixing increased the osmolarity of interstitial fluid, which removed water from taste receptor cells, increased their intracellular Kþ concentration, and hyperpolarized these taste receptor cells. Since rat taste receptor cells express aquaporins on the basolateral membrane of the cell (Watson et al., 2007), bullfrogs may express similar channels, which could account for the transcellular water movement proposed above. Although the mixing antagonized the suppression effect by decreasing the diffusion potential, the
brain research 1556 (2014) 1–9
7
Fig. 7 – Hypertonicity-dependent diffusion of Lucifer Yellow CH into taste disks. (a) A taste disk in a fungiform papilla (left), its fluorescent image (middle), and their overlay (right). Scale bars, 50 μm. Lucifer Yellow CH was applied to this taste disk in the presence of 1.4 M urea (see Section 4). (b) Relative pixel values along a dotted line (inset) as a function of distance from the tongue surface. Negative distances, outside tongues. We exposed taste disks to Lucifer Yellow CH in the absence (open circles) and presence of 1.4 M urea (closed circles), measured the fluorescent intensity of each pixel on the dotted line, averaged them over 80 μm to 0 μm, divided each pixel value by the average to obtain relative pixel values, and plotted the mean and S.D. of relative pixel values. Each pixel size was 0.66 μm 0.66 μm. (c) Relative pixel values at 20 μm apart from the tongue surface in the presence of respective non-taste substances as indicated. Control, relative pixel values in the absence of non-taste substances. nnpo0.01, ANOVA followed by Scheffe's multiple comparison test. We only show significant differences with respect to control.
8
brain research 1556 (2014) 1–9
the council of the Physiological Society of Japan, and were permitted by the Animal Institutional Review Board of Saitama Institute of Technology in accordance with the guidelines of the U.S. National Institutes of Health.
4.1.
Fig. 8 – Calculated diffusion potentials across tight junctions. Diffusion potentials were calculated with Goldman's equation and plotted as a function of the concentration of each taste substance in stimulating solutions. The diffusion potential was null in the presence of 119 mM NaCl, 37 mM CaCl2 or 44 mM quinine in respective stimulating solutions, and changes its polarity on either side. Broken line, the diffusion potential of quinine at concentrations higher than its solubility of 1.6 mM.
mixing-induced hyperpolarization accelerated the suppression. Therefore it is likely that taste nerve responses to any substance at any concentration are suppressed under steady state hypertonic conditions, though taste nerve responses are enhanced or suppressed under transient hypertonic conditions as the present results showed. We have shown the hypertonic effects in terms of diffusion potentials across tight junctions with an equivalent electrical circuit in this (Fig. 1) and a previous study (Beppu et al., 2012). In rodents, the role of diffusion potentials was assumed in term of transephthelial potentials in investigating anion effects on taste nerve responses to salts (Breza and Contreras, 2012; Rehnberg et al., 1993; Ye et al., 1991, 1993, 1994). Although the authors did not investigate associated hypertonic effects, they elucidated the role of transepithelial potentials in greater detail. Also one of these studies used an equivalent electrical circuit similar to ours (Ye et al., 1994). Hypertonicity enhanced the taste nerve responses of bullfrogs to NaCl4 100 mM. Chorda tympani nerve fibers of mammals generate substantial responses to sugar and non-sugar stimulants. Recordings from single non-sugar fibers may be useful in investigating the hypertonic effects in mammals. If a similar effect could enhance the salt taste in human beings, it would greatly improve the quality of life for people who require low Naþ diets.
4.
Experimental procedures
All experiments were carried out in Saitama Institute of Technology. All experimental protocols were conducted in compliance with the Guiding Principles for the Care and Use of Animals in the Field of Physiological Sciences approved by
Neural responses
The taste responses of glossopharyngeal nerves were recorded from anesthetized adult bullfrogs, Rana catesbeiana, as described previously (Beppu et al., 2012). In brief, we anesthetized bullfrogs of 200–300 g with intraperitoneal injection of urethane (0.2 g/100 g body weight), freed the glossopharyngeal nerve from surrounding tissues, placed it on a pair of Ag/AgCl electrodes, and covered the nerve with liquid paraffin. A small amount of urethane was added to maintain anesthetization. These animals were sacrificed by decapitation under anesthetization after the experiments. The neural activity of whole glossopharyngeal nerve thus recorded was amplified, band-pass filtered (150–3000 Hz), integrated with a time constant of 0.3 s, and displayed on a pen-recorder. The stimulus response magnitude was normalized with respect to the magnitude of pre-stimulus neural activity and expressed as relative response magnitude (Beppu et al., 2012). Whenever the magnitude of responses to the same control changed by more than 20% of the magnitude of the preceding one, we discarded the preparation.
4.2.
Stimulation
We adapted frog tongues to a HEPES solution (5 mM HEPES dissolved in deionized water, pH 7.0), applied preincubating solutions for different periods, and applied stimulating solutions. We added the same concentration of the same nontaste substances to the preincubating and stimulating solutions. The flow rate of all solutions was 100 mL/min. The stimulation interval was 410 min. Although water elicited substantial taste nerve responses in the absence of salts (Miyake et al., 1976; Zotterman, 1949), 5 mM HEPES completely suppressed taste nerve responses to them (Beppu et al., 2012). All stimulating and preincubating solutions were freshly prepared with the HEPES solution.
4.3.
Diffusion of Lucifer Yellow CH
We investigated the diffusion of Lucifer Yellow CH into taste disks, taste receptor organs of bullfrogs, in the presence and absence of hypertonicity. We washed the tongue of anesthetized bullfrogs for 30 min with a physiological saline (115 mM NaCl, 2.0 mM KCl, and 2.5 mM CaCl2, 2.0 mM MgCl2, 5 mM HEPES, pH 7.4), applied 0.1 mM Lucifer Yellow CH dissolved in the HEPES solution supplemented with or without 1.4 M urea, ethylene glycol, glycerol, or sorbitol for 30 s, and removed the tongue. The bullfrog was killed by decapitation under the anesthetization. We fixed the tongue with 15% formalin for 2–3 days, sectioned it at 60 μm (LinearSlicer Pro 7, Dosaka EM, Kyoto, Japan), and viewed the distribution of Lucifer Yellow CH in taste disks with a fluorescent microscope (BX60, Olympus, Tokyo) with a 10 objective. Images were recorded with a digital CCD camera and processed with Aquacosmos
brain research 1556 (2014) 1–9
software (ver. 2.5, Hamamatsu Photonics K.K., Hamamatsu, Japan).
Acknowledgments This work was partially supported by the “Nano Project” for Private Universities: 2011–2015, matching fund subsidy from Ministry of Education, Culture, Sports, Science and Technology.
r e f e r e n c e s
Beppu, N., Higure, Y., Mashiyama, K., Ohtubo, Y., Kumazawa, T., Yoshii, K., 2012. Hypertonicity augments bullfrog taste nerve responses to inorganic salts. Pflugers Arch. 463, 845–851. Breza, J.M., Contreras, R.J., 2012. Anion size modulates salt taste in rats. J. Neurophysiol. 107, 1632–1648. Budavari, S., 1996. The Merck Index. Merck Research Laboratories Division of Merck & Co., Inc., Whitehouse Station, N. J.. Caprio, J., 1975. High sensitivity of catfish taste receptors to amino acids. Comp. Biochem. Physiol. A: Mol. Integr. Physiol. 52, 247–251. Erlij, D., Martinez-Palomo, A., 1972. Opening of tight junctions in frog skin by hypertonic urea solutions. J. Membr. Biol 9, 229–240. Hashimoto, S., Murakami, M., Kanaseki, T., Kobayashi, S., Matsuki, M., Shimono, M., Segawa, A., 2003. Morphological and functional changes in cell junctions during secretory stimulation in the perfused rat submandibular gland. Eur. J. Morphol. 41, 35–39. Kamo, N., Kashiwagura, T., Kobatake, Y., Kurihara, K., 1978. Role of membrane-bound calcium in taste reception of the frog. J. Physiol. 282, 115–129. Kamo, N., Kashiwagura, T., Kurihara, K., Kobatake, Y., 1980. A theory of dynamic and steady responses in chemoreception. J. Theor. Biol. 83, 111–130. Kashiwagura, T., Kamo, N., Kurihara, K., Kobatake, Y., 1980. Interpretation by theoretical model of dynamic and steady components in frog gustatory response. Am. J. Physiol. 238, G445–G452. Krouwer, V.J., Hekking, L.H., Langelaar-Makkinje, M., ReganKlapisz, E., Post, J.A., 2012. Endothelial cell senescence is associated with disrupted cell–cell junctions and increased monolayer permeability. Vasc. Cell 4, 12. Kumazawa, T., Kurihara, K., 1990. Large enhancement of canine taste responses to sugars by salts. J. Gen. Physiol 95, 1007–1018. Kumazawa, T., Mashiyama, K., Ohtubo, Y., Yoshii, K., 2012. Bullfrog taste nerve responses under hyperosmotic conditions. In: Eighth Forum of Neuroscience. vol., ed. Barcelona, Spain, Abstract. no. 4173.
9
Mashiyama, K., Bunya, H., Higure, Y., Beppu, N., Yoshii, K., Kumazawa, T., 2007. Effects of osmotic pressure on the phasic component of bullfrog taste responses. Chem. Senses 32 (JASTS Abstract. J3). Miyake, M., Kamo, N., Kurihara, K., Kobatake, Y., 1976. Physicochemical studies of taste reception. III. Interpretation of the water response in taste reception. Biochim. Biophys. Acta 436, 843–855. Mudry, B., Carrupt, P.A., Guy, R.H., Delgado-Charro, M.B., 2007. Quantitative structure–permeation relationship for iontophoretic transport across the skin. J. Controlled Release 122, 165–172. Ogawa, H., Sato, M., Yamashita, S., 1968. Multiple sensitivity of chordat typani fibres of the rat and hamster to gustatory and thermal stimuli. J. Physiol 199, 223–240. Rehnberg, B.G., MacKinnon, B.I., Hettinger, T.P., Frank, M.E., 1993. Anion modulation of taste responses in sodium-sensitive neurons of the hamster chorda tympani nerve. J. Gen. Physiol. 101, 453–465. Segawa, A., 1994. Tight junctional permeability in living cells: dynamic changes directly visualized by confocal laser microscopy. J. Electron. Microsc. (Tokyo) 43, 290–298. Ussing, H.H., 1965. Relationship between osmotic reactions and active sodium transport in the frog skin epithelium. Acta Physiol. Scand 63, 141–155. Ussing, H.H., 1966. Anomalous transport of electrolytes and sucrose through the isolated frog skin induced by hypertonicity of the outside bathing solution. Ann. N. Y. Acad. Sci. 137, 543–555. Watson, K.J., Kim, I., Baquero, A.F., Burks, C.A., Liu, L., Gilbertson, T.A., 2007. Expression of aquaporin water channels in rat taste buds. Chem. Senses 32, 411–421. Ye, Q., Heck, G.L., DeSimone, J.A., 1991. The anion paradox in sodium taste reception: resolution by voltage-clamp studies. Science 254, 724–726. Ye, Q., Heck, G.L., DeSimone, J.A., 1993. Voltage dependence of the rat chorda tympani response to Naþ salts: implications for the functional organization of taste receptor cells. J. Neurophysiol. 70, 167–178. Ye, Q., Heck, G.L., DeSimone, J.A., 1994. Effects of voltage perturbation of the lingual receptive field on chorda tympani responses to Naþ and Kþ salts in the rat: implications for gustatory transduction. J. Gen. Physiol 104, 885–907. Yoshii, K., Kamo, N., Kurihara, K., Kobatake, Y., 1979. Gustatory responses of eel palatine receptors to amino acids and carboxylic acids. J. Gen. Physiol. 74, 301–317. Yoshii, K., Kobatake, Y., Kurihara, K., 1981. Selective enhancement and suppression of frog gustatory responses to amino acids. J. Gen. Physiol. 77, 373–385. Yoshii, K., Yokouchi, C., Kurihara, K., 1986. Synergistic effects of 50 -nucleotides on rat taste responses to various amino acids. Brain Res 367, 45–51. Zotterman, Y., 1949. The response of the frog’s taste fibres to the application of pure water. Acta Physiol. Scand 18, 181–189.