| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
College of Sciences, San Diego State University, San Diego, California
Submitted 22 August 2003; accepted in final form 22 August 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
In taste these issues have not enjoyed definitive resolutions. 1) Taste encodes a variety of physical dimensions, and coordination among them emerges only when they are collapsed into a dimension that relates to the animal's physiological welfare (Scott and Mark 1987
). 2) Although the existence of primary taste qualities is broadly acknowledged, their definition and number remain in doubt (Scott and Giza 2000). 3) The topographic organization of taste qualities remains a tantalizing possibility, reinforced by single-cell recordings (Ishiko and Akogi 1972; Yamamoto et al. 1985, 1989) and functional MRI studies (O'Doherty et al. 2001), yet never so clearly demonstrated as to serve as a reliable coding mechanism. 4) The existence of gustatory neuron types has received strong support, although mainly at the hindbrain level in rodents. This paper addresses this last issue.
There are several criteria according to which neuron types may be identified. The most obvious means of grouping is according to the stimulus to which the neuron responds best (Frank et al. 1988). Although useful as a first approximation, this does not accommodate the emergence of different groups as stimulus concentrations or qualities change. A more complete description of a neuron's sensitivity derives from its profile of responsiveness, either to the acknowledged basic stimuli or to a broader stimulus array (Woolston and Erickson 1979). However, statistical procedures based on these profiles, such as cluster analysis, are unable to define objectively whether a population merely contains discontinuities in sensitivity or is composed of distinct and internally coherent groups (Scott and Giza 2000).
Other criteria may be used as converging evidence for the existence of groups, including cytoarchitecture, anatomical connections, and neurochemistry. These are not independent of function, so this information could be related to the respective neurons' responses to gain insights into the functional organization of taste. Such hypothetical interactions (e.g., all fusiform glutamatergic neurons in the parabrachial nucleus have response profiles distinct from stellate cholinergic cells) could lend further support to the case for discrete functional groups of taste cells.
Finally, distinct classes of taste cells might be identified by altering the environment in which the taste system normally operates, and determining whether only a subset of the neurons is affected by the change. Chang and Scott (1984b) and McCaughey et al. (1997) conditioned rats to avoid sodium saccharin and found that only the sugar-oriented subset of taste neurons in the nucleus of the solitary tract (NTS) was affected. Contreras and Frank (1979) found that sodium deprivation reduced the responsiveness only of sodium-oriented cells in the rat's chorda tympani nerve. Jacobs et al. (1988) reported that sodium deprivation resulted in decreased activity of sodium-oriented neurons, and increased responsiveness of sugar-oriented cells in the NTS, and so a shift in the manner in which the taste of sodium was represented. These results were confirmed by McCaughey and Scott (2000) in the NTS of rats in which the neurochemical response to sodium deficiency was mimicked through exogenous administration of neuromodulators. Both Giza et al. (1997), administering a satiating load of glucose intravenously and recording from NTS cells, and Hajnal et al. (1999), administering a satiating load of intralipid into the duodenum and recording from parabrachial cells, reported that the consequences of satiety are visited almost exclusively on sugar-oriented taste cells. In all 7 of these cases, an alteration of the experiences or the physiological condition of the rats resulted in gustatory modulations that were limited to discrete subsets of taste neurons.
A more direct approach to the issue of how taste neurons respond to altered conditions is afforded by applying a specific sodium channel blocker to the receptors and determining whether the ensuing blockade of sodium transduction causes universal or only limited inhibition of responsiveness to sodium. Amiloride (N-amidino-3,5-diamino-6-chloropyrazine), the sodium channel blocker, has been used for this purpose. The oral administration of amiloride has been found to inhibit amiloride-sensitive ion channels as well as sodium-oriented neurons in both the chorda tympani nerve (Ninomiya 1988) and the NTS (Giza and Scott 1991; Smith et al. 1996) of rat and hamster. There is evidence that this indicates functional specificity of neurons as well because rats cannot identify NaCl at concentrations up to 0.1 M after amiloride treatment (Hill et al. 1990; McCutcheon 1991).
In the current study, we investigated whether this degree of specificity extended to the rat's forebrain. We explored the effects of oral amiloride application on the responses of gustatory cells of the ventroposteromedial nucleus pars parvicellularis (VPMpc), the thalamic taste nucleus, 2 synaptic levels beyond the NTS. We recorded the responses of 42 single neurons in VPMpc before and after oral application of amiloride to determine the relationship between their response characteristics and the extent to which they were affected by amiloride.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Subjects were 12 female and 12 male Wistar rats weighing 494 ± 113 g (mean ± SD; females: 411 ± 25 g; males: 564 ± 82 g). They were maintained on pellet food and tap water that was available ad libitum in their home cages. Room lights were on from 0800 to 2200 h.
Surgery
Each rat was anesthetized to a surgical level with urethane (1.2 g/kg, intraperitoneally [ip]) and 
-chloralose (20 mg/kg, ip). Epileptiform activity may be induced by urethane. As a monitor for this, we took continuous EEG recordings in 6 rats and found normal patterns of synchrony and asynchrony during the recording sessions. The depth of anesthesia was monitored by frequent testing for the presence of leg flexion reflexes whose presence warranted a supplemental dose of urethane. Lidocaine was applied to all wound margins. Body temperature was maintained between 36 and 38°C and heart rate was continuously monitored. All surgery was performed under aseptic conditions. The trachea was cannulated and an esophageal fistula was inserted to prevent taste solutions from reaching the stomach where they might induce postingestive effects. The head was mounted in a nontraumatic head holder (Erickson 1966) and the skull was exposed. A 4 x 4-mm area of skull centered on coordinates 3.6 mm posterior to bregma and 1.0 mm lateral to the midline suture was removed. Sterile saline was used to keep the exposed cortical surface moist. The skull was positioned to be horizontal in the anteroposterior dimension by measuring at bregma and lambda, and in the mediolateral dimension by measuring 4 mm on opposite sides of the midline.
Electrophysiological recording
Borosilicate pipettes, filled with 2.5 M potassium citrate, were used to isolate the activity of individual cells in the VPMpc. The inner diameter of the electrode tips was about 1.0 µm, and the impedance was 1–4 M
at 1.0 kHz. Conventional electrophysiological recording techniques were used for differential amplification and display of the neural signal. Action potentials of a single cell were identified by consistency of amplitude and waveform, and by an interspike interval of at least 1.5 ms. Neural data, voice commentary, and onset marker signal for stimulus delivery were stored on a 4-channel TEAC tape recorder and analyzed off-line.
Stimuli and stimulus delivery
Seventeen sapid stimuli were applied at 23°C (room temperature). These chemicals, their concentrations, abbreviations used to identify them in this paper, their molecular weights, and chemical groups are listed in Table 1. All solutions were prepared with deionized water except that 5% tap water was added to nonionic solutions to activate the stimulus onset marker.
|
), an amount 2 orders of magnitude below the threshold for electric taste (Bujas 1971; Pfaffmann and Pritchard 1980). Each stimulus was followed by a 20-ml distilled water rinse and by a minimum rest period of 30 s. Additional rinses and rest periods were occasionally required for baseline activity to be reestablished as demonstrated by the restoration of spontaneous activity to its prestimulus levels. Stimuli were presented in quasi-random order with the stipulation that chemicals representing similar taste qualities not be applied consecutively. If the neuron remained well isolated at the end of the stimulus series, 180 ml of 0.5 mM amiloride was sprayed into the mouth at a rate of 1 ml/s for 3 min. This was followed by reapplication of the stimulus series. During the post-amiloride period all procedures for stimulus application were identical to those used before, except that the amiloride solution was used for rinses between stimulus applications. If the neural response was still well isolated after this series was complete, the mouth was rinsed with at least 750 ml distilled water over a period of at least 20 min, and responses to the 4 prototypical stimuli were recorded again to compare with the pre-amiloride data. Recording
Initial coordinates to locate VPMpc were 3.6 mm posterior to bregma and 1.0 mm lateral to the midline. During the search for cells, the electrode was advanced in 30-µm steps between a depth of 6.0 and 7.2 mm from the cortical surface.
Responses of taste cells were recorded as the mean impulses/s over a 5-s poststimulus period. This rate was compared with the spontaneous activity distribution of the neuron. An evoked response was defined by a change of at least 1.65 SD (P = 0.05) from mean spontaneous rate, sustained over 5 s. We recognize that behavioral discriminations require only a fraction of this time (Halpern and Tapper 1971), but a 5-s response period has been shown at lower-order levels to yield neural data most in accord with taste-mediated behavior (Doetsch et al. 1969). We also acknowledge that, given the robustness of thalamocortical connections, there are likely to be extragustatory influences on these neurons during the response period.
Data analysis
The neural signal was digitized using a window discriminator. A PC was used to count action potentials for 3 s before (spontaneous) and 5 s after (evoked) stimulus application, and to perform basic statistical analyses. The net discharge rate (evoked minus spontaneous) over 5 s provided the basis for derived analyses, which included calculations of interneuronal and interstimulus Pearson product–moment correlation coefficients, multidimensional scaling (Guttman 1968), ANOVAs, and cluster analyses using the programs Systat 9.0 and SPSS 9.0. The t-tests were performed using Microsoft Excel 2000. Values of P < 0.05 were considered statistically significant, and the level was adjusted using Bonferroni's method for multiple comparisons (/n, where n = number of comparisons). Means are presented with their associated SE values unless otherwise specified.
Histology
The locations of recorded taste cells were verified by bringing the tip of a pipette filled with pontamine sky blue (ID = 5 µm) to the same coordinates as those of the final taste cell in each preparation. The dye was applied iontophoretically at –10 µA, 10 min, 10 s on/off. The rat was then euthanized with an overdose of urethane and the brain was frozen at –40°C in isopentane. Coronal sections were cut at 20 µm and stained with cresyl violet or left unstained and subsequently coverslipped. Brain sections of 4 rats were stained for acetylcholinesterase (Karnovsky 1964), with the modification that all compounds were dissolved in 0.013 M potassium phosphate monobasic sodium hydroxide buffer with a final pH of 6.0–6.5 rather than in tris-maleate buffer.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Acetylcholinesterase stains in 4 rats revealed that each track was in the VPMpc. Recording sites were also verified using cresyl violet–stained sections in 7 rats whose taste cells are included in the present analysis (Fig. 1).
|
We isolated the activity of 115 thalamic taste cells, of which we were able to maintain good isolation on 42 for the duration of the pre-amiloride and post-amiloride series. Across these 42 taste cells, amiloride significantly reduced responses to NaCl. Evoked rates were lower by 69% to 0.03 M NaCl, and by 39% to 0.1 M (P < 0.001 and P = 0.001, respectively; protected = 0.003, n = 17). By this strict criterion, amiloride did not affect the responses to acids, sugars, alkaloid, or nonsodium salts.
A striking distinction was found in the effectiveness of amiloride among neural groups. Three groups of taste cells were identified: sodium-, sugar-, and acid-oriented (N, S, and H groups, respectively). Amiloride significantly reduced responses only in the N and S groups. The N group could be further subdivided into an NaCl-specific subgroup and an NaCl–HCl subgroup. Amiloride reduced responses predominantly among the NaCl-specific cells.
General effects of amiloride on taste cells
The mean spontaneous activity level of taste cells in this study was 6.4 ± 1.1 spikes/s (range = 0.1–29.9, n = 42). This did not change with amiloride application (5.9 ± 0.1 spikes/s; P = 0.42). Nor was there a significant change in the spontaneous firing rate of cells in any of the 3 neural groups.
There were 714 (17 x 42) stimulus–neuron interactions. Of these, 471 (66%) resulted in excitatory responses, 7 (1%) in inhibition, and 236 (33%) in no response. No cell gave fewer than 4 significant excitatory responses. After amiloride, the number of excitatory responses declined and nonresponses increased: 380 (53%) excitatory, 10 (1%) inhibitory, and 324 (45%) no response (P < 0.001, 
2 test). In both conditions, no cell had more than 2 inhibitory responses. The breadth-of-tuning coefficient is the accepted metric for indexing the breadth of a neuron's response profile across the 4 prototypical stimuli (Smith and Travers 1979). The proportional distribution of a cell's response across these stimuli is converted to a coefficient that may range from 0.00, indicating total specificity to one of the 4 chemicals, to 1.00, resulting from equal responsiveness to all 4. Across all 42 taste cells, the breadth-of-tuning metric was 0.77 ± 0.03 pre-amiloride, and 0.75 ± 0.04 post-amiloride (P = 0.50; n.s.).
Mean evoked discharge rates across all cells before and after amiloride are depicted in Fig. 2. The rates elicited by the basic tastants in spikes/s were as follows
 |
|
= 0.003, n = 17). When a nonprotected less-conservative critical significance level of 0.05 was used, all Na- and Li-containing stimuli except the weakest (0.01 M) concentration of NaCl gave significantly reduced responses: 0.03, 0.10, and 0.30 M NaCl, LiCl, MSG, NaSaccharin (with 30 mM Na), and polycose (containing 3.8 mM NaCl, but which appears to have an idiosyncratic taste in rats; Sako et al. 1994). In contrast, the response to citric acid was enhanced (P = 0.05). Thus the effect of amiloride was largely specific to the Na and Li ions. To verify that these changes were attributed to amiloride application, we tested whether the inhibition of the response to 0.1 M NaCl was reversible in 15 neurons. Following the post-amiloride stimulus series, we applied 840 ± 66 ml distilled water over a period of 28 ± 4 min to eliminate any residual effect of amiloride. The subsequent response to 0.1 M NaCl was restored to its pre-amiloride level (P = 0.34; n.s.), and both were significantly different from the response during amiloride application (P = 0.007 and 0.003, respectively). Thus the impact of amiloride on the response to NaCl in the VPMpc was significant and reversible.
Neural characteristics and the effects of amiloride
The current study was designed to evaluate whether there were significant differences among cells in this neural population with respect to their susceptibility to amiloride, and whether these were related to their response profiles. Three approaches were used to address this issue. First, taste cells were classified according to their response profiles and the groups were subsequently evaluated for their differential sensitivity to amiloride. Second, cells and stimuli were organized according to the effects of amiloride to determine whether neurons with particular response profiles would emerge. Third, specificity to NaCl was calculated as a ratio of the response to 0.1 M NaCl divided by the sum of responses to the 4 prototypical stimuli. This measure of specificity was then compared to the impact of amiloride on the response to NaCl in that cell to determine whether greater specificity to NaCl implied greater susceptibility to amiloride blockade.
NEURON GROUPS AND THE EFFECTS OF AMILORIDE. Gustatory neurons were classified according to their best responses and according to their response profiles. Twenty-four cells (57%) responded best to NaCl among the 4 basic stimuli, 9 (21%) to sucrose, 7 (17%) to HCl, and 2 (5%) to quinine HCl.
A more comprehensive categorization was derived from an analysis of the similarity of neural response profiles across these basic stimuli. We generated a similarity matrix based on correlation coefficients between each pair of response profiles [(42 x 41)/2 = 861], then subjected the coefficients of this matrix to a cluster analysis. The results appear as a dendrogram in Fig. 3. Distances are defined as one minus the correlation coefficient. Pairs of cells are interconnected at the level of correlation between their response profiles; groups are fused at the mean level of correlation between their constituent members in an iterative process that was complete when all neurons were interconnected. Each cell is labeled according to the basic stimulus that elicited its largest response, followed by any stimulus that evoked a response at least 80% of that level. Cells are numbered in the order in which they were isolated.
|
The N group was most coherent, its 21 cells being fully intercorrelated at a level of 0.79; the H group was fully intercorrelated at 0.62, and the S group at 0.42. ANOVA and post hoc t-tests confirmed the statistical independence of these groups by differences between their defining stimuli. For example, neurons in the S and N groups gave significantly different responses to NaCl and sucrose, but not to HCl or quinine.
Only the N group could be further divided into significantly different subgroups: N-specialists, from neuron N33 to N18 constituting 12 cells, and NH neurons, from N15 to N09 constituting 9. There was a significant difference between their responses only to HCl, which was greater in the latter subgroup (P = 0.004; protected = 0.013). Even though the first subgroup was more specifically tuned to NaCl, the difference was not sufficient to be revealed in the breadth of tuning between them, which did not differ. Nor were there any significant differences in breadth of tuning among the 3 main groups before amiloride application.
Figure 4 shows the response profiles before and after amiloride for each of the 3 main groups. Amiloride significantly affected activity only in the N and S groups. Among N cells, the response to 0.03 M NaCl was reduced by 80%, to 0.1 M by 54%, and to 0.3 M by 42% (all P < 0.001; protected = 0.003). Using a less-conservative unprotected level of 0.05 revealed additional reductions in response to LiCl (50%; P = 0.02), NaSaccharin (47%; P = 0.04), MSG (65%; P = 0.02), and CaCl2 (37%; P = 0.03), plus an increased response to sucrose (–191%; P = 0.02). S cells showed significant reductions to 0.03 M NaCl (85%; P = 0.01) and to polycose (71%; P = 0.04). There were no significant changes among neurons in the H group.
|
|
ORGANIZATION OF NEURONS BASED ON AMILORIDE'S EFFECTS. The second approach was to organize neurons according to the impact amiloride had on their responses. To do so, the response profile of each cell was taken to be the difference between its pre- and post-amiloride response to each of the 4 basic stimuli. A correlation matrix was then generated based on these difference profiles, and a cluster analysis performed as before. The results appear in Fig. 6.
|
SPECIFICITY TO NACL AND THE EFFECTS OF AMILORIDE.
The third approach demonstrates the relationship between a neuron's sodium specificity and its sensitivity to blockade by amiloride (Fig. 7). The specificity of a cell's response to NaCl was determined by calculating that proportion of the total response (to the basic stimuli) that was attributable to the response to 0.1 M NaCl alone. There was a significant regression (r2 = 0.12; P = 0.026) between specificity to NaCl before amiloride (S) and the percentage reduction of response to NaCl after amiloride (R)
 |
|
Taste qualities
The profile of activity generated by each stimulus across the 42 taste cells was determined, and the correlation coefficient between each pair of profiles was calculated [(17 x 16)/2 = 136 coefficients]. A 3-dimensional representation of the relative similarities among response profiles before amiloride, accounting for 96% of the data variance, is shown in Fig. 8A. Sodium and lithium salts are positioned together near the back of the space, with potassium somewhat below them, and quinine much lower. Sugars are located near the front of the space, with polycose nearby, and NaSaccharin is between the sodium salts and sugars. The acids and bitter salts are at the back, far left.
|
Conversely, it is possible to compute hypothetical response profiles of only the amiloride-sensitive portion of the response by subtracting the residual post-amiloride activity from the pre-amiloride response. A taste space derived from these computed response profiles, shown in Fig. 8C, demonstrates clear differentiation among stimuli. Sodium–lithium salts, including MSG, are at the back right of the space, separated from acids and bitter salts at the front right, and from sugars on the left.
We can calculate the number of chemicals that are statistically discriminable in each of the 3 cases—total response (Fig. 8A), amiloride-insensitive response (8B), and amiloride-sensitive response (8C)—based on whether the correlations between their profiles fail to reach significance (thus are not statistically related at P = 0.05), as computed by regression analysis. The total number of comparisons in each case is (4 x 17) – 4 = 64. This is not meant to imply that a behavioral discrimination between 2 chemicals requires this same degree of difference between their neural profiles, but only that there is a recognized degree of statistical separation.
Before amiloride (Fig. 8A), 33 of the 64 pairs of stimuli had profiles that were not significantly correlated, and thus may be inferred to be discriminable, at least statistically. After amiloride application (Fig. 8B), the number dropped to 20, resulting from the fact that sodium–lithium salts collapsed with sour and bitter stimuli. However, when only the calculated amiloride-sensitive portion of the response is considered (Fig. 8C), 46 stimulus pairs were discriminable. This is greater discriminability than is offered by the total response, and could be interpreted to imply that the amiloride-insensitive portion is imposing noise on an otherwise orderly system.
Sex differences
There was one significant difference between activity from male (n = 12) and female (n = 12) rats, and a second difference that derived from it. The spontaneous rate of neural activity was higher in males, both before (males = 8.6 ± 6.3; females = 4.5 ± 2.5; P = 0.02) and after (males = 8.3 ± 5.7; females = 4.3 ± 2.5; P = 0.03) amiloride. Because spontaneous rate was subtracted from gross evoked activity, which did not differ between the sexes, it follows that the number of the 17 stimuli that elicited excitatory responses was lower in cells derived from males than from females (9.8 ± 2.8 vs. 12.2 ± 2.2; P = 0.04). There were no significant sex differences in any of the other properties reported above or in the effectiveness of amiloride.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
) or the recently proposed amiloride-insensitive apical channel (DeSimone et al. 2001). Moreover, this segregation is maintained through the peripheral nerves to the NTS, and onward to the parabrachial nucleus (by interpolation) to be manifested in the VPMpc of the forebrain. Thus, receptive fields of VPMpc neurons tend either to include amiloride-sensitive or -insensitive taste receptor cells, each largely to the exclusion of the other. It follows that the output of amiloride-sensitive receptors tends to converge on particular peripheral nerve axons to establish a functional organization that is maintained through at least 4 synaptic levels of the gustatory system. One notable feature of thalamic taste cells is their frequent sensitivity to touch and temperature in the oral cavity. Notwithstanding this multimodal responsiveness, within the taste modality, they appear to maintain functional discretion among qualities.
The existence of discrete neuron types does not argue for either labeled-line or pattern coding in gustation. Neuron types are necessary but not sufficient for a labeled-line code, and patterning neither requires nor rejects their existence. Rather than support a theoretical viewpoint, the finding of neuron types through the thalamic level is intended to provide further definition to the functional architecture of the taste system.
Doolin and Gilbertson (1996) reported that the distribution of amiloride-sensitive receptor cells across the rat's gustatory epithelium is skewed toward the anterior. Some 65% of fungiform receptors show amiloride-sensitive sodium transduction, versus 38% of those on the soft palate, 37% of those in foliate, and 0% of those in circumvallate papillae. Thus functional specificity in peripheral nerves, and extending into the CNS, could originate partly from the physical location of amiloride-sensitive receptors.
In hamsters, however, Gilbertson and Troy-Fontenot (1998) reported that the distribution of amiloride-sensitive receptors across the oral cavity is homogeneous. Because amiloride susceptibility is as specific in the chorda tympani nerve of the hamster (Hettinger and Frank 1990) as in the rat (Ninomiya and Funakoshi 1988), it is likely attributable to convergent projections from regions of receptors wherever they exist. Thus, amiloride sensitivity appears to be a functional characteristic of taste receptors that serves as a general organizing principle of the system.
That organization probably comes not from an intrinsic quality of the receptors, but from their connections to peripheral nerves. Ninomiya (1998) crossed and regenerated the chorda tympani and glossopharyngeal nerves in rats. He reported that the regenerated chorda tympani, now connected to the receptor population that formerly innervated the glossopharyngeal nerve, showed its normal proportion of amiloride-sensitive and -insensitive fibers, despite the fact that fibers of the normal glossopharyngeal nerve are almost exclusively amiloride-insensitive. This extends the conclusion reached by Oakley (1998) that trophic factors from the sensory nerve fibers confer sensitivity patterns on receptor cells, and so maintain functional stability in the face of constant receptor turnover.
Contributions to the gustatory neural code
The reduction in evoked activity caused by amiloride leaves only the amiloride-insensitive portion of the response intact. The coding capacity of the VPMpc, bereft of its amiloride-sensitive component, is shown in Fig. 8B (post-amiloride) where differentiation among the sodium salts, bitter salts, and acids is lost. Conversely, the space resulting from the calculated amiloride-sensitive response implies enhanced discriminability among stimuli.
An analysis of behavioral data supports this implication. Yamamoto and colleagues (1985, 1987) used the conditioned taste aversion technique to determine the behavioral similarities among a series of taste stimuli that largely overlapped our stimulus array, including the same concentration series of NaCl. We performed a regression between the level of behavioral similarity of each stimulus pair and the correlation between their neural response profiles based on pre-amiloride (total), post-amiloride (amiloride-insensitive), and pre- minus post-amiloride (amiloride-sensitive) thalamic responses.
The total neural response in VPMpc explains 32% of the variance in behaviorally established taste similarities (P < 0.001), whereas the amiloride-insensitive portion explains only 11% (P = 0.06). However, the amiloride-sensitive component alone accounts for 62% (P < 0.001). When only similarities between the basic stimuli and the 3 other concentrations of NaCl are considered (i.e., the stimuli on which amiloride has its primary impact), the amiloride-sensitive component explains an extraordinary 78% of the behavioral categorization, whereas the whole response explains 42%. Therefore, the amiloride-sensitive portion of the thalamic neural response may relate particularly well to the rat's ability to discriminate among taste stimuli, especially those containing sodium.
The availability of data on behavioral similarity also permitted an analysis of the relationship between a thalamic cell's response profile and its contribution to behavioral taste categorization. As noted above, the neural data from 42 cells accounted for 32% of the total variance in the behavioral data. Responses of the 5 cells with the greatest specificity (i.e., the lowest breadth of tuning coefficients) explained 7% of the behavioral data; the most narrowly tuned 10 cells explained 18%, and the most narrowly tuned 20 cells explained 27%. Therefore, activity in this last group, representing less than half the total neural sample of 42, accounted for 27/32%, or 84% of the variance explained by the entire sample. Those neurons with the greatest specificity are apparently most critical to taste discrimination.
Taste cells in VPMpc may be divided into statistically discrete subgroups based on their responses to the basic taste stimuli. The inhibitory effects of amiloride are largely confined to the sodium- and sucrose-oriented neural groups, implying that functional gustatory neuron types extend through at least the thalamic level of the rat's taste system.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
1 In contrast to our findings, St. John and Smith (2000) report a significant decrease in responsiveness to KCl in the rat nucleus tractus solitarius (NTS) following amiloride application. This reduction was most prominent among N-best (51%) and S-best cells (33%) and nearly non-existent among H-best cells (7%). This is in accord with the consistent finding that the effects of amiloride are visited almost exclusively on N and S neurons. Since the qualities that make NaCl and KCl distinctive were muted by amiloride, it follows that the correlation between the profiles they evoked increased from +0.54 to +0.76 following amiloride (St. John and Smith 2000). This increasing neural similarity corresponded closely to the loss of behavioral discriminability between NaCl and KCl with amiloride administration (Spector et al. 1996). We did not find a loss of KCl sensitivity among N-cells in the thalamus, though we saw a non-significant decrease to KCl among S- and H-cells (Fig. 4). However, we did find an increased correlation between the response profiles generated by three of the four concentrations of NaCl versus that of KCl. Therefore the effect of amiloride on neural activity in the thalamus is consistent with the observed loss of behavioral discriminability between NaCl and KCl in the rat. 
Address for reprint requests and other correspondence: T. R. Scott, College of Sciences, San Diego State University, 5500 Campanile Dr., San Diego, CA 92182-1010 (E-mail: trscott{at}sciences.sdsu.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Chang FCT and Scott TR. A technique for gustatory stimulus delivery in the rodent. Chem Senses 9: 91–96, 1984a.
Chang FCT and Scott TR. Conditioned taste aversions modify neural responses in the rat nucleus tractus solitarius. J Neurosci 4: 1850–1862, 1984b.[Abstract]
Contreras RJ and Frank M. Sodium deprivation alters neural responses to gustatory stimuli. J Gen Physiol 73: 569–594, 1979.
DeSimone JA, Lyall VJ, Heck GL, Phan THT, Alam RI, Feldman GM, and Buch R. A novel pharmacological probe links the amiloride-insensitive NaCl, KCl, and NH4Cl chorda tympani taste responses. J Neurophysiol 86: 2638–2641, 2001.
Doetsch DS, Ganchrow JJ, Nelson LM, and Erickson RP. Information processing in the taste system of the rat. In: Olfaction and Taste, edited by Pfaffmann C. New York: Rockefeller Univ. Press, 1969, vol. III, p. 492–511.
Doolin RE and Gilbertson TA. Distribution and characterization of functional amiloride-sensitive sodium channels in rat tongue. J Gen Physiol 107: 545–554, 1996.
Erickson RP. Nontraumatic headholders for mammals. Physiol Behav 1: 97–98, 1966.
Erickson RP. Stimulus coding in topographic and non-topographic afferent modalities: on the significance of the activity of individual sensory neurons. Psych Rev 75: 447–465, 1968.
Erickson RP, Rogers JL, and Sarle WS. Statistical analysis of neural organization. J Neurophysiol 70: 2289–2300, 1993.
Frank ME, Bieber SL, and Smith DV. The organization of taste sensibilities in hamster chorda tympani nerve fibers. J Gen Physiol 91: 861–896, 1988.
Gilbertson TA and Troy-Fontenot D. Distribution of amiloride-sensitive sodium channels in the oral cavity of the hamster. Chem Senses 23: 495–499, 1998.
Giza BK, McCaughey SA, Scott CL, Zhang L, and Scott TR. Intravenous glucose injections suppress responses of sugar-sensitive taste neurons in the rat NST. Soc Neurosci Abstr 23: 1037, 1997.
Giza BK and Scott TR. The effect of amiloride on taste-evoked activity in the nucleus tractus solitarius of the rat. Brain Res 550: 247–256, 1991.[CrossRef][ISI][Medline]
Guttman L. A general nonmetric technique for finding the smallest coordinate space for a configuration of points. Psychometrika 33: 469–507, 1968.[CrossRef][ISI]
Hajnal A, Takenouchi K, and Norgren R. Effect of intraduodenal lipid on parabrachial gustatory coding in awake rats. J Neurosci 19: 7182–7190, 1999.
Halpern BP and Tapper DN. Taste stimuli: quality coding time. Science 171: 1256–1258, 1971.
Hettinger T and Frank ME. Specificity of amiloride inhibition of hamster taste responses. Brain Res 513: 24–34, 1990.[CrossRef][ISI][Medline]
Hill DL, Formaker BK, and White KS. Perceptual characteristics of amiloride-suppressed sodium chloride taste response in the rat. Behav Neurosci 104: 734–741, 1990.[CrossRef][ISI][Medline]
Ishiko N and Akogi T. Topographical organization of gustatory nervous system. In: Olfaction and Taste, edited by Schneider D. Stuttgart, Germany: Wissenschaftliche Verlagsgesellschafte, 1972, vol. IV, p. 343–349.
Jacobs KM, Mark GP, and Scott TR. Taste responses in the nucleus tractus solitarius of sodium-deprived rats. J Physiol 406: 393–410, 1988.
Karnovsky MJ and Roots L. A "direct-coloring" tricholine method for cholinesterases. J Histochem Cytochem 12: 219–221, 1964.[ISI][Medline]
McCaughey SA, Giza BK, Nolan LJ, and Scott TR. Extinction of a conditioned taste aversion in rats. II. Neural effects in the nucleus of the solitary tract. Physiol Behav 61: 373–379, 1997.[CrossRef][Medline]
McCaughey SA and Scott TR. Rapid induction of sodium appetite modifies taste-evoked activity in the rat nucleus of the solitary tract. Am J Phsyiol Regul Integr Comp Physiol 279: R1121–R1131, 2000.
McCutcheon NB. Sodium-deficient rats are unmotivated by sodium chloride solutions mixed with the sodium channel blocker amiloride. Behav Neurosci 105: 764–766, 1991.[CrossRef][ISI][Medline]
Ninomiya Y. Reinnervation of cross-regenerated gustatory nerve fibers into amiloride-sensitive and amiloride-insensitive taste receptor cells. Proc Natl Acad Sci USA 95: 5347–5350, 1998.
Ninomiya Y and Funakoshi M. Amiloride inhibition of responses of rat single chorda tympani fibers to chemical and electrical tongue stimulations. Brain Res 451: 319–325, 1988.[CrossRef][ISI][Medline]
Oakley B. Taste neurons have multiple inductive roles in mammalian gustatory development. Ann NY Acad Sci 855: 50–57, 1998.[CrossRef][ISI][Medline]
O'Doherty J, Rolls ET, Francis S, McGlone F, and Bowtell R. Representation of pleasant and aversive taste in the human brain. J Neurophysiol 85: 1315–1321, 2001.
Pfaffmann C and Pritchard TC. Ion specificity of electric taste. In: Olfaction and Taste, edited by van der Starre H. Oxford, UK: IRL Press, 1980, vol. VII, p. 175–178.
Sako N, Shimura T, and Yamamoto T. Differences in taste responses to polycose and common sugars in the rat as revealed by behavioral and electrophysiological studies. Physiol Behav 56: 741–745, 1994.[CrossRef][Medline]
Scott TR and Giza BK. Issues of gustatory neural coding: where they stand today. Physiol Behav 69: 65–76, 2000.[CrossRef][Medline]
Scott TR and Mark GP. The taste system encodes stimulus toxicity. Brain Res 414: 197–203, 1987.[CrossRef][ISI][Medline]
Smith DV, Liu H, and Vogt MB. Responses of gustatory cells in the nucleus of the solitary tract of the hamster after NaCl or amiloride adaptation. J Neurophysiol 76: 47–58, 1996.
Smith DV and Travers JB. A metric for the breadth of tuning of gustatory neurons. Chem Senses 4: 215–229, 1979.
Spector AC and Guagliardo NA. Amiloride disrupts NaCl versus KCl discrimination performance: implications for salt taste coding in rats. J Neurosci 16: 8115–8122, 1996.
St. John SJ and Smith DV. Neural representation of salts in the rat solitary nucleus: brain stem correlates of taste discrimination. J Neurophysiol 84: 628–638, 2000.
Woolston DC and Erickson RP. Concept of neuron types in gustation in the rat. J Neurophysiol 42: 1390–1409, 1979.
Yamamoto T, Matsuo R, Kiyomitsu Y, and Kitamura R. Taste responses of cortical neurons in freely ingesting rats. J Neurophysiol 61: 1244–1258, 1989.
Yamamoto T and Yuyama N. On a neural mechanism for cortical processing of taste quality in the rat. Brain Res 400: 312–320, 1987.[CrossRef][ISI][Medline]
Yamamoto T, Yuyama N, Kato T, and Kawamura Y. Gustatory responses of cortical neurons in rats. III. Neural and behavioral measures compared. J Neurophysiol 53: 1370–1386, 1985.
Ye Q, Heck GL, and DeSimone JA. Voltage dependence of the rat chorda tympani response to Na+ salts: implications for the functional organization of taste receptor cells. J Neurophysiol 70: 167–178, 1993.
This article has been cited by other articles:
|
|
 |
|
  L. Mao, Y. K. Cho, and C.-S. Li Modulation of activity of gustatory neurons in the hamster parabrachial nuclei by electrical stimulation of the ventroposteromedial nucleus of the thalamus Am J Physiol Regulatory Integrative Comp Physiol, May 1, 2008; 294(5): R1461 - R1473. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
: P < 0.05; 
: P < 0.002.
