Rapid taste responses in the gustatory cortex during licking

Abstract

Rapid tastant detection is necessary to prevent the ingestion of potentially poisonous compounds. Behavioral studies have shown that rats can identify tastants in approximately 200 ms, although the electrophysiological correlates for fast tastant detection have not been identified. For this reason, we investigated whether neurons in the primary gustatory cortex (GC), a cortical area necessary for tastant identification and discrimination, contain sufficient information in a single lick cycle, or approximately 150 ms, to distinguish between tastants at different concentrations. This was achieved by recording neural activity in GC while rats licked four times without a liquid reward, and then, on the fifth lick, received a tastant (FR5 schedule). We found that 34% (61 of 178) of GC units were chemosensitive. The remaining neurons were activated during some phase of the licking cycle, discriminated between reinforced and unreinforced licks, or processed task-related information. Chemosensory neurons exhibited a latency of 70-120 ms depending on concentration, and a temporally precise phasic response that returned to baseline in tens of milliseconds. Tastant-responsive neurons were broadly tuned and responded to increasing tastant concentrations by either increasing or decreasing their firing rates. In addition, some responses were only evoked at intermediate tastant concentrations. In summary, these results suggest that the gustatory cortex is capable of processing multimodal information on a rapid timescale and provide the physiological basis by which animals may discriminate between tastants during a single lick cycle.

Figure 1.

Histology. Shown is a 50-μm-thick coronal section that was taken through the gustatory cortex and stained with cresyl violet. The cannula track is clearly visible and is indicated by two arrows. The terminations of the electrodes in dysgranular cortex are marked by an arrowhead. CC, Corpus callosum; Pir, piriform cortex; CPu, caudate–putamen; GI, granular insular cortex; DI, dysgranular insular cortex; AI, agranular insular cortex.

Figure 2.

Chemosensitive GC neuronal responses. Shown are examples of four different neurons obtained from the same ensemble. Displayed are the raster plots and PSTHs for the responses of these neurons to multiple stimuli. Zero on the x-axis denotes the time of stimulus delivery. The reinforced lick occurs at time 0 (solid red line), and the previous and subsequent licks are visible at around −150 and 150 ms, respectively. These conventions will be followed throughout this paper unless explicitly noted. One outstanding feature of these responses is that their spike trains often exhibit great temporal precision across multiple tastant deliveries (A–C) and that the reinforced activity is much greater than the activity for the dry licks. The spike trains of these three neurons displayed a latency of 50–70 ms after tastant delivery, peak response times of 80–90 ms, and offset times of ∼100–120 ms. An example of a response to sucrose that continued firing beyond the time of the subsequent lick is shown in D.

Figure 3.

GC neurons exhibit rapid and selective responses to tastants. The activity of this neuron is sufficient to discriminate between reinforced and unreinforced licks as well as between tastants. NaCl and sucrose (both at 0.3 m) elicited the largest response from this unit, whereas 0.01 m citric acid caused a moderate increase in activity. The firing rate of this cell is sparse, and it is also broadly tuned. The bottom right panel depicts the responses to all of the tastants across all trials. The unreinforced lick markers have been omitted from this panel for clarity. Symbols are the same as given in Figure 2.

Figure 4.

Chemosensory GC neurons are broadly tuned. A, A histogram showing the percentage of neurons responsive to a given number of tastants. It is seen that most neurons respond to four or five stimuli, indicating that GC neurons are broadly tuned. B, This graph depicts the percentage of neurons that responded to each tastant. As seen, most tastants elicit statistically significant responses from gustatory neurons. MSG, NaCl, sucrose, and citric acid elicit responses from approximately the same percentage of neurons, whereas fewer neurons respond to quinine (QHCl) and water.

Figure 5.

Non-chemosensory neurons that covary with licking. A, The evoked activity is shown to occur in oscillatory bursts 50–60 ms after each lick. This activity, however, is not sufficient to discriminate between reinforced and unreinforced licks (p > 0.4) or between tastants (p > 0.8). B, This excitatory activity shown was found to be sufficient to discriminate between reinforced and unreinforced licks (p < 0.0001), but not between tastants (p > 0.6). C, The elicited activity of this neuron was found to be sufficient to discriminate between reinforced and unreinforced licks (p < 4 × 10⁻²⁰), but not between tastants (p > 0.23). Here, it is seen that the presence of any tastant causes an inhibition of the firing rate.

Figure 6.

Anticipatory and state-dependent non-chemosensory neurons. A, Shown is a response where the neuron rapidly fires ∼20 ms before each lick but fails to discriminate between reinforced and unreinforced licks (p > 0.3). Neurons exhibiting this type of activity are termed “anticipatory.” B, An example of a neuron whose activity was “state dependent” in that its firing rate was modulated by the presentation of different tastants. The activity of this neuron failed to discriminate between reinforced and unreinforced licks (p > 0.9).

Figure 7.

GC neurons exhibit concentration-dependent responses. Shown is the response profile of a single neuron for four tastants at different concentrations and water. This neuron responded to all tastants but was unresponsive to 0.075 m MSG. Robust, temporally precise responses were elicited by 0.3 m MSG and by 0.075 m sucrose. Note also that an increase in MSG concentration results in an increase in firing rate, whereas an increase in sucrose concentration results in a decrease in firing rate.

Figure 8.

Comparison of evoked responses with the fitted values returned by the glm. The left panels present the observed firing patterns to each of the tastants, and in the right panels the best-fitted values are depicted as given by the glm. The responses to each of tastants are shown over a 150 ms window, which is the approximate interval between licks. The mean spike counts ± SEMs for each 15 ms bin are plotted on the ordinate, and the bin numbers are located on the abscissas. Each row of panels corresponds to the responses of the neuron with respect to different concentrations of a given tastant. The first row of panels corresponds to MSG, the second to NaCl, the third to sucrose, and the fourth to citric acid and water. For each panel, purple diamonds represent the responses to the highest tastant concentration, red squares represent the responses to the middle concentration, and green triangles indicate the responses to the lowest concentration. Note the good agreement between the observed data and the fitted values returned by the glm. The model contains 72.3% of the observed data within the 90% predictive interval. As seen in both the observed and fitted data, the time of maximal separation between 0.3 and 0.1 m MSG is at 105 ms, whereas the point of maximal separation between 0.075 m sucrose and the higher concentrations is at 75 ms.

Figure 9.

GC responses are sparse and exhibit temporally precise firing patterns. Shown is the tuning profile of another GC neuron across five tastants and some of their concentrations. All tastants elicited responses, but this neuron was unresponsive to 0.3 and 0.075 m MSG, whereas 0.1 m MSG did evoke a response. Although all three sucrose concentrations evoked responses, the maximal response is evoked by the middle concentration (0.1 m sucrose).

Figure 10.

Comparison of observed and fitted values for responses shown in Figure 9. These graphs represent the evolution of the response of the neuron to the tastants over the 150 ms window. The observed firing patterns are presented in the panels on the left, and the panels on the right depict the best-fit curves generated by the glm. As seen, there is a good overall agreement between the observed data set and the predictions of the model, because 88.5% of the observed data fell within the 90% confidence interval. Note that the firing pattern for 0.1 m MSG differs from 0.075 and 0.3 m only between 90 and 105 ms.

Figure 11.

Sucrose and quinine evoke responses in the same GC neuron. Shown are the responses of a neuron to four tastants at multiple concentrations. From the responses, it is evident that this neuron is responsive to both sucrose and quinine. The higher concentration of quinine evokes a larger response than that at 0.0001 m, whereas the intermediate concentration of sucrose elicits the largest response compared with the other concentrations. Note the temporal precision of the firing rate in response to 0.075 m MSG as well as to 0.075 m sucrose and 0.0003 m quinine.

Figure 12.

Comparison of observed and fitted values for responses shown in Figure 11. The observed responses of the neuron for each tastant are presented in the graphs on the left, and the right panels depict the fits for each tastant as predicted by the model. The model accounts for 63.0% of the observed data within the 90% confidence interval. Note that, for MSG, 0.1 m elicits a peak at 75 ms, whereas 0.075 m elicits a peak at 90 ms. Hence, a span of 15 ms can be used to discriminate between concentrations of tastants. Whereas the middle concentration of NaCl evokes the smallest response, the times to peak after each concentration of NaCl do not differ among each other. In contrast, different sucrose concentrations elicit peaks at differing times. For 0.025 m sucrose, the maximal firing rate occurs at 75 ms, whereas for 0.075 m and for 0.3 m, the maximal firing time occurs at 90 ms. Increases in the quinine concentration, however, lead to increases in firing rate without causing the peak response time to shift.