TASTE RESPONSES OF NEURONS IN THE NUCLEUS OF THE SOLITARY TRACT OF AWAKE RATS - AN EXTENDED STIMULUS ARRAY

1. Fifty-seven taste neurons were isolated in the nucleus solitary tract (NST) and tested with 15 sapid chemicals. On average, NST neurons responded well to NaCl, sucrose, monosodium L-glutamate MSG), NaNO3, and glycine (mean = 8.2-11.0 spikes/s). Mean responses to KCl, NH4Cl, HCl, malic acid, and quinine HCl (QHCl) were low (mean = 0.7-2.9). The average responses to the other stimuli (citric acid, MgCl2, fruCtose, maltose, and polycose) fell between these extremes (mean = 4.3-5.1 2. On the basis of the largest response to the four standard stimuli, the neurons were classified as follows: 15 NaCl-best, 23 sucrose-best, 17 citric acid-best, and 2 QHCl-best. 3. The NaCl-best neurons responded robustly and nearly equally to the three sodium salts (mean = 15.7-20.8) but much less so and more variably to the nonsodium, chloride salts (mean = -0.1-4.6). Sucrose-best neurons responded strongly to sucrose, glycine, and MSG (mean = 13.7-17.8), but only moderately to the other sugars (fructose and maltose) and to polycose (mean = 8.4, 9.8, and 8.8, respectively). 4. Citric acid-best neurons responded moderately to citric and malic acid (mean = 9.4 and 4.7), but less so to HCl (mean = 3.1). The two QHCl-best neurons responded moderately to QHCI and MgCl2 (mean = 12.0 and 9.5), but weakly or not at all to the other stimuli (mean = 1.1-3.1). 5. Unlike parabrachial taste neurons, none of the medullary taste cells responded specifically to Cl--containing chemicals. The responses that did occur to nonsodium salts were weak and variable and often occurred in either citric acid-best or QHCI-best neurons, rather than in those that responded vigorously to sodium salts. Similar relationships have been observed in anesthetized preparations. 6. A hierarchical cluster analysis for 57 neurons across 15 stimuli produced four second-order clusters that consisted primarily of NaCl-best, sucrose-best, citric acid-best, and QHCI-best neurons, respectively. Although the analysis for neurons produced only four such clusters, a similar analysis for the 15 stimuli separated the sodium salts (NaCl and NaNO3), nonsodium salts (KCL, NH4Cl, and MGCl2), sweeteners (sucrose, maltose, fructose, and glycine), acids (citric acid and malic acid), and QHCl. 7. Monosodium glutamate activated both NaCl-best and sucrose-best neurons, but the stimulus analysis clumped it with the sodium salts. Polycose was as potent a stimulus for sucrose-best neurons as were maltose and fructose, but the stimulus analysis segregated it from the four chemicals that taste sweet to humans. On average, the responses to HCl exhibited no similarity to those of the other stimuli, including the organic acids. 8. The correlation coefficients between stimuli across 57 neurons reinforced the organization imposed by the cluster analysis without exactly duplicating it. The responses to the sodium salts were significantly correlated with one another, as were the responses of the sweet stimuli to each other and to polycose. Monosodium glutamate was correlated with both of these stimulus groups. The responses to citric acid were correlated significantly only with those of malic acid and NH4Cl; those of the nonsodium salts were not always significantly correlated with each other. The responses to HCl and QHCl were not correlated with any of the other stimuli. 9. In a two-dimensional space, multidimensional scaling of neuronal responses to the 15 stimuli produced a ringlike arrangement that was essentially the same as that produced from a previous sample using only four standard stimuli. When the stimuli were scaled similarly, however, the resulting space required three dimensions to account for 80% of the variance. This resulted largely from the three nonsodium salts and HCI being separated from the other stimuli in the third dimension as well as the first two. 10. A principal factor analysis using Pearson's correlations between the 15 stimuli across the responses of 57 NST neurons identified five stimulus factors-one for sweeteners, another for sodium salts, a third for the two organic acids, a fourth for the nonsodium salts, and a fifth for HCl. Monosodium glutamate loaded on both the first and second factors; polycose on both the first and fifth. Quinine also loaded on factor 5, but its sign was negative, whereas those for HCl and polycose were positive. The five factors accounted for 98.7% of the variance among the 15 stimuli.