Taste as a highly discriminative system: a hamster intrapapillar single unit study with 18 compounds

Seventy-nine single units were recorded with glass micropipettes applied onto taste pores of the anterior portion of the tongue in anesthetized hamsters. The receptive field of each recorded unit was located by iontophoretic stimulation applying either anodal or cathodal current (1-20 microA) to a non-stimulating solution of sodium cyclamate (5 mM). Different kinds of responses to iontophoretic stimulation were described. Eighteen chemical stimuli including sweet and bitter tastants for humans, and amino acids were locally applied to one papilla of the receptive field of 63 of these units. Stimulations were applied in a continuous flow (30 ml/min) in a small chamber. Response criterion was chosen as 2 S.D. above the mean activity recorded during the minute preceding the stimulus arrival. The low amplitude of single unit responses to chemicals is discussed by reference to the recording and stimulating techniques and compared to results of control experiments on whole nerve recordings and psychophysical experiments on human subjects. The importance of a high flow rate during continuous flow stimulation was demonstrated. The possible necessity of mechanical stimulation to facilitate taste responses was outlined. The sensitivity of units to stimuli applied either chemically or iontophoretically was not identical. Contrary to authors expectations, localized sensitivities for a few specific chemicals were disclosed. The response reproducibility to chemical stimulation was 84% (S.D. = 1.6%). chi 2 calculation, correspondence analysis and hierarchical clustering showed small distances between the two profiles representing the same stimulus or the two profiles representing the same unit, but great distances between profiles representing either different stimuli or different units. All stimuli are different from one another. Only 6 pairs of similar units were found among 63 units. Unit clusters could be found only if a few stimuli were considered but they vanished when all 18 stimuli were used in the calculation. The breadth of tuning of units ranged from 0.25 to 0.92 with a mean of 0.68. We show that the peripheral taste system is highly discriminative. Each stimulus evokes a distinct sensory image. At least 8 independent factors are needed to describe the peripheral taste space, setting a lower limit to the number of different peripheral information channels (putative acceptors) involved.