Gustatory neural coding in the monkey cortex: L-amino acids

1. Single-neuron activity in the primary gustatory cortex of the alert cynomolgus monkey (Macaca fascicularis) was analyzed in response to a range of taste stimuli. Tastants included the four prototypical stimuli (glucose, NaCl, HCl, and quinine), fruit juice, and 12 amino acids selected for their chemical characteristics, nutritional significance, and biological importance, as well as for the availability of human psychophysical data on their perceived qualities. 2. Taste-evoked responses could be recorded from a cortical area that measured 3.5 mm in its anteroposterior extent, 2.0 mm mediolaterally, and 6.0 mm dorsoventrally. Gustatory cells constituted 4.8% of the 1,129 neurons tested. Nongustatory cells gave responses associated with mouth movements (11.1%), somatosensory stimulation (3.8%), approach or anticipation of the taste stimulus (2.2%), and tongue extension (0.4%). 3. The most effective taste stimuli were those with qualities that humans describe as salty or sweet: NaCl, monosodium glutamate, glucose, proline, glycine, and fruit juice. The least effective tastants were those rated bitter or insipid: tyrosine, tryptophan, phenylalanine, and leucine. Accordingly, 79% of the gustatory neurons responded best to glucose (46%) or NaCl (33%) among the basic stimuli; only 19% responded best to quinine (13%) or HCl (6%). One cell (2%) responded exclusively to fruit juice. 4. Cortical gustatory neurons showed a moderate breadth of sensitivity, with a mean breadth of tuning coefficient of 0.71 across 54 cells. There was no evidence of chemotopic organization in the taste cortex. 5. The taste quality of each stimulus was inferred from the relative similarity of the profiles they evoked. The clearest distinction among stimuli was between those that humans characterize as sweet versus those with other qualities. Several amino acids that have dominant sweet (glycine and proline), salty (arginine and monosodium glutamate), sour (tryptophan), or bitter (phenylalanine) components to humans evoked activity profiles that were associated with those of the appropriate prototypical stimuli. Others (cysteine and lysine) were not closely related to any single prototype. 6. Conclusions based on the responses of cortical cells in the monkey are in close agreement with those that derive from human psychophysical studies of L-amino acids, reinforcing the value of this neural model for human taste perception.