Sweet-bitter and umami-bitter taste interactions in single parabrachial neurons in C57BL/6J mice

Abstract

We investigated sweet-bitter and umami-bitter mixture taste interactions by presenting sucrose or umami stimuli mixed with quinine hydrochloride (QHCl) while recording single-unit activity of neurons in the parabrachial nucleus (PbN) of urethane-anesthetized C57BL/6J mice. A total of 70 taste-responsive neurons were classified according to which stimulus evoked the greatest net response (36 sucrose-best, 19 NaCl-best, 6 citric acid-best, and 9 QHCl-best). Although no neurons responded best to monopotassium glutamate (MPG) or inosine 5'-monophosphate (IMP), the combination of these two stimuli evoked a synergistic response (i.e., response > 120% of the sum of the component responses) in all sucrose-best and some NaCl-best neurons (n = 43). Adding QHCl to sucrose or MPG + IMP resulted in suppression of the response (responses to mixture < responses to the more effective component) in 41 of 43 synergistic neurons. Neurons showing QHCl suppression were classified into two types: an "MS1" type (n = 27) with suppressed responses both to sucrose and MPG + IMP and an "MS2" type (n = 14) that showed suppressed responses only to sucrose. No neuron displayed suppressed responses to MPG or IMP alone. The suppression ratio (1 - mixture response/sucrose or MPG + IMP response) of sucrose and MPG + IMP in MS1 neurons had a weak positive correlation (r = 0.36). The pattern of reconstructed recording sites of neuron types suggested chemotopic organization in the PbN. Although a peripheral basis for QHCl suppression has been demonstrated, our results suggest that convergence in the PbN plays a role in shaping responses to taste mixtures.

Fig. 1.

Response profiles of parabrachial nucleus (PbN) taste neurons. Taste neurons were grouped into best-stimulus categories (gray bars) and arranged within those categories in descending order of response magnitude to the best stimulus [n = 70: sucrose (S)-best, 39; NaCl (N)-best, 19; citric acid (C)-best, 9; quinine hydrochloride (QHCl, Q)-best, 9]. Taste responses are presented as net responses (i.e., responses to stimulus − responses to water). Hatched bars indicate synergistic responses to the mixture of monopotassium glutamate (MPG) and inosine 5′-monophosphate (IMP). Inverted triangles are associated with bars when the responses evoked by the mixture were smaller than those evoked by the more effective component (MEC).

Fig. 2.

Mean (±SE) net taste responses of synergistic (shaded bars, n = 43) and nonsynergistic (open bars, n = 27) neurons to all taste stimuli used in the present study. *P < 0.05. **P < 0.01.

Fig. 3.

Examples of responses of 3 types of neurons to basic and mixture taste stimuli. A: a synergistic neuron whose responses to both sucrose and MPG+IMP were suppressed by addition of 0.05 M QHCl (mixture suppression type 1, MS1). B: a synergistic neuron whose responses to sucrose but not MPG+IMP were suppressed by QHCl (mixture suppression type 2, MS2). C: a nonsynergistic neuron that showed no QHCl suppression. Arrowheads indicate stimulus onset. Scale bars, 1 s.

Fig. 4.

Responses of neurons that showed taste suppression by QHCl (MS1 and MS2 type, n = 41; A) and neurons that did not show QHCl suppression (n = 29; B). The order of neurons is the same across panels and arranged based on their magnitude of responses to sucrose. Data from the only neuron that showed QHCl suppression by sucrose are not included in this figure.

Fig. 5.

Mean (±SE) net taste responses of MS1 (shaded bars, n = 27)- and MS2 (open bars, n = 14)-type neurons to all taste stimuli used in the present study. *P < 0.05.

Fig. 6.

Effect of QHCl on taste responses to sucrose and MPG+IMP in M1-type neurons. A: mean responses to sucrose and MPG+IMP with or without QHCl added. Responses to both stimuli were significantly suppressed by QHCl. QHCl's suppressive effect was greater with sucrose than with MPG+IMP, although there was no significant difference in response magnitude when these stimuli were presented alone. B: comparison of suppression ratio. Sucrose suppression ratio was significantly greater than that of MPG+IMP. C: correlation between sucrose and MPG+IMP suppression ratio. There was a weakly positive correlation (r = 0.36). *P < 0.01.

Fig. 7.

Distribution of 14 taste stimuli in a 3-dimensional taste space resulting from multidimensional scaling. Stress values were 0.16405 for 1 dimension, 0.07809 for 2 dimensions, 0.01161 for 3 dimensions, 0.00122 for 4 dimensions, and 0.00022 for 5 dimensions. S, sucrose; N, NaCl; C, citric acid; Q, QHCl; G, glucose; L, LiCl; P, Polycose; MIX, MPG+IMP. Shaded circles indicate mixtures with QHCl.

Fig. 8.

Anatomical reconstruction of 70 recording sites in the right PbN. A–D: coronal sections are arranged rostral to caudal and +50, −100, −250, and −400 μm separated from the caudal end of the cuneiform nucleus, respectively. Filled symbols, synergistic units; open symbols, nonsynergistic units; circles, S-best units; squares, N-best units; triangles, C-best units; inverted triangles, Q-best units. BC, brachium conjunctivum; cl, central lateral subnucleus; dl, dorsal lateral subnucleus; dm, dorsal medial subnucleus; el, external lateral subnucleus; em, external medial subnucleus; il, internal lateral subnucleus; LC, locus coeruleus; m, medial subnucleus; Me5, mesencephalic trigeminal nucleus; vl, ventral lateral subnucleus. E: photomicrograph of a cresyl violet-stained section in the PbN. Marking lesion in the ventral lateral subnucleus is indicated by an arrowhead. Scale bars, 500 μm.