Y1 receptors modulate taste-related behavioral responsiveness in male mice to prototypical gustatory stimuli

Abstract

Mammalian taste bud cells express receptors for numerous peptides implicated elsewhere in the body in the regulation of metabolism, nutrient assimilation, and satiety. The perturbation of several peptide signaling pathways in the gustatory periphery results in changes in behavioral and/or physiological responsiveness to subsets of taste stimuli. We previously showed that Peptide YY (PYY) – which is expressed in both saliva and in subsets of taste cells – can affect behavioral taste responsiveness and reduce food intake and body weight. Here, we investigated the contributions of taste bud-localized receptors for PYY and the related Neuropeptide Y (NPY) on behavioral taste responsiveness. Y1R, but not Y2R, null mice show reduced responsiveness to sweet, bitter, and salty taste stimuli in brief-access taste tests; similar results were seen when wildtype mice were exposed to Y receptor antagonists in the taste stimuli. Finally, mice in which the gene encoding the NPY propeptide was deleted also showed reduced taste responsiveness to sweet and bitter taste stimuli. Collectively, these results suggest that Y1R signaling, likely through its interactions with NPY, can modulate peripheral taste responsiveness in mice.

INTRODUCTION

Two discoveries in chemosensory research in recent years have realigned our thinking about how taste perception is linked to mechanisms of appetite and satiety. The first was the observation that many cells in the gut express the same molecular machinery required for nutrient detection as that found in taste cells. These gut “taste” receptors detect ingested nutrients and mediate the secretion of peptide hormones implicated in metabolism, nutrient assimilation, and satiety (see Calvo and Egan, 2015; Depoortere, 2014 for a review). The second was the recognition that many of these same peptides, together with their cognate receptors, are expressed in taste receptor cells of the peripheral gustatory system (see Cai et al., 2014b; Dotson et al., 2013 for a review). Together, these two sets of observations suggested a potential regulatory link between the gustatory system and post-ingestive signaling related to nutrient assimilation and metabolism. Indeed, the idea that peripheral taste functions may be modulated by factors associated with metabolic state is supported by a number of studies (e.g., Cai et al., 2014a; Chen et al., 2010; Duca et al., 2014; Ikeda et al., 2013; Maliphol et al., 2013; Sekine et al., 2012; Zhang et al., 2013; Zhou et al., 2009). Even so, the mechanisms by which diet and metabolic state may impact peripheral taste function remains unclear.

In addition to the prospect of hormonal modulation of taste responsiveness by circulating GI peptides, evidence also exists suggesting that cells in the peripheral gustatory system (e.g., taste bud cells, afferent nerve fibers) can be modulated by autocrine and/or paracrine mechanisms (e.g., Herness and Zhao, 2009; Kinnamon and Finger, 2019; Roper and Chaudhari, 2017; Shen et al., 2005; Yang et al., 2020; Yee et al., 2001). Receptors for peptides including GLP-1, glucagon, cholecystokinin, and vasoactive intestinal peptide, as well as the peptides themselves, are expressed in taste receptor cells or the afferent nerves that innervate the taste bud (Dotson et al., 2013; Elson et al., 2010; Herness et al., 2002; Martin et al., 2010; Shin et al., 2008). The anatomical proximity of both agonists and receptor suggest that there is likely paracrine/autocrine signaling in the peripheral gustatory system (see Dotson et al., 2013). Furthermore, many of these peptides and/or peptide receptors have been implicated in the regulation of taste cell physiology or taste behaviors (Brindisi et al., 2019; Cai et al., 2013; De Jonghe et al., 2005; Elson et al., 2010; Hajnal et al., 2005; Hajnal et al., 2007; Herness et al., 2002; Kolodiy et al., 1993; Martin et al., 2009; Martin et al., 2010; Martin et al., 2012; Shin et al., 2008; Swartz et al., 2010; Zhao et al., 2005). For example, we previously reported that peptide tyrosine tyrosine (PYY) is both expressed in taste cells and is transported as an endocrine hormone from circulation to the saliva of both humans and mice (Acosta et al., 2011; La Sala et al., 2013). Augmentation of salivary PYY_3–36 through genetic or pharmacological approaches affects behavioral taste responsiveness and reduces food intake and body weight in diet-induced obese mice (Acosta et al., 2011; Hurtado et al., 2013; La Sala et al., 2013).

NPY family peptides and their receptors have been strongly implicated in the regulation of energy homeostasis (Michel et al., 1998; Nguyen et al., 2011; Zhang et al., 2011). The NPY family consists of three 36-amino acid peptides: NPY, PYY, and pancreatic polypeptide (PP), all of which influence energy balance via their unique interactions with G-protein-coupled Y receptors (Y1, Y2, Y4, and Y5; Michel et al., 1998). Subsets of taste cells express all major Y receptors (Hurtado et al., 2012; Zhao et al., 2005).

Both PYY and NPY are expressed in TRCs (Zhao et al., 2005). NPY expression is restricted to a subset of taste cells, and application of the peptide to dissociated taste cells enhances an inward-rectifying K⁺ conductance (Zhao et al., 2005). Since both NPY and PYY are agonists of the Y1 receptor (Yulyaningsih et al., 2011), it is interesting to speculate whether Y1 receptor-mediated signaling in TRCs may be mediating the loss of behavioral responsiveness observed in animals that have had PYY-mediated signaling disrupted. Here, we detail experiments designed to test the hypothesis that the disruption of Y1 receptor-mediated signaling modulates behavioral responsiveness towards prototypical gustatory stimuli.

MATERIALS AND METHODS

This study was approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Florida. All procedures in the study were carried out in accordance with the principles of the National Research Council’s guide for the care and use of laboratory animals. The mice were housed at 22–24°C in a twelve-hour dark-light cycle with ad libitum access to water and food unless specified otherwise.

Behavioral assays

Mice

All mice used in behavioral experiments were 8–10 weeks old. The experimental groups (n=8–15) included (Npy1r ^–/–) (Howell et al., 2003) and (Npy2r ^–/–) (Sainsbury et al., 2002) mice and 129S-Npy^tm1Rpa/J mice (Jackson Laboratory, Bar Harbor, MA, USA). C57BL/6J (Jackson Laboratory, Bar Harbor, MA, USA) mice (n=10) served as controls for all experimental groups except for the 129S-Npy^tm1Rpa/J mice where the 129S1/SvImJ strain (n=8–15) were used as controls. The control strains were recommended by Jackson labs for each line. To contend with cohort effects, novel control groups were tested contemporaneously with all experimental groups. Mice were housed individually in standard cages with bedding. Mice were habituated to their environment for at least seven days before testing began.

Taste Stimuli

Tastants were prepared with purified water (Elix 10; Millipore, Billerica, MA, USA) and reagent grade chemicals. Tastants were presented to the mice at room temperature. Presented tastants include sucrose, NaCl, denatonium benzoate, and citric acid. Testing stimuli consisted of a “no stimulus” water control and 5 or 6 concentrations of each tastant: two different sucrose concentration ranges, a lower range was used for the NPY receptor knockouts and controls (25, 50, 100, 200, and 400 mM; Fisher Scientific, Atlanta, GA, USA) and a high range for the 129S-Npy^tm1Rpa/J mice and controls (62.5, 125, 250, 500, and 1000 mM); NaCl (30, 100, 300, 600, and 1000 mM; MilliporeSigma, Burlington, MA, USA); denatonium benzoate (DB; 0.05, 0.1, 0.5, and 1, 5 mM; Sigma-Aldrich); and citric acid (CA; 0.3, 1, 3, 10, 30, and 100 mM; Sigma-Aldrich).

The Y1R antagonist BIBO 3304 trifluoroacetate (1000 nM; Tocris Bioscience, Minneapolis, MN, USA) and the Y2R antagonist BIIE 0246 (1000 nM; (Tocris Bioscience, Minneapolis, MN, USA) were also included in some experiments.

Brief-Access Taste Testing

Brief-access taste testing was administered within a Davis Rig Gustometer (Davis MS-160; DiLog Instruments, Tallahassee, FL, USA; Smith, 2001) as previously described (e.g., Crosson et al., 2019; Dotson et al., 2005; Dotson and Spector, 2004, 2005, 2007; Elson et al., 2010; La Sala et al., 2013; Shin et al., 2008). Two testing protocols were used – one for preferred tastants and one for aversive tastants. Testing consisted of 25 min sessions during which mice were presented access to the sipper tubes for 5 sec with 7.5 sec inter-presentation intervals. The animals were first trained to lick a stationary tube of water for 30 min in the Davis rig after being placed on ∼23.5 h restricted water access schedule. Animals then received 2 days of testing with the sucrose stimulus arrays and purified water while maintained on the water-restriction schedule. This was done to familiarize the animals with the testing procedure and the stimulus array. The mice were subsequently tested for three consecutive days. During testing, mice were restricted to 1 g of food and 2 ml of water for 23.5 hours prior to testing.

For aversive stimuli, animals were trained to lick a stationary tube of water for 30 min in the Davis rig after being placed on ∼23.5 h restricted water access schedule. Animals then received 2 days of testing with a ‘water-only’ stimulus array while maintained on the water-restriction schedule. This was done to familiarize the animals with the testing procedure. The mice were subsequently tested for three consecutive days, under a ∼23.5 h restricted water access schedule, with one of the aversive stimulus arrays detailed above and purified water. Following presentation of an aversive tastant, mice were presented with a 1 sec H₂O rinse to minimize crossing over effects. If mice dropped below 85% of their starting body weight, they received 1 ml of water after the testing session was completed.

Statistical Analysis

For the normally avoided stimuli, the average number of licks per trial for each concentration was divided by that animal’s average licks per trial to water yielding a tastant/water lick ratio (Glendinning et al., 2002). To control for the low rate of water licking when animals are tested with normally preferred stimuli, a “tastant minus licks to water” difference score was also derived by taking the mean number of licks to water and subtracting it from the mean number of licks at each concentration (Jiang et al., 2008; Spector et al., 1996; Treesukosol et al., 2009). All scores were analyzed with analyses of variance (ANOVAs). The main variable of interest was genotype. If a main effect of genotype was not observed, genotype X concentration interaction were explored. If a significant interaction was observed, post hoc t tests were conducted to determine which concentrations differed between the experimental groups. The conventional statistic p ≤ 0.05 was applied as the statistical rejection criterion. Effect sizes were estimated using eta squared (η²) or Cohen’s d. Only mice that had at least one trial at every concentration were included in the analysis of a given stimulus. One Y1R KO was excluded from the analysis of denatonium responsiveness. One Y2R KO, one Y2R control, and one mouse tested with the Y1 receptor antagonist were excluded from the analysis of sucrose responsiveness in the deprived condition. Lastly, two NPY KOs were excluded from the analysis of sucrose responsiveness in the food and water restricted condition.

Curves were fitted to the mean data for each genotype using a logistic function of the form:

$$f\left(x\right)=\frac{a-d}{1+{10}^{({(\log (x)-c)}^{\ast }b)}+d}$$

where x log₁₀ concentration, c log₁₀ concentration at the inflection point, b slope, a the asymptotic lick ratio, and d minimum asymptote of lick ratio. These logistic functions help to quantify the differences in stimulus sensitivity between the groups.

RESULTS

Previously, we found that both Y1R and Y2R proteins are expressed in taste receptor cells (TRCs; La Sala et al., 2013). Thus, we next asked if deletion of either receptor would impact behavioral responsiveness to prototypical taste stimuli. When tested in the water-deprived condition, the response of the Y1R germline KOs did not differ from controls (Figure 1A). However, the Y1R KOs showed decreased responsiveness to sucrose as assessed by the tastant minus licks to water difference score [F(4,72) = 5.04, p = 0.001, interaction, η² = 0.07] (Figure 1B) when tested in the food and water restricted condition. Post hoc t-tests revealed that the Y1R KOs showed significantly decreased responsiveness to sucrose at the highest concentration [t(18) = 3.19, p = 0.005, Cohen’s d = 1.43].

Figure 1. Y1R KO mice show reduced behavioral taste responsiveness to sweet, bitter and salty-tasting stimuli.

Brief-access taste testing of Y1R KOs (n = 10; open circles) and WT mice (n = 10; solid circles) in response to (A) sucrose (after water deprivation; p = 0.63), (B) sucrose (after food and water restriction; p = 0.005, interaction), (C) denatonium benzoate (p = 0.006, interaction), (D) NaCl (p = 0.0009), and (E) citric acid (p = 0.50). For mice sampling sucrose after food and water restriction, a “tastant minus licks to water” difference score was derived by taking the mean number of licks to water and subtracting it from the mean number of licks at each concentration. For all other conditions and stimuli, the average number of licks per trial for each concentration was divided by that animal’s average licks per trial to water yielding a tastant/water lick ratio. Data are presented ± SEM. When a significant interaction was observed, stars indicate where that post hoc t-tests determined that a given concentration differed between the experimental groups.

When tested with normally avoided stimuli, Y1R KOs displayed a slight decrease in responsiveness to both the bitter-tasting stimulus denatonium benzoate [F(1,17) = 3.93, p = 0.006, interaction, η² = 0.01] (Figure 1C) and to sodium chloride [F(1,18) = 15.93, p = 0.0009, η² = 0.47] (Figure 1D). Post hoc t-tests revealed that the animals showed decreased responsiveness to denatonium benzoate at the two highest concentrations [1mM - t(17) = 2.35, p = 0.03, Cohen’s d = 1.08; 5mM - t(17) = 2.62, p = 0.02, Cohen’s d = 1.20]. No significant differences were seen between Y1R KOs and control mice (Figure 1E) when sampling citric acid. By contrast, behavioral responses of Y2R KO mice did not differ from controls for any stimuli (Figure 2A–F). The average number of test trials taken by all experimental groups is detailed in Table 1. Y1R KOs took fewer trials, relative to controls, when sampling all tested stimuli (Table 1).

Figure 2. Y2R KO mice do not differ from WT mice in behavioral responsiveness to prototypical taste stimuli.

Brief-access taste testing of Y2R KOs (n = 9; open circles) and WT mice (n = 10; solid circles) in response to (A) sucrose (after water deprivation; p = 0.34), (B) sucrose (after food and water restriction; p = 0.31), (C) denatonium benzoate (p = 0.87), (D) NaCl (p = 0.17), and (E) citric acid (p = 0.37). For mice sampling sucrose after food and water restriction, a “tastant minus licks to water” difference score was derived by taking the mean number of licks to water and subtracting it from the mean number of licks at each concentration. For all other conditions and stimuli, the average number of licks per trial for each concentration was divided by that animal’s average licks per trial to water yielding a tastant/water lick ratio. Data are presented ± SEM.

Table 1:

Average Number of Trials Taken During Testing

	Sucrose (WD)			Sucrose (F&W)			Den			NaCl			Citric Acid
Group	EG ± SD	CG ± SD	P value	EG ± SD	CG ± SD	P value	EG ± SD	CG ± SD	P value	EG ± SD	CG ± SD	P value	EG ± SD	CG ± SD	P value
Y1R KO	28.5 ± 6.1	63.8 ± 16.0	0.000004	35.2 ± 25.6	76.9 ± 25.5	0.002	54.9 ± 9.1	70.6 ± 10.1	0.002	85.6 ± 9.7	98.9 ± 8.4	0.005	56.6 ± 8.3	72 ± 12.1	0.004
Y2R KO	78.4 ± 25.6	80 ±14.9	0.87	140.8± 38.6	139.1± 22.7	0.91	53.1 ± 12.5	62.8 ± 9.3	0.07	80.2 ± 14.1	82.4 ± 8.1	0.68	57.2±16.6	60.± 9.9	0.66
Y1R A	71.8 ± 10.6	81.2 ± 18.4	0.18	101.± 29.2	107.2 ±31.3	0.66	53.6 ± 8.7	74.5 ± 13.8	0.0007	71.2 ± 6.8	67.6 ± 14	0.47	51.5 ± 5.8	51.6 ±10.1	0.98
Y2R A	69.7± 23.2	81.2 ± 18.4	0.26	79.2 ± 30.3	107.2 ±31.3	0.07
NPY KO	40.9 ± 13.1	41.6 ± 10.4	0.88	21.5 ± 13	29 ± 16	0.19	32.1 ± 6.9	35.4 ± 6.3	0.34	79.5 ± 15.5	68.1 ± 10.4	0.11	44.4 ± 14.5	48.3 ±12.7	0.58

To control for the possibility that the Y1R deletion is impacting behavioral taste responses through actions in extraoral tissues, we next assessed the effects of Y1R and Y2R receptor antagonists in the taste solutions. When testing the animals’ responsiveness to the antagonist when presented without the presence of a taste stimulus, animals did not respond to either antagonist differently than they responded to water [p = 0.54; Figure 3]. Mice receiving the Y1R antagonist BIBO 3304 trifluoroacetate in taste solutions showed decreased behavioral responsiveness to sucrose when tested in the water-deprived condition [F(1,17) = 12.42, p = 0.003, η² = 0.42] (Figure 4A). When tested in the food and water restricted condition, the response of these animals did not differ from vehicle-treated controls (Figure 4B).

Figure 3. The response of mice to the NPY receptor antagonists does not differ from their response to water.

Brief-access taste testing of WT mice receiving the antagonist BIBO 3304 trifluoroacetate (n = 10; solid circles) or the antagonist BIIE 0246 (n = 10; open circles). The response of mice to either NPY receptor antagonist did not differ from their response to water. The average number of licks per trial for each concentration was divided by that animal’s average licks per trial to water yielding a tastant/water lick ratio. Data are presented ± SEM.

Figure 4. A Y1R antagonist reduces behavioral taste responsiveness to sweet, bitter and salty-tasting stimuli.

Brief-access taste testing of WT mice receiving the antagonist BIBO 3304 trifluoroacetate (n = 10; open circles) or vehicle (n = 10; solid circles) in the taste stimulus solution: (A) sucrose (after water deprivation; p = 0.003), (B) sucrose (after food and water restriction; p = 0.56), (C) denatonium benzoate (p = 0.004), (D) NaCl (p = 0.00001, interaction), and (E) citric acid (p = 0.80). For mice sampling sucrose after food and water restriction, a “tastant minus licks to water” difference score was derived by taking the mean number of licks to water and subtracting it from the mean number of licks at each concentration. For all other conditions and stimuli, the average number of licks per trial for each concentration was divided by that animal’s average licks per trial to water yielding a tastant/water lick ratio. Data are presented ± SEM. When a significant interaction was observed, stars indicate where that post hoc t tests determined that a given concentration differed between the experimental groups.

When sampling normally avoided stimuli adulterated with the Y1R antagonist, mice displayed decreased responsiveness to denatonium benzoate [F(1,18) = 11.15, p = 0.004, η² = 0.38] (Figure 4C). These animals took few trials, relative to vehicle treated controls, when sampling denatonium (Table 1). Antagonist-treated mice also displayed decreased responsiveness to sodium chloride [F(4,72) = 8.88, p < 0.0001, interaction, η² = 0.42] (Figure 4D). Post hoc t-tests revealed that the animals showed decreased responsiveness to sodium chloride at the 600mM concentration [t(18) = 4.51, p = 0.0003, Cohen’s d = 2.02]. There was no significant difference in responsiveness to citric acid (Figure 4E). Consistent with the results observed with the Y2R KO mice, the Y2R antagonist BIIE 0246 did not affect responses to sucrose (Figure 5A and B; aversive stimuli not tested). These data demonstrate that adulterating stimuli with any receptor antagonist does not, obligatorily, impact upon behavioral responsiveness as assessed in the Davis rig.

Figure 5. A Y2R antagonist does not impact behavioral responsiveness to sucrose.

Brief-access taste testing of WT mice receiving the antagonist BIIE 0246 (n = 10; open circles) or vehicle (n = 10; solid circles) in the taste stimulus solution: (A) sucrose (after water deprivation; p = 0.81), (B) sucrose (after food and water restriction; p = 0.23). When mice sampled sucrose in the water restricted condition, the average number of licks per trial for each concentration was divided by that animal’s average licks per trial to water yielding a tastant/water lick ratio. For mice sampling sucrose after food and water restriction, a “tastant minus licks to water” difference score was derived by taking the mean number of licks to water and subtracting it from the mean number of licks at each concentration. Data are presented ± SEM.

Previously, we found that the hormone PYY is expressed in TRCs and that deletion of the gene encoding for the protein has impacts on taste-related behavioral responsiveness (Acosta et al., 2011; La Sala et al., 2013). NPY, a member of the NPY family of peptides along with PYY and pancreatic polypeptide, is also expressed in TRCs (Zhao et al., 2005). Thus, we next asked if deletion of the gene encoding NPY would impact behavioral responsiveness to prototypical taste stimuli. NPY germline KOs showed decreased responsiveness to sucrose [F(1,26) = 10.87, p = 0.003, η² = 0.30] (Figure 6B) when tested in the food and water restricted condition. When tested in the water-deprived condition, the response of the NPY KOs did not differ from controls (Figure 6A).

Figure 6. Npy null mice show reduced behavioral taste responsiveness to sweet and bitter, but not salty or sour, tasting stimuli.

Brief-access taste testing of 129S-Npy^tm1Rpa/J (n = 15 for sucrose, n = 8 for all other stimuli; open circles) and WT mice (n = 15 for sucrose, n = 8 for all other stimuli; solid circles) in response to (A) sucrose (after water deprivation; p = 0.18), (B) sucrose (after food and water restriction; p = 0.003), (C) denatonium benzoate (p = 0.009, interaction), (D) NaCl (p = 0.18), and (E) citric acid (p = 0.26). For mice sampling sucrose after food and water restriction, a “tastant minus licks to water” difference score was derived by taking the mean number of licks to water and subtracting it from the mean number of licks at each concentration. For all other conditions and stimuli, the average number of licks per trial for each concentration was divided by that animal’s average licks per trial to water yielding a tastant/water lick ratio. Data are presented ± SEM. When a significant interaction was observed, stars indicate where that post hoc t tests determined that a given concentration differed between the experimental groups.

NPY KOs also showed decreased responsiveness to the bitter-tasting stimulus denatonium benzoate [F(1,14) = 4.94, p = 0.04, η² = 0.26] (Figure 6C). No significant differences were seen between NPY KOs and control mice (Figure 6D and E) when sampling two other aversive stimuli, sodium chloride or citric acid.

DISCUSSION

We assessed the contribution of Y receptor signaling on taste behavioral responsiveness using germline Y receptor KOs, as well as with highly specific Y receptor antagonists. The latter approach helped to control for the possibility that the deletion of a Y receptor gene would indirectly impact taste behaviors through effects in extraoral tissues. Moreover, the approach also helped to control for possible developmental compensatory changes in hormonal signaling present in our germline KO mice. These highly specific Y receptor antagonists work with IC₅₀s in the low nM range in vitro (Dumont et al., 2000; Wieland et al., 1998). We conducted experiments flanking this range to select concentrations that do not appear to have a taste to mice.

Both Y1R germline KOs, as well as Y1R antagonist treated WT mice showed decreased responsiveness to sweet-, bitter-, and salty-tasting stimuli. By contrast, neither the Y2R KO mice nor mice receiving the Y2R antagonist differed from controls in their responsiveness for any of the tested stimuli. Interestingly, mice presented with the Y1R antagonist only showed decreased sucrose responsiveness in the water-deprived condition, whereas Y1R KOs display the deficit only in the food and water restricted condition. The source of this unexplained variation is unknown. While it is certainly possible that these differences result from random variation (e.g., technical issues), we speculate that this difference reflects the disruption of normal Y1R-mediated signaling in extraoral tissues in the KOs. For example, it is well known that Y1R is highly concentrated in the nucleus accumbens, a region involved in reward, motivation, and the regulation of palatable feeding (e.g., Hsieh et al., 2013; Kask et al., 1998; Skibicka et al., 2012; van den Heuvel et al., 2015; Wieland et al., 1998; Zheng et al., 2010). These results suggest that the interplay between motivation, sensory processing, and the central circuits that encode perceived food value can subtly impact upon the behavior of an animal. The relationship between these factors is likely complex and will require additional research to disentangle.

The disruption of Y1R-mediated signaling affected behavioral responsiveness in a taste-salient brief-access assay where post-ingestive influences on behavior are minimized (e.g., Nelson et al., 2003). The use of such an assay increases our confidence that the behavior we observed was being influenced by group differences in gustatory signals emanating from the periphery. However, since we know that Y receptor-mediated signaling can also influence appetite and motivation, it is possible that the observed differences in licking in these germline KO mice may be due to changes in the relative appetitiveness of the stimuli resulting from changes in the central nervous system.

Ingestive behavior can be subdivided into discrete segments. Appetitive behaviors are those that bring the animal towards a meal. Consummatory behaviors are those that are elicited after chemical stimuli comprising the meal bind with receptors in the oral cavity (see Craig, 1917; Sherrington, 1906). The Davis rig allows for some segregation of these behavioral components by providing a measure of trial initiation, which can be thought of as appetitive behavior, as well as by measuring unconditioned lick responsiveness, which can be thought of as consummatory behavior (e.g., Li et al., 2014; Loney and Meyer, 2018; Mathes et al., 2012; Mathes and Spector, 2011; Myers et al., 2020; Schier et al., 2019; Schier and Spector, 2016; Smith, 2001; Treesukosol et al., 2013). Not only did the Y1R KOs demonstrate a decrement in consummatory behavior, these animals also displayed a reduction in appetitive behavior indicated by the fact that the animal took few trials when sampling all of the tested stimuli. Collectively, our data are consistent with the known influences of CNS-mediated NPY related signaling on appetitive behavior (e.g., Ammar et al., 2005; Ammar et al., 2000; Sederholm et al., 2002; Seeley et al., 1995; Treesukosol et al., 2013). Interestingly, these same reports also detail a lack of impact of this signaling on consummatory behavior, suggesting that these types of behaviors are neurally dissociable and mediated by different brain circuits. For example, intracerebroventricular administration of NPY increases intake of a sucrose when presented in a one-bottle test, which involves both appetitive and consummatory behavior, but not when the sucrose is infused intraorally, a measure which focuses more on the consummatory behavior (e.g., Ammar et al., 2005; Ammar et al., 2000; Sederholm et al., 2002; Seeley et al., 1995). Indeed, the relative lack of impact that the Y1 receptor antagonist had on appetitive behavior and robust impact that it had on consummatory behavior is consistent with the notion that modulation of consummatory behavior is primarily controlled by orosensory signals emanating from the periphery (e.g., Ferrario et al., 2016; Fu et al., 2021; Rolls, 2005; Schier and Spector, 2019). It should be noted that some reports suggest that the influence of central NPY signaling on appetitive and consummatory behavior is more complex and varied (e.g., Baird et al., 2006; Torregrossa et al., 2006). Indeed, recent data suggest that these varied influence on appetitive and consummatory behaviors may be mediated via different NPY receptors (e.g., Y1R vs Y5R; see Keen-Rhinehart et al., 2013; Schneider et al., 2013 for a review). More research is needed to fully flesh out which centrally expressed receptors and signaling systems, in conjunction with orosensory signals, are influencing consummatory behavior.

Similar to Y1R KOs, NPY germline KOs showed decreased responsiveness to sucrose and denatonium benzoate. It is interesting to speculate that the changes in taste-related behavioral responsiveness seen upon disruption of Y1R-mediated signaling is due to the disruption of NPY-mediated signaling in taste buds. The slight difference in the behavior of these groups could be explained by the loss of NPY-mediated signaling in extra-oral tissues of the NPY germline KOs.

Numerous peptides implicated elsewhere in nutrient metabolism are expressed in TRCs, while their cognate receptors are expressed in taste cells or found in fibers of afferent taste nerves (see Dotson et al., 2013; Shigemura and Ninomiya, 2016 for a review). However, the function of these signaling systems in peripheral taste tissues remains poorly understood. It has long been hypothesized that output from the taste periphery may be modulated in the context of an animal’s metabolic state and nutritional needs (e.g., Cabanac, 1971; Cabanac and Fantino, 1977; Geerling and Loewy, 2008; Han et al., 2017; Kral, 2006; Lee et al., 2019; Richter, 1936; Scherr and King, 1982; Stellar, 1993; Stellar and Epstein, 1991). One purported mechanism for this phenomenon is that circulating gastrointestinal peptides modulate the functioning of the peripheral gustatory system. For example, it has been postulated that alterations in the levels of circulating gastrointestinal peptides mediate the changes in taste perception observed after gastric bypass surgery (e.g., Ahmed et al., 2018; Bueter et al., 2011; Cummings and Shannon, 2003; Le Roux et al., 2011; Münzberg et al., 2015).

The functional significance of peptides expressed as paracrine factors in TRCs is still poorly understood. The anatomical proximity of agonists and receptors suggest that these peptides may play a role in TRC functioning. Paracrine signaling by neuropeptides in TRCs may influence behavioral responsiveness to tastants by affecting cell-to-cell communication within the taste bud, thereby influencing the total output from taste buds to the CNS (see Herness and Zhao, 2009 for a review). We previously showed that the Y1 receptor is expressed in TRCs of the circumvallate papillae (CV) in mice (Hurtado et al., 2012; La Sala et al., 2013). Results from these studies do not provide clarity on whether the behavioral changes resulting from disruption of NPY and/or Y1R signaling reflect changes in endocrine or paracrine signaling. Herness and colleagues also showed expression of the Y1R, along with the NPY peptide, in TRCs of the rat CV (Herness and Zhao, 2009; Zhao et al., 2005). Herness found that the NPY peptide was expressed primarily in a subset of T2r-expressing, bitter tastant-responsive TRCs, while Y1R was found primarily in type 1 taste receptor, T1r2-expressing, sweet tastant-responsive TRCs. These results indicate that cells that express NPY are distinct from those that express Y1R, suggesting that NPY expressing cells may be influencing the response properties of taste buds by paracrine modulation of neighboring TRCs. However, predicting complex systemic responses from isolated cellular data can be risky. For example, the expression patterns of key molecular components that mediate taste transduction between rats and mice (e.g., Ma et al., 2007). These species also differ in their taste-related behavioral responsiveness (e.g., Kolodiy et al., 1993; Sclafani et al., 2010). More importantly, the experiments suggesting that Y1R activation leads to TRC hyperpolarization was done using CV taste buds. The response properties of TRCs from different papillae often differ substantially (e.g., Dana and McCaughey, 2015; Kim et al., 2003; Shingai and Beidler, 1985) and may contribute differentially to the functional aspects of taste (e.g., sensory discriminative vs. affective functioning; Spector, 2003 for a review; Spector and Glendinning, 2009). Consequently, the influence of a given molecular manipulation on taste-related behavior, which results from the processing of sensory input by the entirety of the gustatory system, may not be easily predicted by observing the output of individual TRCs. Lastly, the TRCs assessed by Zhao, Herness and colleagues were not tested in the presence of taste stimuli (Zhao et al., 2005). Indeed, it is most likely that NPY does not operate in isolation but coordinately other peptides and with small molecule neurotransmitters in neuromodulatory roles (Herness and Zhao, 2009).

Highlights.

• Mouse vallate taste buds express all four Y receptor subtypes (Y1R, Y2R, Y4R, Y5R).

• Disruption of Y1R signaling reduces sweet, bitter and salty tastant responsiveness.

• Npy null mice show reduced behavioral responsiveness to sweet and bitter stimuli.

• We conclude NPY-dependent Y1R signaling modulates murine peripheral taste responses.