Temperature effects on taste preferences are influenced by TRPM8

Abstract

Taste perception is influenced by stimulus and oral temperature. Change in oral temperature modulates trigeminal neurons. Whether trigeminal thermal sensing interacts with taste is unknown. Here we studied temperature influences on mouse taste preferences and how they could change following silencing of TRPM8 (transient receptor potential melastatin 8) – a thermoreceptor supporting cool and warm temperature coding by trigeminal neurons. Female and male TRPM8 gene deficient and C57BL/6J (B6) control mice (n = 69) entered thermolickometry tests where they sampled taste solutions at cool (15°C) and warm (30°C) temperatures during brief-access (10-sec long) exposure trials, which capture oral sensory/tongue control of licking behavior. Taste solutions included innately avoided bitter quinine (0.03 and 0.3 mM) and preferred sugars (sucrose or glucose, 100 and 500 mM). Mice were respectively maintained under water restriction or water-replete conditions during quinine and sugar tests, which were conducted separately. Analyses revealed that 15°C enhanced, while 30°C reduced, licks to quinine in both B6 and TRPM8 deficient mice, which responded similarly (p > 0.05). In contrast, licks to static concentrations of sugars trended towards enhancement by 30°C, compared to 15°C, in male B6 mice but were suppressed, and inhibited, by 30°C (p < 0.05) in male TRPM8 deficient mice. This result agrees with prior studies that show warmth normally facilitates sweetness and that 30°C stimulation of oral tissues becomes anomalously aversive in TRPM8 deficient mice. These data provide initial evidence that TRPM8 thermosensory influences interact with sugar taste preferences, which may reflect a trigeminal-taste cross-modal phenomenon.

Introduction

Taste sensation is implicated in the detection of nutrients and chemical toxins in ingesta. The neural processing of taste and gustatory perception can be strongly affected by taste stimulus and oral temperature. For example, rodent peripheral and central gustatory neural responses to oral presence of sucrose (sweet to humans) are markedly enhanced by innocuous oral warming and suppressed by cooling (e.g., Talavera et al., 2005^; Breza et al., 2006^; Wilson and Lemon, 2014^; Lu et al., 2016). Relatedly, warm (30°C) fluid temperatures can facilitate orosensory-guided licking of sucrose solutions in rats compared to cold (10°C) (Torregrossa et al., 2012). Innocuous warming of taste solutions and lingual epithelia can also increase the perceived gustatory sweetness of sugars in humans compared to cooling (Bartoshuk et al., 1982; Calvino, 1986; Green and Frankmann, 1987; Nachtigal and Green, 2020).

Temperature effects on sugar taste responses arise, in part, from the direct actions of temperature on taste receptors. Type II taste bud cells, which initiate the detection of sweet tastes, utilize transient receptor potential (TRP) melastatin 5 (TRPM5) ion channels to support depolarization and transmitter secretion (Huang and Roper, 2010). TRPM5 is activated by warm temperatures, with genetic knockout of TRPM5 impairing the ability of warming to facilitate gustatory nerve responses to sweet taste solutions in mice (Talavera et al., 2005). Additionally, warmth applied to the tongue can elicit a thermal-induced sweet taste perception in humans (Cruz and Green, 2000) that is blocked by pharmacological inhibition of the sweet taste receptor TAS1R2/TAS1R3 (Nachtigal and Green, 2020).

While engaging taste processes, comparative data show that changes in oral temperature more strongly modulate thermosensory neurons of the trigeminal system (Lemon et al., 2016). Trigeminal neurons supply somatosensation to craniofacial and intraoral tissues and are insensitive to oral presence of prototypical taste chemicals (Simons et al., 2003), including under fluid temperature-controlled conditions (Lemon et al., 2016). Primary trigeminal fibers express thermo-TRP channels implicated in pain-related and thermosensory transduction. These include the capsaicin- and heat-sensitive ion channel TRP vanilloid 1 (TRPV1) (Caterina et al., 1997) and the cool and menthol receptor TRP melastatin 8 (TRPM8) (McKemy et al., 2002; Bautista et al., 2007).

TRPM8 contributes to responses to temperature drops below 30°C (McKemy et al., 2002; Tajino et al., 2011). Furthermore, the actions of TRPM8 are evidenced to influence warmth perception, as silencing TRPM8 impairs cutaneous detection of warm temperatures (Paricio-Montesinos et al., 2020) and alters orosensory hedonic responses to warm fluids. For instance, TRPM8 gene deficient mice avoid warm 30°C water in favor of cool (e.g., 15°C) water in brief-access thermolickometry tests (Li et al., 2024), which capture the influence of oral/tongue thermosensation on fluid licking while mitigating post-ingestive cues (Torregrossa et al., 2012; Li et al., 2024). In contrast, wild-type mice display indifference to, and avidly lick, both 30° and 15°C water (Li et al., 2024). This difference between mouse lines implies the normal development of warmth orosensory hedonics partly relies on TRPM8 input (Li et al., 2024).

Here we used brief-access thermolickometry with TRPM8 deficient and control mice to study whether TRPM8-mediated oral temperature sensing interacts with taste-guided fluid licking and ingestive behavior. Whether trigeminal thermal signals affect taste perception and behavior was hitherto unknown. Results showed that while both TRPM8 deficient and control mice increased their licking acceptance of an innately avoided bitter tastant when cooled, TRPM8 deficient mice showed uniquely reduced orosensory responses to normally preferred sugar fluids when warmed. These data suggest that input from TRPM8 on trigeminal fibers contributes to an oral thermosensory context that interacts with the sensory appeal of sugar (sweet) taste. Our results represent an initial account of how thermosensory information mediated by TRPM8 may interact with taste sensation and hedonics.

Methods

Mice

All experiments and procedures were approved by the University of Oklahoma Institutional Animal Care and Use Committee and followed the Guide for the Care and Use of Laboratory Animals by the National Research Council. These studies used adult female and male C57BL/6J (B6; strain 000664, The Jackson Laboratory [JAX], Bar Harbor, ME) and homozygous TRPM8 deficient/knockout (JAX strain 008198) mice. TRPM8 deficient mice are derived from backcrosses with C57BL/6 mice (JAX). Sixty-nine mice entered behavioral tests. At the onset of behavioral training, mice were 17.5 (mean) ± 5.9 (SD) weeks of age. On average, female B6 mice (n = 17) weighed 19.7 ± 1.3 g and males (n = 18) 26.6 ± 2.0 g. Female TRPM8 deficient mice (n = 17) weighed 20.8 ± 1.6 g and males (n = 17) 28.7 ± 2.3 g.

Prior to experiments, mice were group housed in disposable cages with ad libitum access to standard mouse chow and filtered water. Once the experiment started, each study mouse was singly housed in a disposable cage, with ad libitum food access but water availability regulated, as below. Mouse cages were housed in a ventilated (HEPA filtered) rack system (Innovive, San Diego, CA) located in a climate-controlled (~22°C ambient temperature) room maintained on a 12 hr light/dark cycle. Individual cages were prepared with standard bedding and enrichment materials (e.g., paper huts, cardboard tubes). Cage bedding was changed on a regular schedule prior to experiments, and only when needed during studies to mitigate housing disruptions.

Experimenter blinding

Prior to experiments, a unique alphanumeric code was assigned to each study mouse and used to label their cage in lieu of all other identifying information. Cage locations in the colony housing rack were then randomized. These procedures were performed by lab personnel not involved with data collection, which blinded the experimenter handling and running mice to mouse strain (i.e., B6 or TRPM8 deficient). All mice had black coats and indistinguishable outward appearances, facilitating blinding. Blinding was used to collect behavioral data across all training and test sessions. The experimenter was unblinded to the squad undergoing tests only after all data collection was complete.

Water restriction schedule

An overnight water restriction schedule was imposed on mice the day before their lickometer training commenced. This schedule aimed to motivate mice to lick fluids offered during brief-access fluid exposure training sessions. To do this, the water bottle for each home cage was removed and replaced with a marble-weighted bottle with no water. This weighted bottle blocked the water access hole on the cage top. Water restriction conditions continued through tests with quinine. For sugars, water restriction was used during initial sugar orientation sessions but not during sugar tests, as below. Mice consumed their daily fluids in a lickometer when performing in training and test phases under water restriction. During all study phases, individual mice were given an additional 45-min access to water in their home cage following a brief-access session if their daily measured body weight fell below about 80% of their baseline weight.

Lickometer apparatus

Lickometer training sessions were partly carried out using standard “Davis Rig” contact lickometers (DiLog Instruments, Tallahassee, FL; Med Associates, St. Albans, VT). The Davis Rig can record licking responses made by a mouse to different fluids offered on sequential, seconds-long trials while randomizing the order of fluid presentation within each test session. The short trial length and limited number of fluid trials offered during brief-access tests mitigated post-ingestive influences on licking responses, capturing ingestive/licking behavior driven by initial oral sensation (Smith, 2001; Boughter et al., 2002).

Mouse licking responses to temperature-controlled fluids were measured using a custom-modified Davis Rig, as described (Li et al., 2024). Briefly, our “thermolickometer” functioned like a normal Davis Rig contact lickometer, capturing lick rates to different fluid sipper tubes randomly offered to mice on discrete trials, while also holding the temperature of each fluid at a different set point value. During development, this device was established to maintain actual fluid temperatures within 0.2°C SD of set point temperatures.

Lickometer training

All brief-access tests were preceded by a training period, which oriented mice to the apparatus and brief-access procedure. Overnight water-restricted mice were individually trained to receive fluids in a standard Davis Rig over four days. On training days 1 and 2, mice were allowed free access to one sipper tube filled with room temperature purified water for 30 min to habituate them to receiving fluids in the rig chamber. The 30 min session started with the first lick on the sipper tube. Days 3 and 4 familiarized mice with the brief-access fluid exposure procedure, offering them different sipper tubes of room temperature purified water over 20, 10-sec access trials. Once a tube was presented, mice were allowed 30 sec to make a lick, which started the trial. Zero licks were recorded if no licks were made after 30 sec. Inter-trial intervals were 10 sec. After completing the last day of training, mice were returned to their single-housing cage and given ad libitum access to water for approximately two days, prior to starting tests.

Brief-access fluid exposure tests

After this break, mice returned to an overnight water-restriction schedule and entered brief-access fluid exposure tests with temperature-controlled taste stimuli, conducted in the thermolickometer. All chemical stimuli were dissolved in purified water. Three taste stimuli were tested: the bitter taste stimulus quinine-HCl (quinine), the disaccharide sucrose, and the monosaccharide D-glucose (glucose). Quinine, sucrose, and glucose were tested separately, during different weeks, and on different mice. During each daily brief-access test session with each taste stimulus, mice were proffered two different solutions that differed in stimulus concentration and/or fluid temperature, as follows.

Quinine.

Two squads of water-restricted (thirst-motivated) mice received six different quinine concentration/fluid temperature test conditions across six sequential days (Table 1). One test was performed per day, with test order randomized for each squad.

Table 1.

Stimulus concentrations and temperatures used in two-fluid brief-access thermolickometry tests with quinine.

Test	Fluid 1: concentration, temperature	Fluid 2: concentration, temperature
1	0.03 mM (low), 15°C (cool)	0.03 mM (low), 30°C (warm)
2	0.3 mM (high), 15°C (cool)	0.3 mM (high), 30°C (warm)
3	0.03 mM (low), 30°C (warm)	0.3 mM (high), 15°C (cool)
4	0.03 mM (low), 15°C (cool)	0.3 mM (high), 30°C (warm)
5	0.03 mM (low), 15°C (cool)	0.3 mM (high), 15°C (cool)
6	0.03 mM (low), 30°C (warm)	0.3 mM (high), 30°C (warm)

The two fluids used in each test represented different combinations of a low (0.03 mM) and high (0.3 mM) concentration of quinine, and cool (15°C) and warm (30°C) solution temperatures. In one test, the low quinine concentration was proffered at cool (solution 1) and warm (solution 2) temperatures across brief-access trials (Table 1, Test 1). Another test repeated this procedure but using the high quinine concentration (Test 2). Two additional tests proffered mice both quinine concentrations, but the high concentration was cool (Test 3) or warm (Test 4) relative to the low. Finally, two tests presented both low and high quinine at a constant cool (Test 5) or warm (Test 6) fluid temperature. During each test, the two solutions were individually offered on 10-sec trials, with solution order randomized within each of 10 contiguous trial blocks (20 trials total; mice were allowed up to 100 sec of licking access to each solution per session). Each quinine test condition was presented once, for a total of six brief-access sessions.

Sucrose and glucose.

Two squads of mice were tested with sucrose and two additional squads with glucose; mice received either sucrose or glucose, but not both. Brief-access fluid exposure procedures with sugars were performed in two phases. Phase 1 was an orientation session to allow mice to learn that the sipper tubes offered sugar solutions, which are strongly preferred. Here, overnight water-restricted mice were proffered room temperature solutions of 300 mM sucrose or glucose in two brief-access exposure sessions conducted over 2 consecutive days. During each session, mice were allowed to lick either fluid over 20, 10-sec long trials.

Once phase 1 completed, mice were returned to their home cage and given unrestricted access to water. They then entered phase 2, which began the next day. Phase 2 monitored brief-access licking responses to temperature-controlled solutions of sucrose or glucose in water-replete (not thirst-motivated) mice with continuous access to water in their home cage. Mice also had free access to food in their home cage, so it was presumed that the preferred hedonic tone of the sugars, but not thirst or hunger, was the primary motivator of licking in phase 2.

Each test offered mice two temperature-controlled fluids containing one of the sugars. Solutions represented different combinations of low (100 mM) and high (500 mM) sugar concentrations and cool (15°C) and warm (30°C) solution temperatures (Table 2); these conditions were analogous to those used in quinine tests. Sugar test fluids were singly presented in discrete 10-sec trials, with fluid order randomized within 10 contiguous trial blocks (20 trials total; mice were allowed up to 100 sec of licking access to each solution per day). Each sugar test condition was repeated over three consecutive days, for a total of 18 brief-access test sessions. This aimed to allow sufficient data collection using non-thirst-motivated mice. Test condition order was randomized for each mouse squad. All mice within a squad experienced the same sugar (sucrose or glucose) and order of test conditions.

Table 2.

Stimulus concentrations and temperatures used in two-fluid brief-access thermolickometry tests with sugars (sucrose or glucose).

Test	Fluid 1: concentration, temperature	Fluid 2: concentration, temperature
1	100 mM (low), 15°C (cool)	100 mM (low), 30°C (warm)
2	500 mM (high), 15°C (cool)	500 mM (high), 30°C (warm)
3	100 mM (low), 30°C (warm)	500 mM (high), 15°C (cool)
4	100 mM (low), 15°C (cool)	500 mM (high), 30°C (warm)
5	100 mM (low), 15°C (cool)	500 mM (high), 15°C (cool)
6	100 mM (low), 30°C (warm)	500 mM (high), 30°C (warm)

The low and high stimulus concentrations tested here elicit concentration-dependent effects on licking behavior in B6 mice (Boughter et al., 2005; Ascencio Gutierrez et al., 2024). The tested fluid temperatures increase (15°C) or reduce (30°C) water licking in TRPM8 deficient mice but do not affect water licking in B6 mice during brief-access tests conducted under water restriction (Li et al., 2024).

Data analysis and statistics

The number of licks mice emitted on each brief-access trial were calculated by adding 1 to the number of inter-lick intervals that were >50 msec. This criterion removed potential false licks attributable to electrical/RF noise. Trials with 1 or 2 licks were considered non-sampled and zeroed. The mean number of licks each mouse made to a stimulus per trial, including trials captured across multiple test days for sugars, was then calculated. Mean licks per trial is referred to herein as “licks” for convenience.

Statistical analyses coupled inferential with estimation statistics to study the significance and magnitude of differences in licking observed between stimulus conditions, mouse groups, and sexes. Factorial ANOVAs were conducted using R (Team, 2022). Some follow up analyses used Yuen’s dependent samples trimmed mean t-tests (WRS2 R package; Mair and Wilcox, 2020) for robust analysis of effects. Yuen’s trimmed t retained data magnitude information (e.g., licks) as opposed to converting data points to nonparametric ranks, accommodated unequal group variances and outliers that arose in some comparisons, and provides only slightly less power than a Student’s t-test under normality (Yuen, 1974). Ten percent trimming was used. Statistical effect size in Yuen’s test was gauged by $\widehat{\epsilon }$, where 0.10, 0.30, and 0.50 correspond to small, medium, and large effects, respectively (Mair and Wilcox, 2020). Sample sizes for specific tests are detailed in the results.

For estimation statistics, a custom bootstrap method performed in MATLAB (release 2024b, MathWorks, Natick, MA) estimated the 95% confidence interval of the mean difference between two compared groups (i.e., the effect size in lick response units). To do this, each data group was randomly resampled with replacement, with the number of resampled data points equal to the group sample size. A 10% trimmed mean was then calculated. The difference between the resampled group means was stored. This process was repeated 1,000 times. The calculated differences were then ranked to identify the 25th (2.5%) and 975th (97.5%) entries, which defined the 95% confidence interval range for the mean difference between groups. This calculation was repeated 100 times, with the average 95% confidence interval reported in the results as [low value, high value].

Gardner-Altman plots

Gardner-Altman plots (Gardner and Altman, 1986) custom coded in MATLAB were used to visualize the outcome of estimation analyses (Lemon et al., 2025). These plots represented the effect size difference between two groups in the context of effect probability, data unit of measurement, and error/group variances. In the figures, each panel displays a Gardner-Altman plot of licking responses. The left y-axis indexes licks while the right y-axis captures the effect size – the licks difference between the two stimulus conditions represented. On the plot, licks by individual mice are depicted as circles for each stimulus condition. Mean (10% trimmed) licks (horizontal bar) and the 95% confidence interval for the mean (vertical bar) are plotted to the right of each distribution of licks. The difference between the sample means (the effect) is shown to the right of the plot (filled circle) along with its 95% confidence interval (vertical bar). The expected sampling error for the mean difference is represented as a shaded probability distribution. The effect size is stated by a real number, with the 95% confidence interval given in brackets as [low value, high value].

Results

Taste fluid warming enhances bitter taste avoidance in mice

Using brief-access thermolickometry, we measured orosensory responses to temperature-controlled quinine solutions from 12 B6 (6 females, 6 males) and 11 TRPM8 deficient (5 females, 6 males) mice. Mice were maintained on a water restriction schedule to motivate licking during quinine tests. These mice were tested with only quinine and were not included in brief-access tests with sugars to mitigate carryover effects.

Quinine concentration constant, fluid temperature varied

We first asked if cooling (15°C) and warming (30°C) a singly tested low (0.03 mM) or high (0.3 mM) concentration of quinine would change mouse licking acceptance of this fluid. Tests (1 and 2, Table1) conducted with each concentration on separate days revealed that both low and high quinine solutions evoked significantly fewer licks when warmed (main effect of temperature, F_1,18 = 32.8, p = 0.00002). This effect was not influenced by TRPM8 genotype, sex, or quinine concentration (n.s. mouse line × sex × quinine concentration × temperature interaction, p = 0.163). Collapsing data across B6 and TRPM8 deficient mice revealed that responses to 0.03 mM quinine were, on average, 22.6 [−33.2, −12.1] licks lower at 30°C than at 15°C (Figure 1A, t₁₇ = 4.9, p = 0.0001; $\widehat{\epsilon }=0.73$). Licking to 0.3 mM quinine was reduced by 21.1 [−33.7, −8.6] licks at 30°C (Figure 1B, t₁₇ = 4.8, p = 0.0002; $\widehat{\epsilon }=0.67$). The effect size was large ($\widehat{\epsilon }>0.5$) for these drops.

Figure 1.

Warming taste fluids increases orosensory avoidance of static concentrations of quinine. TRPM8 deficient and B6 mice showed fewer licks to 0.03 (A) and 0.3 (B) mM quinine solutions proffered at 30° compared to 15°C (p < 0.05). See methods section on Gardner-Altman plots for details on data presentation.

Quinine concentration and fluid temperature varied

We next asked if taste fluid cooling and warming would affect concentration-dependent avoidance of quinine, where mice emit fewer licks to high compared to low quinine intensities (e.g., St John and Boughter, 2004^; Boughter et al., 2005). In one test session, mice were proffered low (0.03 mM) quinine at a warm (30°C) temperature, and high (0.3 mM) quinine cooled (15°C) (Test 3, Table 1). Under these conditions, responses to the high quinine solution were, on average, 3.6 [−10.5, 17.2] licks greater than responding to low quinine (Figure 2A), implying the fluid temperatures disrupted quinine concentration-dependent avoidance. The difference in licking between concentrations was small ($\widehat{\epsilon }=0.04$) and non-significant (p = 0.806). There was no influence of TRPM8 genotype or sex on licking responses to quinine in this test (n.s. mouse line × sex × quinine concentration, p = 0.391).

Figure 2.

Cooling and warming taste fluids interacts with stimulus concentration to affect orosensory responses to quinine. A, TRPM8 deficient and B6 mice showed increased licking towards high (0.3 mM) quinine when cool (15°C) compared to low (0.03 mM) quinine when warm (30°C), albeit this increase was small and not significant (p > 0.05). Nonetheless, this temperature condition countered the ability of a higher quinine concentration to drive increased avoidance. B, In contrast, TRPM8 deficient and B6 mice showed markedly reduced licking to warm-high compared to cool-low quinine (p < 0.05). See methods section on Gardner-Altman plots for details.

Mice showed strong, concentration-dependent avoidance of quinine when, in an additional brief-access test (Test 4, Table 1), high (0.3 mM) quinine was warmed (30°C), and low (0.03 mM) quinine was cooled (15°C). Here, mice significantly reduced their average response to high quinine by 30.2 [−43.6, −16.0] licks compared to low quinine (Figure 2B, t₁₇ = 6.4, p = 0.00001), which represented a large effect size difference ($\widehat{\epsilon }=0.75$). This outcome was not influenced by TRPM8 genotype or sex (n.s. mouse line × sex × quinine concentration interaction, p = 0.375).

Quinine concentration varied, fluid temperature constant

Finally, we examined if the cool (15°C) and warm (30°C) fluid temperatures, by themselves, disrupted licking avoidance of high (0.3 mM) compared to low (0.03 mM) quinine (Tests 5 and 6, Table 1). Brief-access tests conducted with each temperature on separate days revealed no significant influence of fluid temperature, mouse line, or sex on licking responses to quinine (n.s. mouse line × sex × quinine concentration × temperature interaction, p = 0.630). At both temperatures, there were significant reductions in responding to high compared to low quinine across mice (main effect of quinine concentration, F_1,18 = 43.4, p = 0.000003). Specifically, responses to 0.3 mM quinine were reduced compared to 0.03 mM by, on average, 20.8 [−32.4, −8.8] licks at 15°C (Figure 3A, t₁₇ = 6.7, p = 0.000004; $\widehat{\epsilon }=0.65$), and by 17.2 [−28.8, −4.8] licks at 30°C (Figure 3B, t₁₇ = 4.5, p = 0.0003; $\widehat{\epsilon }=0.52$). The effect size for these decreases was large ($\widehat{\epsilon }>0.5$).

Figure 3.

Concentration drives quinine avoidance at constant cool and warm fluid temperatures. TRPM8 deficient and B6 mice showed fewer licks to 0.3 compared to 0.03 mM quinine solutions when both were proffered at 15° (A) or 30°C (B) (p < 0.05). See methods section on Gardner-Altman plots for details.

Altogether, these data suggest the two fluid temperatures examined here are not sufficient, by themselves, to modulate quinine avoidance (Figure 3). However, when given a choice between them, fluid cooling and warming changed quinine responses. Specifically, warming decreased (Figure 1) and cooling, in certain cases, increased (Figure 2) mouse licking rates to quinine. The greatest effect size was observed when comparing the cool-low and warm-high conditions (Figure 2B), where temperature differences markedly accentuated contrast in licking between quinine concentrations. These data suggest differences in fluid temperature interact with stimulus concentration to affect orosensory responses to quinine. We did not find evidence that TRPM8 thermal input contributed to this result, as no differences were observed between mouse lines.

TRPM8 influences thermal effects on sugar taste preferences

Orosensory responses to temperature-controlled solutions of sucrose or glucose were acquired from 24 B6 (12 females, 12 males) and 24 TRPM8 deficient (13 females, 11 males) mice. These mice had ad libitum access to water and food in their home cage during sugar tests. Thus, it was presumed that their lick responses to sucrose and glucose were primarily guided by the positive hedonic tone of these sugars, not thirst or hunger. For analysis, data were collapsed across sugars because similar results were found between both sucrose and glucose across the six test conditions (n.s. sugar × test interaction on differences in licks between test fluids 1 and 2, p = 0.908). Each mouse was tested with one but not both sugars and represented an independent unit for analyses.

Sugar concentration constant, fluid temperature varied

We studied if mice would change their lick rate to a singly tested low (100 mM) or high (500 mM) concentration of sucrose or glucose if taste fluid temperature varied between cool (15°C) and warm (30°C). Under these conditions, lick responses were influenced by TRPM8 genotype, sex, fluid temperature, and sugar concentration (mouse line × sex × temperature × concentration interaction, F_1,43 = 6.4, p = 0.02).

Follow-up three-way ANOVAs were conducted for each stimulus concentration. Considering 100 mM sugars (Test 1, Table 2), the effect of fluid temperature on licking was conditioned on TRPM8 genotype (mouse line × temperature interaction, F_1,43 = 5.6, p = 0.02). Specifically, B6 mice increased their responding to low sugars by 0.8 [−1.4, 3.1] licks when sampled warm compared to cool (Figure 4A; t₁₈ = 1.4, p = 0.176, $\widehat{\epsilon }=0.25$). On the other hand, TRPM8 deficient mice decreased their responding to warm-low sugars by 1.5 [−3.7, 0.1] licks (Figure 4B; t₁₉ = 1.7, p = 0.104, $\widehat{\epsilon }=0.41$). These differences represented small- to moderate-sized effects that were non-significant when considered alone. There was no influence of sex on these trends (n.s. mouse line × sex × temperature interaction, p = 0.940).

Figure 4.

Effects of cooling and warming on orosensory responses to a low concentration of sugar. The influence of cool (15°C) and warm (30°C) taste fluid temperatures on licking to a low concentration (100 mM) of sucrose or glucose was not the same between mouse lines (interaction, p < 0.05), with warmth tending to increase and decrease responding in B6 control (A) and TRPM8 deficient (B) mice, respectively. See methods section on Gardner-Altman plots for details on presentation.

Interactive effects of sex, mouse line, and fluid temperature emerged in brief-access tests where the high (500 mM) concentration of sucrose or glucose was proffered at cool (15°C) and warm (30°C) temperatures (Test 2, Table 2; mouse line × sex × temperature interaction, F_1,43 = 5.4, p = 0.02). Here, male B6 mice increased their mean responding to high sugars by 8.3 [−2.0, 18.2] licks when sampled warm compared to cool (Figure 5C). This effect was medium in size ($\widehat{\epsilon }=0.44$) and not significant (p = 0.141), with wide variance in licking to 30°C sugars noted across mice. In contrast, male TRPM8 deficient mice significantly dropped their responding to high sugars by 6.9 [−19.1, −1.8] licks (t₈ = 2.7, p = 0.027) when warmed (Figure 5D), which represented a large effect size ($\widehat{\epsilon }=0.65$).

Figure 5.

Warming taste fluids causes orosensory avoidance of a high sugar concentration in male TRPM8 deficient mice. Female C57BL/6J control (A) and TRPM8 deficient (B) mice showed indifference to lick a high (500 mM) concentration of sucrose or glucose at cool (15°C) or warm (30°C) temperatures (p > 0.05). In contrast, male C57BL/6J mice tended to increase their licks to sugars at 30° compared to 15°C (C) whereas TRPM8 deficient mice displayed significant avoidance of 30°C sugars (D) (p < 0.05). See methods section on Gardner-Altman plots for details on presentation.

Responses by B6 females to high sugars decreased, on average, by 1.9 [−9.8, 5.7] licks when warmed (Figure 5A), which was a small ($\widehat{\epsilon }=0.14$) non-significant (p = 0.649) effect. Similarly, female TRPM8 deficient mice decreased their responding to high sugars by only 1.1 [−6.2, 1.9] licks when sampled warm compared to cool (p = 0.477; $\widehat{\epsilon }=0.25$). These mice generally showed uniquely reduced responding at both temperatures (Figure 5B).

Sugar concentration and fluid temperature varied

We next asked if taste fluid cooling (15°C) and warming (30°C) would affect typical B6 mouse preferences to lick more of high (500 mM) compared to low (100 mM) concentration of sucrose or glucose (e.g., Glendinning et al., 2015). In a test where mice were proffered low (100 mM) sugars warm (30°C), and high (500 mM) sugars cool (15°C) (Test 3, Table 2), mice significantly increased their licking to the high sugar concentration in a manner that was conditioned on TRPM8 genotype (mouse line × concentration interaction, F_1,44 = 6.8, p = 0.01). Responses to cool-high sugars were 14.0 [8.7, 19.6] licks greater compared to warm-low sugars in B6 mice (Figure 6A; t₁₉ = 5.3, p = 0.00004; $\widehat{\epsilon }=0.81$), and 5.3 [1.5, 9.9] licks greater in TRPM8 deficient mice (Figure 6B; t₁₉ = 2.6, p = 0.02; $\widehat{\epsilon }=0.50$). While both increases represented large effect sizes, the licking response increase was, on average, more than 2.5 times larger in B6 mice. There was no influence of sex on this result (n.s. mouse line × sex × concentration interaction, p = 0.998).

Figure 6.

TRPM8 deficient mice show reduced preference for high concentrations of sugars presented cool. Both C57BL/6J (A) and TRPM8 deficient (B) mice showed increased licking to a high concentration (500 mM) of sucrose or glucose presented cool (15°C) compared to a low concentration (100 mM) offered warm (30°C) (p < 0.05). However, the increase in responding to cool-high sugars was markedly reduced in TRPM8 deficient mice. See methods section for details on Gardner-Altman plots.

TRPM8 genotype also influenced mouse licking responses in an additional brief access test where mice were proffered low sugars cool, and high sugars warm (Test 4, Table 2; mouse line × concentration interaction, F_1,43 = 22.5, p = 0.00002). Specifically, B6 mice significantly increased their responding to warm-high compared to cool-low sugars by, on average, 14.1 [7.8, 21.1] licks (Figure 7A; t₁₈ = 4.4, p = 0.0004), which represented a large effect size ($\widehat{\epsilon }=0.74$). In contrast, TRPM8 deficient mice displayed only a 1.0 [−1.4, 3.8] lick increase to warm-high sugars, which was a small ($\widehat{\epsilon }=0.16$) non-significant (p = 0.252) effect (Figure 7B). Sex did not impact these results (n.s. mouse line × sex × concentration interaction, p = 0.090).

Figure 7.

Warning impairs preference for high sugars in TRPM8 deficient mice. A, C57BL/6J mice prefer to lick warm (30°C) high (500 mM) sugars over cool (15°C) low (100 mM) (p < 0.05). In contrast, TRPM8 deficient mice do not significantly increase their licks to warm-high sugars (p > 0.05). See methods for Gardner-Altman plot details.

Sugar concentration varied, fluid temperature constant

Finally, we asked if the cool (15°C) and warm (30°C) fluid temperatures, by themselves, changed mouse preferences to lick more of the high (500 mM) compared to low (100 mM) solution of sucrose or glucose (Tests 5 and 6, Table 2). Under these conditions, licking responses were influenced by sugar concentration and TRPM8 genotype (mouse line × concentration interaction, F_1,43 = 10.3, p = 0.002) but not fluid temperature or sex (n.s. mouse line × sex × concentration × temperature interaction, p = 0.133). Collapsing data across fluid temperature and sex identified that B6 mice significantly increased their responding to high compared to low sugars by, on average, 13.4 [8.3, 19.3] licks (Figure 8A; t₁₈ = 5.7, p = 0.00002), which represented a large effect size ($\widehat{\epsilon }=0.82$). In contrast, TRPM8 deficient mice significantly increased their licking to high sugars by 3.8 [0.8, 8.0] licks (Figure 8B; t₁₉ = 3.4, p = 0.003), which was a comparably smaller-sized effect ($\widehat{\epsilon }=0.49$).

Figure 8.

TRPM8 deficient mice show reduced sugar preference at constant taste fluid temperatures. C57BL/6J (A) and TRPM8 deficient (B) mice significantly increased their licking to a high concentration (500 mM) of sucrose or glucose compared to low (100 mM) when both solution temperatures were held constant at 15° or 30°C during tests (p < 0.05). This effect did not differ between temperatures (p > 0.05) and plotted data are collapsed across them. The observed increase in licking to high sugars was reduced in TRPM8 deficient mice. See methods section on Gardner-Altman plots for details on data presentation.

Altogether, these data support the hypothesis that a mild warm, 30°C fluid temperature reduces sugar preference in mice that lack TRPM8 function. Support for this trend arose in tests where mice sampled a single sugar concentration (Figure 4, 5), and sugar concentration steps (Figure 7), at cool and warm fluid temperatures. Moreover, when compared to controls, TRPM8 deficient mice appeared to show a reduced avidity to lick sugars under certain temperature varied and invariant conditions (Figure 6, 8).

Discussion

Mouse lick rates measured in the present brief-access tests with the bitter taste stimulus quinine were primarily driven by two factors: (1) the orosensory valence of quinine solutions and (2) thirst. Water restriction is needed to entice sampling of innately aversive stimuli, like bitter tastants, during brief-access trials but concomitantly stimulates thirst drive. We found that under these conditions, licks emitted to quinine were reduced by warm (30°C) and, in some cases, increased by cool (15°C) fluid temperatures when mice were given a choice between them. These results were similar between TRPM8 deficient and B6 mice, which could reflect a confounding influence of common thirst. Nonetheless, fluid temperatures impacted how the mice reacted to quinine solutions.

We measured lick rates to sugar solutions in mice that were not thirst- or hunger-motivated to respond. Thus, their responses to sugars were presumed to largely reflect only stimulus hedonics, with the brief-access nature of the fluid licking tests capturing orosensory reactions to this stimulus feature. Under this condition, we uncovered evidence that thermal sensing mediated by TRPM8 interacts with taste-guided preferences for sugars. When offered a single concentration of sucrose or glucose at cool (15°C) and warm (30°C) fluid temperatures, B6 mice showed facilitated brief-access licking of the warmed sugar. Variability in this trend was present, possibility reflecting varied motivation of mice to lick fluids while in a water-replete state. In contrast, mice that genetically lacked TRPM8 displayed no preference, and significant avoidance, of warm compared to cool sugars in single concentration but varied fluid temperature tests.

This result agrees with prior brief-access thermolickometry data that demonstrated relative to 15°C, warm 30°C water elicits indifference in B6 mice but avoidance in mice genetically deficient for TRPM8 (Li et al., 2024). This effect associated with analyses of trigeminal neural responses in TRPM8 deficient mice that found anomalous positive correlation between activity to oral stimulation with 30°C and normally avoided warmer/hot temperatures (Li et al., 2024). Together with the present findings, these observations suggest that TRPM8 input contributes to the development of oral thermal sensing by trigeminal circuits in a way that imparts valence on sugar taste preferences. While an innocuous warm sugar was preferred over cool with intact TRPM8 function, warm sugar was avoided in its absence. TRPM8 is expressed by trigeminal fibers that surround, but do not innervate, taste buds (Abe et al., 2005; Dhaka et al., 2008). This arrangement implies oral thermosensation mediated by TRPM8 is initiated in parallel with gustatory transduction and the effects that temperature can have on taste processes. Thus, interactions between TRPM8 and sweet taste messages would appear to happen downstream of receptor events.

During certain tests, both B6 and TRPM8 deficient mice emitted more licks to high compared to low sugars, regardless of fluid temperature, when both concentrations were offered in one session. These results may reflect greater attention to a concentration-mediated difference in perceived intensity between 100 and 500 mM sugar fluids than thermal influences. Accordingly, in B6 mice, oral presence of 500 mM sucrose can induce greater levels of spike discharge in peripheral (Talavera et al., 2005) and central (Wilson and Lemon, 2014) gustatory neurons compared to 100 mM sucrose across different temperature conditions. Nevertheless, unlike B6 mice, TRPM8 deficient mice did not presently increase their licking to the warm-high (30°C, 500 mM) sugar when the other sugar fluid option was cool-low (15°C, 100 mM). This further suggests that without TRPM8, warmth adopts a less preferred tone that competes against sugar taste palatability.

Compared to B6 mice, TRPM8 deficient mice displayed a smaller increase in licking high sugar when high and low concentrations of sucrose and glucose were offered at a constant fluid temperature. A similar result emerged in tests with warm-low and cool-high sugars. TRPM8 deficiency is established to affect energy balance and feeding under certain conditions. For instance, TRPM8 deficient mice show hyperphagia and increased stored fat mass when housed at ambient temperatures near typical mouse colony housing temperatures, such as 21° to 25°C (Reimundez et al., 2018). These effects were accompanied by metabolic changes including increased levels of circulating leptin – an anorexigenic mediator (Reimundez et al., 2018). There is some evidence that increased leptin can suppress taste receptor cell and gustatory nerve responses to sweet substances and decrease licking towards these stimuli in mice (Kawai et al., 2000; Shigemura et al., 2004; Yoshida et al., 2013). Yet leptin effects on mouse gustatory neural and behavioral responses to sweet taste stimuli have not been universally found (Glendinning et al., 2015). Nevertheless, it is curious if some facet of metabolic change in TRPM8 deficient mice is regulating their lowered taste preference for elevated sugars. As a next step, it would be informative to investigate if TRPM8 deficiency impacts gustatory neurophysiological responses to sucrose and glucose.

Human psychophysical data show that across a broad range of fluid temperatures (e.g., from 10° to 40°C), cooling can suppress the perceived bitterness of suprathreshold concentrations of quinine relative to warm temperatures (Green and Andrew, 2017). This result implies cooling dampens, while warming enhances, quinine taste intensity. Accordingly, additional human data that gauged time-dependent dynamics in taste perception suggest that warm (35°C) quinine may be perceived as more bitter than cold (5°C) during the first few seconds of gustatory stimulation (Bajec et al., 2012). These human-based results show some accord with the present findings of fluid temperature effects on brief-access licking to quinine in mice. When given a choice between cool (15°C) and warm (30°C) fluids, mice presently showed substantially fewer licks to warm quinine when concentration was held at a constant low or high value. Furthermore, mice decreased their licking of warm-low quinine, and increased their licking of cool-high, when both fluids were offered in one test session. The effect size for this difference, about 30 licks per 10 sec, was the largest observed. These results suggest warming adds to, and cooling subtracts from, the aversive orosensory tone of quinine experienced by mice. Studies in rodents show cooling and warming taste solutions can, respectively, decrease and increase electrophysiological activity to quinine in subsets of peripheral (Breza et al., 2006) and central (Wilson and Lemon, 2013) gustatory neurons. However, thermal effects on gustatory neural activity to quinine are smaller than those observed for sugars like sucrose (Wilson and Lemon, 2013).

We found that lick rates to quinine did not differ between B6 and TRPM8 deficient mice across fluid temperature conditions. The phenotype of TRPM8 deficient mice to lick more of cool, and less of warm, quinine at a static concentration parallels their known response to similarly chilled and warmed water, where 15°C is preferred and 30°C avoided (Li et al., 2024). In contrast, B6 mice show equally high preference to lick both 15° and 30°C water (Li et al., 2024) but were presently found to prefer 15° over 30°C quinine solutions. This discrepancy indicates that oral temperature preferences in mice are normally context-dependent and changeable by other oral sensations. Notably, the present data also highlight that with intact TRPM8 function, licking avoidance of quinine varies with media temperature, in addition to stimulus concentration, and can be reduced by cooling under certain conditions. This result, along with human data discussed above, may reflect thermal effects on bitter hedonics. Such trends could have practical implications for improving the appeal of bitter taste sensations in applied settings.

Overall, these findings contribute to a growing literature on interactions between somatosensation and gustation and dependencies between these sensory systems. Studying taste and thermosensation together appears critical for understanding the operational principles of these senses individually and how they may intersect in flavor processing. Prior and the present data indicate that taste perception is reliant on temperature and trigeminal thermosensation. However, the evolutionary and functional bases of this link remain speculative at best. Moreover, while the parabrachial nucleus is evidenced to support the integration of aversive taste information with noxious trigeminal thermosensory signals in mice (Li and Lemon, 2019; Li et al., 2022), a neural mechanism that could mediate interplay between appetitive sugar tastes and thermal messages mediated by TRPM8 remains unknown. Further investigations in this realm will be critical to delineate how diverse sensory factors affect food preference and nutritional status, and for understanding gustatory and thermosensory coding.