The Effect of Temperature on Umami Taste

Abstract

The effect of temperature on umami taste has not been previously studied in humans. Reported here are 3 experiments in which umami taste was measured for monopotassium glutamate (MPG) and monosodium glutamate (MSG) at solution temperatures between 10 and 37 °C. Experiment 1 showed that for subjects sensitive to MPG on the tongue tip, 1) cooling reduced umami intensity whether sampled with the tongue tip or in the whole mouth, but 2) had no effect on the rate of umami adaptation on the tongue tip. Experiment 2 showed that temperature had similar effects on the umami taste of MSG and MPG on the tongue tip but not in the whole mouth, and that contrary to umami taste, cooling to 10 °C increased rather than decreased the salty taste of both stimuli. Experiment 3 was designed to investigate the contribution of the hT1R1–hT1R3 glutamate receptor to the cooling effect on umami taste by using the T1R3 inhibitor lactisole. However, lactisole failed to block the umami taste of MPG at any temperature, which supports prior evidence that lactisole does not block umami taste for all ligands of the hT1R1–hT1R3 receptor. We conclude that temperature can affect sensitivity to the umami and salty tastes of glutamates, but in opposite directions, and that the magnitude of these effects can vary across stimuli and modes of tasting (i.e., whole mouth vs. tongue tip exposures).

Introduction

Despite more than a century of research on umami taste (see Yamaguchi and Ninomiya 2000; Beauchamp 2009 for reviews), there have been no published studies on how temperature might affect its perception in humans. This omission is attributable in part to the fact that most studies of the effect of temperature on taste (e.g., Stone et al. 1969; Pangborn and Bertolero 1972; McBurney et al. 1973; Moskowitz 1973; Paulus and Reisch 1980; Bartoshuk et al. 1982; Calviño 1986; Green and Frankmann 1988) were carried out before discovery of candidate glutamate receptors in gustatory epithelia (Chaudhari et al. 1996, 2000; Lin and Kinnamon 1999; Li et al. 2002; Nelson et al. 2002) led to wider acceptance of umami as a basic taste. However, the effect of temperature on whole-nerve and cellular responses to glutamates has been studied in animal models. Nakamura and Kurihara (1991b) found that the responses to monosodium glutamate (MSG) and guanosine 5′-monophosphate (GMP) in the canine chorda tympani nerve were inverted u-shaped functions of temperature with maxima between 10 and 15 °C for MSG and near 30 °C for GMP. The authors noted that the GMP function resembled the response curve for sucrose whereas the MSG function resembled the response curve for NaCl. Because Nakamura and Kurihara (1991a) had previously shown that the sodium channel blocker amiloride (Heck et al. 1984; Brand et al. 1985) suppressed the response of the canine chorda tympani to MSG, they concluded that the response to MSG had a large salt-dependent component.

More recently, Talavera et al. (2005) found that the nonselective Ca²⁺ channel TRPM5, which is in the transduction cascade of G-protein-coupled receptors that are sensitive to sugars, glutamates, and bitter-tasting ligands (Zhang et al. 2003), is sensitive to heat. Based on this finding Talavera et al. (2007) proposed that TRPM5 is the primary source of the thermal sensitivity of sweet taste and might also contribute to thermal effects in bitter and umami taste. Data from a study by Wilson and Lemon (2013) of gustatory neurons in the nucleus of the solitary tract (NTS) in mice are consistent with this hypothesis: activity in a subset of NTS neurons sensitive to both sucrose and a mixture of MSG and inosine-monophosphate (IMP) was increased by heating. These findings in rodents are in general agreement with previous studies of the effect of temperature on the sweetness of sucrose in humans, which tends to increase with temperature (Bartoshuk et al. 1982; Green and Frankmann 1987, 1988; Green and Nachtigal 2015).

In the context of these data, we began the present study with the expectation that umami taste would be sensitive to temperature in a way similar to sweet taste. Of particular interest was whether temperature would affect both the initial perception of umami taste and the rate of umami adaptation, which we had recently found occurs for sweet taste (Green and Nachtigal 2015). Our approach in the first experiment was to measure the effect of temperature on perception of monopotassium glutamate (MPG). We chose MPG because it has been shown to activate both T1R1–T1R3 and mGluR4 receptors in mice (Maruyama et al. 2006; Yasuo et al. 2008), and because unlike MSG, MPG does not have a significant “salt-dependent” gustatory component (Nakamura and Kurihara 1991a; Chen et al. 2009) in rodents. However, after data collection with MPG was completed, another group of subjects was tested with MSG to evaluate the generalizability of the findings with MPG and to learn whether temperature also affects the salty taste component of this prototypical umami stimulus in humans.

In both experiments we chose to focus primarily on the fungiform taste area on the tongue tip, where the temperature of the solution can be controlled more uniformly, and to enable comparison of the results with the recent data on the effect of temperature on sweet taste adaptation, which were also collected on the tongue tip (Green and Nachtigal 2012, 2015). However, data on the effect of temperature in the initial perception of umami taste in the whole mouth were also obtained. Because humans generally report umami taste to be stronger in the back of the mouth (Boudreau 1980; Yamaguchi and Ninomiya 2000; Green and George 2004; Feeney and Hayes 2014), those data could determine whether the effect of temperature measured on the tongue tip was predictive of perception throughout the mouth. A third experiment was designed to explore the contribution of hT1R1–hT1R3 to the thermal effect at the tongue tip by using the sweet and umami taste inhibitor lactisole (2-4-methoxyphenoxy propionic acid). Lactisole, which binds to the transmembrane region of T1R3 (Jiang et al. 2005; Winnig et al. 2005), has been shown to block activation of the mouse T1R1–T1R3 receptor in vitro (Xu et al. 2004), and in humans to reduce sensitivity to l-glutamate + IMP (Xu et al. 2004) and to decrease suprathreshold perception of the umami taste of MSG (Galindo-Cuspinera and Breslin 2006).

Experiment 1: Effect of temperature on the tongue tip versus in the whole mouth

Materials and methods

Subjects

A total of 47 subjects (28 females and 19 males) between 18 and 45 years of age were recruited from public postings on the Yale Medical School and Yale College campuses. Each participant gave informed consent and was paid for their participation. The research protocol was approved by the Human Investigations Committee of the Yale University IRB and complies with the Declaration of Helsinki for Medical Research involving Human Subjects. All subjects were fluent English-speakers, self-reported healthy nonsmokers who had no known taste or smell disorders or deficiencies, were not pregnant, and had no lip, cheek, or tongue piercings. Subjects were asked to refrain from eating or drinking foods or beverages for at least 1h prior to their scheduled session. Five participants did not complete the experiment (3 did not return for a second session; 1 was determined to be a cigarette smoker, and 1 could not follow instructions in the screening session) leaving a total of 42 subjects (25 females, 17 males) whose data were included in the analyses.

Stimuli

The test stimuli were 56, 100, and 200mM l-glutamic acid monopotassium salt monohydrate (MPG) (Sigma–Aldrich). Aqueous solutions of 0.56M sucrose (SUC), 18mM citric acid (CA), 0.32M NaCl, and 0.18mM quinine (QU) were also used as practice stimuli. The concentrations of CA, NaCl, and QU were established in pilot testing to produce taste intensities in most subjects that were similar to the sweetness of 0.56M sucrose, which on average evokes a weak-to-moderate sweetness. MPG concentrations that were chosen during pilot testing produced “weak” to “moderate” umami taste in the whole mouth in most individuals and which covered most of the range of concentrations tested by Chen et al. (2009) in a study of the relationship between individual differences in umami taste and variants of the TAS1R1 receptor gene. All stimuli were prepared weekly with deionized water in 250mL volumes and stored in airtight flasks. The solution temperatures tested were 10, 21, and 37 °C. The stimuli were placed in circulated water baths 30min prior to each testing session to control the temperature of the solutions.

Session 1: Training, practice, and screening

Prior to the first data collection session, all subjects attended a practice and screening session in which they were instructed in how to use the general version of the labeled magnitude scale (gLMS; Green et al. 1993, 1996; Bartoshuk et al. 2003) to rate sensation intensity. The gLMS was displayed on a computer monitor and subjects were shown how to use a mouse to move a cursor to appropriate locations on the scale relative to its verbal descriptors to indicate perceived intensity. The subjects rated the sensation intensity of 15 remembered or imagined sensations (e.g., the sweetness of cotton candy, the weight of a feather in your hand, the pain of biting your tongue) that were read to them by the experimenter. These examples were intended to give subjects experience using the gLMS in the broad context of every day experience.

Subjects were then given practice rating the sensation intensities produced by actual stimuli: 9 different stimuli from different sensory modalities (e.g., the cold sensation from a penny placed on the wrist; the touch sensation from a cotton swab; the brightness of the ceiling light) and 10 taste stimuli, including 0.56M sucrose, 0.32M NaCl, 18mM citric acid, 18mM QHCl, 100mM MPG and 5 binary mixtures (sucrose + QHCl, citric acid + NaCl, NaCl + QHCl, sucrose + citric acid, and MPG + NaCl). Importantly, prior to rating the sample solutions the experimenter described umami taste as having a “savory” or broth-like quality that is most commonly associated with the taste of MSG. Subjects then tasted a 5-mL sample of the 100mM MPG at 21 °C and were asked to focus on its unique taste quality and to describe where on the tongue it was most noticeable. After expectorating and rinsing they then tasted and rated the perceived intensity of the sweet, salty, sour, bitter, and umami tastes evoked by each of the practice taste stimuli.

Tongue tip screening

At the beginning of the first testing session subjects began by sampling and rating the perceived intensity of 3 concentrations (56, 100, and 200mM) of MPG using only the tongue tip. This served primarily as a screening procedure to identify individuals who perceived 200mM MPG to evoke at least a “weak” umami taste and thus would be able to participate in a test of the effect of temperature on umami taste on the tongue tip. The procedure was the following: 2 adjacent plastic weigh boats (41mm × 41mm × 8mm, Fisher Scientific) each filled with 7.5mL of solution were placed 2.5cm apart in a holder that was positioned directly in front of the subject. The weigh boat on the left contained only dH₂O at room temperature (~21 °C) and the weigh boat on the right contained one of the 3 concentrations of MPG, also at ~21 °C. A trial began with the experimenter instructing the subject to lean forward and submerge her/his tongue tip in the liquid in the left weigh boat. Timing began when the subject’s tongue tip touched the solution. Three seconds later the experimenter said “switch” to signal the subject to immediately lift the tongue tip out of the first solution and dip it into the second solution. After another 3s the experimenter instructed the subject to lift the tongue out of that solution and immediately begin rating the umami taste of the second solution on the gLMS. The tongue was not retracted back into the mouth until the intensity rating was completed, after which the subject rinsed at least 3 times with 37 °C dH₂O. The MPG concentrations were presented in ascending order as a precaution against possible confusion with any lingering umami taste in the back of the mouth. Preliminary testing indicated that this sometimes occurred when the higher concentrations of MPG spread to the more sensitive areas in the back of the mouth during rinsing. Ratings of umami intensity ≥ “weak” on the gLMS in response to the 200-mM MPG solution sampled with the tongue tip at 21 °C qualified subjects as MPG “sensitive-tasters” (STs). Subjects who did not meet this criterion were classified as MPG “low-tasters” (LTs).

Following the tongue tip screening, all subjects sampled 5mL of 200mM MPG at 3 different temperatures (37, 21, and 10 °C) in the whole mouth with instructions to rate the intensities of umami, sweet, salty, sour, and bitter taste on the gLMS. Temperatures were presented in descending order and subjects sipped and swished the solutions gently throughout the mouth for 3s before expectorating and making their intensity ratings. Each temperature was presented twice for a total of 6 trials. It was emphasized that the ratings should be based on the tastes experienced while the solution was still in the mouth. After completing their ratings subjects were instructed to rinse vigorously with 37 °C dH₂O at least 3 times, or more if tastes continued to linger. The inter-trial interval was at least 30s or longer depending upon when subjects completed their ratings and water rinses.

Test sessions: Effects of temperature on umami taste and adaptation at the tongue tip

STs were asked back to rate the perceived intensity of umami taste in the fungiform region of the tongue as a function of temperature. The stimuli were presented using the same general procedure as the tongue tip screening task, with 7.5-mL taste solutions contained in 2 adjacent weigh boats. However, to measure the possible effect of temperature on umami taste adaptation to MPG, the left weigh boat contained a pre-exposure stimulus of either pure dH₂O or 200mM MPG at 10, 21, or 37 °C, whereas the right weigh boat always contained a 200mM MPG solution at the same temperature. Pre-exposures were for 0, 3, or 10s. For the 3- and 10-s pre-exposures the experimenter instructed the subject to leave the tongue tip submerged in the adapting solution for those durations before signaling them to “switch” immediately to the test solution, which they sampled for 3s, also timed by the experimenter. Subjects again kept the tongue outside the mouth as they switched to the second sample and while making their taste intensity ratings. The effect of temperature on initial taste sensitivity (0s pre-exposure), which also served as the baseline from which adaptation was measured, was determined by instructing the subject to dip the tongue tip in only one weigh boat containing the 200mM MPG solution at 10, 21, or 37 °C for 3s. After completing the ratings the subject rinsed at least 3 times with 37 °C dH₂O to remove residual taste from the mouth. The experiment comprised 2 sessions (15 trials each) that were completed on separate days. Temperature conditions were intermixed in a pseudorandom order to avoid presenting the same temperature on consecutive trials. All subjects were randomly assigned to one of 8 temperature orders for each session.

Results

Sensitivity to MPG on the tongue tip

Based on their ratings of the umami taste of MPG sampled with the tongue tip, 32 participants (20 F, 12M) were classified as STs and 10 (5 F, 5M) as LTs. Despite using different psychophysical methods and classification criteria (intensity ratings vs. performance in a discrimination task), the frequency of LTs we found agrees well with the 20.2% of combined “hyposensitive” and “insensitive” subjects reported by Chen et al. (2009) for MPG and by Lugaz et al. (2002) for MSG. Figure 1 shows that as a group, tongue tip LTs did not rate even the 200mM MPG solution as more than “barely detectable” (main effect of Status, F _1,40 = 17.63, P < 0.00001). For all 3 concentrations, STs gave umami intensity ratings that were on average +0.87 log₁₀ higher than those given by LTs, which equals a 7.4-fold difference in perceived intensity. Also notable is the significant increase (Tukey HSD, P < 0.05) in umami intensity for STs between the 100 and 200mM solutions (+0.41 log₁₀, which equals 2.63-fold difference in intensity), whereas no difference was found between the 56 and 100mM solutions.

Figure 1.

Log₁₀ mean ratings of the perceived intensity of umami taste on the tongue tip are shown as a function of MPG concentration (56–200mM) for STs (filled triangles) and LTs (open triangles) on the tongue tip. Error bars represent standard errors of the means (SEMs); letters on the right y axis refer to intensity descriptors on the gLMS: BD, barely detectable; W, weak; M, moderate; S, strong.

Umami taste sensitivity: Whole mouth and tongue tip

Figure 2 compares the effect of temperature in the whole mouth and on the tongue tip for STs. The data are based on the initial 3-s exposures to 200mM MPG at the 3 temperatures. As expected, MPG umami taste was perceived more strongly in the whole mouth than it was on the tongue tip, although the difference was not large (+0.20 log₁₀). However, a 3-way ANOVA (condition × temperature × replicate) confirmed that this difference was significant (F _1,31 = 12.04, P < 0.005) as was the main effect of Temperature (F _2,62 = 17.83, P < 0.00005). The parallel functions indicate that temperature affected umami taste very similarly during a 3-s exposure to MPG whether exposure was throughout the mouth or limited to the tongue tip (i.e., there was no condition × temperature interaction; F _2,62 = 0.77, P = 0.47).

Figure 2.

Log₁₀ mean perceived intensity of umami taste for STs on the tongue tip (inverted triangles) and in the whole mouth (squares) as a function of solution temperature. For these subjects the effect of temperature was nearly identical for both conditions of taste exposure. Error bars represent standard errors of the means (SEMs). Letters on the right y axis refer to intensity descriptors on the gLMS: BD, barely detectable; W, weak; M, moderate; S, strong.

Umami taste adaptation on the tongue tip

Figure 3 shows that exposure to 200mM MPG for 3 or 10s produced adaptation of umami taste at all 3 test temperatures compared with pre-exposure to H₂O having the same temperature. Adaptation was nearly identical for the 37 and 21 °C solutions after 10-s pre-exposures. Notably, the precipitous drop in sensitivity caused by exposure to H₂O-alone at 10 °C (also evident in Figure 3) did not lead to a further increase in umami adaptation. These effects were confirmed by a 4-factor ANOVA (treatment × temperature × time × replicate). Main effects of treatment (MPG vs. H₂O-alone; F _1,31 = 35.87, P < 0.00001), temperature (F _2,62 = 24.15, P < 0.00001), and time (F _2,62 = 15.42, P < 0.00001) were modified by a 3-way interaction among the factors (F _4,124 = 4.43, P < 0.005). However, inspection of the data shows that the interaction was driven by the large drop in initial sensitivity at 10 °C which reduced the measured adaptation. Adaptation therefore tended to be greater at warmer temperatures than at cooler temperatures (−0.38 log₁₀ at 37 °C, −0.27 log₁₀ at 21 °C, and −0.20 log₁₀ at 10 °C), which was opposite to the result recently reported for the effect of temperature on sweet taste adaptation (Green and Nachtigal 2015), and Tukey HSD tests showed that after 10-s pre-exposures to MPG, umami taste intensity did not differ significantly across the 3 temperatures (all Ps > 0.05). The latter finding underscores the fact that temperature affected initial sensitivity much more strongly than it affected adaptation.

Figure 3.

Log₁₀ mean perceived intensity of umami taste is shown here for 3-s post-exposures to 200mM MPG after 0-, 3-, or 10-s pre-exposures to either H₂O only (open circles) or 200mM MPG (filled circles) at the same temperature (10, 21, or 37 °C). Note that the data for 0 duration are the same in both conditions at each temperature since they indicate umami taste intensity after 3-s exposures to MPG with no pre-exposure stimulus. Error bars represent standard errors of the means (SEMs). Letters on the right y axis refer to intensity descriptors on the gLMS: BD, barely detectable; W, weak; M, moderate; S, strong.

Experiment 2: MSG compared with MPG

Materials and methods

Subjects

A total of 33 subjects (19 females and 14 males) served in the experiment, none of whom had served in experiment 1. Recruitment and eligibility criteria were the same as in experiment 1 and each participant gave informed consent and was paid for his/her participation. The research protocol was approved by the Human Investigations Committee of the Yale University IRB and complies with the Declaration of Helsinki for Medical Research involving Human Subjects. STs of MSG and MPG on the tongue tip were identified based on ratings of MPG and MSG on the tongue tip, as in experiment 1.

Stimuli

The 2 stimuli used were aqueous solutions of 200mM MPG (Sigma–Aldrich) and 200mM MSG (Sigma–Aldrich). Each stimulus was tested at 10, 21, and 37 °C, and solution temperatures were controlled by circulating water baths.

Procedure

Participants completed training (as described in experiment 1) and participated in a whole-mouth testing protocol, comprising 6 trials during which subjects sampled 5mL of MPG and MSG at 10, 21, and 37 °C. As in experiment 1, subjects rated the intensities of umami, sweet, salty, sour, and bitter taste on the gLMS after 3s of exposure timed by the experimenter. Stimuli were presented in 2 orders: one in which MPG was presented first and the other in which MSG was presented first, both in descending temperature series.

Participants then completed 2 experimental sessions which employed the same tongue tip tasting task as experiment 1. All subjects were randomly assigned to 1 of 8 stimulus temperature orders upon enrollment. MSG and MPG were presented in separate sessions with the order counterbalanced across subjects. As in experiment 1 testing was conducted using 2 weigh boats: the left weigh boat contained a pre-exposure stimulus of either dH₂O or the adapting solution at 10, 21, and 37 °C, whereas the right weigh boat always contained the test solution at the same temperature. The solutions were sampled for 0, 3, or 10s followed by a 3-s exposure to the test solution, after which subjects immediately rated (with the tongue still extended from the mouth) the intensity of umami and salty tastes using the gLMS.

Results

Compared with experiment 1, fewer subjects in this experiment qualified as STs to MPG or MSG on the tongue tip. Of the 33 subjects screened, 19 (57.6%) were sensitive to 200mM MPG and 18 (54.5%) were sensitive to 200mM MSG. Log₁₀-mean umami taste intensity ratings were nearly identical for the 2 stimuli (MPG = 0.97; MSG = 0.94), but not all sensitive subjects (14 of 19) met the criterion for sensitivity to both MPG and MSG on the tongue tip.

Umami and salt taste sensitivity: Tongue tip and whole mouth

Figure 4 compares the ratings from STs of umami and salty taste intensity sensed at the tongue tip and in whole mouth during the initial 3-s exposure. The results for MPG were similar to those in experiment 1 (cf. Figure 2), with umami taste being stronger at warmer temperatures (F _2,36 = 3.91, P < 0.05). However, even though ratings of umami taste were on average weaker when exposure was limited to the tongue tip, the main effect of condition was not quite statistically significant (F _1,18 = 3.92, P = 0.063). On the other hand, the effect of condition was significant for the umami taste of MSG (F _1,17 = 4.95, P < 0.05), whereas the effect of temperature was not. The lack of a significant main effect of temperature for MSG umami was due to the absence of an effect of temperature on umami taste in the whole mouth.

Figure 4.

Log₁₀ mean perceived intensity ratings of umami taste and salty taste are shown for 3-s exposures to 200mM MSG (a, b) and 200mM MPG (c, d) as a function of solution temperature in the whole mouth (filled squares) or on the tongue tip (inverted filled triangles) Error bars represent SEMs; letters on the right y axis refer to intensity descriptors on the gLMS: BD, barely detectable (also indicted with dotted line); W, weak; M, moderate; S, strong.

The salty component of the taste of MSG is evident in Figure 4b, although the effect of temperature was markedly different for the tongue tip and whole mouth conditions, leading to a significant condition × temperature interaction (F _2,34 = 7.09, P < 0.005). In contrast to umami taste, on the tongue tip salty taste was strongest at 10 °C and fell to barely detectable for the 37 °C solution. An opposite but much weaker trend was found with whole mouth exposures to MSG, for which average saltiness ratings exceeded barely detectable only at 37 °C. As expected, saltiness was virtually imperceptible for MPG in both conditions, although a weak trend toward more saltiness at colder temperatures was also visible in the tongue tip condition.

Umami and salt taste adaptation on the tongue tip

Figure 5a shows that adaptation to the umami taste of MSG was approximately the same across the 3 temperatures, which was consistent with the temperature independence of umami adaptation for MPG in experiment 1. The latter result was also replicated in this experiment (Figure 5c). Analyses of the data for the 2 stimuli confirmed there were significant main effects of both temperature (MSG, F _2,34 = 3.82, P < 0.05; MPG, F _2,36 = 5.48, P < 0.001) and time (MSG, F _2,34 = 12.7, P < 0.0001; MPG, F _2,36 = 17.81, P < 0.00001), but no significant temperature × time interaction for either stimulus.

Figure 5.

Log₁₀ mean perceived intensity ratings of umami taste and salty taste for 3-s exposures to 200mM MSG (a, b) or 200mM MPG (c, d) after 0-, 3-, or 10-s exposures to the same stimuli at the same temperature (10°, open circles; 21°, filled squares; or 37 °C, filled triangles). Error bars represent SEMs; letters on the right y axis refer to intensity descriptors on the gLMS: BD, barely detectable (also indicated with a dotted line); W, weak; M, moderate; S, strong.

MSG saltiness ratings over time were significantly affected by temperature, but in the opposite direction (main effect of temperature; F _2,34 = 6.95, P < 0.01), and declined significantly over time (F _2,34 = 5.8, P < 0.01). After the initial 3-s exposure, log₁₀-mean saltiness ratings exceeded barely detectable only for the 10 °C solution. Although the rate of saltiness adaptation did not appear to be consistent across temperatures (in part because of the minimal saltiness of the 37 °C solution), the temperature × time interaction did not reach statistical significance (F _42,68 = 1.6, P = 0.18). As expected, MPG saltiness was on average rated barely detectable or less. Despite its minimal intensity, MPG saltiness also showed a significant tendency to be perceived more “strongly” at the cooler temperatures (main effect of temperature; F _2,36 = 9.3, P < 0.005). The effect of time did not approach statistical significance, nor was there a significant temperature × time interaction for MPG saltiness.

Experiment 3: Effect of lactisole on MPG umami taste on the tongue tip

Materials and methods

Subjects

A total of 23 subjects (19 females and 4 males) completed the study, 16 of which were determined to be STs (12 females and 4 males). Each participant gave informed consent and was paid for her/his participation. The research protocol was approved by the Human Investigations Committee of the Yale University IRB and complies with the Declaration of Helsinki for Medical Research involving Human Subjects. Recruitment, eligibility and MPG sensitivity criteria were the same as in experiment 1, and 11 of the subjects had served in experiment 1.

Stimuli

The stimuli were aqueous solutions 200mM MPG (Sigma–Aldrich) and 8mM Lactisole [2(-4-Methoxyphenoxy) Propanoic Acid] (Chem-Impex International Inc.) buffered in solution to pH 7.0 by NaOH. Solution temperatures tested were 10°, 21°, and 37 °C, controlled as before by circulated water baths. An 8mM solution of lactisole was used based upon evidence from a prior study (Galindo-Cuspinera and Breslin 2006) that as little as 1mM lactisole can significantly reduce the sweetness of sucrose in humans, and upon preliminary testing which confirmed that 8mM lactisole could block 0.56M sucrose sweetness in the whole mouth.

Procedure

The tongue-dipping procedure of experiments 1 and 2 was employed and the experiment consisted of 2 test sessions conducted on separate days. (Whole mouth testing with lactisole was ruled out when we found during preliminary testing that the sweet water taste produced by 8mM lactisole made rinse water taste sweet. It was not uncommon for sweetness to linger for several minutes in the back of the mouth, even after multiple rinses). Each session included 6 trials (2 trials per temperature) with a 1-min intertrial interval during which the subject rinsed at least 3 times with 37 °C deionized water to ensure there was no contamination from the previous stimulus. The stimulus conditions (200mM MPG alone followed by 200mM MPG in mixture with Lactisole) were presented in either ascending or descending temperature series which were counterbalanced across subjects. Subjects again rated the intensities of umami, sweet, salty, sour, and bitter taste on the gLMS experienced when the tongue tip was in the solution. After completing their ratings the subjects retracted their tongue into the mouth and rinsed at least 3 times with 37 °C pure dH₂O to remove the residual taste from the mouth.

Results

The data in Figure 6 show that lactisole had no effect on the sensitivity to MPG sensed at the tongue tip, nor did it disrupt the effect of temperature on the perceived intensity of umami taste. An ANOVA with temperature, condition, taste, and replicate as within-subject factors and taster status as a between-subject factor determined that there were main effects of temperature (F _2,42 = 8.96, P < 0.001), taste (F _4,84 = 34.0, P < 0.0001), and taster status (F _1,21 = 12.66, P < 0.005) but not condition (F _1,21 = 1.86, p = 0.19) or replicate (F _1,21 = 0.37, P = 0.55). A significant condition × taste interaction (F _4,84 = 4.56, P < 0.005) was found that was due to slightly higher mean ratings of umami and salty taste when MPG was mixed with lactisole. This unexpected result was likely due to the use of NaOH to buffer the lactisole solution. The temperature × taste interaction (F _8,168 = 6.53, P < 0.00001) was also significant, which was expected given the effect of cooling on umami taste and the floor effect resulting from the very low intensity ratings for other tastes (not shown), all of which averaged near or below “barely detectable.”

Figure 6.

Log₁₀ mean perceived intensity ratings of umami taste for 3-s exposures to 200mM MPG alone or in mixture with 8mM lactisole (MPG + LAC) at 3 test temperatures. Testing was limited to the tongue tip. Rather than blocking umami taste, adding lactisole resulted in nonsignificant increases in mean intensity ratings, which may have been due to a slight salty taste resulting from the addition of NaOH to buffer the lactisole solutions to pH 7.0. Error bars represent SEMs; letters on the right y axis refer to intensity descriptors on the gLMS: BD, barely detectable; W, weak; M, moderate; S, strong.

It is noteworthy that taster status did not interact with condition (F _1,21 = 0.18, P = 0.66). Even though LTs rated umami taste to be more than ½ log unit weaker than did STs (data not shown), they reported nearly identical effects of temperature, with cooling significantly reducing umami taste only at 10°C.

Discussion

The present data demonstrate that when measured on the tongue tip, umami taste in humans is temperature-dependent between 37 and 10°C, with sensitivity declining at cooler temperatures. This finding is consistent with the results of another recent study in this laboratory which found that the sensitivity of the tongue tip to sweeteners was also significantly reduced by cooling below 21 °C (Green and Nachtigal 2015), and both results are consistent with the TRPM5 explanation of the temperature sensitivity of taste GPCRs (Talavera et al. 2005, 2007). However, Green and Nachtigal (2015) also found that cooling from 37 to 21 °C increased the rate of sweet taste adaptation for 5 of the 6 sweeteners they tested, an effect we did not find for the umami tastes of MPG and MSG. Because in the sweetener study not all stimuli were affected, it was proposed that the thermal effect on initial sensitivity to sweet taste might be attributable to the temperature sensitivity of TRPM5, whereas adaptation more likely occurs upstream of TRPM5 in the hT1R2–hT1R3 transduction cascade. This possibility gains indirect support from the present data, since the absence of a thermal effect on umami adaptation might be explained by the presence of a different monomer (T1R1 vs. T1R3) and/or differences in coupling and interactions between monomers in the 2 different heterodimeric receptors.

The TRPM5 explanation of the present results is complicated by the absence of an effect of temperature on MSG umami taste in the whole mouth. One possible explanation is the evidence mentioned above of multiple glutamate receptors in mice (Maruyama et al. 2006; Yasuo et al. 2008). Although taste mGluRs and T1R1–T1R3 are both type III GPCRs, the intracellular signaling pathway of mGluRs appears not to include TRPM5 (Chaudhari et al. 2000; Abaffy et al. 2003). Evidence that in mice T1R1–T1R3 is expressed more densely in fungiform papillae than are mGluRs (Yasuo et al. 2008) raises the possibility that temperature may have a lesser effect on umami taste when caudal gustatory areas are stimulated. However, no comparable data are available regarding the distribution of glutamate receptors in humans, and this explanation conflicts with the evidence that MPG showed nearly identical effects of temperature for both tongue tip and whole mouth exposures. It must also be considered that the salty component of MSG taste may underlie the difference in results for MSG and MPG. When stimulus exposure is throughout the mouth, MSG saltiness may be more difficult to discriminate from the stronger umami taste typically reported in the back of the mouth (Boudreau, 1980; Yamaguchi and Ninomiya, 2000; Green and George, 2004; Feeney and Hayes, 2014). The perceptual integration of, or confusion between, salty and umami tastes in the whole mouth could lead to higher umami ratings at colder temperatures, where salt taste intensity increases. The much lower ratings for MSG saltiness in the whole mouth compared with the tongue tip at 10 °C is consistent with this confusion hypothesis, as are data from single chorda tympani fibers in nonhuman primates which showed that fibers sensitive to MSG also responded to salts and acids (Hellekant et al. 1997a, 1997b).

In experiment 3, direct assessment of the contribution of hT1R1–hT1R3 to the thermal effect was hindered by the failure of lactisole to block umami taste on the tongue tip. One interpretation of this outcome is that sensitivity to MPG is independent of hT1R1–hT1R3. However, a comprehensive study of the effect of lactisole on umami taste by Galindo-Cuspinera and Breslin (2006) found that although lactisole could significantly reduce the umami taste of MSG in a concentration-dependent manner (including 8mM lactisole), this occurred only when MSG was presented alone. Lactisole did not interfere with umami taste when MSG was mixed with IMP or GMP, nor when IMP or GMP were presented alone. Although the Galindo-Cuspinera and Breslin (2006) findings differed from those of a previous study that found lactisole increased the human threshold for detection of both MSG and a mixture of MSG + IMP (Xu et al. 2004), the earlier result was based on data from just 3 subjects. Galindo-Cuspinera and Breslin speculated that specific ligands may cause changes in conformation of the hT1R1–hT1R3 receptor or affect the linkage between its monomers which prevents lactisole from blocking transduction. The present data raise the possibility that MPG may also be such a ligand.

Finally, in addition to providing the first data on the effect of temperature on umami taste in humans, the present study yielded the first data on the effect of temperature on the salty taste of MSG. The finding that on the tongue tip cooling had an opposite (enhancing) effect on saltiness is consistent with the evidence from Nakamura and Kurihara (1991b) that the response of the canine chorda tympani to MSG peaks below 20 °C. A stronger response to salts at lower temperatures is also consistent with an earlier study by the same authors (Nakamura and Kurihara 1988) and a later report by Ninomiya (1996) which showed that in rodents, some amiloride-sensitive chorda tympani neurons that respond to NaCl are most sensitive at 10–12 °C. However, Nakamura and Kurihara (1988) also identified 2 additional salt-sensitive components of the whole nerve response with maxima at 30 °C, and Lundy and Contreras (1997) reported the effect of temperature on rodent salt-sensitive neurons varied depending upon the cation. Given the complexity of thermal effects on the neurobiology of salt taste in animal models, it is not surprising that the effects of temperature on saltiness taste in humans also appear complex. For example, threshold sensitivity to NaCl flowed over the tongue was shown to follow an inverted u-shaped function of temperature with a maximum near 30 °C (McBurney et al. 1973), whereas suprathreshold perception of NaCl saltiness measured in the whole mouth was found to be unaffected by cooling the tongue to 20 °C (Green and Frankmann 1987). Those results together with the present findings indicate that to understand the effects of temperature on salt taste in humans, measurements will need to be made with a amiloride-sensitive and -insensitive salts under conditions of both whole mouth and tongue tip stimulation.

In summary, the results reported here indicate that when sensed by the tongue tip, umami taste can be attenuated by cooling, but unlike sweet taste, cooling does not significantly affect umami taste adaptation. The results for whole mouth stimulation are more complex both with respect to the effect of temperature on the umami taste of MSG versus MPG and the effect of temperature on MSG saltiness. We speculate that this complexity may owe either to spatial differences in receptor innervation like those that have been reported for glutamate-sensitive receptors in rodents, or to perceptual interactions between the umami and salty components of MSG stimulation.

Funding

This research was supported in part by a grant from the National Institute on Deafness and other Communicative Disorders of the National Institutes of Health (RO1-DC05002).