The authors have declared that no competing interests exist.
Conceived and designed the experiments: KM. Performed the experiments: AM DG KM. Analyzed the data: AM DG KM. Contributed reagents/materials/analysis tools: KM. Wrote the paper: AM DG KM.
Obesity is a growing epidemic that causes many serious health related complications. While the causes of obesity are complex, there is conclusive evidence that overconsumption coupled with a sedentary lifestyle is the primary cause of this medical condition. Dietary consumption is controlled by appetite which is in turn regulated by multiple neuronal systems, including the taste system. However, the relationship between taste and obesity has not been well defined. Growing evidence suggests that taste perception in the brain is altered in obese animals and humans, however no studies have determined if there are altered taste responses in the peripheral taste receptor cells, which is the initiation site for the detection and perception of taste stimuli.
In this study, we used C57Bl/6 mice which readily become obese when placed on a high fat diet. After ten weeks on the high fat diet, we used calcium imaging to measure how taste-evoked calcium signals were affected in the obese mice. We found that significantly fewer taste receptor cells were responsive to some appetitive taste stimuli while the numbers of taste cells that were sensitive to aversive taste stimuli did not change. Properties of the taste-evoked calcium signals were also significantly altered in the obese mice. Behavioral analyses found that mice on the high fat diet had reduced ability to detect some taste stimuli compared to their littermate controls.
Our findings demonstrate that diet-induced obesity significantly influences peripheral taste receptor cell signals which likely leads to changes in the central taste system and may cause altered taste perception.
While obesity is a complex disease, it is fundamentally due to overconsumption
Taste determines whether prospective food items will be ingested or rejected. Sweet and umami tastes are appetitive and are used to detect nutrient rich foods. Bitter taste is aversive and identifies potentially harmful compounds to avoid
If taste signals are altered, feeding behaviors could be significantly impacted which could contribute to the development of obesity. Using calcium imaging to measure taste-evoked signals in peripheral taste receptor cells of obese mice, we found that multiple taste signals are significantly reduced compared to control littermates, suggesting that the responsiveness of the peripheral taste receptor cells is suppressed in obese mice. Behavioral analysis also reveals that obesity can affect some taste preferences. Since nothing is currently known about the relationship between the peripheral taste receptor cells and obesity, this study is the first to demonstrate that diet-induced obesity significantly alters the responsiveness of the peripheral taste cells that are responsible for the initial detection of taste stimuli and for sending that taste information to the brain.
All animal studies were approved by the University at Buffalo Animal Care and Use Committee under protocol number #BIO010174N. C57BL/6 mice were taken from litters that were born within a week of each other. One month after weaning, half of the mice (n = 25) from each litter were placed on high fat mouse chow (60% high fat Kcal feed, Harlan Labs, Inc., Madison, WI, USA; diet is comprised of 60% calories from fat, 22% calories from carbohydrates, 18% calories from protein) while the remaining littermates (n = 25) were kept on normal mouse chow (Harlan labs: diet is comprised of 18% calories from fat, 58% calories from carbohydrates, 24% calories from protein). Initial weights of the mice were taken and mice on the high fat diet were measured once a week for 16 weeks. Weekly measurements of the mice on the normal chow began at week 2. Average values with the standard error of the mean are reported.
To determine if diet-induced obesity affects the taste preferences of the C57BL/6 mice, mice from each group were subjected to two bottle preference tests using the protocol described
Five taste stimuli were tested: acesulfame K (AceK), sucrose, saccharin, monopotassium glutamate (MPG), and denatonium (Den). Each stimulus was presented at four concentrations in ascending order. The concentrations used were as follows (in mM): 1) AceK 1, 2, 20, 50; 2) sucrose 5, 50, 150, 300; 3) saccharin 1, 2, 10, 20; 4) MPG 10, 30, 100, 300; 5) Den 0.1, 0.5, 1, 10. Preference ratios for obese and control mice were compared using repeated measures two-way ANOVA with a Bonferroni’s
All measurements of intracellular calcium were performed in isolated taste cells. Taste receptor cells were harvested from the circumvallate and foliate papillae from adult C57BL/6 mice and were isolated from lingual epithelium as previously described
Experimental results were plotted and analyzed using OriginPro software (OriginLab Corp., Northampton, MA, USA). An evoked response was defined as measurable if the increase in fluorescence was more than two standard deviations above baseline. Calcium increases were calculated as [(peak − baseline)/baseline]×100 and were reported as percentage increases over baseline to determine the response amplitudes. We integrated the area under the curve to obtain a proportional measure of the amount of calcium in a response. The entire area under the curve was analyzed beginning with the initial change in baseline calcium and measuring until the levels returned to comparable baseline values. If the response did not return to baseline calcium levels, it was not included in the analysis. Only evoked responses were included in the analyses, null responses were not included. Statistical comparisons were made using either Student’s
All solutions used for calcium imaging were bath applied using a gravity flow perfusion system (Automate Scientific, San Francisco, CA, USA) and laminar flow perfusion chambers (RC-25F; Warner Scientific, Hamden, CT, USA). The following tastants were diluted into Tyrode’s and were used to stimulate the cells during experiments: 2 mM saccharin, 5 mM denatonium benzoate, 20 mM acesulfame K, 20 mM monopotassium glutamate.
C57Bl/6 mice (n = 25) were placed on a high fat (60% high fat kcal feed) diet within one month of weaning while littermate controls (n = 25) were kept on normal mouse chow. Weights were measured and recorded weekly for all mice and are shown in
After two weeks, mice on the high fat diet (black line) began to show rapid weight gain compared to their littermate controls on normal diet (gray lines), n = 25 for each group. Weights were recorded weekly and the average weights are reported here with standard error bars (SEM) for both female (A) and male (B). By week two, males on the high fat diet are significantly heavier than their male littermate controls (p<0.01) and continued to be significantly heavier throughout the rest of the study. By week three, female mice on the high fat diet were significantly heavier than their female littermate controls (p<0.01) and remained significantly heavier for the rest of the study.
After 10 weeks on the high fat diet, mice were between 30% (males) and 40% (females) heavier than controls. Beginning at week 10, we isolated taste receptor cells and analyzed the taste-evoked calcium signals from both obese and control mice. We tested three appetitive taste stimuli, including two sweet stimuli (2 mM saccharin-SAC and 20 mM Acesulfame K-Ace K) and one umami stimulus (20 mM monopotassium glutamate-MPG). We did not test any sugars since their effective concentrations could potentially have non-specific osmotic effects on the isolated taste cells. While the artificial sweeteners we used have been reported to generate a bitter aftertaste at higher concentrations
For each taste stimulus, we generated response profiles for the obese and control mice. The goal of these experiments was to determine if there were any differences in the responsiveness of the taste receptor cells to the taste stimuli as well as to detect any differences in the taste-evoked responses between the obese and normal mice. Thus we did not determine which taste cell type was responding to the taste stimuli, we only measured the taste-evoked calcium signals in each responsive taste cell. The exception to this was the umami (20 mM MPG) responses. Since MPG could potentially activate any neurotransmitter glutamate receptors expressed in the Type III cells
Chi-square analysis was used to compare the percentage of responsive taste receptor cells between normal and obese mice. (A) Bar graphs represent the percentages of taste-evoked calcium responses to each taste stimuli. The number of responsive taste cells for the sweet stimuli (2 mM Sac, 20 mM AceK) were significantly reduced for the obese mice (black bar) compared to controls (gray bar) (***, p<0.001; **, p<0.01; *, p<0.05). The numbers of taste-evoked calcium responses to the umami taste stimuli (20 mM MPG) and the aversive bitter stimuli (5 mM Den) were not significantly different between the two groups. (B) When obese mice were analyzed by sex, both males and females were significantly less responsive to the sweet taste stimuli tested compared to control mice (***, p<0.001; **, p<0.01; *, p<0.05).
Not only were the numbers of responsive taste cells reduced, but the properties of the calcium signals were also significantly altered in the obese mice compared to controls. Representative traces of the different taste-evoked calcium responses are shown in
Representative taste evoked calcium signals are shown from both control (A–D) and obese (E–H) mice for multiple taste stimuli: umami (20 mM MPG, A, E); sweet (2 mM Saccharin, B, F and 20 mM AceK, C, G); and bitter (5 mM denatonium, D, H).
(A) Amplitudes of the taste-evoked calcium responses were measured as a percent increase over baseline. Amplitudes were significantly reduced in obese mice for the sweet stimuli tested (2 mM Sac, 20 mM AceK) and one aversive stimulus, denatonium (5 mM). Response amplitudes of the measured umami (20 mM MPG) stimuli were not different between obese mice and their littermate controls. (B) Response amplitudes were significantly lower in both obese males and obese females for the sweet and bitter stimuli tested (***, p<0.001; **, p<0.01) with the exception of the AceK responses in obese males (p = 0.06). Average peak values (% increase over baseline) for each stimulus as shown in B: MPG, Ctl = 18.6, males = 15.7, females = 17; SAC, Ctl = 74.2, males = 19.1, females = 8.2; ACEK, Ctl = 37.8, males = 21.1, females = 7.9; DEN, Ctl = 59.8, males = 12.4, females = 22.1. (C) A proportional measurement of calcium signals were represented by the integrated area under the curve. The overall calcium responses to the sweet stimuli were significantly smaller in obese mice compared to controls. There was no difference found for the umami stimulus (20 mM MPG) or the bitter stimulus (5 mM denatonium). (D) The integrated areas of taste evoked responses for the obese males and obese females were significantly reduced for the sweet taste stimuli tested compared to control responses (***, p<0.001; **, p<0.01; *, p<0.05). No differences were found for the umami stimulus, MPG or the bitter stimulus, denatonium. Average AUC values (arbitrary units) for each stimulus as shown in D: MPG, Ctl = 14.1, males = 11.9, females = 10.8; SAC, Ctl = 22.8, males = 5.9, females = 6.6; ACEK, Ctl = 29.2, males = 4, females = 10.3; DEN, Ctl = 7.4, males = 6.4, females = 9.3.
We also integrated the area under the curve to get a proportional measurement of the overall calcium signals using the same cells that were analyzed in
After 8 weeks, mice on the high fat food were about 35% heavier than their littermate controls (
Since we used two bottle preference tests, there is potential for post-ingestive effects in these experiments. In particular, the nutritive sweet stimulus, sucrose is noted for having post-ingestive effects. For some experiments, the short-exposure lickometer test is preferred to avoid any potential post-ingestive effects
(A) For the sweet stimulus AceK, the ability to detect this sweetener at higher concentrations was impaired in obese mice compared to controls (**, p<0.01). (B) For saccharin, obese mice were significantly impaired in their ability to detect this sweet stimulus at the lower concentrations tested (*, p<0.05) but were not significantly different in their preferences from controls at higher saccharin concentrations. (C) Control mice showed an increased preference for the umami stimulus, MPG, at the highest concentration but no significant differences between normal and obese mice were measured (p = 0.056). (D) The ability to avoid denatonium at the highest concentration was compromised in obese mice (*, p<0.05) compared to littermate controls. (E) Obese mice were comparable to wild type control mice in their preference for sucrose except for the highest concentration of sucrose tested (300 mM). At this concentration, control mice had a very strong significantly higher preference for sucrose compared to the obese mice.
Our data clearly show that a high fat diet has a significant impact on the peripheral taste system, either directly or as a consequence of the subsequent weight gain. We placed age-matched C57Bl/6 mice on either a high fat diet or regular mouse chow. Regardless of the diet, all mice gained weight during the experiment since they were relatively young (7–8 weeks old) when the experiment began. However, the mice on the high fat diet rapidly gained weight and became obese (
Many studies have established that processing of taste information in the central nervous system is altered as obesity develops
After 10 weeks on the high fat diet, we analyzed the taste-evoked calcium signals in isolated taste receptor cells. Our data reveal significant changes in these peripheral taste signals associated with obesity. Isolated taste cells from obese mice were compared to taste cells from controls to determine their overall responsiveness to different types of taste stimuli. We also measured the response amplitude and integrated the area under the curve to get a measure of how the overall taste-evoked calcium signal was impacted. What we found was quite striking. The number of taste cells that were sensitive to the appetitive sweet taste stimuli was significantly reduced in the obese mice but their responsiveness to the aversive taste stimulus was unaffected. Responsiveness to the umami stimulus tested was also unchanged in the obese mice. Obese female mice had significantly fewer responsive cells to the sweet stimuli compared to the obese male mice (
Not only were fewer taste cells from the obese mice sensitive to the sweet stimuli, but within the responsive cells, the evoked signals were significantly reduced in both peak amplitude and overall area. Thus, the taste receptor cells appear to lose the ability to respond appropriately to these types of stimuli. Similar results were seen in mice fed a high fat diet that were exposed to long chain fatty acids. The overall calcium signal to the lipids was also reduced in the obese mice compared to control
Thus, these data clearly illustrate that at least some of the taste-evoked calcium signals in taste receptor cells are significantly affected in obese mice. These findings are remarkable because the only difference between the obese C57Bl/6 mice and their control littermates was the high fat diet which caused significant weight gain. It is also important to note the selectivity of the obesity effect on the peripheral taste cells, at least for the taste stimuli that we tested. In this study, we only analyzed one bitter compound and one umami compound in addition to the two artificial sweeteners that were tested. For these taste stimuli, the overall responsiveness to bitter and umami taste stimuli was not affected, supporting the conclusion that obesity is not generating a complete suppression of the peripheral taste cell sensitivity but is instead having specific effects on the ability of the cells to respond to certain appetitive taste stimuli. Further comprehensive studies that analyze more taste stimuli, in particular more bitter and umami stimuli, may find different effects. Indeed, our data for denatonium suggests that bitter stimuli can be affected by obesity. The effects were not as severe for the bitter stimulus as was seen for the sweet stimuli tested, but further studies are needed. Taken together, our data demonstrate that obese mice have significant reductions in their responsiveness for at least some taste stimuli.
Different taste stimuli activate distinct signaling pathways in taste cells. Regardless of the signaling mechanism, however, an increase in cytosolic calcium is needed to cause neurotransmitter release and is a critical component in the stimulus response from all taste cells. Recent studies have shown that cytosolic calcium changes in taste receptor cells are directly responsible for transducing all five taste qualities and directly affect the subsequent cranial nerve response
After the mice had been on the high fat food for 8 weeks, we performed a two bottle preference test to determine if the ability of the mice to detect different tastants was affected by their obesity. Obese mice had less preference for the sweet stimuli tested (AceK, sucrose and saccharin) compared to controls. In another study, diet-induced obese rats were also less sensitive to saccharin at higher concentrations when compared to control rats
In our study, obese mice were generally insensitive to the umami stimulus tested (MPG) but these values were not significantly different from controls. These findings agree with earlier studies that reveal a general indifference to MSG in naïve mice
While we cannot conclude that these behavioral changes are directly due to the changes in the taste-evoked calcium signals that we measured in the obese mice compared to controls, our findings do reveal a strong correlative relationship between the responsiveness of the peripheral taste receptor cells and the behavioral responses. The behavioral differences we found between the obese and control mice for the different taste stimuli correlate with the concentrations of those tastants that we used in the calcium imaging experiments. While future studies are required to determine if a causative relationship exists, we can conclude that diet-induced obesity causes a significant repression of the peripheral taste system’s ability to respond to sweet stimuli and that behavioral preferences to these sweet stimuli are also significantly reduced. Thus future studies on the effects of obesity on central taste processing should consider that the initial peripheral taste signals are impacted by obesity which are likely affecting those central pathways and may ultimately be affecting taste perception.
The authors wish to thank Dr. Sue Kinnamon for her insightful comments.