Physiology&Behavior,Vol. 52, pp. 385-388, 1992
0031-9384/92 $5.00 + .00 Copyright @ 1992 Pergamon Press Ltd.
Printed in the USA.
BRIEF COMMUNICATION
Abnormal Taste Preference for Saccharin in Hypothyroid Rats B. H. J. G O R D O N , ~ G E O R G E Y. W O N G ,
JOHN
LIU AND
RICHARD
S. R I V L I N
Hunter College, C.U.N.Y., N e w York, N Y 10010, and Memorial Sloan-Kettering Cancer Center and Cornell University Medical College, New York, N Y 10021 R e c e i v e d 15 J u l y 1991 GORDON, B. H. J., G. Y. WONG, J. LIU AND R. S. RIVLIN. Abnormal taste pre/i,rence[br saccharin in h.vpothyroid rats. PHYSIOL BEHAV 52(2) 385-388, 1992.--Taste preferences for saccharin in concentrations ranging from 0.16 mM to 50 mM were determined in rats made hypothyroid with radioactive iodine and in their littermate controls. Hypothyroid rats demonstrated taste preferences for saccharin which were similar to those of controls only at very low (0.016 mM) or very high (49.0 raM) saccharin concentrations. At these concentrations of tastant, the preferences for tastant and water were similar to one another. At a concentration of 5.1 raM. preferences were also very similar in both groups but were very high. At intermediate saccharin concentrations of 1.1 and 3.0 raM, hypothyroid animals showed significantly lower percent preferences for the sweet tastant than did controls, mean +_ SEM (62.48 ___5.97 vs. 82.92 _+ 4.60, p = 0.0002) for the 1.1 mM concentration and (74.98 + 5.12 vs. 89.40 _+ 2.54, p = 0.0029) for the 3.0 mM concentration. These changes in taste preference for saccharin in hypothyroid rats were similar in direction and magnitude to those previously published by this laboratory using sucrose as the tastant. Thus, hypothyroid rats demonstrate abnormalities in taste preference for both the nonnutritive sweetener, sodium saccharin, as well as for the nutritive sweetener, sucrose. Hypothyroidism
Taste abnormalities
Saccharin
Sweet tastant
D I S T U R B A N C E S of taste preference have been reported in hypothyroidism, both in experimental animals (8-11,15,35) and in humans (7,13,36,37). In humans, elevated taste thresholds for sweet, sour, salty, or bitter substances have been observed in as many as 83% of hypothyroid patients (27). Measurements of suprathreshold taste functions demonstrate that abnormalities occur early in the course of evolving hypothyroidism (26). Compared to controls, hypothyroid patients display flatter intensity and preference functions for suprathreshold concentrations of salty and bitter stimuli throughout a wide concentration range, disturbances that may be a factor contributing to the decrease in food intake known to occur in this disorder (1). In one study, hypothyroid patients reported alterations in taste sensations and their alleviation by treatment with thyroid hormones, but the investigators were unable to demonstrate an objective defect in taste function (22). In an early report, no consistent alteration in taste thresholds was observed in hypothyroid patients, but the range of thresholds appeared to enlarge (32).
Quinine sulfate
Sodium chloride
The fact that hypothyroid animals consume relatively less of a sweet stimulus (sucrose) and relatively more of a bitter stimulus (quinine) than do controls is consistent with a defect of taste function. The disturbances in sucrose consumption in hypothyroidism, however, could possibly be attributable to other factors, such as associated disturbances in glucose homeostasis, which may be expected to influence sucrose intake. It is, therefore, of importance to compare taste preferences of hypothyroid rats with those of euthyroid controls for the nonnutritive sweetener, sodium saccharin. METHOD
Animals Individually caged male Holtzman rats were used in all studies fed Purina Chow Pellets ad lib from weaning until l month of age. Animals were then placed for 1 month on an iodine-deficient diet (Iodine Deficient Diet, Teklabs, Madison, WI). Half the rats
This investigation was supported by research grants IP30CA 29502 and CA 08748 from the National Institutes of Health, and by grants from the Stella and Charles Guttman Foundation and the General Foods Fund. 1 Requests for reprints should be addressed to Barbara H. J. Gordon, Ph.D., Hunter College, C.U.N.Y., Nutrition and Food Science Program, 425 East 25 Street, New York, New York 10010, phone: 212-481-7563
385
386
( ; O R D ( ) N EI AI..
were rendered hypothyroid by a single IP injection of 300/xCi J~l per 100 g of body weight. Forty-eight hours after this dose was administered, the iodine-deficient diet was replaced by Purina Chow Pellets. These rats were studied 2 months after receiving the thyroid-ablating dose of radioactive iodine. Hypothyroidism is produced reliably by this technique with resultant reduction of serum T 4 and T 3 levels, as reported previously from this laboratory (35). Room temperature was maintained at 21.5°C throughout.
Seh'ction h~r Right-L¢lt Prd~'rence All animals were tested initially for right-left drinking preference to eliminate bias due to the location of drinking bottles on the cage. Two identical 250 ml rubber-stoppered glass bottles containing distilled water were attached, one on the right, the other on the left side of the cage, with both bottles equally accessible to the animal. After 48 hours, the amount of water consumed from each bottle was measured and compared. Animals were discarded from the study if consumption from one bottle exceeded that from the other by 25% or more during each of 34 successive trials. Overall, 10-15% of the total number of rats tested initially were eliminated.
l"aste Pre/erem'e Tests For taste preference tests, two 250 ml rubber-stoppered glass bottles were filled with either distilled water or tastant dissolved in distilled water. During a 48-hour test period the animals drank ad lib. At the end of this period, measurements of the contents remaining in each bottle were made, the bottles were refilled, and their positions on the cage were reversed. Two consecutive 48-hour periods constituted one taste test. After the elimination of animals exhibiting significant rightleft preference, each study animal was tested with a quinine solution compared to water to validate the model. Our previous studies showed that hypothyroid rats reject this bitter stimulus less well than do controls throughout a wide concentration range (35). In the present study, a single concentration of quinine (0.67 mM) was employed. Subsequently, preference tests were pertbrmed in the order of progressively increasing concentrations with the sweet tastant saccharin, at 0.016 raM, 1.1 raM, 3.0 mM, 5.1 mM, and 49 mM. Taste preferences, expressed as percentages, were calculated according to the following formula: percent preference vol (ml) test solution consumed × 100. vol (ml) test solution + vol (ml) water consumed The data presented are the means of two consecutive taste tests. The number of animals per group varied between 8-10 for each taste test. The same animals were used in all experiments.
of ANCOVA. the contounding factors, concentration and total fluid intake, were linearly adjusted in the comparison of taste preference between the two groups of animals. R EStJ I l'S
Hypothyroid rats had taste preferences for quinine sulfate (0.67 mM) which were nearly twice those of the controls (22 _+ 1% vs. 13 +_ 1%, p < 0.001 ), respectively. These data are virtually identical to those obtained in hypothyroid animals investigated previously in this laboratory. (35). Thus, when faced with a choice between water and a bitter stimulus, hypothyroid rats consumed relatively more of the bitter stimulus than did animals with normal thyroid function. Studies were next performed with the nonnutritive sweetener, sodium saccharin. The comparison between percent preference and concentration revealed a linear relationship for the hypothyroid group. Therefore, a linear regression line was fitted to the individual data using all available replications at each concentration. However, there was a systematic nonlinear relationship between percent preference and concentration in the control group. Instead of fitting a complicated nonlinear statistical model to the control data, the control data for preference were compared at each concentration with the mean hypothyroid preference value as predicted by the linear regression model for the hypothyroid group. In terms of possible learned preference or aversion to the tastants, between the first and second 48-hour trials, the control group appeared to have a greater preference in the first of the two taste experiences with saccharin at concentrations of 0.016 (n - 10), and 1.1 m M (n - 8) (p = 0.035 and p - 0.019 by paired l-test). The same was true for hypothyroid animals at 0.016 m M (it - 9, p - 0.039), but was reversed at 3.0 m M (n = 8, p -: 0.013). When preference was compared to total fluid intake, a suggestion of weak statistical significance existed for higher preferences with greater volumes of intake in control rats. No such trend was apparent for the hypothyroid subjects. At a low concentration of the tastant, 0.016 raM, which appears to be very close to the taste threshold for the rat, percent preferences in control and hypothyroid rats were similar. Neither group consumed significantly more saccharin than water. As the concentrations of saccharin increased to 1.1 m M and 3.0 raM. hypothyroid rats had lower increments in taste preference than did euthyroid controls (p = 0.0002 and p = 0.0029, respectively). At a still higher concentration of sodium saccharin (5.1 raM), taste preference was maximal and nearly identical in both groups of rats. When the saccharin concentration was raised even further to the very high concentration of 49 raM, taste percent preferences began to decline and were comparable for the hypothyroid (n = 11.66.94 _+ 6.59%) and control animals (n .... 14, 60.59 _+ 7.94~ ). The comparison for this last, and very high, concentration was made using a two-sample t-test instead of the regression method to avoid the problem of extrapolation. Such a test was feasible because there were enough rats in both the hypothyroid and control groups (see Fig. 1).
Statistical Analysis All comparisons of percent preference were made between control (euthyroid) and hypothyroid animals. There is a possible dependence between preference and concentration or preference and total fluid intake; therefore, the analysis ofcovafiance (ANCOVA) technique (6) was used to compare taste preference between control and hypothyroid animals. The A N C O V A technique is an extension of the usual A N O V A technique which includes adjustments for confounding factors. In our application
DISCUSSION
A number of investigators have reported changes in the consumption of various tastants in hypothyroid animals that in their aggregate are consistent with the hypothesis that hypothyroidism may cause disturbances of taste function. Thus, Hoshishima et al. observed elevated thresholds in mice for detecting saccharin, acetic acid, and PTU at low concentrations (15). Fregly found an increase in salt consumption in hypothyroid rats (8-11).
TASTE FOR SACCHARIN IN HYPOTHYROID RATS
387
100
9(] 80
70 W
o z w nc 60, w ,J i.i. w ¢c 5( O.
4O
30
N
5'o
SACCHARIN CONCENTRATION (mM)
FIG. 1. Taste preference for sodium saccharin in hypothyroid rats and in euthyroid littermate controls, expressed as percentages. Data shown are mean _+ SE of two 48-h trials with 8-10 animals per group (O = hypothyroid rats; O = controls). Statisticalsignificanceof the differences is shown by an asterisk. **p = 0.0029: ***p = 0.0002.
Similar findings have been reported for adrenalectomized rats (10,33,34). In previous studies from this laboratory, decreased preference for sucrose and elevated preferences for salt and quinine were clearly noted in rats rendered hypothyroid with radioactive iodine (35). The mechanisms underlying the changes in consumption patterns for various tastants in hypothyroidism are unknown, and several possibilities need to be considered. Changes in adrenal structure and function reported in hypothyroid rats (2,5,14,21,28) result in diminished mineralocorticoid secretion (25), which would be expected to increase salt intake. Abnormalities in salivary flow and composition have been reported in hypothyroidism after radioiodine ablation (38), thyroidectomy and treatment with propylthiouracil (39), or methylthouracil (17,37,39). Many patients with hypothyroidism complain of dry mouth (44). Disturbances in salivary function may contribute to abnormal taste preferences observed in patients with hypothyroidism of any etiology, although another report suggests that despite altered thresholds, patients are not aware of any abnormality (43,44). It is important to note that changes in consumption patterns consistent with abnormalities of taste have been observed in animals with hypothyroidism, whether produced by radioactive iodine ablation (35), surgical thyroidectomy (39), or antithyroid drugs (12,23). While damage to the salivary glands certainly may result from treatment with large doses of radioactive iodine, this mechanism is unlikely to completely explain the changes in consumption patterns that occur under these conditions, inasmuch as treatment with thyroxine can return elevated thresholds for quinine to normal levels (35). Much evidence indicates that salivary gland dysfunction influences taste perception in general (44); and that desalivated rats have marked sensory loss (3). Salivary calmodulin, a membrane-bound protein which regulates c-AMP phosphodiesterase (PDE), is believed to influence
taste sensations. This enzyme is present in significantly lower concentrations in saliva from patients with a variety of taste dysfunctions (20). Thyroid hormone was previously reported to inhibit PDE in isolated taste bud membranes (19), an effect that could contribute to the dependence of taste sensations upon thyroid function. In addition, reversible inhibition of renin mRNA synthesis is noted after the administration of the antithyroid drug, propylthiouracil, to mice (17), which could contribute to changes in salt consumption. Thyroid hormones, furthermore, facilitate development ofaversive responses to quinine (40). In neonatal animals, low doses of thyroid hormones accelerate the development of unconditioned taste responses (43). In view of these considerations, it was of interest to study taste preference for the nonnutritive sweetener, sodium saccharin. This tastant is known to be approximately 100 times as sweet as sucrose (31,41) and to stimulate the chordi tympani nerve and superficial petrosal nerve responses, the latter being relatively more important in mediating responses to saccharin (16,30). These effects are positively correlated with behavioral responses in many mammals, including rats (29,31). The abnormalities in taste preference which we report here for saccharin are in the same direction and are of similar magnitude to those we previously observed with sucrose (35). At low concentrations of saccharin, taste preferences are decreased in hypothyroid rats. These findings lend further support to our hypothesis that defects of taste per se are likely to occur in experimental hypothyroidism as well as in patients with this disorder. At high concentrations of saccharin, hypothyroid and control animals exhibit similarly increased preferences. We postulate that at these high concentrations of tastant, receptors may be saturated with taste stimuli and, therefore, taste preferences in hypothyroid rats would attain the same high level as in euthyroid controls. At still higher concentrations of tastant there appears to be an inhibitory response (29,31). Such a behavioral effect may possibly be due to stimuli binding to a second population of receptors with lower affinity constants (31), or to the perception of the stimuli as being less pleasant. In the case of saccharin, a bitter aftertaste could explain decreased taste preference at high concentrations. Similarly, in streptozotocin-induced diabetic rats, there is also a decreased preference for high concentrations of saccharin and sucrose (42). The present findings do not exclude the possibility that saccharin may have direct metabolic effects that can influence taste preference. For example, in the presence of saccharin, 50% inhibition of the glycolytic enzymes, hexokinase and glyceraldehyde-3-phosphate dehydrogenase, occurred in cell-free extracts of streptococcus mutants (24). Inhibition of lactate dehydrogenase and overall utilization of glucose also may be decreased in the presence of saccharin (4). Furthermore, saccharin can attenuate the volume of food-stimulated secretion of pancreatic juice, lowering the amounts of specific digestive enzymes, such as pepsin and amylase, in the intestinal tract (18). It is likely that the mechanisms producing the abnormal taste behaviors recorded both in hypothyroid animals and in hypothyroid patients are complex and multifactorial. Further research is required to elucidate the molecular mechanisms underlying these apparent abnormalities.
ACKNOWLEDGEMENTS
The authors would like to acknowledge the encouragement and scientific advice generously provided during the course of these studies by the late Martha Osnos, M.S. and John Pinto, Ph.D.
388
(i()RI)t)N
1 I
,\i
REFERENCES 1. Abraham, G.: Fakou, R., Rozen, R.: Mandenofl, A.: Autissier, N.: Apfelbaum, M. The effect of a constant T3 level and thermoneutrality in diet-induced hyperphagia. Horm. Metab. Res. 19:96-100: 1987. 2. Baumann, E. J.: Marine, D. Involution of the adrenal cortex in rats fed with thiouracil. Endocrinology 36:400-405; 1945. 3. Brosvic, G. M.; Hoey, N. E. Taste detection and descrimination performance of rats following selective desalivation. Physiol. Behav. 48:617-623: 1990. 4. Brown, A. T.: Best, G. M. A proposed mechanism for the effects of saccharin in glucose metabolism by streptococcus mutans. Caries Res. 20:406-418; 1986. 5. Deane, H. W.: Greep, R. O. A cytochemical study of the adrenal cortex in hypo- and hyperthyroidism. Endocrinology 41:243-257; 1947. 6. Dunn, O. J.: Clark, V. A. Applied statistics: Analysis of variance and regression. New York: John Wiley; 1974:387p. 7. Fisher, R.: Griffin, F. Taste-blindness and variations in taste threshold in relation to thyroid metabolism. J. Neuropsychol. 3:98-104: 1961. 8. Fregly, M. J. The role of hormones in the regulation of salt intake in rats. In: Kare, M. R.: Mailer, O., eds. The chemical senses and nutrition. Baltimore: Johns Hopkins University Press; 1967:115138. 9. Fregly, M. J.: Brimhall, R. L.; Galindo, O. J. Effect of the antithyroid drug propylthiouracil in sodium balance of rats. Endocrinology 71: 693-700: 1962. 10. Fregly, M. J.: Galindo, O.: Cook, K. M. Spontaneous sodium chloride appetite of goitrogen-treated rats: Effect of hypophysectomy and adrenalectomy. Endocrinology 69:1060-1067:1961. 11. Fregly, M. J.: Waters, 1. W. Effect of propylthiouracil on preference threshold of rats for NaCI solutions. Proc. Soc. Exp. Biol. Med. 120: 637-640: 1965. 12. Green, W. L. Mechanisms of action of antithyroid compounds. In: Werner, S. C., Ingbar, S. H., eds. The thyroid. 4th edition. New York: Harper and Row, 1978:57-65. 13. Hallman, B. L.; Hurst, J. W. Loss of taste as toxic effect ofmethimazole (Tapazole) therapy. JAMA 152:322: 1953. 14. Holmes, E. W., Jr.: Di Scala, V. A. Studies on exaggerated natnurectic response to a saline infusion in the hypothyroid rat. J. Clin. Invest. 49:1224- t 236:1971). 15. Hoshishima, K.: Yokoyama, S.; Seto, K. Taste sensitivity in various strains of mice. Am. J. Physiol. 202:1200-1204: 1962. 16. Jakinovich, W. Stimulation of the gerbil's gustatory.' receptors by saccharin. J. Neurosci. 2:49-56; 1982. 17. Karen P.: Morris, B. J. Stimulation by thyroid hormone of renin mRNA in mouse submandibular gland. Am. J. Physiol. 25 I:E290E293; 1986. 18. Kondo, M.: Ohe, M.: Ono, A.; Magihira, M. Effect of sodium saccharin on rat pancreatic enzyme secretion. J. Nutr. Sci. Vitaminol. 30:56%576: 1984. 19. Law, J. S.; Henkin, R. 1. Thyroid hormone inhibits purified taste bud membrane adenosine 3'5' monophosphate phosphodiesterase activity. Res. Comm. Chem. Pathol. Pharmacol. 43:449-462; 1984. 20. Law, J. S.: Henkin, R. I. Low parotid saliva calmodulin in patients with taste and smell dysfunction. Biochem. Med. Metab. Biol. 36: 118-124: 1986. 21. Leblond, C. P.: Huff; H. E. Effects of sulfonamides and thiourea derivatives on heart rate and organ morphology. Endocrinology 35: 229-233: 1944.
22. I.ev~ilt. M. S.; l.aing. D. G.: F'anhubei, M.; ('orbctl. A.: ('artcr. J. M. Sensory perception and h,,polhyroidism. Chem. Senses 14: 537-546: 1989. 23. l.eys, D. ttyperthyroidism treated ~ith methylthiouracil, lancel !: 461-464:1945. 24. kinke, H. A. B.; Kohn, J. S. Inhibitory effect of saccharin on glycolytic enzymes in cell-free extracts of streptococcus mutans. Caries Res. 18:12-16; 1984. 25. Marks, P.: Anderson, J,: Vincent, R. Aldosterone m myxedema. Lancet 2:1277-1278: 1978. 26. Mattes. R. D.: Heller, A. D.; Rivlin, R. S. Abnormalities in suprathreshold taste function in early hypothyroidism in humans. In: Meisleman, M. l_.: Rivfin, R. S., eds. Clinical measurement of taste and smell. New York: MacMillan: 1986:467-486. 27. McConnell, R. J.: Menendez, (7. E.: Smith, F. R,: Henkin, R. 1.: Rivlin, R. S. Defects of taste and smell in patients with hypothyroidism. Am. J. Med. 59:354-363: 1975. 28. Michael, V. F.: Barenberg, R. 1.: Chavez, R.: Vaamonde. C. A.: Papper, S. Renal handling of sodium and water in the hypothyroid rat. J. Clin. Invest. 51:1405-1412: 1972. 29. Mook. D. G. Saccharin preference in the rat: Some unpalatable findings. Psychol. Rev. 81:475-490: 1974. 30. Norgren, R.: Nishijo, H.: Travers, S. P. Taste responses from the entire gustatory apparatus. Ann. NY Acad. Sci. 575:246-263: 1989. 31. Ogawa. M.: Sato, M.: Yamashita, S. Gustatory impulse discharges in response to saccharin in rats and hamsters. J. Physiol. 204:31 I329; 1969. 32. Pittman, J. A.: geschi, R. J. Taste thresholds in hyper- and hypothyroidism. J. Cfin. Endocrinol. Metab. 27:895-896; 1967. 33. Richter, C P. Increased salt appetite in adrenalectomized rats. Am. J. Physiol. 115:155-161: 1936. 34. Richter. C. P. Salt taste thresholds of normal and adrenalectomized rats. Endocrinology 24:367-37 I: 1939. 35. Rivlin, R. S.: Osnos, M.: Rosenthal, S.: Henkin, R. I. Abnormalities in taste preference in hypothyroid rats. Am. J. Physiol. 232:ES0E84: 1977. 36. Scfiaupp, H. Geruch und Geschmack bet endokrinen Erkankungen. Arch. Klin. Exp. Ohr. Nas. Kefikopfheilk. 195:179-191: 1969. 37. Schneeberg, N. G. Loss of sense of taste due to methylthiouracil therapy. JAMA 149:1091-1093: 1952. 38. Schneyer, k.: Tanchester, D. Some oral aspects of radioactive iodine therapy for thyroid disease. NY J. Dent. 24:308-309: 1954. 39. Shafer, W. G.: Clark, P. G,: Bixler, D.: Muhler, J. C. Salivary gland function in rats, I1. Effect of thyroid function on salivary flow and viscosity. Proc. Soc. Exp. Biol. Med. 98:245-247: 1958, 40. Shapiro, E. G.: Johnson, I. B. The development ofaversive responses to quinine in hyperthyroid rats. Behav. Neural Biol. 43:274-286: 1985. 41. Shimazaki, K.: Sato, M.: Takegami, T. Binding of 3~S-saccharin to a protein fraction of rat tongue epithelia. Biochim. Biophys. Acta 677:331-338: 1981. 42. Smith, J. C.; Gannon, K. S. Ingestion patterns of food, water, saccharin and sucrose in streptozotocin-induced diabetic rats. Physiol. Behav. 49:181-199: 1991. 43. Vogt, M. B.: Rudy, J. W. Neonatal hyperthyroidism in the rat: Thyroxine accelerates the development of unconditioned but not learned responses to taste. Behav. Neural Biol. 46:358-371; 1986. 44. Weiffenbach, J. M.: Fox, P. C.: Baum, B. J. Taste and salivary hmction. Proc. Natl. Acad. Sci. 83:6103-6106: 1986.
0031-9384/92 $5.00 + .00 Copyright @ 1992 Pergamon Press Ltd.
Printed in the USA.
BRIEF COMMUNICATION
Abnormal Taste Preference for Saccharin in Hypothyroid Rats B. H. J. G O R D O N , ~ G E O R G E Y. W O N G ,
JOHN
LIU AND
RICHARD
S. R I V L I N
Hunter College, C.U.N.Y., N e w York, N Y 10010, and Memorial Sloan-Kettering Cancer Center and Cornell University Medical College, New York, N Y 10021 R e c e i v e d 15 J u l y 1991 GORDON, B. H. J., G. Y. WONG, J. LIU AND R. S. RIVLIN. Abnormal taste pre/i,rence[br saccharin in h.vpothyroid rats. PHYSIOL BEHAV 52(2) 385-388, 1992.--Taste preferences for saccharin in concentrations ranging from 0.16 mM to 50 mM were determined in rats made hypothyroid with radioactive iodine and in their littermate controls. Hypothyroid rats demonstrated taste preferences for saccharin which were similar to those of controls only at very low (0.016 mM) or very high (49.0 raM) saccharin concentrations. At these concentrations of tastant, the preferences for tastant and water were similar to one another. At a concentration of 5.1 raM. preferences were also very similar in both groups but were very high. At intermediate saccharin concentrations of 1.1 and 3.0 raM, hypothyroid animals showed significantly lower percent preferences for the sweet tastant than did controls, mean +_ SEM (62.48 ___5.97 vs. 82.92 _+ 4.60, p = 0.0002) for the 1.1 mM concentration and (74.98 + 5.12 vs. 89.40 _+ 2.54, p = 0.0029) for the 3.0 mM concentration. These changes in taste preference for saccharin in hypothyroid rats were similar in direction and magnitude to those previously published by this laboratory using sucrose as the tastant. Thus, hypothyroid rats demonstrate abnormalities in taste preference for both the nonnutritive sweetener, sodium saccharin, as well as for the nutritive sweetener, sucrose. Hypothyroidism
Taste abnormalities
Saccharin
Sweet tastant
D I S T U R B A N C E S of taste preference have been reported in hypothyroidism, both in experimental animals (8-11,15,35) and in humans (7,13,36,37). In humans, elevated taste thresholds for sweet, sour, salty, or bitter substances have been observed in as many as 83% of hypothyroid patients (27). Measurements of suprathreshold taste functions demonstrate that abnormalities occur early in the course of evolving hypothyroidism (26). Compared to controls, hypothyroid patients display flatter intensity and preference functions for suprathreshold concentrations of salty and bitter stimuli throughout a wide concentration range, disturbances that may be a factor contributing to the decrease in food intake known to occur in this disorder (1). In one study, hypothyroid patients reported alterations in taste sensations and their alleviation by treatment with thyroid hormones, but the investigators were unable to demonstrate an objective defect in taste function (22). In an early report, no consistent alteration in taste thresholds was observed in hypothyroid patients, but the range of thresholds appeared to enlarge (32).
Quinine sulfate
Sodium chloride
The fact that hypothyroid animals consume relatively less of a sweet stimulus (sucrose) and relatively more of a bitter stimulus (quinine) than do controls is consistent with a defect of taste function. The disturbances in sucrose consumption in hypothyroidism, however, could possibly be attributable to other factors, such as associated disturbances in glucose homeostasis, which may be expected to influence sucrose intake. It is, therefore, of importance to compare taste preferences of hypothyroid rats with those of euthyroid controls for the nonnutritive sweetener, sodium saccharin. METHOD
Animals Individually caged male Holtzman rats were used in all studies fed Purina Chow Pellets ad lib from weaning until l month of age. Animals were then placed for 1 month on an iodine-deficient diet (Iodine Deficient Diet, Teklabs, Madison, WI). Half the rats
This investigation was supported by research grants IP30CA 29502 and CA 08748 from the National Institutes of Health, and by grants from the Stella and Charles Guttman Foundation and the General Foods Fund. 1 Requests for reprints should be addressed to Barbara H. J. Gordon, Ph.D., Hunter College, C.U.N.Y., Nutrition and Food Science Program, 425 East 25 Street, New York, New York 10010, phone: 212-481-7563
385
386
( ; O R D ( ) N EI AI..
were rendered hypothyroid by a single IP injection of 300/xCi J~l per 100 g of body weight. Forty-eight hours after this dose was administered, the iodine-deficient diet was replaced by Purina Chow Pellets. These rats were studied 2 months after receiving the thyroid-ablating dose of radioactive iodine. Hypothyroidism is produced reliably by this technique with resultant reduction of serum T 4 and T 3 levels, as reported previously from this laboratory (35). Room temperature was maintained at 21.5°C throughout.
Seh'ction h~r Right-L¢lt Prd~'rence All animals were tested initially for right-left drinking preference to eliminate bias due to the location of drinking bottles on the cage. Two identical 250 ml rubber-stoppered glass bottles containing distilled water were attached, one on the right, the other on the left side of the cage, with both bottles equally accessible to the animal. After 48 hours, the amount of water consumed from each bottle was measured and compared. Animals were discarded from the study if consumption from one bottle exceeded that from the other by 25% or more during each of 34 successive trials. Overall, 10-15% of the total number of rats tested initially were eliminated.
l"aste Pre/erem'e Tests For taste preference tests, two 250 ml rubber-stoppered glass bottles were filled with either distilled water or tastant dissolved in distilled water. During a 48-hour test period the animals drank ad lib. At the end of this period, measurements of the contents remaining in each bottle were made, the bottles were refilled, and their positions on the cage were reversed. Two consecutive 48-hour periods constituted one taste test. After the elimination of animals exhibiting significant rightleft preference, each study animal was tested with a quinine solution compared to water to validate the model. Our previous studies showed that hypothyroid rats reject this bitter stimulus less well than do controls throughout a wide concentration range (35). In the present study, a single concentration of quinine (0.67 mM) was employed. Subsequently, preference tests were pertbrmed in the order of progressively increasing concentrations with the sweet tastant saccharin, at 0.016 raM, 1.1 raM, 3.0 mM, 5.1 mM, and 49 mM. Taste preferences, expressed as percentages, were calculated according to the following formula: percent preference vol (ml) test solution consumed × 100. vol (ml) test solution + vol (ml) water consumed The data presented are the means of two consecutive taste tests. The number of animals per group varied between 8-10 for each taste test. The same animals were used in all experiments.
of ANCOVA. the contounding factors, concentration and total fluid intake, were linearly adjusted in the comparison of taste preference between the two groups of animals. R EStJ I l'S
Hypothyroid rats had taste preferences for quinine sulfate (0.67 mM) which were nearly twice those of the controls (22 _+ 1% vs. 13 +_ 1%, p < 0.001 ), respectively. These data are virtually identical to those obtained in hypothyroid animals investigated previously in this laboratory. (35). Thus, when faced with a choice between water and a bitter stimulus, hypothyroid rats consumed relatively more of the bitter stimulus than did animals with normal thyroid function. Studies were next performed with the nonnutritive sweetener, sodium saccharin. The comparison between percent preference and concentration revealed a linear relationship for the hypothyroid group. Therefore, a linear regression line was fitted to the individual data using all available replications at each concentration. However, there was a systematic nonlinear relationship between percent preference and concentration in the control group. Instead of fitting a complicated nonlinear statistical model to the control data, the control data for preference were compared at each concentration with the mean hypothyroid preference value as predicted by the linear regression model for the hypothyroid group. In terms of possible learned preference or aversion to the tastants, between the first and second 48-hour trials, the control group appeared to have a greater preference in the first of the two taste experiences with saccharin at concentrations of 0.016 (n - 10), and 1.1 m M (n - 8) (p = 0.035 and p - 0.019 by paired l-test). The same was true for hypothyroid animals at 0.016 m M (it - 9, p - 0.039), but was reversed at 3.0 m M (n = 8, p -: 0.013). When preference was compared to total fluid intake, a suggestion of weak statistical significance existed for higher preferences with greater volumes of intake in control rats. No such trend was apparent for the hypothyroid subjects. At a low concentration of the tastant, 0.016 raM, which appears to be very close to the taste threshold for the rat, percent preferences in control and hypothyroid rats were similar. Neither group consumed significantly more saccharin than water. As the concentrations of saccharin increased to 1.1 m M and 3.0 raM. hypothyroid rats had lower increments in taste preference than did euthyroid controls (p = 0.0002 and p = 0.0029, respectively). At a still higher concentration of sodium saccharin (5.1 raM), taste preference was maximal and nearly identical in both groups of rats. When the saccharin concentration was raised even further to the very high concentration of 49 raM, taste percent preferences began to decline and were comparable for the hypothyroid (n = 11.66.94 _+ 6.59%) and control animals (n .... 14, 60.59 _+ 7.94~ ). The comparison for this last, and very high, concentration was made using a two-sample t-test instead of the regression method to avoid the problem of extrapolation. Such a test was feasible because there were enough rats in both the hypothyroid and control groups (see Fig. 1).
Statistical Analysis All comparisons of percent preference were made between control (euthyroid) and hypothyroid animals. There is a possible dependence between preference and concentration or preference and total fluid intake; therefore, the analysis ofcovafiance (ANCOVA) technique (6) was used to compare taste preference between control and hypothyroid animals. The A N C O V A technique is an extension of the usual A N O V A technique which includes adjustments for confounding factors. In our application
DISCUSSION
A number of investigators have reported changes in the consumption of various tastants in hypothyroid animals that in their aggregate are consistent with the hypothesis that hypothyroidism may cause disturbances of taste function. Thus, Hoshishima et al. observed elevated thresholds in mice for detecting saccharin, acetic acid, and PTU at low concentrations (15). Fregly found an increase in salt consumption in hypothyroid rats (8-11).
TASTE FOR SACCHARIN IN HYPOTHYROID RATS
387
100
9(] 80
70 W
o z w nc 60, w ,J i.i. w ¢c 5( O.
4O
30
N
5'o
SACCHARIN CONCENTRATION (mM)
FIG. 1. Taste preference for sodium saccharin in hypothyroid rats and in euthyroid littermate controls, expressed as percentages. Data shown are mean _+ SE of two 48-h trials with 8-10 animals per group (O = hypothyroid rats; O = controls). Statisticalsignificanceof the differences is shown by an asterisk. **p = 0.0029: ***p = 0.0002.
Similar findings have been reported for adrenalectomized rats (10,33,34). In previous studies from this laboratory, decreased preference for sucrose and elevated preferences for salt and quinine were clearly noted in rats rendered hypothyroid with radioactive iodine (35). The mechanisms underlying the changes in consumption patterns for various tastants in hypothyroidism are unknown, and several possibilities need to be considered. Changes in adrenal structure and function reported in hypothyroid rats (2,5,14,21,28) result in diminished mineralocorticoid secretion (25), which would be expected to increase salt intake. Abnormalities in salivary flow and composition have been reported in hypothyroidism after radioiodine ablation (38), thyroidectomy and treatment with propylthiouracil (39), or methylthouracil (17,37,39). Many patients with hypothyroidism complain of dry mouth (44). Disturbances in salivary function may contribute to abnormal taste preferences observed in patients with hypothyroidism of any etiology, although another report suggests that despite altered thresholds, patients are not aware of any abnormality (43,44). It is important to note that changes in consumption patterns consistent with abnormalities of taste have been observed in animals with hypothyroidism, whether produced by radioactive iodine ablation (35), surgical thyroidectomy (39), or antithyroid drugs (12,23). While damage to the salivary glands certainly may result from treatment with large doses of radioactive iodine, this mechanism is unlikely to completely explain the changes in consumption patterns that occur under these conditions, inasmuch as treatment with thyroxine can return elevated thresholds for quinine to normal levels (35). Much evidence indicates that salivary gland dysfunction influences taste perception in general (44); and that desalivated rats have marked sensory loss (3). Salivary calmodulin, a membrane-bound protein which regulates c-AMP phosphodiesterase (PDE), is believed to influence
taste sensations. This enzyme is present in significantly lower concentrations in saliva from patients with a variety of taste dysfunctions (20). Thyroid hormone was previously reported to inhibit PDE in isolated taste bud membranes (19), an effect that could contribute to the dependence of taste sensations upon thyroid function. In addition, reversible inhibition of renin mRNA synthesis is noted after the administration of the antithyroid drug, propylthiouracil, to mice (17), which could contribute to changes in salt consumption. Thyroid hormones, furthermore, facilitate development ofaversive responses to quinine (40). In neonatal animals, low doses of thyroid hormones accelerate the development of unconditioned taste responses (43). In view of these considerations, it was of interest to study taste preference for the nonnutritive sweetener, sodium saccharin. This tastant is known to be approximately 100 times as sweet as sucrose (31,41) and to stimulate the chordi tympani nerve and superficial petrosal nerve responses, the latter being relatively more important in mediating responses to saccharin (16,30). These effects are positively correlated with behavioral responses in many mammals, including rats (29,31). The abnormalities in taste preference which we report here for saccharin are in the same direction and are of similar magnitude to those we previously observed with sucrose (35). At low concentrations of saccharin, taste preferences are decreased in hypothyroid rats. These findings lend further support to our hypothesis that defects of taste per se are likely to occur in experimental hypothyroidism as well as in patients with this disorder. At high concentrations of saccharin, hypothyroid and control animals exhibit similarly increased preferences. We postulate that at these high concentrations of tastant, receptors may be saturated with taste stimuli and, therefore, taste preferences in hypothyroid rats would attain the same high level as in euthyroid controls. At still higher concentrations of tastant there appears to be an inhibitory response (29,31). Such a behavioral effect may possibly be due to stimuli binding to a second population of receptors with lower affinity constants (31), or to the perception of the stimuli as being less pleasant. In the case of saccharin, a bitter aftertaste could explain decreased taste preference at high concentrations. Similarly, in streptozotocin-induced diabetic rats, there is also a decreased preference for high concentrations of saccharin and sucrose (42). The present findings do not exclude the possibility that saccharin may have direct metabolic effects that can influence taste preference. For example, in the presence of saccharin, 50% inhibition of the glycolytic enzymes, hexokinase and glyceraldehyde-3-phosphate dehydrogenase, occurred in cell-free extracts of streptococcus mutants (24). Inhibition of lactate dehydrogenase and overall utilization of glucose also may be decreased in the presence of saccharin (4). Furthermore, saccharin can attenuate the volume of food-stimulated secretion of pancreatic juice, lowering the amounts of specific digestive enzymes, such as pepsin and amylase, in the intestinal tract (18). It is likely that the mechanisms producing the abnormal taste behaviors recorded both in hypothyroid animals and in hypothyroid patients are complex and multifactorial. Further research is required to elucidate the molecular mechanisms underlying these apparent abnormalities.
ACKNOWLEDGEMENTS
The authors would like to acknowledge the encouragement and scientific advice generously provided during the course of these studies by the late Martha Osnos, M.S. and John Pinto, Ph.D.
388
(i()RI)t)N
1 I
,\i
REFERENCES 1. Abraham, G.: Fakou, R., Rozen, R.: Mandenofl, A.: Autissier, N.: Apfelbaum, M. The effect of a constant T3 level and thermoneutrality in diet-induced hyperphagia. Horm. Metab. Res. 19:96-100: 1987. 2. Baumann, E. J.: Marine, D. Involution of the adrenal cortex in rats fed with thiouracil. Endocrinology 36:400-405; 1945. 3. Brosvic, G. M.; Hoey, N. E. Taste detection and descrimination performance of rats following selective desalivation. Physiol. Behav. 48:617-623: 1990. 4. Brown, A. T.: Best, G. M. A proposed mechanism for the effects of saccharin in glucose metabolism by streptococcus mutans. Caries Res. 20:406-418; 1986. 5. Deane, H. W.: Greep, R. O. A cytochemical study of the adrenal cortex in hypo- and hyperthyroidism. Endocrinology 41:243-257; 1947. 6. Dunn, O. J.: Clark, V. A. Applied statistics: Analysis of variance and regression. New York: John Wiley; 1974:387p. 7. Fisher, R.: Griffin, F. Taste-blindness and variations in taste threshold in relation to thyroid metabolism. J. Neuropsychol. 3:98-104: 1961. 8. Fregly, M. J. The role of hormones in the regulation of salt intake in rats. In: Kare, M. R.: Mailer, O., eds. The chemical senses and nutrition. Baltimore: Johns Hopkins University Press; 1967:115138. 9. Fregly, M. J.: Brimhall, R. L.; Galindo, O. J. Effect of the antithyroid drug propylthiouracil in sodium balance of rats. Endocrinology 71: 693-700: 1962. 10. Fregly, M. J.: Galindo, O.: Cook, K. M. Spontaneous sodium chloride appetite of goitrogen-treated rats: Effect of hypophysectomy and adrenalectomy. Endocrinology 69:1060-1067:1961. 11. Fregly, M. J.: Waters, 1. W. Effect of propylthiouracil on preference threshold of rats for NaCI solutions. Proc. Soc. Exp. Biol. Med. 120: 637-640: 1965. 12. Green, W. L. Mechanisms of action of antithyroid compounds. In: Werner, S. C., Ingbar, S. H., eds. The thyroid. 4th edition. New York: Harper and Row, 1978:57-65. 13. Hallman, B. L.; Hurst, J. W. Loss of taste as toxic effect ofmethimazole (Tapazole) therapy. JAMA 152:322: 1953. 14. Holmes, E. W., Jr.: Di Scala, V. A. Studies on exaggerated natnurectic response to a saline infusion in the hypothyroid rat. J. Clin. Invest. 49:1224- t 236:1971). 15. Hoshishima, K.: Yokoyama, S.; Seto, K. Taste sensitivity in various strains of mice. Am. J. Physiol. 202:1200-1204: 1962. 16. Jakinovich, W. Stimulation of the gerbil's gustatory.' receptors by saccharin. J. Neurosci. 2:49-56; 1982. 17. Karen P.: Morris, B. J. Stimulation by thyroid hormone of renin mRNA in mouse submandibular gland. Am. J. Physiol. 25 I:E290E293; 1986. 18. Kondo, M.: Ohe, M.: Ono, A.; Magihira, M. Effect of sodium saccharin on rat pancreatic enzyme secretion. J. Nutr. Sci. Vitaminol. 30:56%576: 1984. 19. Law, J. S.; Henkin, R. 1. Thyroid hormone inhibits purified taste bud membrane adenosine 3'5' monophosphate phosphodiesterase activity. Res. Comm. Chem. Pathol. Pharmacol. 43:449-462; 1984. 20. Law, J. S.: Henkin, R. I. Low parotid saliva calmodulin in patients with taste and smell dysfunction. Biochem. Med. Metab. Biol. 36: 118-124: 1986. 21. Leblond, C. P.: Huff; H. E. Effects of sulfonamides and thiourea derivatives on heart rate and organ morphology. Endocrinology 35: 229-233: 1944.
22. I.ev~ilt. M. S.; l.aing. D. G.: F'anhubei, M.; ('orbctl. A.: ('artcr. J. M. Sensory perception and h,,polhyroidism. Chem. Senses 14: 537-546: 1989. 23. l.eys, D. ttyperthyroidism treated ~ith methylthiouracil, lancel !: 461-464:1945. 24. kinke, H. A. B.; Kohn, J. S. Inhibitory effect of saccharin on glycolytic enzymes in cell-free extracts of streptococcus mutans. Caries Res. 18:12-16; 1984. 25. Marks, P.: Anderson, J,: Vincent, R. Aldosterone m myxedema. Lancet 2:1277-1278: 1978. 26. Mattes. R. D.: Heller, A. D.; Rivlin, R. S. Abnormalities in suprathreshold taste function in early hypothyroidism in humans. In: Meisleman, M. l_.: Rivfin, R. S., eds. Clinical measurement of taste and smell. New York: MacMillan: 1986:467-486. 27. McConnell, R. J.: Menendez, (7. E.: Smith, F. R,: Henkin, R. 1.: Rivlin, R. S. Defects of taste and smell in patients with hypothyroidism. Am. J. Med. 59:354-363: 1975. 28. Michael, V. F.: Barenberg, R. 1.: Chavez, R.: Vaamonde. C. A.: Papper, S. Renal handling of sodium and water in the hypothyroid rat. J. Clin. Invest. 51:1405-1412: 1972. 29. Mook. D. G. Saccharin preference in the rat: Some unpalatable findings. Psychol. Rev. 81:475-490: 1974. 30. Norgren, R.: Nishijo, H.: Travers, S. P. Taste responses from the entire gustatory apparatus. Ann. NY Acad. Sci. 575:246-263: 1989. 31. Ogawa. M.: Sato, M.: Yamashita, S. Gustatory impulse discharges in response to saccharin in rats and hamsters. J. Physiol. 204:31 I329; 1969. 32. Pittman, J. A.: geschi, R. J. Taste thresholds in hyper- and hypothyroidism. J. Cfin. Endocrinol. Metab. 27:895-896; 1967. 33. Richter, C P. Increased salt appetite in adrenalectomized rats. Am. J. Physiol. 115:155-161: 1936. 34. Richter. C. P. Salt taste thresholds of normal and adrenalectomized rats. Endocrinology 24:367-37 I: 1939. 35. Rivlin, R. S.: Osnos, M.: Rosenthal, S.: Henkin, R. I. Abnormalities in taste preference in hypothyroid rats. Am. J. Physiol. 232:ES0E84: 1977. 36. Scfiaupp, H. Geruch und Geschmack bet endokrinen Erkankungen. Arch. Klin. Exp. Ohr. Nas. Kefikopfheilk. 195:179-191: 1969. 37. Schneeberg, N. G. Loss of sense of taste due to methylthiouracil therapy. JAMA 149:1091-1093: 1952. 38. Schneyer, k.: Tanchester, D. Some oral aspects of radioactive iodine therapy for thyroid disease. NY J. Dent. 24:308-309: 1954. 39. Shafer, W. G.: Clark, P. G,: Bixler, D.: Muhler, J. C. Salivary gland function in rats, I1. Effect of thyroid function on salivary flow and viscosity. Proc. Soc. Exp. Biol. Med. 98:245-247: 1958, 40. Shapiro, E. G.: Johnson, I. B. The development ofaversive responses to quinine in hyperthyroid rats. Behav. Neural Biol. 43:274-286: 1985. 41. Shimazaki, K.: Sato, M.: Takegami, T. Binding of 3~S-saccharin to a protein fraction of rat tongue epithelia. Biochim. Biophys. Acta 677:331-338: 1981. 42. Smith, J. C.; Gannon, K. S. Ingestion patterns of food, water, saccharin and sucrose in streptozotocin-induced diabetic rats. Physiol. Behav. 49:181-199: 1991. 43. Vogt, M. B.: Rudy, J. W. Neonatal hyperthyroidism in the rat: Thyroxine accelerates the development of unconditioned but not learned responses to taste. Behav. Neural Biol. 46:358-371; 1986. 44. Weiffenbach, J. M.: Fox, P. C.: Baum, B. J. Taste and salivary hmction. Proc. Natl. Acad. Sci. 83:6103-6106: 1986.
