The Otsuka Long-Evans Tokushima Fatty (OLETF) rat, an outbred strain of Long- Evans Tokushima Otsuka rat (LETO) that lacks CCK-1 receptor expression, is hyperphagic and develops obesity and type-2 diabetes. The present study sought to assess how OLETF rats alter intake, preference, and conditioned preference of palatable solutions after acute food deprivation. Our results show that after 24 h chow restriction, LETO rats increase both sucrose intake and two-bottle sucrose preference relative to their free-fed baseline, whereas OLETF rats do not increase sucrose intake (0.3 M or 1.0 M sucrose) or preference (1.0 M vs. 0.3 M sucrose) when they are food deprived. In contrast, OLETF rats exhibit a higher conditioned flavor preference when sucrose is used as unconditioned stimulus (US) relative to LETO rats, whether overnight food restricted (81% vs. 71% for OLETF and LETO rats, respectively) or free fed (82% vs. 54% for OLETF and LETO rats, respectively) during the test. When a noncaloric saccharin solution is used as US, OLETF rats show a higher preference for the saccharin-associated flavor relative to LETO rats when nondeprived (76% vs. 58% for OLETF and LETO rats, respectively); however, neither strain shows differential conditioned flavor preference for saccharin in the deprivation state during the test. These findings suggest that OLETF rats fail to integrate postabsorptive and orosensory effects of sucrose in a conditioning setting to influence intake. Thus, it appears that OLETF rats form preferences for sucrose based largely on orosensory and hedonic properties of the solution, rather than caloric value.
- food intake
- CCK-1 receptor
- diet-induced obesity
the otsuka long-evans tokushima fatty rat (OLETF) is an outbred strain of Long-Evans Tokushima Otsuka rat (LETO) that has been previously characterized by the lack of CCK-1 receptor expression due to a spontaneous mutation spanning the promoter and first two exon regions of the CCK-1 receptor gene (45). In addition to the use of these animals as a model of insulin resistance, because of their natural manifestation of hyperglycemia and noninsulin-dependent diabetes mellitus relative to age-matched LETO rats (28), OLETF rats are also currently under investigation as a model of obesity. OLETF rats exhibit an increased rate of weight gain relative to controls across their life span, with a marked elevation in body weight seen as early as 2–4 days postnatal (5, 34, 38). It is also known that OLETF rats are hyperphagic via increased meal size. This behavior has been attributed not only to deficits in CCK-related satiation mechanisms (34) but also to gastric mechano-detection (16), intestinal nutrient satiation signaling (12, 17, 38), and more recently, enhanced oral responsiveness to palatable stimuli (15).
It is well known that the amount consumed within a meal is a function of both oral and postoral nutrient properties of foods (42). Heightened orosensory responses to preferred tastants in OLETF rats (17, 21) suggest that altered oral or taste sensation may additionally compound hyperphagia attributed to known peripheral satiation defects in these animals. In this context, learned associations between the flavor of food and its postingestive consequences can influence food intake and preference (27, 39). Compelling evidence exists that learned preferences can be mediated by associations of nutritive feedback. Indeed, rats can acquire preference from odors and tastes that have been paired with intragastric or intraduodenal infusions of sugars (1, 18), as well as when given prior access to an arbitrary cue flavor solution paired with a specific palatable food (10).
It has also been established that rats subjected to periods of acute food deprivation increase subsequent consumption of palatable foods (11, 44). The exact mechanism responsible for this increase, however, is not entirely clear. For instance, food-restricted rats can show heightened intake of both caloric sucrose and noncaloric saccharin solutions (44). It appears that an increase in motivation to eat due to caloric deprivation does not comprise the sole impetus for increased consumption under restrictive conditions but rather increases the overall responsiveness to gustatory properties of the food. Indeed, there are reports of enhanced hedonic response (2), as well as lower taste thresholds for sweet solutions under deprived conditions (6, 51) that support this notion.
Bi and Moran (4) have shown that in addition to overconsuming more chow over 24 h under ad libitum fed conditions, OLETF rats also overconsume chow to a higher degree than LETO rats following a 24-h fast (4). The effect of food restriction on the intake of preferred liquid solutions in these animals has not been tested. Therefore, the first aim of the present study was to test food deprivation effects on intake and preference for normally preferred sucrose solutions in OLETF and LETO rats.
Conditioning a flavor preference has been used to dissociate positive orosensory vs. postingestive effects in the consumption or preference for a particular food. Several studies have provided evidence that the formation and expression of flavor-taste and flavor-calorie associations are regulated differentially by levels of hunger (10, 19, 24). For example, hungry rats show increased preference for a conditioned stimulus (CS) flavor associated with sucrose relative to sated animals. In contrast, there appears to be no differential effect of hunger on flavor preference for a flavor associated with saccharin (8, 19, 24). Furthermore, more recent data suggest that food deprivation affects the expression, but not acquisition, of conditioned flavor preference (48, 49).
We have previously shown that OLETF rats exhibit increased intake, as well as preference, for palatable sucrose in both real and sham feeding paradigms (17, 21). This increased sensitivity to oral stimulation by sweet tastes suggests that orosensory components of sucrose solutions impart enhanced contributions to sucrose intake relative to LETO controls. To investigate effects of food deprivation on learned preferences for sweet solutions, we used a conditioned flavor-calorie procedure using palatable sucrose or saccharin solutions as unconditioned stimuli (US) and tested intake of paired conditioned stimulus (CS) flavors under both deprived and free-fed conditions. If OLETF rats are more sensitive to the orosensory properties of sweet solutions and less sensitive to the postabsorptive effects, we would expect these rats to be less responsive to potential differences in expression of conditioned sucrose flavor preference induced by acute food restriction relative to LETO rats, when tested in a conditioned sweet preference paradigm. We also hypothesized that if OLETF rats are indeed more sensitive to orosensory associations of sweet solutions, then saccharin should produce a greater conditioned flavor preference in OLETF rats compared with LETO rats when animals are nondeprived. Therefore, our second aim was to assess possible alterations in the expression of conditioned flavor preferences for both caloric and noncaloric sweet solutions in OLETF and LETO rats under food-deprived and ad libitum-fed conditions.
MATERIALS AND METHODS
Male OLETF and LETO rats were obtained as a generous gift of the Tokushima Research Institute, Otsuka Pharmaceutical, Tokushima, Japan. All rats were housed individually in mesh-floored, stainless-steel hanging cages and maintained in a temperature-controlled vivarium on a constant 12:12-h light-dark cycle (lights on at 0600). Animals were handled daily for a minimum of 1 wk before the onset of experiments. Tap water and pelleted rat chow (Purina 5001) were available ad libitum throughout experiments, except where indicated otherwise. All protocols used were approved by The Pennsylvania State University Institutional Animal Care and Use Committee.
Intake and preference of 0.3 M and 1.0 M sucrose in OLETF and LETO rats under ad libitum and food-deprived conditions.
Twenty-four-week-old OLETF and LETO rats (n = 5 per strain, weighing 537 ± 29.6 g and 435 ± 6.7 g, respectively) were used for the following experiments. In sucrose intake studies, all rats received daily 1-h access to calibrated glass drinking burettes containing 0.3 M sucrose between 1000 and 1100 every day. Before daily sucrose presentation, 24-h chow intake and spillage were measured. After the establishment of a stable baseline sucrose intake under ad libitum fed conditions (6–8 days), rats were food deprived across two levels of chow restriction. In the low-restriction condition, animals were tested 24 h after presentation of a food ration limited to 75% baseline daily ad libitum chow intake. Given the well-documented differences in daily chow intake between OLETF and LETO rats, the low restriction was rationed in this manner to provide comparable degrees of food restriction as a function of daily intake within strain. In the high-restriction condition, rats had no access to food for 24 h prior to testing (i.e., both strains were deprived of 100% ad libitum daily chow intake). One hour daily sucrose intake under either ad libitum or food-restricted conditions was compared with test how differences in caloric motivation would affect intake of a palatable sucrose solution. Daily sucrose access time and duration remained unaltered throughout experimentation.
Food restriction conditions were imposed a minimum of 2 days apart both to allow for compensatory chow intake and to assess possible shifting in nondeprived sucrose baseline intake as a result of intermittent access. Every deprivation condition was repeated for a minimum of two occasions. This study was first completed using 0.3 M sucrose as test solution, followed by an identical experiment using a more concentrated 1.0 M sucrose solution in the same groups of rats. Sucrose intake was measured to the nearest 0.1 ml.
In a further experiment, using the same deprivation schedule and sucrose access periods, daily 1-h, two-bottle choice tests were performed using 0.3 M vs. 1.0 M sucrose to assess sucrose concentration preference changes as a function of motivational state. To control for side preference, the position of each sucrose burette was alternated daily and counter-balanced within animals on the same testing day.
Conditioned preference for caloric and noncaloric sweet solutions in food-deprived or nondeprived OLETF and LETO rats.
Twenty, 10-wk-old, naïve prediabetic OLETF and LETO rats were used in this experiment (n = 10 per strain, weighing 287.2 ± 6.8 g and 220.4 ± 3.8 g, respectively). A two-day habituation period preceded experimentation, during which all of the animals were given brief access to 3 ml of an unflavored 0.2% saccharin solution at 1000 to familiarize the animals with one-bottle acceptance. As shown in Fig. 1, over the next eight calendar days all rats were exposed to a 3-ml volume of a cue-flavored solution (conditioned stimulus, CS) of either grape- or cherry-flavored saccharin [0.2% saccharin (wt/vol) and 0.05% Kool-Aid (wt/vol)] immediately preceding a 30-min presentation of an assigned unconditioned stimulus (US) solution. One group of OLETF and LETO rats (n = 5 per strain) received 3 ml grape saccharin (CS+) paired with 30-min ad libitum exposure to 0.3 M unflavored sucrose (US+) and 3 ml cherry-flavored saccharin (CS−) with 30-min access to plain water (US−), while an additional group of rats (n = 5 per strain) were conditioned using the opposite CS-US pairings. All animals were given CS+/US+ and CS−/US− pairings on alternate calendar days for a total of four exposures to each pairing prior to CS preference testing. All solutions were presented in standard drinking bottles attached to the front of the home cage. Pelleted rat chow was continuously available except during conditioning procedures. To account for the possibility of inducing stronger preferences in OLETF due to greater reinforcement as a result of higher US+ intakes during training, in a separate group of rats (OLETF, n = 13; LETO, n = 10), OLETF rats’ sucrose intake (0.3 M) during conditioning training was clamped to the amount consumed by LETO. This was done in both deprived and nondeprived conditions.
On days 9 and 10, 30-min two-bottle choice tests between CS+ and CS− flavors were conducted in food-deprived or ad libitum-fed OLETF and LETO rats to assess conditioned preference for the CS solutions in the absence of US(+/−) presentation. Food-restricted and ad libitum-fed choice tests were counterbalanced between groups such that, on day 8, the two experimental groups within each strain were split randomly: half of the animals within each group had chow removed at 1800 before testing on day 9, while the other half of the animals within each group were not food restricted before day 9 preference testing. After testing on day 9, plain water and rat chow were returned. Chow was then removed at 1800 from rats not food restricted before two-bottle choice testing on day 9. On day 10, two-bottle choice testing was performed in rats under the opposite restriction conditions imposed before day 9 choice testing. The left- and right-side orientation of the two CS flavor solutions was counterbalanced equally between animals. In the experiment, in which OLETF rats' intake during conditioning was limited to the amount consumed by LETO rats, clamping did not occur during the preference test; thus rats had ad libitum access to 0.3 M sucrose solution.
A parallel experiment was conducted simultaneously with the experiment described above in a separate group of animals (twenty, 10-wk-old naïve rats, weighing 275.4 ± 5.0 and 228.7 ± 6.2 g, respectively), in which 0.3 M sucrose was replaced with 0.2% noncaloric saccharin solution as the US+ solution. All other protocols were identical. Our final experiment expanded upon the previous two conditioning studies by using both saccharin and sucrose solutions as US solutions. Therefore, in this last experiment, another group of rats (twenty, 10-wk-old, naïve rats, weighing 265.9 ± 3.3 and 222.5 ± 3.7 g, respectively) were conditioned to associate a CS with both a caloric and noncaloric sweet-tasting US (0.3 M sucrose: US+; 0.2% saccharin: US–). All daily access and testing regimens were the same as described in the previous two studies.
For sucrose intake studies, one-way repeated-measures ANOVA (rmANOVA) was performed to test within strain effects of food deprivation on subsequent sucrose intake and preference in OLETF or LETO rats. These results were analyzed post hoc by Tukey's honestly significant difference test when necessary. For sucrose preference testing, a two-way rmANOVA was performed with strain and chow deprivation as main effects. Total intake of both solutions (0.3 M and 1.0 M sucrose) was used to calculate preference percentages according to the following formula: Preference percentage = [volume of 1.0 M sucrose solution (ml) × 100/ total volume of 0.3 M sucrose and 1.0 M sucrose (ml)].
For all conditioned preference studies, US intake across the conditioning period was analyzed separately according to US solutions using two-way rmANOVA with strain and time as main effects. Conditioned preference testing results were analyzed by appropriate one-way ANOVA planned comparisons with strain or deprivation state on test as the main factor.
All data were expressed as means ± SE. Differences were considered statistically significant if P < 0.05. Statistical analyses were computed with PC-SAS (version 8.02, SAS Institute, Carey, NC).
Intake of 0.3 M and 1.0 M sucrose in OLETF and LETO rats under food-deprived conditions.
Results of ANOVA testing for a food deprivation effect on 0.3 M sucrose intake revealed a significant increase in sucrose intake in LETO rats [F(2,12) = 12.94; P < 0.05], while OLETF rats showed no such effect (P = 0.124). Figure 2 depicts results of post hoc analyses showing that after high food restriction, LETO rats significantly increased their 0.3 M sucrose intake relative to nondeprived sucrose intake (16.3 ± 1.1 vs. 20.4 ± 1.4 ml for baseline and low-restriction intakes, respectively; P < 0.05), while both OLETF and LETO rats did not increase intake when under low deprivation conditions (P = 0.391). Similarly, when 1.0 M sucrose was tested, LETO rats significantly increased their sucrose intake relative to nondeprived sucrose baseline after food deprivation [F(2,12) = 18.42; P < 0.05], while OLETF rats again did not (P = 0.917). Post hoc results depicted in Fig. 3 show an increase in consumption to have occurred only after high-deprivation conditions (9.1 ± 0.9 vs. 12.5 ± 0.4 ml for baseline and low-restriction intakes, respectively; P < 0.01).
Preference for 1.0 M vs. 0.3 M sucrose in two-bottle choice test in OLETF and LETO rats under food-deprived conditions.
Both OLETF and LETO animals preferred 1.0 M sucrose over 0.3 M sucrose in two-bottle choice testing. One-way ANOVA assessing strain effects on sucrose preference showed significantly higher preference for 1.0 M sucrose in OLETF rats compared with LETO animals [F(1,8) = 10.24; P < 0.05]. As shown in Fig. 4, OLETF rats showed higher preference for 1.0 M sucrose than LETO rats when non-food deprived (96.0 ± 1.7% vs. 63.7 ± 12.1% for OLETF and LETO rats, respectively; P < 0.05), or under low-deprivation conditions (96.1 ± 1.2% vs. 60.5 ± 12.4% for OLETF and LETO rats, respectively; P < 0.05), while no strain difference was noted under the higher-restriction condition. Consistent with our preceding results, indicating motivational state increases sucrose intake in LETO rats, LETO rats also showed increased 1.0 M sucrose preference [F(2,12) = 8.83; P < 0.05] when food restricted, while no such response was detected in OLETF rats (Fig. 4). Post hoc testing showed a significant increase in sucrose preference in LETO rats only after high food restriction (P < 0.05).
Conditioned preference for sweet solutions in OLETF and LETO rats.
There were no overall within-group differences in US(+/–) intake during training, or CS(+/–) intake, or preference during testing, according to whether grape or cherry was used as CS+ flavor in all three conditioning studies. In all experiments, rats consumed all of the 3 ml clamped CS(+/−) flavored solutions before US(+/–) access. Additionally, no significant differences between results of day 9 or day 10 preference tests were found within deprivation condition across days. Therefore, these data were pooled for subsequent analyses and presented in the following section accordingly.
Conditioned sucrose vs. water preference in OLETF and LETO rats.
Fig. 5A illustrates intake of US solutions during the 8-day training period. OLETF rats consumed more sucrose (US+) than LETO rats during the 30-min access period [F(4,18)=33.78; P < 0.05]. Post hoc analyses showed this effect across all 4 days of presentation (largest P < 0.05). Water (US–) intake was not significantly different between strains. Figure 5B shows results of conditioned preference testing after all rats were given a 30-min, two-bottle choice test between CS+ and CS–-flavored solutions under either food-deprived or nondeprived conditions. OLETF rats preferred the CS+ flavor to a greater extent than LETO rats under both deprived [80.8 ± 4.9% and 70.8 ± 5.4% for OLETF and LETO rats, respectively: F(1,18) = 13.08; P < 0.05] and nondeprived conditions [82.1 ± 5.3% and 53.6 ± 7.5% for OLETF and LETO rats, respectively: F(1,18) = 39.81; P < 0.01]. When examining the effect of motivational state on conditioned sucrose preference, we observed that LETO rats exhibited a significantly increased CS+ preference when food deprived during the test [F(1,18) = 19.08; P < 0.01], in contrast to OLETF rats that did not show altered preference for CS+ after overnight food deprivation (P = 0.726). A similar effect was observed when OLETF rats' sucrose intake during conditioning was clamped to the amount of LETO rats for either nondeprived or food-deprived condition (data not shown).
Conditioned saccharin vs. water preference in OLETF and LETO rats.
Figure 6A shows intake of US solutions across the 8-day training period. OLETF rats consumed more of the noncaloric sweet saccharin (US+) than LETO rats during the 30-min access period directly following CS+ consumption [F(4,18) = 26.78; P < 0.05]. Post hoc analyses revealed intake to be persistently higher in OLETF rats compared with LETO rats each day of presentation (largest P < 0.05). Water (US–) intake was again not significantly different between strains.
Results of conditioned saccharin preference testing are presented in Fig. 6B. Under conditions identical to those imposed in the previous experiment, OLETF rats preferred the CS+ flavor to a greater extent than LETO rats under only nondeprived conditions [76.4 ± 4.3% and 57.5 ± 5.9% for OLETF and LETO rats, respectively: F(1,18) = 28.66; P < 0.01], whereas no significant difference was noted between strains under deprived conditions (P = 0.894). In addition, LETO rats also exhibited no significant increase in CS+ preference when food deprived during the test (P = 0.087) relative to nondeprived preference. Similarly, OLETF rats showed comparable preference for CS+ regardless of motivational state on test (P = 0.219).
Conditioned sucrose vs. saccharin preference in OLETF and LETO rats.
Results of raw intakes during sucrose (US+) and saccharin (US–) conditioning periods are shown in Fig. 7A. ANOVA results showed that OLETF rats consumed significantly greater amounts of both sucrose (US+) [F(4,18) = 24.96; P < 0.05] and saccharin (US–) [F(4,18) = 17.45; P < 0.05] solutions during training than LETO rats during the 30-min access period directly following CS(+/–) consumption. Post hoc analyses revealed intake to be persistently higher in OLETF rats compared with LETO rats each day of presentation for both US+ and US− solutions (largest P < 0.05 for both US+ and US–). Conditioned preference testing revealed the sucrose (US+)-associated flavor (CS) to be the most preferred. Thus, the results of conditioned preference testing are presented in terms of CS+ preference. Figure 7B depicts results showing OLETF rats preferred the CS+ flavor to a greater extent than LETO rats under only nondeprived conditions [87.6 ± 4.4% and 57.9 ± 4.5% for OLETF and LETO rats, respectively: F(1,18) = 31.19; P < 0.01], while no significant difference was noted between strains under deprived conditions (P = 0.919). In addition, LETO rats exhibited a significantly increased CS+ preference when food deprived during the test [F(1,18) = 25.39; P < 0.01] relative to nondeprived preference, whereas OLETF rats showed similar preference for CS+ regardless of motivational state during the test (P = 0.622).
The present results show that OLETF rats exhibit clear deficiencies in modifying sucrose intake after periods of acute food deprivation compared with LETO controls. For both a relatively dilute and concentrated sucrose concentration, OLETF rats show no significant alterations in either sucrose intake or preference, while LETO rats show both increased sucrose intake and preference under food deprivation conditions. Similarly, OLETF rats exhibit a higher conditioned flavor preference for sucrose relative to LETO, regardless of deprivation state. When OLETF and LETO rats are conditioned to prefer a saccharin solution, OLETF rats show a higher preference for the saccharin-associated flavor relative to LETO rats when nondeprived; however, neither strain shows differential flavor preference for saccharin according to deprivation state during the test. These results suggest that orosensory effects of palatable solutions impart a greater role in governing learned food preferences in OLETF rats relative to control, nonmutant LETO rats.
The size of a meal is partly dictated by the manifestation of postingestive inhibitory feedback signals. Disruptions of these signals can lead to overconsumption and weight gain. Indeed, we, as well as others, have shown that OLETF rats suppress intake less than LETO controls in response to a variety of intestinal nutrient infusions (12), as well as gastric fat preloads (38). Also, OLETF rats reduce sucrose sham feeding less than LETO rats after gastric balloon distention (16), suggesting that mechano-detection may be also impaired in these animals. Although there is no doubt that aberrations in these signaling mechanisms play a major role in leading to excessive meal size (34), the amount consumed within a meal can also be modulated by the feed-forward effects resultant from orosensory stimulation (43). In this regard, we have recently reported ingestive abnormalities in the OLETF rats characteristic of both enhanced taste sensitivity and orosensory controls of sucrose intake, which are manifest in the absence of gastric/postgastric satiation effects (17, 21). In fact, non-food-deprived, prediabetic, OLETF rats will sham feed more of a concentrated sucrose solution than age-matched LETO rats, thus overconsuming sucrose in the absence of postabsorptive effects (17). Furthermore, recent data from our laboratory using an automated gustometer have shown OLETF rats to exhibit increased licking of multiple concentrations of sucrose, fructose, and saccharin when granted only short 10-s access to solutions, pointing to a generalized increase in sweet taste preference (21). The current experiments complement these existing studies by allowing the animals to experience and associate sucrose as both an oral and postingestive stimulus. By testing rats under food-deprived and free-feeding conditions, we are able to assess how these associations may be weakened or strengthened, without modifying exposure to the stimulus itself. In fact, in the conditioning studies, the rats are tested solely on conditioned associations, as no sucrose is presented during testing.
The possibility that central taste functions are impaired in the OLETF rat, including an increased sensitivity to the incentive motivational effect (i.e., palatability) of ingestive stimuli is further supported by the results reported here. The observation that the magnitude of sucrose intake is similar in both nondeprived and 24-h food-deprived OLETF rats suggests that an increase in orosensory sensitivity that is already present in ad libitum-fed conditions could be a major contributor to enhanced sucrose intake in addition to known satiation deficits. Specifically, free-feeding OLETF rats overconsume rat chow by 30% relative to age-matched LETO rats (34). However, when given access to a palatable sucrose solution, the degree of hyperphagia observed in non-food-deprived OLETF rats increases by almost 60% for 0.3 M sucrose and 146% for 1.0 M sucrose relative to freely fed LETO controls. Barring differences between liquid and solid foods for this effect, the augmented magnitude of sucrose hyperphagia relative to chow consumption suggests that other factors than caloric properties of the sucrose, such as palatability, exacerbate the hyperphagic behavior of the OLETF rat. It could be argued that the failure of OLETF rats to increase sucrose intake when deprived may have been due to a ceiling effect as nondeprived intake was large. Similarly, the failure to observe an increase in 1.0 M sucrose preference in OLETF rats might be due to their high preference when nondeprived. However, inspection of data from Fig. 4 shows that this was not the case. Specifically, under the “high” restriction condition, OLETFs consumed ∼17 ml of 1.0 M sucrose, while in the nondeprived condition (“none”), they consumed ∼24 ml of 1.0 M sucrose. In fact, OLETFs showed a moderately decreased intake of the 1.0 M sucrose despite an extra “7-ml room” that could have allowed them to increase intake further, at least to the levels of controls. However, they did not. This rejects the argument that the ceiling effect might have contributed to their failure to increase intake when deprived. Although intake is not limited by mere volume and is concentration dependent, our data demonstrate that OLETF rats are capable of drinking larger quantities. For instance, in Fig. 2, OLETF rats drank ≥35 ml of 0.3 M sucrose. Furthermore, in an earlier report, we showed that OLETF rats drank ≥26 ml of 0.7 M glucose in half the time (30 min) compared with the current study (12). Nevertheless, one condition where OLETF might have encountered a ceiling effect is when tested for preference under the “low” deprivation. In this condition, total intake was approaching, but not reaching, the intake during the nondeprived condition. Therefore, we can cautiously assume that in our study during the 60-min recording period, OLETF rats would have been able to increase their intake beyond the amount consumed under food deprivation.
It is well known that the postingestive, or nutritive, properties of nutrients are able to condition flavor preference (40, 42). Previous work has focused on the effects of food deprivation on conditioned flavor preference in normal (i.e., nonmutant) rats using similar degrees of food deprivation employed in the current design (7, 9, 10). However, several of these studies have examined specifically how flavor preferences are modulated by food deprivation using food exposures during conditioning phases that have been previously associated with a specific caloric need state before preference testing, i.e., mild or high food deprivation conditions. In all three of our flavor conditioning tests, CS solutions and the associated tastants (US) were presented when the animal was under ad libitum fed conditions prior to testing. Thus, our design aimed at the effects of differential caloric motivational state specifically on the expression of conditioned flavor preferences and not the acquisition of these preferences (i.e., ad libitum chow availability throughout the conditioning phase).
In our first flavor conditioning test, we showed that food-deprived OLETF rats exhibit no difference in conditioned flavor preference for sucrose when either ad libitum fed or food deprived during the test, while LETO rats showed significantly increased flavor preference for sucrose when food deprived during the test. These results support our hypothesis that while in LETO rats, a calorie association controls preference for sucrose over taste association under deprived conditions, in OLETF rats, the performance during the test was not enhanced by increasing caloric drive. We also showed that when comparing between strains, OLETF rats exhibit a higher flavor preference for sucrose relative to LETO rats regardless of deprivation state. Such results may be indicative of a generalized stronger degree of sucrose preference formation in OLETF relative to LETO rats. One possible explanation for this effect is that, among OLETF rats, sucrose possesses an inherently greater reinforcing effect as an unconditioned stimulus (37, 41). Such an effect may be related to differences in gustatory functions or the sensitivity of the reward systems, both of which aspects are currently under investigation in our laboratory. Although beyond the scope of the present design, investigation of the reinforcing effects of sucrose through operant conditioning procedures may support this notion more directly. Nonetheless, we are still able to provide clear evidence that when tested for conditioned preference, increased caloric need due to food deprivation appears to play a minimal role in the expression of conditioned sucrose preference in the OLETF rat. In other words, orosensory effects of sucrose in hyperphagic OLETF rats are stronger to form preferences than in LETO rats, irrespective of deprivation state.
Our results of enhanced conditioned saccharin preference in nondeprived OLETF rats compared with LETO controls further support this hypothesis. As saccharin is a noncaloric sweet tastant, nutrient satiation deficiencies in the OLETF rat could not account for increased saccharin intake or preference in these animals. We have recently published evidence of enhanced licking of saccharin solutions in the OLETF rat in brief 10-s access testing (21). Thus, one possible explanation for increased saccharin intake during our conditioning phase is an enhanced gustatory sensitivity for the sweet taste of saccharin. Finally, neither strain showed significant alterations in conditioned preference for saccharin in the deprivation state during the test consistent with previous reports of conditioned saccharin preference (19). This finding lends support to the notion that an increased potential of sucrose to form conditioned preferences is based on its caloric predictive value or metabolic effects. Such a possibility can be addressed by examining sham feeding of sucrose during conditioning.
Our final study showed that when replacing water (US–) with a more palatable noncaloric sweet saccharin solution, we obtained similar results of conditioned sucrose (US+) preference as that of our earlier experiment. Specifically, OLETF rats showed a higher conditioned preference under nondeprived conditions relative to LETO rats. Secondly, OLETF rats did not show intake differences by motivational state during the test, while LETO rats showed a significant increase in conditioned sucrose preference in the food deprivation test. This study suggests that although control LETO rats increase their conditioned preference for a caloric sweet taste over one that is noncaloric when food-deprived, OLETF rats do not appear to display such effects.
In conditioning experiments in which US intakes were not clamped during training, we readily observed increased sucrose intake in OLETF rats relative to LETO rats. One can argue that greater reinforcement might lead to stronger preferences, and therefore OLETF rats' test results cannot be attributed solely to strain differences in learning about sucrose associates. The existing literature on the influence of the strength of the reinforcement stimulus on forming subsequent preferences is inconclusive. Some suggest that the magnitude of conditioned preference increases after additional training (42), while others show no such correlations (32). In our tests, when OLETF sucrose intakes were clamped to the levels of LETO during conditioning training, OLETF rats did not increase CS+ when deprived compared with nondeprived rats. Thus, the increased sucrose intake during training seems to have no effect on our results. However, even this method has its limitations (see Ref. 50). In particular, OLETF rats drank significantly less CS+/sucrose solution in the clamped study than they normally did when intakes were unlimited. This may have resulted in a “frustration” effect that could reduce their attraction to the CS+ flavor during preference tests. Furthermore, OLETF rats overconsume, in part, because of reduced satiation signaling, suggesting that increased US intake under ad libitum access conditions is not directly due to increased US exposure effects, but rather may be an artifact of reduced negative feedback signals normally limiting intake.
Overall, the present findings are likely due to deficiencies in the integration of postingestive and orosensory signals determining meal size. Recent work from our laboratory has demonstrated that neurons expressing c-fos immunoreactivity in response to CCK are virtually absent in the OLETF rat (13). These data confirm that CCK-related satiation deficits in the OLETF extend to regions of the hindbrain known to relay satiation signals from the gut to the brain via CCK-1 receptor activation. A separate body of work has also shown that the absence of CCK-1 receptors in the dorsomedial hypothalamus may be responsible for altered NPY expression in the obese OLETF rat (3, 35). Of particular interest to the current results are our recent reports that OLETF rats exhibit enhanced intake of sucrose when postingestive effects of the solution are eliminated or largely minimized (17, 21). Furthermore, it must also be mentioned that the observed deficits in OLETF rats may be due to the absence of CCK-1 receptor-mediated effects on specific taste receptors encoding for sweet stimuli. Recent evidence has supported a possible role for CCK, acting through CCK-1 receptor activation, in taste receptor cells (26). Specifically, it has been shown that many CCK-responsive taste cells are also sensitive to the two bitter stimuli quinine and caffeine (26, 31). The possible impact of CCK-sensitive cells and sweet stimuli sensitivity is unknown; however, such studies may help to determine why the CCK-1 receptor-deficient OLETF rat may exhibit altered orosensory effects of sweet solutions. In addition, the CCK-1 receptor-deficient OLETF rat responding to the gustatory relays from taste buds to the NTS and/or projections to the gustatory centers of the parabrachial nucleus could also be implicated in these results.
It is also possible that deficits in processing food reward could help explain enhanced orosensory control of sweet preference. For example, dopamine activity (DA), particularly within the mesolimbic pathway, increases after ingestion of palatable sucrose solutions and when orosensory components of these tastants were isolated (22, 46). With relevance to the current model, CCK stimulates DA release by acting on CCK-1 receptors (14). Central CCK-1 receptors are not ubiquitous; in fact, only specific populations are known to exist, within limited regions such as the dorsomedial hypothalamus (36, 47), striatum (20), and the caudo-medial shell of the NAcc (25, 30). In the caudal nucleus accumbens, DA and CCK are coreleased in vivo after administration of drugs that increase DA neuronal firing rate (14, 29). A similar corelease of DA and CCK in the ventral tegmental area has been shown to affect DA cell firing rate (23). Moreover, we have shown an altered D2 dopamine receptor function related to sucrose intake in addition to lower D2 receptor binding in the striatum of prediabetic OLETF rats of similar age (33). These neural alterations could function to enhance the effects of the hyperphagia in the OLETF rat through not only independent mechanisms, but also collectively, resulting in an overall augmentation of meal size.
Taken together, the present studies indicate that among OLETF rats, both intake of and preference for sucrose, as well as a flavor-conditioned sucrose preference, has a relatively stronger orosensory component compared with control LETO animals. Furthermore, food deprivation conditions, which lead to increased consumption of sucrose or sucrose-associated flavors in LETO rats, appear to have little impact on consumption of these same fluids in OLETF rats. Thus, it appears that OLETF rats form preferences for sucrose based largely on orosensory and hedonic properties of the solution, rather than postingestive associations.
This research was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-065709.
The authors wish to thank Otsuka Pharmaceutical (Tokushima, Japan) for the generous donation of the OLETF and LETO animals used to perform this research. We also wish to thank Dr. Anthony V. Azzara for his critical reading of an earlier version of the manuscript.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society