nutrients Article Behavioral and Metabolic Effects of a Calorie-Restricted Cafeteria Diet and Oleuropein Supplementation in Obese Male Rats Alex Subias-Gusils 1,2 , Adam Álvarez-Monell 1,3 , Noemí Boqué 4 , Antoni Caimari 4,5 , Josep M. Del Bas 4 , Roger Mariné-Casadó 4 , Montserrat Solanas 1,3, * and Rosa M. Escorihuela 1,2, * 1 Institut de Neurociències, Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain; Alex.Subias@uab.cat (A.S.-G.); Adam.Alvarez@uab.cat (A.Á.-M.) 2 Departament de Psiquiatria i Medicina Legal, Facultat de Medicina, Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain 3 Department of Cell Biology, Physiology and Immunology, Faculty of Medicine, Universitat Autònoma de Barcelona, 08913 Bellaterra, Spain 4 Eurecat, Centre Tecnològic de Catalunya, Technological Unit of Nutrition and Health, 43204 Reus, Spain; noemi.boque@eurecat.org (N.B.); antoni.caimari@eurecat.org (A.C.); josep.delbas@eurecat.org (J.M.D.B.); roger.marine@eurecat.org (R.M.-C.) 5 Eurecat, Centre Tecnològic de Catalunya, Biotechnology Area and Technological Unit of Nutrition and Health, 43204 Reus, Spain * Correspondence: Montserrat.Solanas@uab.cat (M.S.); RosaMaria.Escorihuela@uab.cat (R.M.E.);   Tel.: +34-93-5811373 (M.S.); +34-93-5813296 (R.M.E.) Citation: Subias-Gusils, A.; Abstract: Diet-induced obesity models are widely used to investigate dietary interventions for Álvarez-Monell, A.; Boqué, N.; Caimari, A.; Del Bas, J.M.; treating obesity. This study was aimed to test whether a dietary intervention based on a calorie- Mariné-Casadó, R.; Solanas, M.; restricted cafeteria diet (CAF-R) and a polyphenolic compound (Oleuropein, OLE) supplementation Escorihuela, R.M. Behavioral and modified sucrose intake, preference, and taste reactivity in cafeteria diet (CAF)-induced obese Metabolic Effects of a Calorie- rats. CAF diet consists of high-energy, highly palatable human foods. Male rats fed standard Restricted Cafeteria Diet and chow (STD) or CAF diet were compared with obese rats fed CAF-R diet, alone or supplemented Oleuropein Supplementation in with an olive tree leaves extract (25 mg/kg*day) containing a 20.1% of OLE (CAF-RO). Biometric, Obese Male Rats. Nutrients 2021, 13, food consumption, and serum parameters were measured. CAF diet increased body weight, food 4474. https://doi.org/10.3390/ and energy consumption and obesity-associated metabolic parameters. CAF-R and CAF-RO diets nu13124474 significantly attenuated body weight gain and BMI, diminished food and energy intake and improved biochemical parameters such as triacylglycerides and insulin resistance which did not differ between Academic Editor: Lindsay Brown CAF-RO and STD groups. The three cafeteria groups diminished sucrose intake and preference Received: 29 October 2021 compared to STD group. CAF-RO also diminished the hedonic responses for the high sucrose Accepted: 12 December 2021 concentrations compared with the other groups. These results indicate that CAF-R diet may be an Published: 15 December 2021 efficient strategy to restore obesity-associated alterations, whilst OLE supplementation seems to have an additional beneficial effect on sweet taste function. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in Keywords: diet-induced obesity; energy restriction; food intake; hedonic response; leptin; polyphe- published maps and institutional affil- nols; sucrose preference; sweet taste iations. 1. Introduction Copyright: © 2021 by the authors. The prevalence of overweight and obesity are escalating worldwide, along with Licensee MDPI, Basel, Switzerland. Metabolic Syndrome (MetS) and the development of type-2 diabetes and cardiovascular This article is an open access article diseases (CVD) [1,2]. Among the alterations associated with obesity, changes in taste distributed under the terms and function are supported by a robust body of evidence. Psychophysical studies have demon- conditions of the Creative Commons strated that sensory and hedonic properties of sweet and fat vary with body mass index in Attribution (CC BY) license (https:// humans, in a manner such that obese people experience reduced sweetness and intensified creativecommons.org/licenses/by/ fat sensations, liking both sweet and fat more than the non-obese do [3]. Moreover, sensory 4.0/). Nutrients 2021, 13, 4474. https://doi.org/10.3390/nu13124474 https://www.mdpi.com/journal/nutrients Nutrients 2021, 13, 4474 2 of 17 taste (including sweetness, saltiness, sourness, umaminess, bitterness, and oleic acid, a fatty stimulus), olfactory function, and eating behavior have been shown to improve after bariatric surgery [4]. In a recent meta-analysis investigating how physiological factors, pathologies and acquired habits influence taste sensitivity, the only factor increasing the sucrose detection threshold was a high body mass index (BMI), while aging and type 2 dia- betes mellitus patients were found to exhibit an increased sucrose recognition threshold [5]. However, whether these changes are associated with the metabolic consequences of obesity or are the consequence of the sucrose consumption per se remain unclear [6]. The possi- bility that the taste system can increase or decrease its sensitivity with diet composition and influence food preference, choice, and overall energy intake, through diet-dependent chemosensory mechanisms of plasticity has been proposed [7]. The hormone leptin has been associated with the transmission of sweet taste and with the mechanisms of satiety and neural reinforcement. In humans and rodents with normal body weight, the behavioral response to sweet solutions of different concentrations changes with diurnal variations of plasma leptin levels, while in obese humans and rodents, those diurnal leptin variations were not found [8]. Leptin selectively suppressed sweet taste via the leptin receptor Ob-Rb in sweet taste cells [9]. In humans, it has been also proved that leptin selectively modulates sweet taste neuronal responses and taste perception through activation of oral leptin receptors in taste receptor cells [6]. The most widely used interventions to counteract the impact of overweight and obesity in humans are based on the administration of hypocaloric diets and lifestyle modi- fications (moderate exercise programs) often combined with bioactive dietary compound supplementation [10,11]. Konstantinidi and Koutelidakis [12] reviewed the contribution of functional foods such as coffee, green tea, berries, nuts, or olive oil, among others, and their bioactive compounds such as caffeine, catechins, gallic acid, anthocyanins, ascorbic acid or oleuropein, among others, to weight management, obesity prevention, and obesity’s metabolism along with their possible mechanisms. They concluded that functional foods, as part of a balanced diet, can be useful for weight management and can decrease obesity’s metabolic consequences, although scientific evidence remains unclear and controversial. The fruits and oil of the olive tree (Olea europaea) are essential components of the Mediterranean diet. Oleuropein (OLE) is a phenolic constituent of olives consisting of three structural subunits: hydroxytyrosol, elenolic acid, and a glucose molecule. OLE and its related compounds have shown to be beneficial against dyslipidemia and to have antiobe- sity, antidiabetic, antihypertensive, and hepatoprotective actions [13]. The antioxidant, anti-inflammatory, and immunomodulatory properties of secoiridoids from the olive tree such as OLE have potential applications in many inflammatory and reactive oxygen species (ROS)-mediated diseases [14]. Among the animal models developed to study human obesity, the rodent diet-induced obesity (DIO) models have considerable face validity with human obesity and are widely used. More specifically, the cafeteria diet (CAF) better mimics the hedonic hyperphagia that is observed in obese humans and promotes more severe diabetic symptoms than other high fat diets [15]. CAF diet consisting of exposing the rodents to high-energy, highly palatable human foods like cakes, savory snacks, cheese, and sugared milk, induces a rapid weight gain [16–19] and metabolic disorders associated with obesity and MetS in humans (hyperleptinemia, hypertension, hypertriglyceridemia, hyperglycemia, and insulin resistance) [20,21]. OLE administration prevented body weight gain, body fat accumulation, hyperinsu- linemia, and hyperleptinemia in obese mice fed with CAF diet [22] and attenuated visceral adiposity in high-fat DIO mice [23]. OLE also improved the cardiovascular, hepatic, and metabolic signs in high-carbohydrate high-fat fed rats [24] and enhanced fat-oxidation and optimized cardiac energy metabolism in obese rats [25]. In humans, olive leaf tea decreased serum levels of triacylglycerides and low-density lipoprotein cholesterol in prediabetic individuals between 40 and 70 years of age with a body mass index of 23.0–29.9 kg/m2 and Nutrients 2021, 13, 4474 3 of 17 changed body weight, waist circumference, and insulin levels after a 12-week intervention consisting of consuming 330 mL of the test beverage (consisting of tea leaves steeped in boiled water) three times daily during mealtimes [26]. The current study was aimed to evaluate whether a dietary intervention based on a previously characterized calorie-restricted cafeteria diet (CAF-R) [27], alone or supple- mented with OLE, modified sucrose preference, and sweet taste reactivity in cafeteria diet-induced obese rats. We also analyzed the effects on biometric, food and energy intake, and metabolic parameters including leptin levels. 2. Materials and Methods 2.1. Animals and General Procedures Forty male Sprague–Dawley rats (23–25 days old) were used (Harlan Laboratories, Barcelona, Spain). The animals were housed 2 per cage at 22 ◦ C under 12 h/12 h light/dark cycle (lights on at 08:00 a.m.) in standard conditions of temperature (21 ± 1 ◦ C) and humidity (50 ± 10%) for the duration of the experiment with free access to food and water. After one week of habituation to the animal facility, the animals were assigned to two treatment groups with equivalent averages of initial body weight: the control group, fed with standard chow ad libitum (STD, n = 10), and the CAF group, fed with cafeteria diet ad libitum (CAF, n = 30) to induce obesity. After 8 weeks, all animals were housed individually. The STD group was maintained on an ad libitum chow diet, whereas CAF-fed animals were subdivided into three subgroups depending on the dietary treatment received: the remaining cafeteria diet ad libitum group (CAF, n = 10); the CAF-R group (n = 10), fed with calorie-restricted cafeteria diet; and the CAF-RO group (n = 10), fed with CAF-R diet supplemented with olive tree leaves extract (25 mg/kg*day). Once a week, cages were changed and after the cleaning, body weight and food consumption were recorded. Nose:anal length (NAL) was measured once a week from the end of the obesity induction period (week 9) to the end of the experiment. Two blood samples were collected, one immediately before the dietary treatments started (week 9) and the second one immediately before the behavioral tests (week 19). Serum was obtained by centrifugation at 2000× g for 15 min at 4 ◦ C and stored at −80 ◦ C until further analyses. After 22 weeks of dietary treatments, the animals were sacrificed by decapitation after 8 h of fasting. All animals received human care under an institutionally approved experimental animal protocol, following the legislation applicable in Spain. The experimental protocol was approved by the Generalitat de Catalunya (DAAM 9978), following the ‘Principles of laboratory animal care’, and was carried out in accordance with the EU Directive 2010/63/EU for animal experiments. 2.2. Diets The animals were fed with the corresponding diets daily for 22 weeks, from weaning until the end of the study. Food consumption was calculated as the difference between the weight of each dietary component provided and the amount unconsumed after 24 h. During the obesity-induction period, the CAF diet daily included (average quantity administered per cage/day) all the following items: bacon (10 g), biscuit with pâté (12 g), biscuit with cheese (13 g), muffins (16 g), carrots (10 g), jellied sugared milk (72 g, 18% sucrose w/w), and standard chow (25 g). CAF items were placed in a clay pot inside the cage and prepared fresh each day. Mean daily total food and energy administered per cage/day were 158 g/day and 367 kcal/day respectively. The caloric distribution of the CAF diet was 10% protein, 36% fat, and 54% carbohydrates. During the dietary treatments period, the CAF diet included (average quantity ad- ministered per rat/day): bacon (6 g), biscuit with pâté (7 g), biscuit with cheese (7 g), muffins (11 g), carrots (5 g), jellied sugared milk (45 g, 18% w/w), and standard chow (25 g). Mean daily total food and energy administered per rat/day were 106 g/day and 256 kcal/day, respectively. The caloric distribution of the CAF diet was 11% protein, 34% Nutrients 2021, 13, 4474 4 of 17 fat, and 55% carbohydrates. Both CAF-R and CAF-RO diets were based on the same items and had a very similar qualitative composition to the CAF diet, but the amount of each food item administered was readjusted every week with a 30% calorie restriction relative to the energy intake consumed by the CAF group. Averaged over the 14 weeks, CAF-R and CAF-RO diets included (average quantity administered per rat/day): bacon (3.2 g), biscuit with pâté (3.4 g), biscuit with cheese (6.4 g), muffins (3.2 g), carrots (3.8 g), jellied sugared milk (12.1 g, 18% w/w), and standard chow (13.2 g). Mean daily total food and energy administered per rat/day were 45.3 g/day and 115 kcal/day, respectively. The caloric distribution of this diet was 12% protein, 35% fat, and 53% carbohydrates. The STD group was fed with standard chow (Teklad Global 14% Protein Rodent Diet 2014, Harlan) ad libitum during the entire experiment. Mean daily total food and energy administered per rat/day were 25 g/day and 72.5 kcal/day. The caloric distribution of the STD diet was 20% protein, 13% fat, and 67% carbohydrates. 2.3. Oleuropein Supplementation A 25 mg/kg daily dose of an Olea europaea leaf extract (Benolea® , Frutarom Health BU, Switzerland), containing a 20.1% of OLE, was administered to the CAF-RO-fed animals. The extract was diluted with sugared (0.7% w/w) strawberry gelatin to a concentration of 3 mg/mL gelatin and was prepared and administered daily for the 14 weeks of the dietary treatments to the CAF-RO group. Simultaneously, a placebo strawberry gelatin without OLE was prepared and administered daily for the rest of the groups (STD, CAF and CAF-R) as a control procedure. During the 14 weeks of OLE/placebo sugared gelatin supplementation the animals were fed with their corresponding diets. 2.4. Body Mass Index Body length was measured as the NAL in order to estimate the BMI using the following formula: weight (g)/body length (cm)2 . 2.5. Serum Analyses An enzymatic colorimetric kit was used to assay serum triacylglycerides (QCA, Barcelona, Spain). Serum insulin and leptin levels were measured using a mouse/rat insulin ELISA kit (Millipore, Barcelona, Spain) and a rat leptin ELISA kit (Millipore), respectively. Insulin resistance was estimated using the homeostatic model assessment (HOMA-IR), following the formula: HOMA-IR = Glucose × Insulin/22.5 [28]. 2.6. Behavioural Procedures 2.6.1. Two-Bottle Preference Test To determine the preference for sucrose solutions, a two-bottle preference test was performed from weeks 19 to 21. It consisted of placing two bottles at the opposing sides of the food container cover of the animals’ cage, one containing a sucrose solution and the other one containing tap water. First, the animals were presented with tap water from two bottles to habituate them to the procedure and as a control to measure water intake for 48 h. After the habituation period seven concentrations of sucrose solutions were tested (0.01 M, 0.03 M, 0.06 M, 0.1 M, 0.3 M, 0.6 M, 1 M), presented in ascending order. Sucrose solutions were prepared daily using tap water. Each sucrose solution was presented for 2 days/48 h for a total of 14 days together with a bottle of water. Water and sucrose intakes were recorded daily and bottle positions were switched every 24 h to avoid laterality preference [29]. Sucrose preference ratio was calculated according to the formula: sucrose solution intake (g)/water + sucrose solution intake (g). All animals were fed with their corresponding diets throughout the behavioral test. 2.6.2. Taste Reactivity Test To evaluate hedonic responses to different sucrose solutions, we performed the taste reactivity test for 4 days at week 22 of the experiment. On the first day of the test, the Nutrients 2021, 13, 4474 5 of 17 animal was placed in a cylindrical cage for 5 min without solution to habituate it to the test conditions. The next day the animal was placed in the cylindrical cage and after 2.5 min of habituation to the testing place, 1 mL of tap water was deposited on the floor to habituate it to the presence of a drop of solution. After these acclimation days, the third and the fourth day 1 mL of sucrose solution (0.1 M and 1 M) was deposited on a corner of the transparent floor after an acclimation period of 2.5 min. The orofacial expressions were recorded from below using a video camera (JVC, HD Everio GZ-HD10) during and after voluntary sucrose intake for 2.5 min. Sucrose solutions were prepared daily using tap water. Positive hedonic responses were measured as number of episodes (n), and tongue protrusion (TP), paw licking (PL) and mouthing (M) were measured. Aversive responses were measured as the number of episodes, and included head shaking (HS), forelimb flailing (FF), snout cleaning (SC) and gagging (G). Both types of responses were considered by examination of slow-motion recorded videos [30]. This test was performed similarly as described by Shin et al. 2011 [31]. 2.7. Statistical Analyses Statistical analyses were performed using SPSS Statistics 22 (SPSS, Inc., Chicago, IL, USA). The homoscedasticity among groups was measured using Levene’s test. The normal distribution of the values was measured using the Kolmogorov-Smirnov test. A Mann- Whitney U test was performed to evaluate differences between groups on biometric, food intake and serum parameters at the end of the obesity-induction period due to the different sample size of STD and CAF groups. One-way analysis of variance (ANOVA) followed by Tukey post-hoc or Student’s t tests were performed to evaluate differences among groups on biometric, food intake and serum parameters at the end of the dietary treatments period. Triacylglycerides and leptin data were log transformed to homogenize variances. The intake data for sucrose solution, total fluid and total sucrose were analyzed by a Repeated Measures ANOVA, with the sucrose solution concentration as the within-subjects factor and the diet as the between-subjects factor. Wilks’ Lambda of Multivariate tests was used to detect significant sucrose solution effects and significant ‘sucrose solution × diet’ interactions. Sucrose preference was analyzed with a Kruskal-Wallis test followed by a Mann-Whitney U test. Data from taste reactivity test was analyzed using Student’s t-test for comparisons between groups. Spearman’s rho correlation was performed to determine the relationship between serum leptin levels and sucrose preference and total sucrose consumption. All the results were expressed as mean ± SEM. The level of statistical significance was set at bilateral 5%. 3. Results 3.1. Effects of Diets on the Biometric, Food Intake, and Serum Parameters Biometric, average daily intake of food, energy and simple sugars, and serum parame- ters (Table 1) were measured at the end of the obesity-induction period and after 10 weeks of dietary treatments (weeks 9 and 19, respectively). At week 9 the CAF animals showed a higher increase of body weight and BMI (p < 0.01) than the STD animals. The CAF group also showed higher values of daily food and energy intake than the STD group during that period (p < 0.001). Only the CAF group consumed simple sugars due to the presence of high-sugar food components in this diet. As for the serum metabolic parameters, the CAF diet induced an increase in circulating levels of triacylglycerides, insulin, insulin resistance indicated by the HOMA-IR (p < 0.05), and leptin (p < 0.001), compared to the STD group. At week 19 the results showed that CAF-fed animals had a greater body weight gain than the other three groups (p < 0.001). CAF-R and CAF-RO groups showed a significantly lower weight gain compared with the CAF group, whereas no differences were observed with the STD group or between them. Similarly, the CAF group showed a significantly increased BMI compared to the other three groups at this time, while no differences were observed between the STD and CAF-R or CAF-RO groups (p < 0.001). In addition, the CAF-fed animals showed the highest values of food and energy intake, the STD the lowest Nutrients 2021, 13, 4474 6 of 17 values, and CAF-R and CAF-RO animals intermediate values (p < 0.001). Both CAF-R and CAF-RO-fed animals consumed less than half the amount of simple sugars consumed by CAF-fed animals (p < 0.001). Table 1. Biometric, food intake and serum parameters. CAF Diet-Induced Obesity Period Dietary Treatments Period STD CAF STD CAF CAF-R CAF-RO Biometric parameters Weeks 1–8 Weeks 9–19 Body weight gain (g) 255 ± 6 313 ± 5 +++ 121 ± 5 158 ± 9 ** 124 ± 5 $$ 132 ± 6 $$ BMI (g/cm2 ) 1 0.64 ± 0.01 0.70 ± 0.01 ++ 0.74 ± 0.01 0.83 ± 0.01 *** 0.78 ± 0.01 $ 0.78 ± 0.02 $ Daily food intake parameters Food intake (g/kg) 101 ± 1 309 ± 2 +++ 46.3 ± 0.6 140 ± 4 *** 77.2 ± 2.1 *** $$$ 75.9 ± 1.3 *** $$$ Energy intake (kcal/kg) 292 ± 3 667 ± 5 +++ 134 ± 2 289 ± 6 *** 183 ± 4 *** $$$ 177 ± 4 *** $$$ Chow intake (kcal/kg) 292 ± 3 123 ± 3 +++ 134 ± 2 44.1 ± 2.9 *** 58.8 ± 4.2 *** $ 53.8 ± 4.5 *** Simple sugars (kcal/kg) 196 ± 2 93.7 ± 3.6 36.8 ± 1.3 $$$ 36.7 ± 1.1 $$$ Serum parameters Week 9 Week 19 Triacylglycerides 0.93 ± 0.06 1.31 ± 0.09 + 1.07 ± 0.08 1.79 ± 0.22 ** 1.58 ± 0.14 * 1.54 ± 0.13 (mmol/L) Insulin (ng/mL) 0.51 ± 0.07 0.70 ± 0.05 + 0.61 ± 0.08 1.03 ± 0.17 * 0.82 ± 0.06 0.83 ± 0.10 HOMA-IR 3.20 ± 0.54 4.73 ± 0.37 + 3.68 ± 0.57 6.92 ± 1.31 * 5.65 ± 0.58 * 5.36 ± 0.76 Leptin (ng/mL) 4.23 ± 0.40 12.5 ± 1.0 +++ 7.05 ± 0.78 23.7 ± 4.3 *** 19.1 ± 1.9 *** 17.0 ± 2.0 *** Average daily food, energy and simple sugars intakes are relativized to the body weight (BW, kg). 1 BMI was calculated at week 9 and at week 19. STD, standard group; CAF, cafeteria diet group; CAF-R, calorie-restricted cafeteria diet group; CAF-RO, calorie-restricted cafeteria diet with OLE; HOMA-IR, homeostatic model assessment of insulin resistance. Data are expressed as the mean ± SEM. + p < 0.05, ++ p < 0.01, +++ p < 0.001 vs. STD group (obesity-induction period). * p < 0.05, ** p < 0.01, *** p < 0.001 vs. STD group; $ p < 0.05, $$ p < 0.01, $$$ p < 0.001 vs. CAF group (dietary treatments period). Regarding the serum parameters, the CAF and CAF-R groups showed higher levels of circulating triacylglycerides compared with the STD group (p < 0.001 and p < 0.05, respectively), whereas CAF-RO showed no differences from STD. The CAF group also showed a significant increase in insulin levels (p < 0.05), whereas no differences appeared in CAF-R and CAF-RO groups, compared to the STD group. Both CAF and CAF-R diets increased insulin resistance compared to STD diet (p < 0.05), whereas the CAF-RO diet did not change the HOMA-IR compared with the STD diet. Finally, the CAF, CAF-R and CAF-RO diets increased leptin levels compared to the STD diet (p < 0.001). 3.2. Two-Bottle Preference Test 3.2.1. Sucrose Solution Intake and Preference To determine the preference for different sucrose solutions, a two-bottle preference test using several concentrations of sucrose (0.01 M, 0.03 M, 0.06 M, 0.1 M, 0.3 M, 0.6 M, 1 M) was performed (Figure 1). Results revealed a significant effect of concentration on the sucrose solution intake (p < 0.001), and that this effect was dependent on the dietary treatment received (‘sucrose concentration × diet’, p < 0.05). The sucrose solution intake followed an inverted-U pattern in all experimental groups. They gradually increased the intake over 0.01 M–0.3 M solutions, reaching the maximum of the inverted curve at 0.3 M. From that point on, all groups gradually decreased the intake over the 0.6 M and 1 M solutions with a steep or gentle slope depending on the group. The STD group showed the highest intakes at all sucrose solution concentrations. In fact, the sucrose solution intake of the STD rats almost doubled the intake shown by the cafeteria rats, not showing differences among them at any of the sucrose concentrations. the highest intakes at all sucrose solution concentrations. In fact, the sucrose solution in- take of the STD rats almost doubled the intake shown by the cafeteria rats, not showing differences among them at any of the sucrose concentrations. the highest intakes at all sucrose solution concentrations. In fact, the sucrose solution in- Nutrients 2021, 13, 4474 take of the STD rats almost doubled the intake shown by the cafeteria rats, not showing 7 of 17 differences among them at any of the sucrose concentrations. Figure 1. Effects of the dietary treatments on sucrose solution intakes in the two-bottle preference test. Sucrose solution intakes were relativized to the body weight (BW, kg). Data are expressed as theFigure mean1.±Effects SEM. *ofpthe < 0.05, ** ptreatments dietary < 0.01, ***onp