Sex and Feeding Status Differently Affect Natural Reward Seeking Behavior in Olfactory Bulbectomized Rats

Ruda-Kucerova, Jana; Zanda, Mary Tresa; Amchova, Petra; Fratta, Walter; Fattore, Liana

doi:10.3389/fnbeh.2018.00255

ORIGINAL RESEARCH article

Front. Behav. Neurosci., 30 October 2018
Sec. Motivation and Reward
Volume 12 - 2018 | https://doi.org/10.3389/fnbeh.2018.00255

Sex and Feeding Status Differently Affect Natural Reward Seeking Behavior in Olfactory Bulbectomized Rats

Jana Ruda-Kucerova¹

Mary Tresa Zanda²

Petra Amchova¹

Walter Fratta^2,3

Liana Fattore^3,4*

¹Department of Pharmacology, Faculty of Medicine, Masaryk University, Brno, Czechia
²Department of Biomedical Sciences, Section of Neuroscience and Clinical Pharmacology, University of Cagliari, Monserrato, Italy
³Center of Excellence “Neurobiology of Addiction”, University of Cagliari, Monserrato, Italy
⁴CNR Institute of Neuroscience-Cagliari, National Research Council, Rome, Italy

Substance abuse and depression are common psychiatric disorders with a high rate of comorbidity. Both conditions affect differently men and women and preclinical research has showed many sex differences in drug addiction and depression. The most common approach for modeling depression-addiction comorbidity is the combination of the intravenous drug self-administration and the olfactory bulbectomy (OBX) models in rats. Such a combination has revealed enhanced drug-taking and drug-seeking behaviors in OBX rats, but no study has investigated so far potential sex differences in operant responding and motivation for natural reinforcers in OBX rats. This study investigated for the first time operant self-administration of palatable food pellets in male and female OBX rats under different feeding status, i.e., ad libitum vs. restricted food, and schedules of reinforcement, i.e., a continuous ratio schedule fixed ratio 1 (FR1) vs. a complex (FR5_(x)) second order schedule of reinforcement. In the FR1 experiment, OBX rats of both sexes exhibited lower operant responding and intake of palatable food pellets than sham-operated controls, with food restriction leading to increased operant responding in both OBX and SHAM groups. Female rats showed higher responding than males but this effect was abolished by the OBX lesion. Similarly, in the (FR5_(x)) second order schedule of reinforcement both male and female OBX rats showed lower responding and food intake, with SHAM and OBX females showing higher operant responding than corresponding male groups. Overall, our findings showed that: (i) responding for food was lower in OBX than in SHAM rats under both FR1 and (FR5_(x)) schedules of reinforcement; (ii) sex and food restriction affect operant responding for palatable food; and (iii) the suppressing effect of OBX lesion on food intake was consistently present in both sexes and represents the most robust factor in the analysis. This may represent anhedonia which is associated with depressive-like phenotype and palatable food self-administration may serve as a robust behavioral index of anhedonia in the OBX model.

Introduction

Substance use disorder and depression are common in the general population and display a high frequency of comorbidity (Torrens and Rossi, 2015). In humans, gender differences have been consistently described in substance use disorder (Fattore et al., 2008; Fattore and Melis, 2016) and depression (Marcus et al., 2005) as well as in other forms of addictive behaviors (Fattore et al., 2014) and emotional disturbances (Seney and Sibille, 2014). In depressed substance users it is hard to both determine if one disorder (e.g., depression) may affect predisposition to the other (e.g., drug abuse) and identify the underlying mechanisms. Animal studies experimentally reproduce behavioral traits and features relevant to depression and drug intake and allow the study of the underlying mechanisms. Despite inevitable limitations, animal models are unanimously acknowledged as essential for understanding the biological factors that contribute to addiction and mood disorders and for developing new pharmacotherapies (Micale et al., 2013; Müller, 2017; Robinson, 2018). They also represent an essential step in the study of sex differences (Becker and Koob, 2016) and motivational brain system (Gonen et al., 2012) as well as in the development of tailored therapeutic strategies for male and female patients (Buoncervello et al., 2017).

Reliable and validated animal models are currently available to study addiction and depression. Yet, only relatively few preclinical studies have examined addictive behaviors using animal models of depression (Filip et al., 2013) and even less are those that performed the experiments in both male and female animals to investigate potential sex differences (Ng et al., 2017). Operant behavior to obtain palatable food strongly impacts upon the brain reward circuit (Guegan et al., 2013) and, in animals, operant responding for food reinforcers has been shown to be affected by sex (Ward and Walker, 2009). Food intake is strictly associated to emotional states by complex physiological and behavioral interactions (Ulrich-Lai et al., 2015), and eating behavior can be significantly altered in depressed subjects (Paans et al., 2018).

The olfactory bulbectomy (OBX) in rats is a widely recognized animal model that displays a high degree of neurochemical similarity to human depression, helps to identify potential antidepressant drugs and gives important insights into their possible mechanisms of action (Leonard, 1984; van Riezen and Leonard, 1990; Kelly et al., 1997; Harkin et al., 2003; Song and Leonard, 2005). A number of self-administration studies have been conducted in OBX rats to verify whether depressive-like animals displayed altered voluntary intake of drugs of abuse. Findings have shown that, when compared to corresponding sham-operated animals, OBX rats exhibit a higher intake of amphetamine (Holmes et al., 2002), nicotine (Vieyra-Reyes et al., 2008), methamphetamine (Kucerova et al., 2012), synthetic cannabinoid (Amchova et al., 2014) and ketamine (Babinska and Ruda-Kucerova, 2017), but not of cocaine (Frankowska et al., 2014), suggesting that OBX rats may have a different sensitivity to the pharmacological effects of drugs of abuse. Notably, neural circuits regulating drug-induced reward extensively overlap with those regulating natural reward and reinforcement, including food (Blum et al., 2011). Interestingly, bulbectomized rats were found to display pronounced deficits in brain reward functions (Slattery et al., 2007; Grecksch and Becker, 2015) associated to a dysfunctional dopaminergic signaling (Amchova et al., 2014; Jastrzębska et al., 2015; Ruda-Kucerova et al., 2015a), which suggests that the reward mechanisms may be altered in this animal model as a common phenomenon associated with depression. The hypothesis of a brain hypodopaminergic function induced by bilateral OBX is strengthened by the finding that sucrose preference is reduced, i.e., anhedonia (Romeas et al., 2009; Sato et al., 2010) and the expression of cocaine-induced place preference is disrupted in OBX rats (Calcagnetti et al., 1996). OBX was also shown to modulate feeding pattern in rats (Meguid et al., 1993), with functional adaptations of food intake regulatory mechanism occurring over time after olfactory bulbs ablation (Meguid et al., 1997). Yet, whether operant responding for palatable food and underlying motivation is altered in OBX rats remains an open question. Equally unexplored are the potential factors that may affect self-administration of palatable food in OBX rats, such as the diet regimen and the amount of effort required to obtain a single pellet of palatable food.

This study was designed to: (i) assess if alterations in the drug self-administration reported in OBX rats extend to operant responding and motivation for natural reinforcers, i.e., palatable food; (ii) verify whether a depression-like state may affect operant behavior differently in male and female rats; and (iii) determine the potential modulating role of the feeding status (i.e., restricted vs. ad libitum) and the reinforcement schedule complexity fixed ratio 1 (FR1) vs. (FR5_(x)) second order schedule) in operant responding for food in OBX rats.

Materials and Methods

Animals

Thirty-six male and 36 female Lister-Hooded rats (weight range of 250–300 g at the beginning of the experiment) were purchased from Harlan-Nossan (Italy) and housed four per cage at the Animal Facility of the Department of Biomedical Sciences, University of Cagliari, Italy. Males and females were housed in different rooms. Environmental conditions during the whole study were constant: relative humidity 50%–60%, room temperature 23°C ± 1°C, inverted 12-h light-dark cycle (6 a.m. to 6 p.m. darkness). Food and water were available ad libitum unless otherwise specified below. All procedures were performed in accordance with EU Directive no. 2010/63/EU and approved by the local Animal Care Committee at the Department of Biomedical Sciences, University of Cagliari, Italy.

Olfactory Bulbectomy Surgery

At the beginning of the experiments, both male and female rats were randomly divided into two groups (n = 18 per group): bulbectomized (OBX) and sham-operated (SHAM) rats. The bilateral ablation of the olfactory bulbs was performed as previously described (Amchova et al., 2014; Ruda-Kucerova et al., 2015a). Animals were anesthetized with isofluran 2%, the top of the skull was shaved and swabbed with an antiseptic solution. Then, midline frontal incision was made on the skull and the skin was retracted bilaterally. Two burr holes, 2 mm in diameter, were drilled in the frontal bone 7 and 7.5 mm anterior from the Bregma, 1.5 and 2 mm lateral to Bregma suture for rats weighing 230 ± 10 g and 260 ± 10 g, respectively. Both olfactory bulbs were removed by aspiration paying particular attention to not damage the frontal cortex. Prevention of blood loss from the ablation cavity was achieved by filling the dead space with a hemostatic sponge. The skin above the lesion was closed with suture. Finally, bacitracin plus neomycin powder was applied to prevent bacterial infection. Sham-operated rats underwent identical anesthetic and drilling procedures but their bulbs were left intact. A period of at 3 weeks was allowed for the recovery from the surgical procedure and the development of the characteristic phenotype. During this period, animals were handled daily for few minutes to eliminate aggression, which could otherwise arise (Kelly et al., 1997; Song and Leonard, 2005). At the end of the experiment, rats were euthanized by an anesthetic overdose and the brains were dissected for confirmation of the successful surgical lesion. Three animals (1 F, 2 M) were excluded from analysis due to incomplete removal of the OB, while one female rat was excluded because of damage to the prefrontal cortex.

Food Self-Administration Protocols

Food self-administration was conducted in 12 operant chambers (29.5 × 32.5 × 23.5 cm, Med Associates, Fairfax, VT, USA) using lever-pressing as operandum. Each chamber was encased in a sound and light attenuating cube. In addition, chambers had a ventilation fan, and a front panel equipped with two retractable levers (each 4 cm wide) positioned 12 cm apart, 8 cm from the grid and extending 1.5 cm into the box. A white stimulus light was placed above each lever and a red house light was located on the opposite wall. All self-administration sessions were conducted at the same time daily during the dark period of the inverted light-dark cycle.

There were two experiments performed. Experiment 1 (FR1) aimed to determine consummatory behavior and evaluated the effect of OBX and feeding status (ad libitum vs. food restriction to 80% of free-feeding body weight). For this part of the study the rats were divided into following groups (n = 6 per group): SHAM fed ad libitum, SHAM restricted, OBX fed ad libitum and OBX restricted. Experiment 2 (FR5_(x)) was designed to assess appetitive behavior. In this experiment, the number of groups was reduced due to equal minimal effect of the food restriction observed in the Experiment 1. Therefore, rats were only divided into two groups, n = 6 per group: SHAM and OBX. This allowed a reduction of rats used and refinement of the procedures by including only ad libitum fed rats.

Experiment 1 (FR1)

The training was conducted under a FR1 schedule of reinforcement, i.e., the animal had to press to the active lever once to obtain a single palatable food pellet (BioServ, palatable dustless rodent food pellets, F0021-Purified Casein Based Formula—45 mg). The session lasted 30 min and the house-light was on throughout the session. The length of the training was 10 consecutive days. All animals consumed the vast majority of the gained pellets of food. There were four experimental groups of male rats (n = 6) and four groups of female rats (n = 6) divided by the OBX surgery and feeding status. Data are expressed as mean of active/inactive lever presses per group and session; gained food pellets were analyzed as a cumulative count, e.g., “day 5 cumulative intake” equals the sum of the number of food pellets self-administered on days 1–5).

Experiment 2 (FR5_(x))

A different batch of SHAM and OBX rats was used in the Experiment 2. The training was conducted under a (FR5_(x)) second order protocol as described by Thornton-Jones et al. (2005) with slight modifications. Briefly, animals were first trained to lever press on a FR1 schedule for 7–8 days. A single press on the active lever led to the illumination of a house-light and delivery of a single food pellet after 4 s. Inactive lever presses were not rewarded but the house-light was on, indicating that the FR requirement has been fulfilled. The house-light remained on for 4 s prior to food pellets delivery and 4 s after delivery to enhance the cue effect of the light. Once animals reached more than 100 active lever presses on FR1, they were transferred to an FR5 schedule. The protocol was analogous to FR1 but rats were required to press five times the active lever to get a single food pellet and lasted 4 days. After completion rats were transferred to a FR5(2) schedule and were required to obtain two consecutive house-light presentations in order to receive two food pellets (4 days) followed by FR5(3) schedule in a similar manner, i.e., three house-light presentations to receive three food pellets (4 days) and FR5(4) schedule (4 days). Finally, the animals underwent the final training schedule FR5(5) for 7 days. There were two experimental groups of male rats (SHAM and OBX, n = 6) and two groups of female rats (SHAM and OBX, n = 6). Data are expressed as mean of active/inactive lever presses and number of food pellets per group and session.

Statistical Data Analysis

Primary data were summarized using arithmetic mean and standard error of the mean (SEM) estimate. The analyses were calculated using Statistica 12 (StatSoft, Inc. Tulsa, OK, USA). A value p < 0.05 was recognized as boundary of statistical significance in all applied tests.

Experiment 1

Food self-administration data were analyzed by ANOVA for repeated measures (RM) with factors: OBX model, feeding status, repeated variable: day; followed by Bonferroni post hoc test for analysis of significant interactions of factors. Sex differences in the food self-administration were also assessed RM ANOVA (factors: OBX model, sex, repeated variable: day) followed by Bonferroni post hoc test for analysis of significant interactions of factors.

Experiment 2

Food self-administration data were analyzed by RM ANOVA with factors: OBX model, repeated variable: FR schedule; followed by Bonferroni post hoc test for analysis of significant interactions of factors. Sex differences were assessed by RM ANOVA (factors: OBX model, sex, repeated variable: FR schedule) followed by Bonferroni post hoc test for analysis of significant interactions of factors.

Results

In line with previous study conducted on hamsters (Pieper et al., 1992) and rats (Meguid et al., 1993), bilateral OBX did not alter food intake and body weight gain, and these parameters did not differ significantly through the study between OBX and SHAM groups of the same sex.

Experiment 1 (FR1 Schedule)

Experiment 1 assessed operant self-administration of palatable food pellets over 10 days under FR1 schedule of reinforcement in both male and female Lister Hooded rats. The data are presented as daily mean numbers of active lever presses, inactive lever presses and cumulative number of delivered food pellets.

Male Rats

In active lever pressing (Figure 1A) RM ANOVA revealed a highly significant effect of OBX model (F_(1,20) = 123.65, p < 0.001), feeding status (F_(1,20) = 22.37, p < 0.001), day (F_(9,180) = 49.76, p < 0.001) and day*OBX model interaction (F_(9,180) = 3.57, p < 0.001). Bonferroni post hoc test for day*OBX model interaction indicated significantly decreased responding in the OBX rats from day 2 onwards (day 2: t_(158,38) = 4.20, p = 0.009; day 3: t_(158,38) = 4.45, p = 0.003; day 4: t_(158,38) = 5.49, p < 0.001; day 5: t_(158,38) = 5.30, p < 0.001; day 6: t_(158,38) = 6.13, p < 0.001; day 7: t_(158,38) = 6.37, p < 0.001; day 8: t_(158,38) = 7.32, p < 0.001; day 9: t_(158,38) = 7.68, p < 0.001; day 10: t_(158,38) = 7.09, p < 0.001). Inactive lever pressing did not differ among the groups (Figure 1B) and RM ANOVA indicated only a significant effect of the repeated factor—day (F_(9,180) = 2.81, p = 0.004) which reflects the learning process during acquisition.

FIGURE 1

Figure 1. Self-administration of palatable pellets in MALE rats under fixed ratio 1 (FR1) schedule. The graphs present the mean ± standard error of the mean (SEM) of numbers of active lever presses (A), inactive lever presses (B) and cumulative numbers of self-administered pellets (C) in male rats during the 10-days training. The tables on the side summarize results of repeated measures (RM) ANOVA and Bonferroni post hoc test for each variable (**p < 0.01, ***p < 0.001 for day*olfactory bulbectomy (OBX) interaction, ^#p < 0.05, ^##p < 0.01, ^###p < 0.001 for day*feeding interaction).

Cumulative number of delivered food pellets (Figure 1C) showed similar trends as active responding. RM ANOVA showed a highly significant effect of OBX model (F_(1,20) = 81.42, p < 0.001), feeding status (F_(1,20) = 20.99, p < 0.001), day (F_(9,180) = 1175.89, p < 0.001), day*OBX model interaction (F_(9,180) = 81.90, p < 0.001) and day*feeding status interaction (F_(9,180) = 14.80, p < 0.001). Bonferroni post hoc test for day*OBX model interaction indicated significantly suppressed food pellet intake in the OBX rats from day 4 onwards (day 4: t_(41,118) = 4.55, p = 0.009; day 5: t_(41,118) = 6.0, p < 0.001; day 6: t_(41,118) = 7.85, p < 0.001; day 7: t_(41,118) = 9.54, p < 0.001; day 8: t_(41,118) = 11.70, p < 0.001; day 9: t_(41,118) = 13.96, p < 0.001; day 10: t_(41,118) = 16.19, p < 0.001). Bonferroni post hoc test for day*feeding status interaction revealed higher food pellet intake in the food restricted animals from day 6 onwards (day 6: t_(41,12) = 4.30, p = 0.02; day 7: t_(41,12) = 4.97, p = 0.002; day 8: t_(41,12) = 5.63, p < 0.001; day 9: t_(41,12) = 6.31, p < 0.001; day 10: t_(41,12) = 7.12, p < 0.001). All variables are shown in Figure 1 together with an overview of statistical results (right).

Female Rats

In active lever pressing (Figure 2A) RM ANOVA revealed a highly significant effect of OBX model (F_(1,20) = 275.57, p < 0.001), feeding status (F_(1,20) = 11.25, p = 0.003) and OBX model*feeding status interaction (F_(1,20) = 5.36, p = 0.031). Bonferroni post hoc test showed increased responding in SHAM restricted rats compared to ad libitum fed animals (t_(20,00) = 4.01, p = 0.004) while the comparison of OBX restricted and ad libitum fed animals renders non-significant result. Furthermore, RM ANOVA showed a significant effect of day (F_(9,180) = 34.00, p < 0.001) and day*OBX model interaction (F_(9,180) = 6.08, p < 0.001). Bonferroni post hoc test for day*OBX model interaction indicated significantly lower responding in the OBX rats in all 10 consecutive sessions (day 1: t_(109,41) = 5.19, p < 0.001; day 2: t_(109,41) = 8.94, p < 0.001; day 3: t_(109,41) = 9.55, p < 0.001; day 4: t_(109,41) = 9.62, p < 0.001; day 5: t_(109,41) = 10.66, p < 0.001; day 6: t_(109,41) = 11.93, p < 0.001; day 7: t_(109,41) = 12.50, p < 0.001; day 8: t_(109,41) = 11.46, p < 0.001; day 9: t_(109,41) = 10.78, p < 0.001; day 10: t_(109,41) = 10.74, p < 0.001). As in male rats, inactive lever pressing did not differ among groups (Figure 2B) and RM ANOVA indicated only a significant effect of the repeated factor—day (F_(9,180) = 47.16, p < 0.001).

FIGURE 2

Figure 2. Self-administration of palatable pellets in FEMALE rats under FR1 schedule. The graphs present the mean ± SEM of numbers of active lever presses (A), inactive lever presses (B) and cumulative numbers of self-administered pellets (C) in female rats during the 10-days training. The tables on the side summarize results of RM ANOVA and Bonferroni post hoc test for each variable (***p < 0.001 for day*OBX interaction, ^##p < 0.01 for day*feeding status interaction, ^$p < 0.05, ^$$p < 0.01 for day*OBX feeding status interaction—SHAM restricted vs. SHAM ad libitum).

In the analysis of cumulative number of delivered food pellets (Figure 2C) RM ANOVA showed a highly significant effect of OBX model (F_(1,20) = 249.55, p < 0.001), feeding status (F_(1,20) = 8.80, p = 0.007), day (F_(9,180) = 883.33, p < 0.001), day*OBX model interaction (F_(9,180) = 192.26, p < 0.001), day*feeding status interaction (F_(9,180) = 6.03, p < 0.001) and day*OBX model*feeding status interaction (F_(9,180) = 2.17, p = 0.026). Bonferroni post hoc test for day*OBX model interaction indicated significantly decreased food pellet intake in the OBX rats from day 3 onwards (day 3: t_(42,207) = 6.03, p < 0.001; day 4: t_(42,207) = 8.45, p < 0.001; day 5: t_(42,207) = 11.15, p < 0.001; day 6: t_(42,207) = 14.16, p < 0.001; day 7: t_(42,207) = 17.15, p < 0.001; day 8: t_(42,207) = 20.08, p < 0.001; day 9: t_(42,207) = 22.74, p < 0.001; day 10: t_(42,207) = 25.49, p < 0.001). Bonferroni post hoc test for day*feeding status interaction revealed higher food pellet intake in the food restricted animals on day 9 and 10 (day 9: t_(42,21) = 4.19, p = 0.026; day 10: t_(42,21) = 4.70, p = 0.005) and Bonferroni post hoc test for day*OBX model*feeding status interaction showed higher food pellet intake in the food restricted SHAM animals compared to ad libitum fed SHAM control on days 9–10 (day 9: t_(42,21) = 4.73, p = 0.020, day 10: t_(42,21) = 5.37, p = 0.002). There was no effect of feeding status in the OBX animals. All variables are shown in Figure 2 together with an overview of statistical results (right).

Evaluation of Sex Differences

For the analysis of sex differences only the ad libitum fed groups were used and therefore the analyzed groups were the following: males SHAM, males OBX, females SHAM, females OBX (Figure 3). Variables included in the analysis were active lever presses and cumulative food pellets intake, responding on the inactive lever not differing among groups.

FIGURE 3

Figure 3. Sex differences in the self-administration of palatable pellets under FR1 schedule. The graphs present the mean ± SEM of daily numbers of active lever presses (A) and cumulative numbers of self-administered pellets (B) in ad libitum fed experimental groups of both sexes during the 10-days training. The tables on the side summarize results of RM ANOVA and Bonferroni post hoc test for each variable (*p < 0.05, **p < 0.01, ***p < 0.001). There is no sex difference in the repeated design. For the main effect of sex, see the “Results” section.

In active lever pressing (Figure 3A) RM ANOVA revealed a highly significant effect of the OBX model (F_(1,20) = 196.45, p < 0.001) and OBX model*sex interaction (F_(1,20) = 6.33, p = 0.021). Bonferroni post hoc test for OBX model*sex interaction indicated significantly lower responding in the male OBX rats compared to male SHAM rats (t_(20,00) = 8.13, p < 0.001) and female OBX rats responding less than female SHAM rats (t_(20,00) = 11.69, p < 0.001). Furthermore, RM ANOVA showed a significant effect of the day (F_(9,180) = 43.94, p < 0.001), day*OBX model (F_(9,180) = 5.90, p < 0.001) and day*sex interaction (F_(9,180) = 2.69, p = 0.006). Bonferroni post hoc test for day*OBX model interaction revealed lower responding in the OBX animals from day 2 onwards (day 2: t_(137,08) = 5.66, p < 0.001; day 3: t_(137,08) = 6.56, p < 0.001; day 4: t_(137,08) = 7.18, p < 0.001; day 5: t_(137,08) = 8.10, p < 0.001; day 6: t_(137,08) = 8.79, p < 0.001; day 7: t_(137,08) = 9.03, p < 0.001; day 8: t_(137,08) = 9.83, p < 0.001; day 9: t_(137,08) = 9.88, p < 0.001; day 10: t_(137,08) = 9.05, p < 0.001) and Bonferroni post hoc test for the day*sex interaction did not show any significant result.

In the cumulative number of delivered food pellets (Figure 3B) RM ANOVA showed a highly significant effect of OBX model (F_(1,20) = 202.73, p < 0.001), sex (F_(1,20) = 4.45, p = 0.048), and OBX model*sex interaction (F_(1,20) = 13.55, p = 0.001). Bonferroni post hoc test for OBX model*sex interaction indicated significantly higher food pellets intake in the female SHAM rats compared to male SHAM rats (t_(20,00) = −4.09, p = 0.003) but not in the OBX groups: female OBX rats compared to male OBX rats (n.s.). Furthermore, RM ANOVA detected a significant effect of day (F_(9,180) = 888.39, p < 0.001), day*OBX model interaction (F_(9,180) = 131.27, p < 0.001), and day*OBX model*sex interaction (F_(9,180) = 4.83, p < 0.001). Bonferroni post hoc test for day*OBX model interaction revealed higher responding in the SHAM animals from day 3 onwards (day 3: t_(52,23) = 4.58, p < 0.001; day 4: t_(52,23) = 6.78, p < 0.001; day 5: t_(52,23) = 9.18, p < 0.001; day 6: t_(52,23) = 11.70, p < 0.001; day 7: t_(52,23) = 14.35, p < 0.001; day 8: t_(52,23) = 17.41, p < 0.001; day 9: t_(52,23) = 20.27, p < 0.001; day 10: t_(52,23) = 23.01, p < 0.001) and Bonferroni post hoc test for the day*OBX model*sex interaction did not show any significant results.