Atorvastatin reduces alloxan-induced impairment of aversive stimulus memory in mice

After obtaining approval from Ondokuz Mayıs University Experimental Animal Research Committee (Hadyek_2020/19), https://ebyssorgu.omu.edu.tr (document No. 9203, verification No. 81DI-HHV0-0ES7), we obtained 35 male BALB/c mice (Mus musculus, 30–40 g) raised under specific pathogen free conditions in the Ondokuz Mayıs Vivarium. All experiments were conducted according to the regulations of Ondokuz Mayıs University Experimental Animal Research Committee that at least meet the standards of the National Institutes of Health Guide for the Care and Use of Laboratory Animals (Washington, DC: National Academy Press; 1996). Studies are reported in compliance with the ARRIVE 2.0 guidelines [25]. The mice were maintained in the standard conditions (22 ± 2 °C, 55% humidity, 12–12 night–day cycle, and 3–4 mice per cage) and allowed free access to water and standard laboratory mouse food pellets, except that food pellets were withheld without water restriction for 12 h before alloxan administration, and mice were fed with a 25% glucose solution overnight after alloxan administration to avoid hypoglycemia-induced deaths. Before and every 7 d of the experimental schedule, mice were weighed. A power analysis performed using G*Power software [26] indicated that a sample of 35 mice would be needed to observe the effects with 95% power, which was determined by one-way analysis of variance (ANOVA) between means with an α of 0.05. Assuming possible animal losses based on the mortality rate of the experimental model of diabetes model in previous work would increase the number of mice required to 40. Ultimately, we used 7 mice per group; fortunately, we had no losses and did not have to use extra mice for our experiments. Mice were monitored for behavioral and physical signs of animal welfare [27] and observed for at least 1 h each day by two investigators blinded to treatments, and the mice remained within acceptably humane endpoint criteria with no indication of pain and distress outside of the experimental stimulus and obesity because of treatments. At the end of the experiments, all mice were humanely euthanized using high-dose anesthesia (pentobarbital, 100 mg/kg, i.p.) followed by cervical dislocation.

Alloxan monohydrate was obtained from Sigma. Atorvastatin (Ator; Sanovel) and liraglutide (Saxenda; Novo Nordisk-Danimarka) were purchased from a local pharmacy in Turkey. All drugs were freshly dissolved in the saline (0.9% NaCl) before administration. Mice were administered 100 μL/kg of each drug at the same time each day.

Standard mouse food pellets were withheld from the mice for 12 h before alloxan administration. An experimental model of diabetes was induced with 150 mg/kg and 100 mg/kg intraperitoneal (i.p.) alloxan injection at a 48 h interval [28]. Control mice were injected with saline alone as the alloxan vehicle control. Mice were fed with a 25% glucose solution overnight after alloxan injection to avoid hypoglycemia-induced deaths. After 72 h, fasting glucose was measured with the drop of blood collected from the tail vein using a glycosometer (Accu-Chek, Roche). Mice with a blood glucose level of ≥250 mg/dL were accepted as having alloxan-induced diabetes mellitus, and without any specific selection divided into 5 equal groups for treatment: control (n = 7), diabetes mellitus (DM, n = 7, DM + atorvastatin (n = 7), DM + liraglutide (n = 7), and atorvastatin (n = 7). Atorvastatin (10 mg/kg, i.p.) or liraglutide (200 μg/kg, i.p.) was administered daily for 28 d. Atorvastatin and liraglutide doses were selected based on previous reports [29,30,31,32,33]. Liraglutide was used as a positive control for its effect in reducing alloxan-induced hyperglycemia. The 10 mg/kg atorvastatin treatment dose selected in these studies was the lowest neuroprotective dose suggested by the reports [29,30,31,32,33]. All behavioral tests were observed by two investigators blinded to the treatments.

According to methods previously described, an oral glucose tolerance test (OGTT) was performed to investigate glucose homeostasis and the effects of drug treatments on glucose levels [34, 35]. After 12 h fasting without water restriction, mice were treated with atorvastatin, liraglutide, or saline vehicle. After 30 min, 2.0 g/kg glucose was administered, and then blood glucose levels were measured 0.5 h, 1 h, and 2 h later. To evaluate the effects of drug treatments on motor coordination the locomotor activity was measured 24 h after the final drug treatments, after determining the blood glucose levels. The locomotor activity test was conducted using an activity cage (Locomotor Activity Cage; Ugo Basile). The cage consists of horizontal and vertical grids that record and count mouse movements in the selected period. Mice were placed in the cage and their activity was recorded for 10 min, and then the distance that mice traveled was recorded. After the tests, the apparatus was cleaned with super hypochlorous water and 70% ethanol to avoid any bias due to olfactory cues.

Mice were tested to determine any deficit in their learned avoidance of an adverse stimulus as a model of memory and cognition using a passive avoidance device (Passive Avoidance Apparatus, catalog No. 7550, Ugo Basile). The apparatus consists of an opaque black and an illuminated opaque white plexiglass compartment with an adjoining floor, but separated by a gate. The floor of the apparatus is made of a stainless-steel grid that can conduct electrical current with an indicated program to provide an adverse stimulus to the planter surface of the mouse paws (0.5–1.5 mA, 1–3 s). The test was repeated for 2 d to assess the memory and cognitive capacity of the mice [36, 37]. A mouse was placed in the white compartment and allowed to explore freely for 20 s before opening the gate. Mice that passed through the gate to the black compartment were given an electrical shock to the planter surface of their paws (0.5 mA, 1 s). The time that mice spent in the white compartment before crossing to the dark compartment where they were given the electrical shock was recorded as the initial latency for the first day, and retention period latency for the second day. Additionally, the time that mice spent in the black compartment on the second day was also recorded. This passive avoidance behavior was learned after one trial and normally results in a marked increase in crossover latency. Mice that better “remembered” the adversive stimulus from the first trial spent longer in the white compartment and had a longer retention period latency.

All data were analyzed using IBM SPSS Statistics for Windows (version 21.0). After determination of the data distribution, a one-way ANOVA or a Kruskal–Wallis test was performed. No data were excluded. A general linearized model used to analyze weight data. Multiple comparisons were made by Tukey or Bonferroni post hoc tests. All data are represented as mean ± standard deviation, and 95% confidence interval (CI) and P < 0.05 was considered significant.

Mice in each experimental group were weighed weekly (Figure 1A). Time, treatments, and interaction of time and treatments affected mouse weights significantly. Mouse weight was significantly affected by the various treatments (P < 0.001 and η_p² = 0.35 for effect of time, P < 0.001 and η_p² = 0.52 for effect of group, and P < 0.001 and η_p² = 0.44 for effect of interaction of group*time). Alloxan caused significant decrease in mouse weight in the first week, followed by a significant increase in weight in the third and fourth weeks compared with saline vehicle control (Figure 1A). Atorvastatin did not affect mouse weights significantly. Liraglutide significantly inhibited the alloxan-induced weight increase (P < 0.001, Figure 1A). We found no significant difference in the weight of mice between atorvastatin treatment alone and the vehicle control groups (P > 0.05, Figure 1A). Atorvastatin did not affect blood glucose levels at any time point (F_4,10 = 0.80, P = 0.55, Figure 1B). Liraglutide decreased blood glucose levels after 0.5 h (F_4,10 = 11.7, P < 0.001, Figure 1B). Atorvastatin treatment alone did not significantly affect blood glucose levels (F_4,10 = 0.80, P > 0.99, Figure 1B). We found no significant changes in locomotor activity after diabetes in any drug treatment group (F_4,30 = 2.62, P = 0.81, Figure 1C).

Figure 1

We found no significant differences in initial latencies between the treatment groups (F_4,30 = 0.89, P = 0.89, Figure 2A). On the second day, alloxan-induced diabetes significantly decreased the retention time (79.40 ± 30.25 s) compared with the vehicle control (270.75 ± 66.74 s; F_4,30 = 38.0, P < 0.001, Figure 2B). Atorvastatin significantly attenuated the decrease (140.30 ± 14.33 s; F_4,30 = 38.0, P = 0.02, Figure 2B). Liraglutide also significantly attenuated the decrease (197.50 ± 9.73 s; F_4,30 = 38.0, P < 0.001, Figure 2B). Atorvastatin treatment alone had no significant effect on the latency (288.00 ± 61.30 s) compared with control (270.75 ± 66.74 s; F_4,30 = 38.0, P > 0.05, Figure 2B).

Figure 2

For each experimental group, time spent in the dark compartment was measured (Figure 3). Alloxan-induced diabetes significantly increased the time mice spent in the dark compartment (144.83 ± 29.19 s) compared with the vehicle control group (52.83 ± 13.45 s, F_4,30 = 53.9, P < 0.001). Atorvastatin significantly decreased the time mice with alloxan-induced diabetes spent in the dark compartment (112.50 ± 13.01 s) compared with mice in the diabetes group without atorvastatin treatment (F_4,30 = 53.9, P = 0.046, Figure 3). Liraglutide significantly reduced the time mice with alloxan-induced diabetes spent in the dark compartment compared with untreated mice with alloxan-induced diabetes (80.66 ± 11.00 s, F_4,30 = 53.9, P < 0.001, Figure 3). Atorvastatin treatment alone had no significant effect on the time mice spent in the dark compartment (54.4 ± 9.45 s) compared with the control group (52.83 ± 13.45 s, F_4,30 = 53.9, P > 0.05).

Figure 3