Autism-Like Behaviours and Memory Deficits Result from a Western Diet in Mice

Abstract

Nonalcoholic fatty liver disease, induced by a Western diet (WD), evokes central and peripheral inflammation that is accompanied by altered emotionality. These changes can be associated with abnormalities in social behaviour, hippocampus-dependent cognitive functions, and metabolism. Female C57BL/6J mice were fed with a regular chow or with a WD containing 0.2% of cholesterol and 21% of saturated fat for three weeks. WD-treated mice exhibited increased social avoidance, crawl-over and digging behaviours, decreased body-body contacts, and hyperlocomotion. The WD-fed group also displayed deficits in hippocampal-dependent performance such as contextual memory in a fear conditioning and pellet displacement paradigms. A reduction in glucose tolerance and elevated levels of serum cholesterol and leptin were also associated with the WD. The peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PPARGC1a) mRNA, a marker of mitochondrial activity, was decreased in the prefrontal cortex, hippocampus, hypothalamus, and dorsal raphe, suggesting suppressed brain mitochondrial functions, but not in the liver. This is the first report to show that a WD can profoundly suppress social interactions and induce dominant-like behaviours in naïve adult mice. The spectrum of behaviours that were found to be induced are reminiscent of symptoms associated with autism, and, if paralleled in humans, suggest that a WD might exacerbate autism spectrum disorder.

Figure 1

Experimental design. (a) Composition of the diets and percentage of total calories. Food intake (b) and body weight (c) normalized to control on day 21 of a dietary challenge (control: control diet, WD: high-cholesterol diet; ^∗p < 0.05 versus that in the control diet, t-test). (d) Study of the effects of cholesterol-enriched diet exposure on mouse behaviour in fear conditioning test, home cage, and food competition test. (e) Study of the effects of cholesterol-enriched diet exposure on mouse performance in pellet displacement tube test, glucose tolerance, and blood biochemical parameters. (f) Study of the effects of cholesterol-enriched diet exposure on gene expression in the brain and liver.

Figure 2

Dietary challenge with cholesterol results in aberrant home cage social behaviour. In comparison to control mice, the dietary-challenged group displayed (a) a significant decrease in the duration of group huddle, (b) a significant prolongation of the duration of “sitting alone” behaviour, and (c) a significant increase in the time spent in motion in the cage (^∗p < 0.05 versus that in the control group, t-test). As compared to control animals, a group fed with the cholesterol-enriched diet showed a significant elevation of (d) the percentage of mice displaying digging behaviour (^∗p < 0.05 versus that in the control group, Fisher's exact test) and (e) the duration of digging behaviour (^∗p < 0.05 versus that in the control group, Mann-Whitney test). Control—standard diet, WD—high-cholesterol diet. Data are shown as mean ± SEMs.

Figure 3

The high-cholesterol diet potentiates dominant-like behaviour and suppresses sociability in a food competition test. In comparison to the control group, mice housed on the high-cholesterol diet showed (a) a significant decrease in the latency to crawl-over behaviour and (b) a significant increase in a number of crawl-overs, as well as (c) prolonged total duration of crawl-over behaviour (^∗p < 0.05 versus that in the control group; Mann-Whitney test). In comparison to the control group, mice exposed to the cholesterol-enriched diet had (d) no significant changes in the latency to body-body contacts and had (e) a significant increase in a number of body-body contacts (^∗p < 0.05 versus that in the control group; Mann-Whitney test) and (f) a significant decrease in the duration of body-body contacts (^∗p < 0.05 versus that in the control group; Mann-Whitney test). No significant group differences were observed between the groups in the parameters of nose-anal contacts: (g) the number of nose-anal contacts, (h) latency to nose-anal contacts, and (i) total duration of nose-anal contacts (^∗p > 0.05 versus that in the control group; Mann-Whitney test). There were no significant group differences in the parameters of nose-nose contacts: (j) latency of nose-nose contacts, (k) number of nose-nose contacts, and (l) total duration of nose-nose contacts (^∗p > 0.05 versus that in the control group; Mann-Whitney test). Control—standard diet, WD—high-cholesterol diet. Data are shown as mean ± SEMs.

Figure 4

The cholesterol-enriched diet enhances horizontal and vertical activity during food competition test. In comparison to control mice, animals exposed to the cholesterol-enriched diet showed (a) a significant increase in the total number of line crossing (^∗p < 0.05 versus that in the control group, t-test), (b) significant elevation of the total number of rearings (^∗p < 0.05 versus that in the control group, Mann-Whitney test), (c) and a significant decrease in the latency of rearings (^∗p < 0.05 versus that in the control group, t-test). Control—standard diet, WD—high-cholesterol diet. Data are shown as mean ± SEM (a, c) and median with interquartile range (b).

Figure 5

The high-cholesterol diet compromises the hippocampus-dependent performance. In the fear conditioning test, in comparison to control mice, the dietary-challenged group showed (a) significant decreased number of “good learners” (^∗p < 0.05 versus that in the control group, Fisher's exact test), (b) a strong tendency to a reduced time spent with freezing during recall of conditioning (p = 0.055 versus that in the control group, t-test), and (c) a significant decrease in time spent with freezing in a memory extinction protocol (^∗p < 0.05 versus that in the control group, t-test). In the pellet displacement tube test, in comparison to control animals, mice exposed to the cholesterol-enriched diet showed significantly prolonged (d) latency of a displacement of the 1st pellet and (e) the duration of displacement of 50% pellets (^∗p < 0.05 versus that in the control group, t-test) and (f) did not differ in the duration of a displacement of 100% pellets (p > 0.05 versus that in the control group, t-test). Control—standard diet, WD—high-cholesterol diet. Data are shown as mean ± SEM.

Figure 6

Effects of high-cholesterol diet exposure on glucose tolerance, biochemical blood parameters, and PPARGC1a gene expression. (a) In comparison to that in the control, there was an increase in blood glucose level at 15 and 30 min after glucose load in the glucose tolerance test (^∗p < 0.05 versus that in the control group, 2-way ANOVA and Bonferroni post hoc test). (b) There was a significant increase in the area under curve in glucose concentration in dietary-challenged mice in comparison to controls (^∗p < 0.05 versus control group, t-test). (c) There was no significant difference in basal glucose levels between mice housed on the standard and high-cholesterol diets. Mice housed on the high-cholesterol diet, as compared to the control group, showed significantly increased (d) blood leptin levels, (e) blood cholesterol levels (^∗p < 0.05 versus that in the control group, t-test), and (f) unaltered blood level of triglycerides (p > 0.05 versus that in the control group, t-test). (f) In comparison to control mice, animals fed with the cholesterol-enriched diet had significantly decreased PPARGC1a mRNA in all brain areas (^∗p < 0.05 versus that in the control group, 2-way ANOVA and Bonferroni post hoc test) but not in the liver (p > 0.05 versus that in the control group, 2-way ANOVA and Bonferroni post hoc test). Control—standard diet, WD—high-cholesterol diet. Data are shown as mean ± SEM.