The Effect of Dietary Fat and Sucrose on Cognitive Functioning in Mice Lacking Insulin Signaling in Neuropeptide Y Neurons

Abstract

Obesogenic diets can produce hippocampal insulin resistance and impairments to hippocampal-dependent cognition. This study investigated the effect of disrupted insulin signaling in Neuropeptide Y (NPY) neurons on diet-induced deficits in hippocampal-dependent memory. Wild-type mice and mice that had a targeted knockout of insulin receptors on NPY cells (IR^lox/lox;NPY^Cre/+) were given ad libitum access to a high-fat diet (high fat; HF), 10% sucrose solution (high sugar; HS), both high-fat diet and sucrose solution (high fat, high sugar; HFHS), or a normal fat control chow for 12 weeks. Mice were tested in the Morris Water Maze (MWM), a hippocampal-dependent spatial memory task. Glucose homeostasis was assessed via a glucose tolerance test. Independent of genotype, consumption of HF, but not HS, diet increased energy intake, body weight, and plasma leptin, and impaired glucose tolerance. Disrupted insulin signaling in NPY cells and dietary interventions did not significantly affect the ability of mice to learn the location of the platform in the MWM. However, for IR^lox/lox control mice, consumption of HF, but not HS, diet resulted in reduced time spent in the target quadrant during the probe trial, suggesting a hippocampal-dependent memory deficit. IR^lox/lox;NPY^Cre/+ mice had poor performance in the probe trial regardless of diet, suggesting a floor effect. This study did not find adverse effects of chronic sucrose intake on metabolic outcomes or hippocampal-dependent memory. These data also suggest that the effects of HF diet on hippocampal-dependent memory may be dependent on insulin signaling in hippocampal NPY cells.

Keywords: NPY (neuropeptide Y); diet; hippocampus; morris water maze; spatial cognition.

FIGURE 1

Body weight, energy intake and plasma leptin. Body weights of IR^lox/lox and IR^lox/lox;NPY^Cre/+ mice increased after 12 weeks of diet intervention in all treatment groups. Body weight was further upregulated after 12 weeks of diet in HF and HFHS groups compared to CON and HS diet group (A, B). Energy intake of IR^lox/lox and IR^lox/lox;NPY^Cre/+ mice was higher in HF and HFHS treatment groups compared to CON and HS diet group (C, D). Plasma leptin recordings of IR^lox/lox and IR^lox/lox;NPY^Cre/+ mice were higher in HF and HFHS treatment groups compared to CON and HS group. (E, F). CON = chow diet, HF = high fat diet, HS = high sugar diet, HFHS = high fat and high sugar diet. Values are expressed as mean ± SEM. * = p < 0.05. Analysed by a mixed ANOVA for body weight and a factorial ANOVA for energy intake and plasma leptin. IR^lox/lox; n = 48 (12/diet group). IR^lox/lox;NPY^Cre/+; n = 48 (12/diet group).

FIGURE 2

Blood glucose and insulin. Blood glucose levels in IR^lox/lox control mice in response to peripheral glucose injection over 120 min. HF and HFHS treatment groups displayed reduced glucose tolerance in both IR^lox/lox and IR^lox/lox;NPY^Cre/+ mice compared to CON and HS groups. Insets represent arbitrary units (AU) for area under the curve (AUC) values in IR^lox/lox and IR^lox/lox;NPY^Cre/+ mice. AUC results show a main effect of diet, with HF and HFHS treatment groups displaying increased AUC following peripheral glucose injection compared to CON and HS diet groups (A, C). There was a within-subjects effect of time on insulin sensitivity within all dietary treatment groups in both IR^lox/lox and IR^lox/lox;NPY^Cre/+ mice (B, D). CON = chow diet, HF = high fat diet, HS = high sugar diet, HFHS = high fat and high sugar diet. Values are expressed as mean ± SEM. + = main effect of time, * = p < 0.05. Analysed by a mixedANOVA (A–D) and factorial ANOVA (insets of A, C) followed by Tukey’s honest significance difference (HSD) test. IR^lox/lox; n = 48 (12/diet group). IR^lox/lox;NPY^Cre/+; n = 48 (12/diet group).

FIGURE 3

Morris Water Maze performance–Mean Escape Latencies. Mean escape latencies were collected for each trial day to assess performance over time. Across all dietary treatment groups, mean escape latencies were significantly decreased on Day 2 of visible platform trial for IR^lox/lox control mice and IR^lox/lox;NPY^Cre/+ mice (A, B). IR^lox/lox mice and IR^lox/lox;NPY^Cre/+ mice reached the platform faster across the hidden platform days in all treatment groups (C, D). CON = chow diet, HF = high fat diet, HS = high sugar diet, HFHS = high fat and high sugar diet. Values are expressed as mean ± SEM. + = main effect of time, p < 0.05. Analysed by a mixed ANOVA followed by Tukey’s honest significance difference (HSD) test. IR^lox/lox; n = 48 (12/diet group). IR^lox/lox;NPY^Cre/+; n = 48 (12/diet group).

FIGURE 4

Morris Water Maze Performance - Probe trial. Time spent in the target quadrant was measured to assess hippocampal-dependent memory. IR^lox/lox control mice and IR^lox/lox;NPY^Cre/+ mice across all dietary groups swam similar path lengths in the target quadrant during the probe trial (A, B). IR^lox/lox mice fed a HF or HFHS diet spent less time in the target quadrant relative to IR^lox/lox mice fed a CON or HS diet (C). There was a main effect of genotype on time spent in the target quadrant, with IR^lox/lox;NPY^Cre/+ mice in all dietary treatment groups spending less time in the target quadrant than IR^lox/lox CON mice (D). CON = chow diet, HF = high fat diet, HS = high sugar diet, HFHS = high fat and high sugar diet. Values are expressed as mean ± SEM. * = p < 0.05. Analysed by a factorial ANOVA. IR^lox/lox; n = 48 (12/diet group). IR^lox/lox;NPY^Cre/+; n = 48 (12/diet group).