Fructose-induced synaptic and neuronal adaptations at neuropeptide Y/agouti-related peptide neurons

Abstract

Fructose is a naturally-occurring sugar, consumed in excess as sweeteners, and is linked to the development of obesity. Fructose is consumed with glucose (dextrose) in added sugars, but while dextrose produces satiety, excessive fructose intake promotes hyperphagia through the brain. However, the neurological effects of dietary fructose are not clearly defined. We fed male and female mice standard chow, a 60% high fructose, or 60% high dextrose diet and found that fructose- and dextrose-fed mice ate more calories and gained more body fat despite increasing fat oxidation and energy expenditure. Furthermore, their metabolic syndromes were more prominent in male mice, who also developed glucose intolerance. To define the neurological effects underlying the obesogenic actions of fructose, we performed ex vivo patch-clamp recordings from orexigenic Neuropeptide Y/agouti-related peptide (NPY/AgRP) neurons in the arcuate nucleus. Fructose feeding uniquely increased synaptic excitation at NPY/AgRP neurons, which remained elevated with sustained fructose exposure; this excitation may arise from glutamatergic neurons in the dorsomedial hypothalamic nucleus. Terminating fructose feeding reversed this synaptic excitation at male but not female NPY/AgRP neurons. Furthermore, chronic but not acute fructose feeding in male mice also irreversibly activated NPY/AgRP neurons even following fructose withdrawal. Interestingly, despite sex-dependent fructose-mediated plasticity at NPY/AgRP neurons, a prolonged fructose withdrawal increased innate fructose preference in both male and female mice. Taken together, these findings showed that fructose elicited synaptic and neuronal excitation at NPY/AgRP neurons that can be long-lasting. These actions are consistent with that seen during hunger and may thus promote hyperphagia in the expression of fructose-mediated obesity.

Keywords: Electrophysiology; Fructose; Hypothalamus; Obesity; Plasticity; Sugar.

Graphical abstract

Figure 1

Fructose and dextrose feeding promoted hyperphagia and body fat accumulation. Calorie intake (A), body weight (B), and body fat percentage (C) of chow- (CD; black), 60% fructose- (FrD; green), and 60% dextrose (DxD; orange)-fed male (i) and female mice (ii). Data are presented as mean ± SEM. Diet and sex comparisons from repeated measures three-way ANOVA: ˆ, p < 0.06; ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Figure 2

Fructose and dextrose feeding promoted glucose intolerance in male mice only. Oxygen consumption (A), respiratory exchange ratio (RER; B), and blood glucose levels in response to glucose challenge (C) of chow- (CD; black), 60% fructose- (FrD; green), and 60% dextrose (DxD; orange)-fed male (i) and female mice (ii). Data are presented as mean ± SEM. Diet and sex comparisons from repeated measures three-way ANOVA: ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Figure 3

Acutefructose feeding did not alterthe excitability of NPY/AgRP neurons. Whole-cell recordings from acute brain slices (i) containing fluorescently labeled NPY-hrGFP neurons (ii; A). Representative sample traces of action potential firing elicited by a 3-s current step to determine the rheobase (B) from male (C) and female mice (D) after a 1-week feeding of chow (CD; black), 60% fructose (FrD; green), and 60% dextrose diet (DxD; orange). Input resistance of NPY-hrGFP neurons from male (E) and female mice (F) fed CD, FrD, or DxD for one week. Data are presented as mean ± SEM. Diet and sex comparisons from two-way ANOVA. Scale: 100 μm (Ai), 10 μm (Aii). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Figure 4

Rapid synaptic excitation at NPY/AgRP neurons following fructose intake. Representative sample traces of spontaneous excitatory postsynaptic current (sEPSC) events at NPY-hrGFP neurons (A). sEPSC frequency (B, C) and amplitude (D, E) from male and female mice after 1-week of chow (CD), 60% fructose (FrD), and 60% dextrose (DxD) feeding. Representative sample traces of spontaneous inhibitory postsynaptic current (sIPSC) events at NPY-hrGFP neurons (F). sIPSC frequency (G, H) and amplitude (I, J) from male and female mice fed CD, FrD, and DxD for 1 week. Data are presented as mean ± SEM. Diet and sex comparisons from two-way ANOVA with Bonferroni post-test, where applicable: ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001.

Figure 5

Persistent synaptic excitation at NPY/AgRP neurons with fructose intake. Rheobase (A, B) and input resistance (C, D) at the soma of NPY/AgRP neurons, and sEPSC frequency (E, F) and amplitude (G, H) from afferent input to NPY-hrGFP neurons from male and female mice fed chow (CD), 60% fructose (FrD), and 60% dextrose (DxD) for 4 weeks. Data are presented as mean ± SEM. Diet and sex comparisons from two-way ANOVA with Bonferroni post-test: ^, p < 0.07; ∗, p < 0.05; ∗∗∗, p < 0.001.

Figure 6

Resistance to diurnal variation and distal origins of fructose-mediated synaptic excitation at NPY/AgRP neurons. Relationship between sEPSC frequency detected at male and female NPY/AgRP neurons and time that chow- (CD; gray circles) and fructose (FrD; light green circles)-fed NPY-hrGFP mice were sacrificed (A). Time was expressed as Zeitgeber time (ZT) where ZT 0–12 marks the light cycle period. Lines of the correlation plot represent simple linear regressions among CD- (black line) and FrD-fed (green line) mice, and numerals represent bivariate Pearson correlations (r): ∗, p < 0.05 (A). Schematic of freshly prepared coronal brain sections from fructose-naïve NPY-hrGFP mice incubated in bath ACSF or ACSF containing fructose for 90 min (B). sEPSC frequency (C) and amplitude (D) following treatment with 0.5 mM and 10 mM fructose. Data are presented as mean ± SEM. Drug comparisons from two-way ANOVA (C, D). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Figure 7

Fructose withdrawal reversed synaptic excitation at male but not female NPY/AgRP neurons. Rheobase (A, B), input resistance (C, D), sEPSC frequency (E, F), and sEPSC amplitude (G, H) from male and female mice fed chow for five weeks (CD; black), 60% fructose diet for five weeks (FrD; green), or fructose for four weeks and returned to chow for one week (Rev; striped). Rheobase (I, J), input resistance (K, L), sEPSC frequency (M, N), and sEPSC amplitude (O, P) from male and female mice fed CD for eight weeks, FrD for eight weeks, or FrD for four weeks and CD for four weeks. Diet and sex comparisons from two-way ANOVA with Bonferroni post-hoc testing: ˆ, p < 0.08; ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Figure 8

Fructose feeding increased neuronal activation at hypothalamic regions distal to NPY/AgRP neurons. Schematic of known hypothalamic network contributing to the orexigenic actions of NPY/AgRP neurons. Colocalization (white arrowheads; i) of FosB immunoreactivity (ii) with native fluorescence of NPY-hrGFP neurons following chow (CD) or 60% fructose (FrD) feeding (iii; A). Parcellations (dashed white lines) from stitched photomicrographs of NeuroTrace-labeled cells (i) defined the regions of FosB-immunoreactive cells (ii) to compare the number of FosB cells (iii) within the paraventricular hypothalamic nucleus (PVH; B), dorsomedial hypothalamic nucleus (DMH; C), and suprachiasmatic nucleus (SCH; D) outlined (dashed yellow area) from CD- and FrD-fed mice. Data are presented as mean ± SEM. Diet comparisons from unpaired t test: ∗, p < 0.05. Scale: 100 μm (A, Bii–Dii), 400 μm (Bi–Di). ARH, arcuate hypothalamic nucleus; DMH, dorsomedial hypothalamic nucleus; fx, fornix; opt, optic tract; PVH, paraventricular hypothalamic nucleus; subparaventricular zone, SCH, suprachiasmatic nucleus; V3, third ventricle; VMH, ventromedial hypothalamic nucleus.

Figure 9

Prolonged fructose withdrawal increased innate fructose preference. Fructose preference, expressed as calories (kcal) consumed from a high fructose diet (FrD) as a percentage of total caloric intake when given the choice of chow and FrD before (Pre) and after (Post) chow feeding for five weeks (CD; black), fructose feeding for five weeks (FrD; green), or fructose feeding for four weeks followed by a 1-week abstinence period on chow (Rev; striped) in male (A) and female mice (B). Comparison of fructose preference before (Pre) and after (Post) eight weeks of CD, eight weeks of FrD, or 4 weeks of FrD and 4 weeks of CD in male (C) and female mice (D). Comparison fructose preference from repeated measures three-way ANOVA with Tukey post-test: ˆ, p < 0.06; ∗∗, p < 0.01; ∗∗∗, p < 0.001. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)