High-fat diet intake induces neuroinflammation and alters the dopaminergic system in the amygdala of juvenile rats

Abstract

High-fat diet (HFD) intake is increasingly common during adolescence and is associated with neuroinflammation and amygdala-dependent behavioral alterations. Because the amygdala continues to mature throughout adolescence, this period represents a window of vulnerability to dietary and inflammatory challenges. In addition to its role in stress responses, the amygdala contributes to appetitive behaviors and receives monoaminergic inputs modulating feeding-related processes. Although HFD-induced amygdalar neuroinflammation may alter these functions, how its effects differ across developmental stages remains poorly understood. We compared the effects of 4-week HFD exposure in juvenile and adult rats (jHFD and aHFD). Both groups exhibited moderate hyperphagia. In juveniles, HFD decreased circulating corticosterone and glucocorticoid-receptor immunoreactivity in basolateral amygdala (BLA), which may reduce glucocorticoid-mediated regulation of inflammation and contribute to heightened neuroinflammatory response. Accordingly, jHFD rats showed elevated circulating IL-1β, increased amygdalar proinflammatory cytokine levels, and increased Iba-1-positive myeloid reactivity in BLA. Moreover, jHFD rats displayed reduced dopamine turnover and increased D2R immunoreactivity in BLA. In adults, HFD was associated with glial morphological alterations without the molecular profile of active neuroinflammation observed in juveniles. Thus, juvenile rats exhibited greater susceptibility to HFD-induced alterations in amygdalar neuroinflammatory and dopaminergic signaling. Despite these neurobiological differences, both age groups developed comparable hyperphagia, suggesting that these amygdalar alterations may not be sufficient to account for increased food intake under these conditions. These findings highlight the neuroimmune vulnerability associated with HFD exposure during development and reveal age-dependent effects of HFD on amygdalar function during adolescence, a period of heightened vulnerability to emotional related disorders.

1. Introduction

Adolescence is a period of profound biological and psychological transformation that can have lasting consequences for adult health. During this critical developmental window, the amygdala undergoes substantial maturation, including the refinement of its connections with other brain regions, which may increase its vulnerability to environmental challenges (Saygin et al., 2015; Selleck et al., 2018). Importantly, in humans, adolescence is characterized by increased consumption of palatable, energy-dense foods, including high-fat foods (Jair Vidal et al., 2018; Tsan et al., 2021). This dietary pattern is particularly relevant because exposure to an obesogenic diet during periadolescence can induce persistent neurobehavioral alterations and increase susceptibility to psychiatric disorders later in life (Rocha-Gomes et al., 2026; Vega-Torres et al., 2018).

High-fat diet (HFD) intake is associated with neuroinflammation in multiple brain regions (Miller and Spencer, 2014; Sharma, 2021). Interestingly, even in the absence of increased body weight, saturated fatty acids promote neuroinflammation, a condition characterized by an increase in the central levels of proinflammatory cytokines, mainly interleukin 1-beta (IL-1β), tumor necrosis factor-alpha (TNF-α) and interleukin 6 (IL-6), which are produced by reactive glial cells (Guillemot-Legris et al., 2016; Miller and Spencer, 2014). Brain regions involved in feeding and emotional regulation, including the amygdala, are also susceptible to these HFD-induced inflammatory changes (Almeida-Suhett et al., 2017; Guillemot-Legris and Muccioli, 2017). However, most studies examining the functional consequences of amygdalar neuroinflammation have focused on aversive memory formation (Ziemens et al., 2022), leaving its potential impact on feeding-related processes comparatively unexplored.

The amygdala is increasingly recognized as an important component of the neural circuitry underlying food motivation and reward. In particular, the basolateral amygdala (BLA) contributes to the motivational and learning processes that influence food-related decision making (Keefer et al., 2021; Phillips et al., 2003). BLA projections to the lateral hypothalamus are involved in promoting feeding in the absence of homeostatic need (Petrovich et al., 2002; Sun et al., 2015), while activation of the BLA has also been associated with food addiction in humans (Pursey et al., 2019). Thus, alterations in BLA function may have important consequences for the regulation of food intake beyond homeostatic control.

In addition to neuroinflammation, HFD exposure has been associated with alterations in amygdalar monoaminergic signaling, including dopamine (DA), serotonin (5-HT), and noradrenaline (NA), which have been linked to behavioral and aversive memory changes in rodents (Akter et al., 2020; Osorio-Gómez et al., 2024). Because monoaminergic signaling within the amygdala contributes to the regulation of feeding-related behaviors, these alterations may influence susceptibility to overeating (Coker et al., 2021; Marcos et al., 2023). Importantly, amygdalar monoaminergic and inflammatory processes are also closely related to stress-regulatory mechanisms. Juvenile HFD exposure alters physiological and behavioral responses to stress, while the amygdala participates in the regulation of hypothalamic–pituitary–adrenal (HPA) axis activity (Boukouvalas et al., 2008, Norden et al., 2016). Palatable food can enhance BLA activity through glucocorticoid receptor (GR)-dependent mechanisms (Egan and Ulrich-Lai, 2015), and glucocorticoid-mediated BLA activation contributes to enhanced aversive memory following periadolescent HFD exposure (Boitard et al., 2015; Naneix et al., 2021). These effects are accompanied by increased expression of local inflammatory markers, raising the possibility that altered interactions between stress signaling and neuroinflammation contribute to the persistent consequences of HFD exposure during development (Spencer et al., 2017). Moreover, stress and drugs of abuse can increase DA release in reward-related regions, including the amygdala, highlighting the potential interaction between stress-related and dopaminergic signaling in this region (De Jong and De Kloet, 2004; Del Arco et al., 2015; Inglis and Moghaddam, 1999).

Despite growing evidence that HFD consumption alters amygdalar inflammatory, stress-related, and monoaminergic processes, it remains unclear whether these effects differ across developmental stages and whether they are associated with altered feeding behavior. Determining the extent to which the juvenile amygdala is particularly susceptible to HFD-induced neuroinflammatory and dopaminergic alterations may provide insight into the mechanisms through which early-life exposure to an obesogenic diet affects brain function.

2. Materials and methods

2.1. Animals

Procedures were conducted following approval from the local Ethics Committee on Animal Experimentation of the National Institute of Psychiatry (INPRFM, CEI/C/004/2020) and in compliance with the Mexican Official Norm NOM-062-ZOO-1999.

Twenty juvenile (j, 3 weeks old) and 20 adult (a, 10 weeks old) male Wistar rats (INPRFM vivarium) were housed 2-3 rats/cage and maintained under controlled temperature (23 ± 1°C) and light (lights on 7:00-19:00 h) conditions for 1 week prior to experimentation. Rats were divided into control (jC, aC) or HFD (jHFD, aHFD) groups (n = 10/group). jHFD and aHFD animals were fed ad libitum with a commercial diet containing 60% Kcal from fat, 20% Kcal from protein and 20% of Kcal from carbohydrates (D12492; energy density: 5.2 kcal/g) whereas control animals were offered a diet containing 10% Kcal from fat, 20% Kcal from protein and 70% Kcal of carbohydrates, matched in sucrose to the HFD (D12450J; energy density of 3.8 kcal/g) (Research Diets, New Brunswick, NJ). Body weight (b.w.) was registered weekly, and food intake was quantified 3 days/week. We normalized food intake of animals by estimating the amount of Kcal consumed per Kg of b.w.

Animals were euthanized after 4 weeks of dietary manipulations, at this time juvenile rats were 7 weeks old, which is considered as the adolescent period in rats (Crews et al., 2007). To collect perfused brains, 4 animals/group were anesthetized with a mixture of ketamine (Anesket, 1000 mg/ml Pisa Salud Animal, México, 100 mg/kg b.w.) and xylazine (Procin, 20 mg/mL Pisa Salud Animal, México, 10 mg/kg b.w.) and perfused via the ascending aorta with 100 mL ice-cold 0.9% NaCl followed by 500 mL of ice-cold 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS, pH 7.4), delivered via a peristaltic pump (Masterflex L/S MasterSense Pump, Barrington, IL) over 20 min. Brains were removed and postfixed in the same fixative overnight + 15% sucrose, then transferred to a 30% sucrose solution in 0.1 M PBS for 48 h at 4°C. Brains were sectioned into 30 μm slices on a freezing microtome (Leica Biosystems, SM2010 R, Deer Park, IL); tissue was obtained in a 1-in-4 series and maintained in a cryoprotectant solution (0.05 M PBS, 30% ethylene glycol, 20% glycerol) at −20°C until processing for immunofluorescence analyses.

The remaining animals (n = 6/group) were euthanized by decapitation, brains were immediately removed, frozen in dry ice and stored at −70°C until analyzed. Trunk blood was collected and centrifuged at 3000 g for 30 min at 4°C. Sera were aliquoted and stored at −70°C and adrenal glands were dissected and weighed. Frozen brains were hand dissected to obtain the amygdaloid nuclei (comprising the BLA, medial and central areas) between −1.92 and −3.36 mm from bregma (Paxinos and Watson, 2013). Tissue was obtained from each hemisphere with a 2 mm inner diameter sample corer; a subset of the brains was used for HPLC (n = 6/group; right hemispheres) and flow cytometry analyses (n = 5/group; left hemispheres). Animal euthanasia was planned at the midpoint of the light phase (12:00-13:00h) to control for possible circadian effects on corticosterone levels and serum cytokines (Fig. 1).

Fig. 1.

Experimental design. (1) Groups and timeline. Juvenile (3 weeks) or adult (10 weeks old) male rats were fed with a control (C) or a high-fat diet (HFD) for four weeks. (2) Euthanasia was performed after 4 weeks of experimentation by either decapitation or transcardial perfusion. Brains were dissected to obtain the amygdala and analyze (3) dopamine (DA), dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), serotonin (5-HT), 5-hydroxyindoleacetic acid (5-HIAA) and noradrenaline (NA) levels using HPLC; and (4) Monocyte chemoattractant protein-1 (MCP-1), interferon-γ (IFN-γ), tumor necrosis factor-α (TNF-α), interleukin-1α (IL-1 α), interleukin-1β (IL-1 β), and interleukin-6 (IL-6) levels in both amygdala and serum by flow cytometry. Brain coronal sections from perfused animals were used to perform (5) immunofluorescence for identifying glucocorticoid receptor (GR), dopamine receptor type 2 (D2R), IL-1β, glial fibrillary acidic protein (GFAP) and ionized calcium-binding adaptor protein-1 (Iba-1) expression and glial morphology in the basolateral amygdala (BLA). (6) Corticosterone concentration was measured in serum using an ELISA kit. Image created with BioRender.

2.2. Corticosterone ELISA

Levels of corticosterone (ng/mL) were measured in 20 μL of serum using a commercial ELISA kit (Enzo Life Sciences Farmingdale, NY, cat. ADI-900-097), according to manufacturer's instructions. A μQuant microplate reader (BioTek Instrument Inc, Winooski, VT) and the BioTek's Gen5 Data Analysis Software (BioTek ELx800, Fisher HealthCare) were used to determine corticosterone levels. Euthanasia of animals was planned at the midpoint of the light phase to control for possible circadian effects on corticosterone and serum cytokines' levels. Samples were run in duplicates.

2.3. Fluorescent immunohistochemistry

Free-floating brain sections (3 sections/animal) spanning representative and equivalent sections of the BLA (−1.92 to −2.04 mm from bregma) from each group (n = 4) were subjected to immunofluorescence (IF) in order to analyze the immunoreactivity of GR, IL-1β, type 2 dopamine receptor (D2R), ionized calcium-binding adaptor protein-1 (Iba-1, as a marker of microglia) or glial fibrillary acidic protein (GFAP, as a marker of astrocytes). Briefly, sections were washed in tris-buffer saline (TBS, pH 7.4), blocked for 2h at room temperature (TBS with 2% normal donkey serum and 0.3% Triton X-100) and incubated for 72 h at 4°C in blocking solution containing the primary antibodies (rabbit polyclonal to IL-1β 1:1K [Invitrogen Thermo Fisher Scientific, Waltham, MA, cat. PA5-88078], goat polyclonal to Iba-1 1:1K [Abcam, Cambridge, MA, cat. ab5076], mouse monoclonal to GFAP 1:2K [Invitrogen, cat. MA5-12023], rabbit polyclonal to D2R 1:1.5K [Invitrogen, cat. PA5-115141]) or rabbit polyclonal to GR 1:1K (kindly provided by Prof. Paul Sawchenko). Sections for IL-1β expression determination were incubated for 6h in biotinylated donkey anti-rabbit antisera 1:500 after primary antibody incubation (Jackson ImmunoResearch, West Grove, PA, cat. 711-067-003). After washing in TBS, sections were incubated overnight with fluorescent secondary antibodies (goat anti-mouse IgG Alexa Fluor 568 1:1K [Jackson ImmunoResearch, cat. A-11004], donkey anti-goat IgG Alexa Fluor 594 1:1K [Invitrogen, cat. A-11058], streptavidin Alexa Fluor 488 1:1K [Invitrogen, cat. S32354] or donkey anti rabbit Alexa 488 1:1K [Abcam, cat. ab150073]). Tissues were rinsed in TBS, counterstained with DAPI nuclear stain (Invitrogen, cat. D1306) in TBS for 5 min at room temperature, washed, mounted in gelatin-coated slides (Masterflex, cat. 069-CP) and coverslipped (0.1 M PBS, 50% glycerol).

2.4. Imaging and data analysis

Sections containing the BLA were acquired using a LSM900 Zeiss with Airyscan 2 confocal microscope (Carl Zeiss Microscopy Deutschland, Germany), photomicrographs were obtained using the 40× oil immersion objective with constant illumination settings and analyzed with semi-automated densitometry and a cell area (Ca)/projection area (Pa) ramification index to minimize operator bias using ImageJ software (NIH, version 1.51k). The photomicrographs were processed using a fixed background subtraction with a rolling ball radius of 50 pixels, then transformed to grayscale images and the threshold was manually set by converting every pixel into black or white. Relative staining intensities were semi-quantified in binary converted images, and the results transformed into percentage of positive area staining. To analyze glial morphology, Z-stack-maximum projection images were generated with a 0.5 μm step size.

2.4.1. Ramification index of microglia and astrocytes

During microglial response to a metabolic challenge, the cell body becomes hypertrophic, exhibiting enlarged size and retracted processes (deramification), to measure this morphological response we used the ramification index according to (Wittekindt et al., 2022).

Microglia or astrocyte cells (at least 7 cells/animal) were randomly selected by superimposing a rectangular grid with uniform distances between the lines in the directions X and Y and analyzing all cells crossing the grid lines. The ramification index was calculated as the ratio of the Ca and Pa. To measure Ca, the Iba-1 or GFAP-positive area of an individual cell was outlined using the polygon tool. To measure Pa, a Convex Hull Analysis was performed; this measures the size of the cellular field by considering a branched structure as a solid object controlling for a given amount of physical space.

Quantification of the number of GR, Iba-1 and GFAP positive cells were manually counted using the multi-point tool and divided by area, thus results are expressed as the number of cells/mm².

2.5. Proinflammatory cytokines measurements using flow cytometry

Left hemisphere amygdala punches were generated with a fixed 2 mm corer and homogenized in 120 μl radioimmuno-precipitation assay (RIPA) lysis buffer (25 mM Tris−HCl pH 7.6, 150 mM NaCl, 5 mM EDTA, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS) containing 1:100 protease inhibitor (complete mini; Roche Merck-Millipore, Burlington, MA) to standardize input. Samples were maintained on a shaker for 2 h at 4°C, then centrifuged at 14,800 g at 4°C for 20 min and the supernatant was obtained for subsequent analyses.

Sera and brain supernatants were defrosted on ice, spun at 600 g and immediately processed using a custom LEGENDplex rat inflammation panel, containing the Rat IFN-γ Capture Bead A5, 13X (Biolegend, San Diego, CA, Cat. 740253), Rat CCL2 (MCP-1) (Biolegend, Cat. 740255), Rat TNF-α Capture Bead A8 (Biolegend, Cat. 740256), Rat IL-1β Capture Bead B4 (Biolegend, Cat. 740260), Rat IL-1α Capture Bead B7 (Biolegend, Cat. 740263), and Rat IL-6 Capture Bead B9 (Biolegend, Cat. 740264), according to manufacturer's recommendations. Samples were immediately acquired on a BD® FACSCelesta flow cytometer fitter with 405-nm, 488-nm, and 633-nm lasers, operated through the BD® FACSDiva software v.8. Cytometer settings were validated as within the range of manufacturer's recommendations using CS&T beads (BD®, Franklin Lakes, NJ, cat. 642412) prior to each acquisition. Data were analyzed and cytokine concentrations were obtained using the Cloud-based LEGENDplex Data Analysis Software. Concentrations are expressed as pg/mL of lysate.

2.6. Neurotransmitters quantification using high-performance liquid chromatography (HPLC)

Measurements were performed as previously described (Puga et al., 2016). After weighing, the tissue was homogenized with a sonicator in 200 μL of ice-cold 0.1 M perchloric acid (Merck-Millipore); samples were left undisturbed for 10 min on ice and then centrifuged at 14,800 g for 10 min at 4°C. The supernatant was recovered and filtered for impurities (0.45-μm pore filters [Merck-Millipore]). Samples were analyzed by HPLC consisting of an autosampler (Waters Alliance 2695, Milford, MA) equipped with a Phenomenex® precolumn to filter impurities, an analytical Phenomenex C-18 column (150 × 1 mm with 5-μm silica particles), and a Waters 2465 Electrochemical Detector. The composition of the mobile phase was as previously reported (Patel et al., 2005). A stock buffer was prepared as follows: 25 mM sodium dihydrogen orthophosphate, 27 μM EDTA, 50 mM sodium citrate, 10 mM diethylamine, 10 mM NaCl, and 2 mM sodium decanesulfonate in 1 L of deionized, distilled water, and pH adjusted to 3.1 with H₃PO₄. Mobile phase: buffer 92:8 acetonitrile; flow rate: 0.1 mL/min. The working electrode potential was set at +750 mV vs an Ag/AgCl reference electrode. The sensitivity of the detector was set at 5 nA full-scale detection and the volume of injection was 5 μL at 35°C. Standard solutions were prepared from a 1 mg/mL standard stock solution and further diluted in ice-cold 0.1 M perchloric acid. Retention time was 3.2 min for dihydroxyphenylacetic (DOPAC) standards, 5 min for homovanillic acid (HVA) standards, 8.8 min for DA, 19.3 min for 5-HT, 4.2 min for 5-hydroxyindoleacetic acid (5-HIAA) and 3.6 min for NA standards. For data collection, Waters’ Empower software for use with HPLC was in place. DA and 5-HT turnover rates were calculated as ratios of metabolites to neurotransmitters. Concentrations are expressed as pg/mg wet tissue weight.

2.7. Statistical analysis

Data are expressed as the mean ± standard error of the mean (SEM). Data distribution was evaluated with the Shapiro–Wilk test. Datasets were examined for outliers using the ROUT method prior to statistical analysis. Variables were analyzed by two-way ANOVA (age [adult or juvenile] and type of diet [control or HFD] as the two factors). Body weight and food intake were analyzed by two-way repeated measures ANOVA. When P < 0.05, a Tukey post-hoc test was performed. The analyses were performed with GraphPad Prism version 8.0.2.

3. Results

3.1. HFD increased caloric intake but not body weight

To determine a possible differential energy balance in juveniles and adults following HFD exposure, we evaluated b.w. and caloric intake across experimental groups. B.w. increased in all groups over time, with no changes between groups. The weight gains were attributed to normal growth during the experiment. The two-way repeated measures ANOVA showed a significant effect of time for juveniles (F_(4,72) = 1328, P < 0.001) and adults (F _(4,72) = 98.63, P < 0.001, Fig. 2A and B).

Fig. 2.

Body weight and energy intake of juvenile and adult rats fed with HFD. (A) Body weight of juvenile and (B) adult rats, (C) caloric intake of juvenile and (D) adult rats fed with control (solid lines, circles) or high-fat diet (HFD, dashed lines, triangles) during 4 weeks. Data is presented as the mean ± SEM, expressed in grams of body weight or Kcal per Kg of body weight. *P < 0.05 between depicted groups. n = 10 rats/group.

Although the pattern differed over time, both HFD-fed groups exhibited an overall increase in energy intake throughout the experiment. The two-way repeated measures ANOVA of caloric intake in juvenile rats showed a significant effect of time (F_{(3, 54)} = 72.8, P < 0.001) and diet (F_{(1, 18)} = 4.8, P < 0.05). Post-hoc analyses revealed that jHFD consumed more Kcal than those on a control diet (jC) during the last week of the experiment (321.3 ± 9.5 vs. 285.5 ± 6.3 kcal/kg b.w., P < 0.05). In adult rats, there was an effect of time (F_(3,54) = 23.74, P < 0.001) and an interaction between time and diet (F_{(3, 54)} = 5.518, P < 0.01, Fig. 2C and D).

3.2. Animal age modulated the impact of HFD on stress-related parameters

Given the known interaction between diet and stress regulation, we assessed HPA axis-related parameters, including corticosterone levels and GR expression in the BLA.

Basal corticosterone levels were affected by HFD intake in juveniles only, with a reduction of more than two-fold in jHFD by vs. jC (55.5 ± 17 vs. 191.6 ± 52 ng/ml, P < 0.05), with no significant changes in adult rats. The two-way ANOVA showed a significant effect of diet (F_(1,17) = 7.2, P < 0.05) and interaction between diet and age (F_(1,17) = 4.9, P < 0.05) (Fig. 3A).

Fig. 3.

Effects of a high-fat diet intake on stress-related markers in juvenile and adult rats. (A) Serum corticosterone (CORT) concentration, expressed as log-transformed levels (ng/mL). (B) Density of glucocorticoid receptor (GR)-positive cells (cells/mm²) in the basolateral amygdala (BLA). For both (A) and (B), data are shown for juvenile and adult rats fed a control (white bars; individual values overlaid as circles) or a high-fat diet (HFD; hatched bars; individual values overlaid as triangles). (C–F) Representative photomicrographs of GR immunoreactivity in the BLA of juvenile and adult rats fed a control diet (C, E) or HFD (D, F). Scale bar: 50 μm. Data are presented as mean ± SEM. *P < 0.05 for comparisons between control and HFD within each age group. n = 4–5 rats per group.

GR expression in the BLA exhibited a significant 25% decrease in jHFD vs. jC (P < 0.05), while adult animals displayed no change. Two-way ANOVA showed a significant effect of diet (F_(1,12) = 13.5, P < 0.01) (Fig. 3B–F).

3.3. HFD intake was associated with age-dependent alterations in monoaminergic markers of the amygdala

To evaluate whether HFD affects reward-related neurocircuitry, we quantified monoamine levels and D2R immunoreactivity(ir) in the amygdala. DA levels in the amygdala were significantly affected by the diet (F_(1,20) = 8.7, P < 0.01) and by the interaction between diet and age (F_(1,20) = 16.5, P < 0.001). The jHFD group exhibited an approximately two-fold increase in DA levels than the jC group (22.9 ± 2.6 vs. 8.1 ± 0.9 pg/mg wet tissue, P < 0.001; Table 1). Additionally, in control animals, DA content was lower in juveniles compared to adults (jC: 8.1 ± 0.9 vs. aC: 17 ± 2.3 pg/mg wet tissue, P < 0.05, Table 1). DA turnover was also significantly affected by diet (F_(1,20) = 7.4, P < 0.05); jHFD rats showed a 70% reduction in DA turnover (decreased DOPAC/DA ratio) compared to age-matched control animals (0.0075 ± 0.002 vs. 0.028 ± 0.006, P < 0.05, Table 1).

Table 1.

Monoamine and metabolite levels in the amygdala of juvenile and adult rats.

	Juveniles		Adults
	Control	HFD	Control	HFD
DA	8.1 ± 0.9#	22 ± 2.6*	17 ± 2.3	14.7 ± 2.2
DOPAC	0.2 ± 0.05	0.18 ± 0.06	0.3 ± 0.04	0.25 ± 0.07
HVA	0.12 ± 0.08	0.32 ± 0.13	0.04 ± 0.008	0.07 ± 0.03
DOPAC/DA	0.03 ± 0.006	0.007 ± 0.002*	0.024 ± 0.005	0.02 ± 0.004
HVA/DA	0.013 ± 0.008	0.013 ± 0.005	0.004 ± 0.001	0.004 ± 0.001
NA	8.1 ± 0.7	9.3 ± 0.73	11.6 ± 1.2	9.9 ± 1.6
5-HT	20.9 ± 1.9	26 ± 1.8	27 ± 1.7	23.9 ± 2.1
5-HIAA	8.5 ± 0.63	9.7 ± 0.6	11.2 ± 1.7	9.8 ± 0.97
5-HIAA/5-HT	0.41 ± 0.02	0.38 ± 0.02	0.41 ± 0.05	0.41 ± 0.006

Table 1. Monoamine and metabolite levels in the amygdala of juvenile and adult rats fed with control or high-fat diet (HFD). Dopamine (DA), homovanillic acid (HVA), 3,4-dihydroxyphenylacetic acid (DOPAC), noradrenaline (NA), serotonin (5-HT), 5-hydroxyindoleacetic acid (5-HIAA). Data is expressed as the mean ± SEM in pg/mg of wet tissue of each monoamine or metabolite. Two-way repeated measures ANOVA, followed by Tukey's post hoc test. *P < 0.05 jHFD vs. jC, ^#P < 0.05 jC vs. aC. n = 6 rats/group.

For 5-HT, there was an interaction between age and diet (F_{(1, 20)} = 4.725, P < 0.05), whereas no significant differences were observed between groups in the content of NA or other metabolites including DOPAC, HVA, HVA/DA, 5HIAA or 5HIAA/5-HT (Table 1).

We observed differential effects on amygdalar D2R-ir depending on age. In juvenile animals, the jHFD group showed a 46% increase in D2R-ir compared to jC (11 ± 0.3 vs. 7.5 ± 0.7%, P < 0.05, Fig. 4A–C). Additionally, age increased D2R-ir in control animals, (14.4 ± 1.2 in aC vs. 7.5 ± 0.7% in jC, P < 0.001). aHFD animals showed a 30% reduction in D2R-ir compared to aC (9.6 ± 0.7 vs. 14.4 ± 1.2%, P < 0.01, Fig. 4A–D-E). Two-way ANOVA showed an effect of age (F_{(1, 12)} = 11.7, P < 0.01) and a significant interaction between age and diet (F_{(1, 12)} = 26.3, P < 0.01).

Fig. 4.

Type 2 dopamine receptor (D2R) expression in the basolateral amygdala. (A) Percentage of area covered by D2R immunofluorescent signal in control (white bars, circles) and high-fat diet (HFD)-fed rats (hatched bars, triangles). (B-E) Representative photomicrographs of D2R-ir of juvenile and adult rats fed with control (B, C) or HFD (D, E), scale bar: 20 μm. n = 4/group. *P < 0.05, **P < 0.01, ***P < 0.001 between depicted groups.

3.4. HFD induced a proinflammatory profile in the amygdala of juvenile rats despite limited systemic inflammation

To determine whether HFD elicits peripheral and central inflammatory responses, we measured serum cytokines and local cytokine levels in the amygdala. Peripheral cytokine profile was not affected by HFD consumption in adult rats within the six-cytokine serum panel (Fig. 5). On the other hand, HFD intake correlated with a significant two-fold increase in IL-1β serum levels of jHFD vs. jC rats (Fig. 5D). The two-way ANOVA showed an effect of diet (F_{(1, 15)} = 9.1, P < 0.01). In contrast, more evident inflammation than in serum was observed in the amygdala of jHFD vs. jC (Fig. 6), with nearly all measured cytokines showing significantly elevated levels. These increases in cytokine levels were also significant when compared to aHFD rats (Fig. 6A–F). The two-way ANOVA showed for MCP1 an effect of diet (F_{(1, 16)} = 16.3, P < 0.01) and an interaction between diet and age (F_{(1, 16)} = 14.2, P < 0.01); for TNF-ɑ an effect of diet (F_{(1, 16)} = 12, P < 0.01) and an interaction between diet and age (F_{(1, 16)} = 11.5, P < 0.01); for IL-1β an effect of diet (F_{(1, 15)} = 22.8, P < 0.01); for IL-1ɑ an effect of diet (F_{(1, 16)} = 15.7, P < 0.01) and an interaction between diet and age (F_{(1, 16)} = 13.4, P < 0.01); and for IL-6 an effect of diet (F_{(1, 16)} = 12.1, P < 0.01), age (F_{(1, 16)} = 5.3, P < 0.05) and an interaction between variables (F_{(1, 16)} = 16.3, P < 0.01).

Fig. 5.

Serum cytokine levels in juvenile and adult rats fed with control (white bars, circles) or high-fat diet (hatched bars, triangles). (A) IFN-γ; (B) MCP-1; (C) TNF-α; (D) IL-1β; (E) IL-1α; (F) IL-6. Data is expressed as the mean ± SEM in pg/mL. *P < 0.05 between depicted groups. n = 5 rats/group, samples ran in duplicate.

Fig. 6.

Amygdalar cytokine levels in juvenile and adult rats fed with control (white bars, circles) or high-fat diet (hatched bars, triangles). (A) IFN-γ; (B) MCP-1; (C) TNF-α; (D) IL-1β; (E) IL-1α; (F) IL-6. Data is expressed as the mean ± SEM in pg./mL. *P < 0.05, **P < 0.01, ***P < 0.001 between depicted groups. n = 5 rats/group.

Given the central role of IL-1β in neuroinflammation, we further evaluated its regional expression in the BLA. IL-1β-ir was markedly increase in jHFD, showing a 62% elevation compared with jC (P < 0.001) and more than a two-fold increase vs. aHFD (P < 0.001, Fig. 7A–E). The two-way ANOVA showed significant effects of diet (F_{(1, 12)} = 12.4, P < 0.01), age (F_{(1, 12)} = 78.8, P < 0.001) and an interaction between diet and age (F_{(1, 12)} = 36.4, P < 0.001).

Fig. 7.

Interleukin-1β (Il-1β) expression in the basolateral amygdala. (A) Percentage of area covered by Il-1β immunofluorescent signal in control (white bars, circles) and high-fat diet (HFD)-fed rats (hatched bars, triangles), n = 4/group. *P < 0.05 between depicted groups. (B-E) Representative photomicrographs of Il-1β -ir of juvenile and adult rats fed with control (B, D) or HFD (C, E), scale bar: 20 μm. White arrowheads show positive IL-1β-ir signal.

3.5. Glial activation was differentially regulated by age, with iba-1-positive myeloid cells showing greater sensitivity to HFD in juvenile rats

To assess neuroinflammatory cellular responses, we analyzed astrocytic and microglial markers together with their morphological characteristics in the BLA. HFD feeding produced a robust age-independent reduction in astrocyte density, resulting in an approximately two-fold decrease in the number of GFAP-positive cells in both juvenile and adult animals (Fig. 8A), with significant differences only in juveniles (jHFD vs jC P < 0.05). The two-way ANOVA showed a significant effect of diet (F_{(1, 12)} = 12.8, P < 0.01, Fig. 8A). Similarly, the percentage of area covered by the GFAP-ir followed the same pattern, showing a significant interaction between age and diet (F_{(1, 12)} = 7.097, P < 0.05); post-hoc analyses showed a 40% reduction in jHFD vs. jC (P < 0.05) (Fig. 8B). Regarding morphology, we observed deramification in astrocytes of aHFD animals (Fig. 8C–G), as indicated by a 10% higher ramification index compared to both aC and jHFD. The two-way ANOVA revealed significant effects of age (F_{(1, 12)} = 6.9, P < 0.05) and diet (F_{(1, 12)} = 6.7, P < 0.05).

Fig. 8.

Expression of glial fibrillary acidic protein (GFAP) in the basolateral amigdala. (A) GFAP-positive cells/mm². (B) Percentage of area covered by GFAP-positive cells. (C) Ramification index of GFAP labeled cells of juvenile and adult rats, fed with control (white bars, circles) or high-fat diet (HFD) (hatched bars, triangles). (D-G) Representative photomicrographs of GFAP-ir in juvenile and adult rats fed with control (D, F) or HFD (E, G). D1-G1 are magnifications of one cell to exemplify ramification index measures in each group. Scale bar: 20 μm. n = 4 rats/group. *P < 0.05 between depicted groups.

Similar to astrocytes, there was an overall decrease in microglial cells in HFD-fed rats (Fig. 9A). Specifically, aHFD rats had significantly fewer Iba-1-positive cells compared to aC (P < 0.05), the two-way ANOVA showed a significant effect of diet (F_{(1, 12)} = 13.68, P < 0.01). The jHFD rats showed a 66% increase in the percentage of Iba-1-ir compared to jC, while aHFD rats exhibited a 37% decrease compared to aC (P < 0.05 for both comparisons) (Fig. 9B). The two-way ANOVA showed a significant interaction between age and diet (F_{(1, 12)} = 21.09, P < 0.001). Morphological analyses showed that the aHFD group exhibited decreased microglial ramification (Fig. 9C–G), reflected by a higher ramification index compared to both aC and jHFD groups. The two-way ANOVA revealed significant effects of age (F_{(1, 12)} = 18.6, P < 0.01) and diet (F_{(1, 12)} = 6.5, P < 0.05).

Fig. 9.

Expression of ionized calcium-binding adapter molecule 1 (Iba-1) in the basolateral amygdala. (A) Iba-1-positive cells/mm². (B) Percentage of area covered by Iba-1-positive cells. (C) Ramification index of Iba-1 labeled cells of juvenile and adult rats, fed with control (white bars, circles) or high-fat diet (HFD) (hatched bars, triangles). (D-G) Representative photomicrographs of Iba-1-ir in juvenile and adult rats fed with control (D, F) or HFD (E, G). D1-G1 are magnifications of one cell to exemplify ramification index measures in each group. Scale bar: 20 μm. n = 4 rats/group. *P < 0.05, **P < 0.01 between depicted groups.

4. Discussion

HFD significantly increased neuroinflammation and induced dopaminergic alterations in the amygdala of juvenile rats, while adult rats were largely unaffected. This age-dependent vulnerability is clinically relevant given the rising prevalence of HFD consumption among youth (Tsan et al., 2021), the fact that the amygdala continues to mature through adolescence (Arruda-Carvalho et al., 2017; Ferrara et al., 2021), and its impact on behavior regulation.

Both HFD-fed groups exhibited comparable levels of hyperphagia, although the temporal pattern of caloric intake differed between juveniles and adults. These differences may reflect age-dependent adaptations in energy homeostasis and hypothalamic regulation between the developing and mature brain (Alvarez-Salas et al., 2025). While the amygdala plays an important role in food motivation and the regulation of feeding behavior (Areias and Prada, 2015), the more pronounced neuroinflammatory response observed in juvenile HFD-fed rats, despite similar hyperphagia in adults, suggests that amygdalar neuroinflammation is unlikely to be the primary driver of the increased food intake under these experimental conditions. Interestingly, in our previous study using the same experimental model (Alvarez-Salas et al., 2025), circulating leptin levels were elevated only in jHFD rats, whereas aHFD animals displayed comparable food intake without hyperleptinemia. Orexigenic signals, particularly ghrelin, which interact with hypothalamic and mesolimbic reward circuits, may contribute to the increased caloric intake and warrant investigation in future studies.

HFD exposure did not significantly alter b.w. gain, consistent with evidence that HFD-induced neuroinflammation can precede obesity onset (Almeida-Suhett et al., 2017; Vinuesa et al., 2016). We have already demonstrated that four weeks of HFD exposure activates different parameters of the thyroid axis regardless of age (Alvarez-Salas et al., 2025), potentially increasing energy expenditure and limiting b.w. gain despite hyperphagia.

Elevated peripheral IL-1β levels were detected exclusively in jHFD rats, supporting a systemic proinflammatory state characteristic of HFD intake (Thaler et al., 2012). Nevertheless, because serum cytokines were measured at a single mid-light-phase time point, transient inflammatory fluctuations may have been underestimated. Interestingly, inflammatory changes were even more pronounced in the amygdala of jHFD than in serum. Previous reports in mice have described such an increase in IL-1β expression in the amygdala after adolescent HFD, along with alterations in the expression of genes related to neuroinflammation and neurotransmission (Fülling et al., 2020).

Brain cytokines modulate behavior through multiple mechanisms, including modulating synaptic and non-synaptic plasticity and by altering the expression of neurotransmitters and neuropeptides that regulate food intake and mood regulation (Buchanan and Johnson, 2007; Zheng et al., 2021). In agreement with reports showing that prolonged HFD exposure increases amygdalar IL-1β and anxiety-like behavior in adolescent mice (Almeida-Suhett et al., 2017), jHFD rats showed elevated IL-1β protein expression and increased Iba-1-immunoreactive area in the BLA, indicative of reactive microglia. In contrast, jHFD reduced GFAP immunoreactivity may reflect astrocytic filament remodeling. This pattern aligns with evidence showing that HFD exposure initially triggers microglial activation, followed by later astroglial responses (Baufeld et al., 2016).

Thus, microglia likely contributed to the inflammatory profile observed in jHFD rats, although the cellular source of cytokines was not directly determined. Both HFD and palmitate induce microglial activation in the amygdala (Butler et al., 2020), and glial cytokine release due to an immune challenge often precedes overt glial morphological changes (Biesmans et al., 2015; Norden et al., 2016). Therefore, the elevated amygdalar cytokine levels in jHFD rats may reflect sustained microglial activation and impaired inflammatory resolution.

In contrast, changes in astrocytic morphology and microglial inflammatory markers in aHFD rats were consistent with a later remodeling or resolution phase and in agreement with their basal amygdalar cytokine levels, although pro-resolving markers were not assessed. Similarly, chronic HFD exposure in adult mice is known to alter amygdalar glial activity primarily after an immune challenge (Huang et al., 2020). These divergent inflammatory profiles support the idea that adolescence represents a period of increased neuroimmune vulnerability to HFD intake.

One neuronal pathway particularly affected by HFD consumption is the dopaminergic circuit (Naneix et al., 2017), which is especially relevant for two reasons: it matures throughout adolescence and is highly susceptible to neuroinflammation. Also, jHFD rats appeared more vulnerable regarding coordinated alterations in dopaminergic signaling. Specifically, the lower DOPAC/DA ratio suggested reduced DA release, since DOPAC is synthesized inside the dopaminergic neuron. However, as DA release and uptake were not directly assessed, conclusions regarding functional signaling remain inferential.

Indicators of dopaminergic pathway downregulation were consistent with previous reports from our laboratory regarding low D2R protein expression (García-Luna et al., 2023), along with decreased DA metabolism in the nucleus accumbens (NAc) after ad libitum conditions of palatable food intake (García-Luna et al., 2025), as we observed here in aHFD and jHFD respectively. In contrast, specifically for jHFD a similar decrease in DA release and metabolism in face of heightened neuroinflammation in the NAc are reported (Plaza-Briceño et al., 2023), suggesting that microglial reactivity may contribute to amygdalar dopaminergic dysfunction. Supporting this interpretation, treatments that reduce amygdalar inflammatory signaling can reverse DA alterations and attenuate HFD-induced behavioral changes (Akter et al., 2020).

The low D2R expression in juvenile control rats, contrasts with age-related decline in D2R density typically observed in brain regions such as the prefrontal cortex and striatum, though results from the NAc remain inconsistent (Andersen et al., 2000; Spear, 2000). D2R changes may represent a compensatory postsynaptic adaptation rather than reflect changes in DA content itself. This interpretation is consistent with previous reports showing that D2R regulation may occur independently of measurable changes in tissue DA concentrations (Hukema et al., 2025; Kalinowski et al., 2023).

Evidence regarding the effects of HFD on amygdalar monoaminergic signaling during development is mixed. Juvenile mice exposed to a 12-week HFD exhibit increased amygdalar levels of DA, 5-HT, and NA (Lama et al., 2022), whereas other studies have detected monoaminergic alterations only following stimulation such as conditioned odor aversion (Osorio-Gómez et al., 2024). Conversely, prolonged HFD exposure did not affect basal or stimulated DA release in the BLA of young rats (Osorio-Gómez et al., 2024). Here, jHFD rats showed selective alterations in DA and its metabolism without changes in other monoamines: downregulation of the dopaminergic pathway in jHFD rats may represent a compensatory adaptation to repeated palatable food intake, similar to accumbal neuroadaptations during overeating (García-Luna et al., 2025; Rada et al., 2010; Reichelt and Rank, 2017).

Although the effects were more pronounced in jHFD, aHFD rats also exhibited dopaminergic alterations, their reduced BLA D2R expression may indicate diminished reward sensitivity, as previously associated with stress-related states (Kołosowska et al., 2019).

Given the impact of HFD on stress-related responses and the critical role of the amygdala in facilitating the HPA axis activity (Ziegler and Herman, 2002), we examined stress-associated parameters across ages. We found that jHFD rats displayed reduced circulating corticosterone, similar to rats fed with palatable diets during the perinatal period and as young adults (Sasaki et al., 2013; Marcos et al., 2023), while as previously reported (Gergerlioglu et al., 2016), aHFD during 4 weeks did not change corticosterone basal levels.

The alterations observed in the BLA of jHFD rats suggest a disruption of glucocorticoid–glial regulation that may contribute to the heightened neuroinflammatory profile associated with HFD exposure during this developmental stage. The reduction of circulating corticosterone together with decreased GR expression in the BLA, indicated a potential reduction in glucocorticoid signaling in this region. Glucocorticoids normally participate in restraining inflammatory responses through GR-dependent mechanisms in glial cells (Walkera et al., 2013); therefore, reduced corticosterone availability and GR expression could diminish this regulatory influence and contribute to the increased IL-1β immunoreactivity observed in the BLA. In parallel, the marked reduction in GFAP-positive astrocytes suggests an alteration of astrocytic homeostasis, which may further compromise the capacity of the local glial network to regulate inflammatory signaling. Importantly, these changes were more pronounced in juvenile than in aHFD rats despite comparable hyperphagia, suggesting that the enhanced neuroimmune alterations observed in juveniles are not simply a consequence of greater food intake but may reflect an increased vulnerability of the developing brain to the metabolic and endocrine consequences of HFD exposure. Although the present findings do not establish a causal relationship between reduced corticosterone/GR signaling and glial alterations, the concurrent disruption of the glucocorticoid system, increased IL-1β, and altered astrocytic GFAP expression supports the possibility that impaired glucocorticoid regulation contributes to the altered neuroimmune homeostasis induced by HFD during juvenile development. Furthermore, because suppression of HPA axis activity has been shown to alter the dopaminergic signaling within the BLA during stress (De Oliveira et al., 2013), the reduced corticosterone levels observed in jHFD may have contributed to the lower dopaminergic tone detected in this group.

5. Conclusion

The mechanisms linking HFD-induced neuroinflammation to the dopaminergic alterations observed in juvenile rats remain to be fully elucidated; however, accumulating evidence supports a close interplay between activated microglia, stress responsiveness and dopaminergic neurotransmission. jHFD rats exhibited increased microglial activation and elevated IL-1β-ir, together with increased D2R-ir and altered DA metabolism in the BLA. The upregulation of D2R expression may represent a compensatory adaptation to maintain dopaminergic signaling in the presence of inflammation-induced alterations in DA neurotransmission. In contrast, aHFD rats exhibited a comparatively attenuated neuroinflammatory response and no significant alterations in DA metabolism, suggesting that the mature brain may be less susceptible to inflammation-induced dysregulation of the amygdalar dopaminergic system. Finally, given the established association between amygdalar neuroinflammation and psychiatric disorders (Zheng et al., 2021), these findings emphasize adolescence as a vulnerable developmental window during which HFD consumption may adversely affect brain function.

CRediT authorship contribution statement

J.L. Dolores-San Juan: Formal analysis, Investigation, Visualization. C. García-Luna: Formal analysis, Writing – review & editing, Visualization. O. Jaimes: Investigation. D. Bonilla: Formal analysis, Investigation. M.E.G. Brunck: Resources, Writing – review & editing, Formal analysis. P. de Gortari: Resources, Writing – review & editing. E. Alvarez-Salas: Conceptualization, Formal analysis, Writing – original draft, Funding acquisition.

Funding

This study was supported by A1-S-27322 grant from the SECIHTI (EAS).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.