Sex Differences on Social and Anxiety-Related Responses in Low-Density Lipoprotein Receptor Knockout Mice

TL;DR

This study finds that LDL receptor deficiency affects anxiety and social behaviors differently in male and female mice, linked to changes in brain dopamine regulation.

Contribution

The study reveals sex-specific behavioral and neurochemical effects of LDL receptor deficiency in mice.

Findings

Female LDLr⁻/⁻ mice showed hyperlocomotion and reduced anxiety-like behavior compared to controls.

Both male and female LDLr⁻/⁻ mice exhibited increased sociability with sex-dependent variations.

Reduced COMT levels in the prefrontal cortex of LDLr⁻/⁻ females suggest sex-specific dopaminergic modulation.

Abstract

Familial hypercholesterolemia, caused by mutations in the low-density lipoprotein receptor (LDLr), is associated with cognitive and affective disturbances. However, sex differences in the impact of LDLr deficiency on anxiety and social behaviors remain poorly characterized. Male and female LDLr⁻/⁻ and wild-type (WT) C57BL/6 mice (4–5 months old) underwent a battery of behavioral tests assessing locomotion, anxiety-like behavior (open field, elevated plus maze, marble-burying test), and sociability (social interaction and three-chamber test). Serum cholesterol levels and catechol-O-methyltransferase (COMT) levels in the prefrontal cortex (PFC) and amygdala (AMY) were subsequently analyzed by Western blotting. Both male and female LDLr⁻/⁻ mice displayed marked hypercholesterolemia relative to WT controls. Female LDLr⁻/⁻ mice exhibited hyperlocomotion and reduced anxiety-like behavior in multiple tasks. Both sexes demonstrated increased sociability, although with test- and sex-dependent variations. Neurochemical analysis revealed a selective reduction of COMT in the PFC of LDLr⁻/⁻ females, suggesting sex-specific dopaminergic modulation underlying the behavioral phenotypes. LDLr deficiency induced sex-dependent changes in locomotor, anxiety-related, and social behaviors, accompanied by altered COMT expression in corticolimbic regions. These findings reveal an association between LDL receptors, dopaminergic regulation, and socio-emotional behaviors, emphasizing the relevance of social characterization in the neurobehavioral assessment of dyslipidemia.

The online version contains supplementary material available at 10.1007/s12035-026-05742-x.

Figures

Experimental design and hyperlocomotion in female LDLr⁻/⁻ mice. (A) Schematic representation of the experimental design. Male and female wild-type (WT) and LDL receptor knockout (LDLr⁻/⁻) mice (4–5 months old) underwent a behavioral test battery assessing anxiety-related parameters (elevated plus maze, open field, and marble-burying tests) and sociability (social interaction and three-chamber tests). After 24 h, mice were euthanized for serum cholesterol quantification and Western blot analysis of the prefrontal cortex (PFC) and amygdala (AMY). (B) Serum cholesterol levels (mg/dL) were signifi

Reduced anxiety-like behavior in female LDLr⁻/⁻ mice. (A) Time spent in the center of the open field did not differ among groups. (B–C) Female LDLr⁻/⁻ mice exhibited reduced anxiety-like behavior, indicated by a higher percentage of open-arm entries (B) and increased time spent in open arms (C) in the elevated plus maze (EPM). (D) Frequency of head-dipping in the EPM was elevated in female LDLr⁻/⁻ mice, consistent with enhanced risk-assessment and exploration. (E) Rearing behavior did not differ between groups. (F) Representative images of the marble-burying test showing typical marble distrib

Enhanced social approach in LDLr⁻/⁻ mice depends on the behavioral context. (A) The overall social interaction index did not differ between genotypes. (B) LDLr⁻/⁻ mice spent more time in proximity to the social stimulus during the sociability test, reflecting greater social interest. (C) In the three-chamber test, LDLr⁻/⁻ mice showed a preference for the social chamber over the non-social (object) chamber. (D) Male LDLr⁻/⁻ mice spent significantly more time exploring the social chamber than predicted by chance (33%), whereas female WT mice displayed a preference for the non-social chamber. Dat

Region- and sex-dependent regulation of COMT expression in LDLr⁻/⁻ mice. (A) Western blot quantification revealed reduced COMT expression in the prefrontal cortex (PFC) of female LDLr⁻/⁻ mice compared with WT controls. (B) No significant differences in COMT levels were detected in the amygdala (AMY). Representative immunoblots are shown above each quantification. Data are expressed as mean ± SEM (*n* = 6–9). Two-way ANOVA (B) followed by Newman-Keuls post hoc test (A). *p* < 0.05 for genotype; #*p* < 0.05 for sex differences

Full text

Introduction

Familial hypercholesterolemia (FH) is one of the most prevalent inherited metabolic disorders worldwide. Epidemiological studies indicate considerable variability in its global prevalence, largely influenced by population genetics and ethnicity. The heterozygous form of FH occurs in approximately 1:311 individuals in the general population, whereas the homozygous form is much rarer, affecting about 1:250,000 to 1:1,000,000 individuals depending on the country [1, 2]. FH results from an impaired clearance of low-density lipoprotein (LDL) particles from the bloodstream due to loss-of-function mutations in the LDL receptor (LDLr), leading to elevated circulating LDL-cholesterol levels [3, 4]. In heterozygous FH, partial LDLr dysfunction results in two- to threefold higher LDL-cholesterol levels compared to the general population, whereas complete loss of receptor function in homozygous individuals can increase cholesterol levels by up to sixfold [5].

Although FH is classically recognized as a major risk factor for atherosclerosis and cardiovascular disease, growing evidence indicates that its consequences extend beyond the periphery, affecting brain function and mental health. Several clinical studies have reported associations between hypercholesterolemia and cognitive impairment [6–8] as well as a higher prevalence of depressive symptoms [9]. Conversely, the relationship between cholesterol levels and anxiety disorders remains controversial. Some studies suggest that FH patients may display fewer anxiety symptoms compared to normolipidemic individuals [10], while others report that low cholesterol and LDL levels are associated with manic or generalized anxiety disorders [11]. These inconsistencies underscore the need for experimental studies investigating how disturbances in lipid metabolism influence emotional regulation.

Mice lacking LDLr (LDLr⁻/⁻) are widely used as a preclinical model of FH, displaying elevated serum cholesterol levels and recapitulating several features of the human condition [12]. LDLr⁻/⁻ mice have been reported to exhibit cognitive impairments [13–15], depressive-like behaviors [14, 16, 17], and hyperlocomotion [13, 18]. Notably, anxiety-related changes have been predominantly observed in female LDLr⁻/⁻ mice in unpublished studies, suggesting potential sexual dimorphism, although this remains insufficiently explored in males.

The neurobiological mechanisms underlying these behavioral alterations likely involve dysregulation of monoaminergic neurotransmission. Anxiety and mood disorders are strongly associated with dysfunction of monoaminergic pathways, which are also the primary targets of most anxiolytic and antidepressant treatments [19]. In LDLr⁻/⁻ mice, depressive-like behavior has been linked to increased monoamine oxidase-A (MAO-A) activity in the hippocampus [17], whereas low MAO-A activity has been associated with antisocial behaviors in humans [20] and in MAO-A knockout mice [21]. Catechol-O-methyltransferase (COMT), another major enzyme responsible for catecholamine degradation, particularly dopamine and norepinephrine, is also implicated in emotional and social regulation. Variations in COMT activity influence dopamine signaling in the prefrontal cortex, affecting empathy, sociability, and executive control. In humans, the COMT Val158Met polymorphism is linked to social impairments and altered emotional reactivity [22, 23]. In rodents, COMT deficiency or reduced enzymatic activity alters social dominance and interaction patterns [24]. Beyond its role in neurotransmitter metabolism, COMT polymorphisms have been associated with cardiovascular risk markers [25, 26], suggesting that dopamine catabolism and lipid metabolism may share mechanistic intersections.

Despite accumulating evidence linking LDLr deficiency to affective and cognitive disturbances, no studies have systematically examined sex differences or their underlying neurochemical correlates in anxiety-like and social behaviors in LDLr⁻/⁻ mice. Thus, the present study aimed to (i) characterize locomotor and anxiety-related behaviors in male and female LDLr⁻/⁻ mice; (ii) investigate potential sex-specific alterations in social behavior; and (iii) assess whether these behavioral patterns are associated with changes in COMT expression in corticolimbic regions, particularly the prefrontal cortex and amygdala. By integrating behavioral and molecular analyses, this study seeks to clarify how cholesterol metabolism interacts with monoaminergic systems to modulate sex-dependent emotional and social responses.

Experimental Procedures

Animals and Experimental Design

Male and female C57BL/6 wild-type (WT) and LDL receptor knockout (LDLr⁻/⁻) mice (4–5 months old) were bred and maintained at the Federal University of Santa Catarina (UFSC, Florianópolis, Brazil). The breeding colony originated from animals obtained from the State University of Campinas (UNICAMP, São Paulo, Brazil), originally purchased from Jackson Laboratory (Bar Harbor, ME, USA). Animals were group-housed (4–8 per cage; 42 × 34 × 17 cm) in ventilated chambers (INSIGHT®) under controlled environmental conditions (22 ± 2 °C; 60–80% humidity) with a 12 h light/dark cycle (lights on at 7:00 a.m.) and ad libitum access to food and water. Mice were randomly assigned to four experimental groups (n = 6–8 per group): (i) WT females, (ii) WT males, (iii) LDLr⁻/⁻ females, (iv) LDLr⁻/⁻ males.

A schematic overview of the experimental timeline is shown in Fig. 1A. After behavioral testing, trunk blood was collected for biochemical assays, and brains were rapidly dissected to isolate the prefrontal cortex (PFC) and amygdala (AMY). Samples were flash-frozen in liquid nitrogen and stored at − 80 °C until analysis. All procedures were conducted in accordance with the Brazilian Guidelines for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee (CEUA/UFSC; protocol no. 7816220421).Fig. 1. Experimental design and hyperlocomotion in female LDLr⁻/⁻ mice. (A) Schematic representation of the experimental design. Male and female wild-type (WT) and LDL receptor knockout (LDLr⁻/⁻) mice (4–5 months old) underwent a behavioral test battery assessing anxiety-related parameters (elevated plus maze, open field, and marble-burying tests) and sociability (social interaction and three-chamber tests). After 24 h, mice were euthanized for serum cholesterol quantification and Western blot analysis of the prefrontal cortex (PFC) and amygdala (AMY). (B) Serum cholesterol levels (mg/dL) were significantly elevated in LDLr⁻/⁻ mice of both sexes. (C–H) Female LDLr⁻/⁻ mice exhibited hyperlocomotion in multiple behavioral tasks, as evidenced by total distance traveled in the open field (C–D), sociability test without (E) and with (F) the social stimulus, total distance in the three-chamber test (G), and increased entries into closed arms in the elevated plus maze (H). Data are presented as mean ± SEM (n = 6–9). Two-way ANOVA followed by Newman-Keuls post hoc test. p < 0.05 for genotype effect; #p < 0.05 for sex differences

Behavioral Testing

Behavioral experiments were conducted during the light phase in a temperature-controlled room (22 °C) under low light intensity (12 lx). Mice were acclimated to the testing room for 1 h before each session. Between trials, apparatuses were cleaned with 30% ethanol between same-sex animals and 70% ethanol between sexes. All tests were recorded for subsequent analysis. Parameters from the open field and sociability tests were analyzed using AnyMaze® software, whereas elevated plus maze and marble-burying data were manually scored by an experimenter blind to group allocation.

Elevated Plus Maze (EPM)

The EPM test was used to evaluate anxiety-like behavior [27]. The apparatus consisted of two open arms (18 × 6 cm) and two closed arms (18 × 6 cm) arranged in a plus shape and elevated 60 cm above the floor. Each mouse was placed in the central platform facing a closed arm and allowed to explore for 5 min. Time spent and entries into open and closed arms were recorded. The percentage of open-arm exploration was calculated as:

• % Time = (time in open arms × 100)/(time in open + closed arms)
• % Entries = (entries in open arms × 100)/(entries in open + closed arms)

Additional behaviors (grooming, head-dipping, and rearing) were quantified as ethological indicators of anxiety and exploratory activity.

Open Field Test

The open field test assessed spontaneous locomotion and exploratory anxiety [28]. The apparatus was a transparent acrylic box (40 × 40 cm) divided into 12 squares (10 × 10 cm). Each animal was placed in the center and allowed to explore freely for 5 min. Total distance traveled (locomotor activity) and time/distance spent in the central area (anxiety-like index) were quantified using AnyMaze® software.

Marble-Burying Test

The marble-burying test assessed repetitive and anxiety-related behavior [29, 30]. The test box (17.5 × 10 × 5.5 cm) contained 5 cm of wood shavings with 20 glass marbles (1.4 cm diameter) arranged in 5 × 4 rows. Mice were allowed to explore for 20 min, and the number of buried marbles (≥ two-thirds covered) was recorded at 5, 10, 15, and 20 min [31].

Sociability Test

The sociability test was conducted in an open field arena (40 × 40 cm) with a small wire-mesh cage positioned against one wall [32, 33]. Session 1 (habituation): the experimental mouse explored the empty cage for 2.5 min, followed by a 1 min intersession interval. Session 2 (social phase): an unfamiliar conspecific of the same sex, age, and weight was placed in the mesh cage, and the subject explored for another 2.5 min. Time spent in the interaction zone (7 × 7 cm around the cage) and in the opposite peripheral zones was quantified as an index of sociability.

Three-Chamber Test

Sociability was further evaluated using the three-chamber test [34]. The wooden apparatus (60 × 40 cm) was divided into three chambers (20 × 40 cm each) separated by partitions with small doors (5 × 8 cm). After a 5 min habituation without stimuli, the test phase was conducted 1 h later: one side chamber contained an unfamiliar conspecific enclosed in a wire cage (social stimulus), while the opposite chamber contained a neutral object (Lego® block). The subject was placed in the central chamber and allowed to explore for 10 min. Sociability was quantified as the relative time and number of entries into each chamber compared to chance level (33%).

Biochemical Analyses

Serum Cholesterol and Corticosterone

Trunk blood was collected during euthanasia (2:00–4:00 p.m.) and centrifuged at 10,000 rpm for 12 min to obtain serum, which was stored at − 80 °C. Total cholesterol was measured using a commercial colorimetric assay (Gold Analisa®, Brazil), and corticosterone levels were determined using an ELISA kit (IBL International®) according to manufacturers’ instructions. Absorbance was read on a Multireader Infinite M200® (Tecan®) at the Multiuser Laboratory for Biology Studies (LAMEB, UFSC).

Western Blotting

The PFC and AMY were homogenized in RIPA buffer (Merck, Cat. #20–188) and centrifuged (12,000 g, 20 min, 4 °C). Protein concentrations were determined by the Lowry method (Lowry et al., 1951). Samples (60 μg) were prepared in Laemmli buffer (Bio-Rad, #161–0737), denatured (70 °C, 10 min), separated by SDS-PAGE, and transferred to nitrocellulose membranes (Amersham™). Membranes were blocked (5% BSA in TBS) and incubated overnight (4 °C) with primary antibodies: COMT (Santa Cruz, sc-135872, 1:200), β-actin (Cell Signaling, #8H10D10, 1:1000). After washing (TBS-T), membranes were incubated with HRP-conjugated secondary antibodies: goat anti-mouse (Sigma-Aldrich #AP308P, 1:10,000) or goat anti-rabbit (Invitrogen #31460, 1:10,000) for 1 h at room temperature. Immunoreactive bands were visualized using Super ECL (GE Healthcare®) on a ChemiDoc system (Bio-Rad®, LAMEB/UFSC).

Statistical Analysis

Data were analyzed using Statistica® 8.0 (StatSoft, Tulsa, OK, USA). Outliers were identified using the Grubbs test, and data distribution was assessed by the Kolmogorov–Smirnov test. Comparisons to chance levels (e.g., 50% in olfactory discrimination, 33% in three-chamber) were performed with Student’s t-tests. Two-way ANOVA or repeated-measures ANOVA (genotype vs. sex), followed by Newman-Keuls post hoc tests, were used for multiple comparisons in behavioral and biochemical measures. The Kruskal–Wallis test was applied for nonparametric datasets (e.g., marble burying). A complete description of the ANOVA interactions is presented in Supplementary Tables S1 and S2. Data are expressed as mean ± SEM, with statistical significance set at p < 0.05. Graphs were generated using GraphPad Prism 7® (GraphPad Software, San Diego, CA, USA).

Results

Sex-Dependent Differences in Locomotor Activity of LDLr⁻/⁻ Mice

As expected, serum cholesterol levels were markedly elevated in LDLr⁻/⁻ mice of both sexes compared with WT controls (Fig. 1B; two-way ANOVA, genotype: F(1,29) = 179.86, p < 0.0001), confirming the metabolic phenotype associated with LDLr deficiency. Body weight was comparable between WT and LDLr⁻/⁻ mice in both males and females (see Supplementary Table S3).

Locomotor activity was first evaluated in the open field test. Female LDLr⁻/⁻ mice displayed a significant increase in total distance traveled compared with all other groups (Fig. 1C-D; genotype: F(1,29) = 17.80, p < 0.001; sex: F(1,29) = 5.64, p = 0.024; Newman-Keuls post-hoc, p < 0.05). This hyperlocomotor pattern was replicated in the habituation phase of the sociability test, when no social stimulus was present (Fig. 1E; genotype: F(1,28) = 22.77, p < 0.0001; genotype vs. sex: F(1,28) = 13.14, p = 0.0011). Female LDLr⁻/⁻ mice traveled greater distances than all other groups.

When a social stimulus was introduced in the sociability test, locomotor activity increased in male LDLr⁻/⁻ mice (Fig. 1F; genotype: F(1,29) = 12.31, p = 0.0015). In this condition, both male and female LDLr⁻/⁻ mice showed enhanced exploration compared with their WT counterparts (Newman-Keuls post-hoc p < 0.05). Similar results were obtained in the three-chamber test (Fig. 1G; genotype: F(1,29) = 14.91, p < 0.001), in which LDLr⁻/⁻ females traveled significantly longer distances than WT controls.

Finally, in the elevated plus maze (EPM), the number of entries into closed arms, an index of general locomotor activity, was higher in LDLr⁻/⁻ mice than in WT animals (Fig. 1H; genotype: F(1,29) = 21.24, p < 0.001), with no sex differences detected. Together, these findings indicate a hyperlocomotor phenotype primarily expressed in female LDLr⁻/⁻ mice, consistent across multiple behavioral paradigms.

Sex-Dependent Modulation of Anxiety-Like Behavior in LDLr⁻/⁻ Mice

Anxiety-like behavior was assessed using the open field, EPM, and marble-burying tests. In the open field, the time spent in the center of the arena did not differ among groups (Fig. 2A; genotype: F(1,28) = 3.70, p = 0.065; sex: F(1,28) = 0.72, p = 0.40; interaction: F(1,28) = 2.23, p = 0.15), suggesting that basal exploratory anxiety was not significantly altered.Fig. 2. Reduced anxiety-like behavior in female LDLr⁻/⁻ mice. (A) Time spent in the center of the open field did not differ among groups. (B–C) Female LDLr⁻/⁻ mice exhibited reduced anxiety-like behavior, indicated by a higher percentage of open-arm entries (B) and increased time spent in open arms (C) in the elevated plus maze (EPM). (D) Frequency of head-dipping in the EPM was elevated in female LDLr⁻/⁻ mice, consistent with enhanced risk-assessment and exploration. (E) Rearing behavior did not differ between groups. (F) Representative images of the marble-burying test showing typical marble distribution (upper panel) and group comparisons after 20 min (lower panel). (G) LDLr⁻/⁻ mice buried fewer marbles than WT controls, indicating reduced anxiety- and compulsive-like behavior. Data are expressed as mean ± SEM (n = 6–9). Two-way ANOVA followed by Newman-Keuls post hoc when significant test for panels A–F, and Kruskal–Wallis test for panel H. p < 0.05 for genotype; #p < 0.05 for sex differences

In contrast, the EPM revealed robust sex-dependent effects. Female LDLr⁻/⁻ mice entered the open arms more frequently and spent a higher percentage of time there compared with all other groups (Fig. 2B-C; % entries: genotype: F(1,28) = 10.04, p = 0.0037; sex: F(1,28) = 12.06, p = 0.0017; % time: genotype: F(1,28) = 18.28, p < 0.001; sex: F(1,28) = 10.95, p = 0.0026; Newman-Keuls post hoc p < 0.05). These results indicate reduced anxiety-like behavior specifically in females.

Ethological parameters further supported this interpretation. The frequency of head-dipping, an exploratory behavior associated with risk assessment, was significantly higher in female LDLr⁻/⁻ mice (Fig. 2D; genotype: F(1,29) = 9.26, p = 0.0049; sex: F(1,29) = 6.71, p = 0.015). In the other hand, the % of grooming was less in females LDLr-/- mice, comparing to WT (Fig. 2E; genotype: F(1,26)=10.39, p=0.0034, p<0,05). No significant differences were observed in rearing behavior (Fig. 2F; all p > 0.4).

In the marble-burying test, both male and female LDLr⁻/⁻ mice buried fewer marbles compared with their WT counterparts at all time points (Fig. 2G-H Kruskal–Wallis, 5 min: H(3,32) = 12.03, p = 0.007; 10 min: H(3,32) = 22.86, p < 0.0001; 15 min: H(3,32) = 23.86, p < 0.0001; 20 min: H(3,32) = 21.28, p < 0.001), confirming a reduction in anxiety-related and stereotyped behaviors in both sexes. Overall, these results reveal a clear anxiolytic-like profile in LDLr⁻/⁻ mice, with the strongest effects in females.

Social Behavior Alterations in LDLr⁻/⁻ Mice

Sociability was examined using both the social interaction and three-chamber tests. In the social interaction task, the overall sociability index did not differ across groups (Fig. 3A; genotype: F(1,28) = 0.03, p = 0.86; sex: F(1,28) = 0.09, p = 0.77). However, when spatial preference was analyzed, both female and male LDLr⁻/⁻ mice spent significantly more time in the interaction zone and less time in the peripheral zones relative to WT controls (Fig. 3B; genotype vs. zone interaction: F(1,29) = 5.29, p = 0.029; Newman-Keuls post-hoc, p < 0.05), indicating enhanced social approach and reduced avoidance.Fig. 3. Enhanced social approach in LDLr⁻/⁻ mice depends on the behavioral context. (A) The overall social interaction index did not differ between genotypes. (B) LDLr⁻/⁻ mice spent more time in proximity to the social stimulus during the sociability test, reflecting greater social interest. (C) In the three-chamber test, LDLr⁻/⁻ mice showed a preference for the social chamber over the non-social (object) chamber. (D) Male LDLr⁻/⁻ mice spent significantly more time exploring the social chamber than predicted by chance (33%), whereas female WT mice displayed a preference for the non-social chamber. Data are presented as mean ± SEM (n = 6–9). Two-way ANOVA followed by Newman-Keuls post hoc test for panel A, two-way repeated-measures ANOVA followed by Newman-Keuls post hoc test for panels B, three-way repeated-measures ANOVA for panel C, and Student’s t-test for panel D. p < 0.05 versus control or baseline condition

In the three-chamber test, three-away repeated-measures ANOVA revealed a main effect of chamber (Fig. 3C; F(2,56) = 25.09, p< 0.0001). LDLr⁻/⁻ and WT males spent less time in the empty chamber compared with the social chamber (p< 0.05), suggesting a preference for social interaction. When analyzed against the 33% baseline, all groups spent less time in the central (middle) chamber than expected by chance (male WT: t = 6.53, p < 0.001; female WT: t = 3.19, p = 0.033; male LDLr⁻/⁻: t = 5.42, p< 0.001; female LDLr⁻/⁻: t = 6.297; p = 0.0002). Moreover, male WT and LDLr⁻/⁻ mice spent more time in the social chamber compared with the non-social one (male WT: t = 2.59, p = 0.032; male LDLr⁻/⁻: t = 2.49, p = 0.038), while female WT mice preferred the non-social chamber (t = 4.37, p = 0.012; Fig. 3D).

These findings indicate that LDLr deficiency promotes an enhanced preference for social stimuli, particularly in males, while females display context-dependent variations in sociability.

COMT Expression in Corticolimbic Regions

To explore potential molecular correlates of the observed behavioral phenotypes, COMT protein levels were quantified in the PFC and AMY by Western blotting. In the PFC, female LDLr⁻/⁻ mice exhibited significantly reduced COMT expression compared with female WT controls (Fig. 4A; genotype: F(1,22) = 8.25, p = 0.0089; sex: F(1,22) = 5.52, p = 0.028; Newman-Keuls post-hoc p < 0.05). Interestingly, female WT mice showed higher COMT expression than WT males, supporting a sex-dependent baseline difference.Fig. 4. Region- and sex-dependent regulation of COMT expression in LDLr⁻/⁻ mice. (A) Western blot quantification revealed reduced COMT expression in the prefrontal cortex (PFC) of female LDLr⁻/⁻ mice compared with WT controls. (B) No significant differences in COMT levels were detected in the amygdala (AMY). Representative immunoblots are shown above each quantification. Data are expressed as mean ± SEM (n = 6–9). Two-way ANOVA (B) followed by Newman-Keuls post hoc test (A). p < 0.05 for genotype; #p < 0.05 for sex differences

In contrast, COMT levels in the AMY did not differ among groups (Fig. 4B; genotype: F(1,19) = 4.29, p = 0.089; sex: F(1,19) = 0.04, p = 0.84; interaction: F(1,19) = 1.19, p = 0.29). These data indicate a region- and sex-specific reduction of COMT expression in the PFC of female LDLr⁻/⁻ mice, potentially underlying the observed behavioral modulation of anxiety and sociability.

Discussion

The relationship between hypercholesterolemia and emotional regulation remains incompletely understood. While FH is primarily characterized by elevated circulating cholesterol and cardiovascular risk, increasing evidence suggests that lipid metabolism can also modulate neural circuits involved in mood and social behavior. In this study, we demonstrate that LDLr deficiency produces sex-dependent alterations in locomotion, anxiety-like responses, and sociability, accompanied by reduced COMT expression in the PFC of female LDLr⁻/⁻ mice. These findings provide new insights into the neurobehavioral consequences of altered cholesterol homeostasis and highlight potential monoaminergic mechanisms underlying these effects.

The current biochemical analysis confirmed the expected hypercholesterolemia in LDLr⁻/⁻ mice, consistent with previous studies using this model [13, 17, 35]. Behaviorally, LDLr⁻/⁻ mice exhibited marked hyperlocomotion, predominantly in females, as shown by greater distances traveled in the open field test, the sociability test, and in the three chamber test. This observation contrasts with previous findings [18], which also noted hyperlocomotion in male LDLr-/- mice, but corroborates results from our group showing identified hyperlocomotion only in female LDLr-/- mice. Notably, the higher impairment observed in females might suggest sex-specific vulnerabilities in response to metabolic changes. Findings [18, 36] demonstrated an even greater increase in locomotion in animals fed a cholesterol-rich diet, indicating that elevated cholesterol levels may be the key factor contributing to hyperlocomotion.

Hyperlocomotion is an important confounding variable in anxiety tests. However, when locomotor activity was assessed in the plus maze using closed-arm entries [37], both male and female LDLr-/- show an increased locomotion. This indicates that the anxiolytic-like phenotype observed in female LDLr⁻/⁻ mice in this test cannot be explained solely by their hyperlocomotion. In the plus maze the female LDLr-/- exhibited reduced anxiety-like behavior, as indicated by increased time and entries into the open arms of the elevated plus maze, and significant increase in head-dipping behavior, a response that has been interpreted as enhanced risk assessment and exploratory activity [56]. Moreover, prior reports in other mouse models [38, 39] have shown that elevated time and entries in the open arms can reflect genuine anxiolytic-like behavior even when locomotion is altered. This finding further supports the anxiolytic-like profile observed in conventional measures of the task. In contrast, no differences were detected in rearing behavior across groups. Since rearing is often regarded as a general measure of vertical exploration and locomotor activation [40], the lack of genotype effect on this parameter suggests that the anxiolytic-like phenotype in female LDLr-/- mice cannot be explained simply by altered exploratory drive. Together, the increased open-arm exploration and head-dipping, coupled with unchanged rearing, strengthen the interpretation of a true anxiolytic-like behavioral phenotype in these animals.

Additionally, there were no significant differences in the time spent in the center of the open field apparatus between LDLr-/- and WT animals, in line with earlier reports [18]. This finding is particularly important given that increased locomotion in female LDLr-/- mice could confound measures of anxiety-like behavior. The absence of differences in center exploration suggests that the reduced anxiety-like behavior observed in the elevated plus maze is not simply a byproduct of hyperactivity in the open field. Previous studies have highlighted that center time in the open field does not always correlate directly with elevated plus maze measures, reflecting the involvement of partially distinct neural circuits and motivational factors [28, 38]. Thus, our results reinforce the idea that the anxiolytic-like phenotype in female LDLr-/- mice is task-dependent and cannot be fully explained by changes in general locomotor activity.

The marble burying test, often used as an index of stereotypical anxiety-like behavior, revealed that LDLr-/- mice buried fewer marbles than controls at multiple time points, indicating reduced anxiety-like behaviors in both sexes. Some authors state that the marbles trigger the rodents' innate burying behavior and that the animals do not necessarily enjoy burying the marbles [41]. In humans, Obsessive–Compulsive Disorder is characterized by two components: obsession (repetitive thoughts related, for example, to contamination, sexual, or religious matters) and compulsion (repetitive conflict-assessing behaviors, such as washing one’s hands multiple times). These are quantifiable behavioral manifestations that, therefore, involve compulsive behavior with repetitive actions [57], a behavior that is also observed in rodents when exposed to marbles [41].

Thus, our finding of reduced marble burying in LDLr-/- mice complements the elevated plus maze and open field results, reinforcing the evidence of a robust anxiolytic-like phenotype while also raising questions about whether LDLr deletion influences repetitive or compulsive-like behaviors. Cholesterol plays a critical role in steroid hormone synthesis and synaptic membrane dynamics, influencing neurotransmitter receptor function and plasticity [42]. Altered cholesterol availability can impact hypothalamic–pituitary–adrenal (HPA) axis activity and glucocorticoid signaling, both closely linked to anxiety regulation [43]. However, the literature presents mixed results, while some studies associate hyperlipidemia with greater anxiety and stress sensitivity, others report the opposite or no change [10]. Interestingly, our results contrasts with reports that high-fat diets, which also elevate serum cholesterol, often increase anxiety-like behaviors [44, 45]. These divergent effects suggest that the absence of LDLr may produce neurobehavioral changes distinct from those caused by dietary hyperlipidemia, emphasizing that the receptor itself, not merely cholesterol elevation, contributes to behavioral outcomes.

Our study also provides new evidence that LDLr deficiency affects social behaviors, a largely unexplored domain in dyslipidemia research. In both the sociability and three-chamber test, LDLr⁻/⁻ mice displayed increased approach toward social stimuli, although this effect was modulated by sex and task conditions. Male LDLr⁻/⁻ mice showed a consistent preference for the social chamber, while females exhibited context-dependent variations, possibly reflecting differential motivation or arousal states. However, given the pronounced anxiolytic-like phenotype and hyperlocomotion observed particularly in females, enhanced social approach may reflect reduced anxiety-related avoidance or increased exploratory drive rather than a selective increase in social motivation [63]. To address this limitation, sociability was evaluated using two complementary paradigms with distinct motivational and spatial demands. Although locomotion and anxiety may influence performance in the social interaction test, the three-chamber task partially dissociates social preference from general exploration by contrasting social versus non-social stimuli [34]. In this paradigm, male LDLr⁻/⁻ mice showed a consistent preference for the social chamber, supporting a genuine bias toward social stimuli at least in this subgroup, whereas females exhibited context-dependent patterns. The lack of a clear social preference in the three-chamber test in WT females likely reflects known sex- and context-dependent variability in sociability in C57BL/6 mice, particularly in paradigms using a restrained stimulus animal [67]. WT females also fail to show baseline social preference. In fact, reduced investigation of a confined conspecific may reflect diminished ethological salience rather than impaired sociability [64–66]. Importantly, this baseline pattern was consistent within the group and does not compromise genotype comparisons, as all animals were tested under identical conditions.

Although no direct clinical or preclinical studies link hyperlipidemia with social behavior differences, [47] research showed that male C57BL/6 mice on a high-fat diet, which increases cholesterol metabolism in the liver, do not exhibit social avoidance behavior. Other studies have found that lower cholesterol levels are associated with violent and antisocial behavior in humans [60] and increased criminal violence [48], primarily in men [46]. The mechanisms behind these behaviors are still unclear; however, altered dopaminergic activity has been observed in LDLr-/- mice [17], which is also implicated in the regulation of social behavior regulation.

The reduction of COMT expression in the PFC of female LDLr⁻/⁻ mice offers a compelling molecular link between cholesterol metabolism and altered emotional behavior. COMT is a key enzyme in the catabolism of catecholamines as dopamine, especially in the PFC, where dopamine transporter density is low [49, 61]. Genetic variation in COMT, alters enzymatic activity and thereby influences prefrontal dopamine levels [50, 51]. These differences have been linked to variation in social behavior through modulation of PFC-amygdala circuitry [52]. Reduced COMT activity leads to elevated extracellular dopamine, which can enhance prefrontal control over limbic structures such as the amygdala. In turn, this may decrease anxiety-related avoidance, consistent with the reduced anxiety-like profile observed in female LDLr⁻/⁻ mice. Our data therefore align with work showing that decreased COMT activity can exert anxiolytic effects depending on sex and hormonal background [62]. Sex differences are particularly relevant because estrogen downregulates COMT transcription and activity, thereby amplifying the functional impact of COMT reduction in females [58]. This interaction may explain why the anxiolytic phenotype was pronounced in LDLr⁻/⁻ females but not in males, despite both sexes showing trends toward reduced COMT levels. It also highlights how hormonal regulation of COMT intersects with genetic and metabolic backgrounds to shape behavior.

COMT-mediated regulation of prefrontal dopamine also influences social behavior, as demonstrated in both human genetic studies [53, 54] and animal models [55]. Reduced COMT expression may alter the balance between social approach and avoidance. Social interaction depends on both the rewarding value of social stimuli and the anxiety elicited by novelty or potential threat. Enhanced dopamine signaling in the PFC and striatum has been linked to changes in social reward valuation [59]. Accordingly, the present findings suggest that the LDLr-COMT interaction may represent a novel pathway through which lipid metabolism influences dopaminergic function and socio-emotional behavior. The mechanistic link between hypercholesterolemia and reduced COMT expression in female LDLr⁻/⁻ mice is likely multifactorial. Altered brain cholesterol homeostasis can modify neuronal membrane composition and fluidity, thereby affecting the localization and signaling efficiency of dopamine receptors [75, 76] and downstream regulation of dopamine-metabolizing enzymes such as COMT. In addition, cholesterol is the substrate for steroid hormone synthesis, and dysregulation of estrogen signaling, an established transcriptional regulator of COMT, may contribute to reduced COMT expression in females [77, 78]. Finally, LDLr deficiency and aberrant cholesterol accumulation may promote neuroinflammatory or oxidative processes that secondarily influence COMT levels, as altered brain cholesterol has been shown to drive microglial activation [82] microgliosis and morphological changes [83]. Reduced COMT expression in the PFC would be expected to decrease dopamine degradation, thereby enhancing dopaminergic signaling within prefrontal circuits that regulate social and emotional behavior [79–81]. The PFC acts as a central hub integrating inputs from limbic and thalamic regions to regulate social behaviors, a process heavily modulated by dopaminergic pathways [79–81]. Therefore, cholesterol-driven downregulation of COMT could lead to a localized increase in synaptic dopamine, altering the excitation-inhibition balance within these critical behavioral circuits.

Some methodological considerations should be noted. First, this study employed homozygous LDLr⁻/⁻ mice, which model a severe form of LDL receptor dysfunction that is less prevalent clinically [12]. Because heterozygous LDLr⁺/⁻ mice often do not develop marked hypercholesterolemia under standard dietary conditions due to species-specific differences in lipoprotein metabolism, homozygous deletion is required to elicit robust and reproducible phenotypes [71–74]. Accordingly, the present findings demonstrate the impact of complete LDLr loss on emotional and social behavior and should not be interpreted as a direct quantitative model of the more common heterozygous form of familial hypercholesterolemia. Second, all behavioral testing was conducted during the light phase to maintain consistency with prior work [18, 68–70]. Given that circadian timing modulates locomotor activity and emotional reactivity in nocturnal rodents, behavioral outcomes in assays such as the open field, elevated plus maze, and social interaction tests may differ if assessed during the dark phase. Finally, the estrous cycle was not monitored in female mice. As fluctuations in ovarian hormones, particularly estrogens, are known to influence dopaminergic signaling, COMT expression, and behavior [75], future studies incorporating estrous cycle tracking will be important to refine the interpretation of sex-specific effects.

Taken together, our results indicate that LDLr deficiency leads to sex-specific behavioral and molecular alterations, characterized by hyperactivity, reduced anxiety-like behavior, and enhanced sociability, particularly in females. These changes coincide with a selective reduction in COMT expression in the PFC, suggesting a dopaminergic mechanism linking cholesterol metabolism to emotional regulation. This integrative model may help reconcile the heterogeneous literature on cholesterol and mental health. While hypercholesterolemia has been variably associated with depression, aggression, or anxiety, our data suggest that the direction of behavioral change depends on the underlying neurochemical context, including sexual hormones and monoaminergic enzyme regulation. Furthermore, by identifying sociability as a relevant behavioral domain, this study expands the conceptual framework of LDLr-related phenotypes beyond cardiovascular and cognitive dimensions, introducing a new perspective on the neuropsychiatric burden of lipid metabolism disorders.

Conclusion

This study demonstrates that LDL receptor (LDLr) deficiency produces sex-dependent alterations in emotional and social behaviors, revealing that the impact of dyslipidemia extends beyond peripheral metabolism into neural domains regulating affect and sociability. Female LDLr⁻/⁻ mice exhibited hyperlocomotion, reduced anxiety-like behavior, and altered sociability, accompanied by a selective reduction in COMT expression in the PFC. These results suggest that dopaminergic dysregulation in corticolimbic circuits may underlie the behavioral consequences of LDLr deficiency. By integrating behavioral and molecular analyses, this work provides the first evidence that LDLr loss affects social behavior and identifies COMT as a potential molecular mediator linking cholesterol metabolism to monoaminergic signaling and emotional regulation. The pronounced effects in females highlight sex as a crucial biological variable in the study of neurobehavioral outcomes of metabolic dysfunction.

Collectively, our findings expand the understanding of familial hypercholesterolemia-related phenotypes beyond cardiovascular risk, revealing novel neural and behavioral dimensions of LDLr deficiency. These insights open new avenues for exploring how lipid metabolism influences neuropsychiatric vulnerability and underscore the importance of sex-specific approaches in preclinical and clinical studies of metabolic disorders.

Supplementary Information

Below is the link to the electronic supplementary material.Supplementary file1 (PDF 134 KB)Supplementary file2 (PDF 66 KB)Supplementary file3 (PDF 2239 KB)