跳到正文
原文
Frontiers in Psychiatry· Julie Zhao·· 2 小时前AI 评分36

Frontiers in Psychiatry：活动性厌食模型中的易感性与心理韧性不存在性别差异

Vulnerability and resilience to activity-based anorexia are not sex-dependent

AI 导读

一项发表于 Frontiers in Psychiatry 的小鼠研究显示，活动性厌食（ABA）模型中的易感与韧性表型不存在性别差异。研究以年轻成年雄性 C57BL/6N 小鼠复现了雌性中已发现的两种表型，两性在各表型比例及行为特征上均无差异。易感小鼠的灾难性体重下降与光周期过度跑轮相关，而韧性小鼠则通过摄食适应性变化实现体重稳定。

正文

Abstract

Introduction:

Anorexia nervosa (AN) is more prevalent in women than men, although rates in men are rising. Animal models can provide insight into whether this differential prevalence is rooted in biological mechanisms, but prior studies have yielded conflicting findings. Using the activity-based anorexia (ABA) model, we previously demonstrated that female mice exhibit distinct vulnerable and resilient phenotypes. Here, we use this phenotypic framework to investigate sex differences in ABA susceptibility.

Methods:

We tested young adult male C57BL/6N mice using the same ABA protocol used to test age-matched C57BL/6N females. Individual differences in daily bodyweight, food intake, water intake, and wheel running were analyzed and compared across sexes.

Results:

Males exhibit the same vulnerable and resilient phenotypes as females, with no sex difference in the proportion of mice exhibiting each phenotype or the repertoire of behaviors characterizing them. In both sexes, vulnerable mice exhibit catastrophic weight loss that is associated with excessive light cycle running, while resilient mice exhibit weight stabilization that is associated with adaptive changes in consumption. When food is available, vulnerable mice do not run more than resilient mice, indicating that vulnerability is not characterized by a decision to run instead of eat in either sex.

Conclusions:

Our findings indicate that there are no sex differences in ABA in C57BL/6N mice. The shared behavioral responses to starvation observed across sexes demonstrate that ABA can be used to model adaptive and maladaptive responses to food restriction in men and women.

Introduction

Anorexia nervosa (AN) is an eating disorder that has the highest mortality rate of any psychiatric disorder (, ). It is characterized by an intense fear of gaining weight, a distorted body image, and self-starvation, leading to life-threatening weight loss. Although not part of the formal diagnostic criteria, individuals with AN commonly engage in excessive exercise, which has been implicated in both the development and maintenance of the disorder (). High levels of premorbid activity increase the risk of developing AN () and prolong treatment duration (, ), while excessive exercise after discharge is associated with faster relapse (, ). Women are reported to be more susceptible to AN than men, with lifetime prevalence rates as high as 4% in women compared to 0.3% in men (), although rates in men are increasing (). This disparity may reflect sex differences in the physiological mechanisms underlying AN and different sociocultural pressures faced by men and women to be thin (, ).

Activity-based anorexia (ABA) is a widely used rodent model of AN that combines restricted access to food with unlimited access to a running wheel. Studies dating back to the 1950s have shown that rats tested under these conditions exhibit a paradoxical increase in wheel running and decrease in voluntary food intake, leading to extreme weight loss and death (, ). Investigators have long questioned whether there are sex differences in ABA. It was initially expected that females would be more vulnerable than males (), because they run more at baseline () and higher baseline running had been associated with greater ABA vulnerability (). In support of this, there are several studies reporting that vulnerability is higher in females (, –), but others report that females are actually less vulnerable than males (, ) or that there is no difference between the sexes (–). These discrepant findings may be attributable to numerous methodological differences across studies, including variations in the feeding schedule used, number of days of food restriction, and the rodent species and strain used. Perhaps most importantly, studies differed in how ABA vulnerability was determined. While some studies used weight loss and wheel running (–, , , , ), others used survival time (, ), with the latter depending on complex physiological changes underlying mortality. Furthermore, few studies identified which behaviors drive weight loss and whether this differs between the sexes.

We previously tested ABA in young adult female C57Bl/6N mice and identified vulnerable and resilient phenotypes, each of which was characterized by distinct behavioral responses and changes in body weight (–). While vulnerable mice exhibited the expected failure to eat enough to compensate for energy expenditure, we found that it was their dramatic increase in running during the light cycle that was associated with their catastrophic weight loss. In contrast, resilient mice initially lost weight, but then adapted by progressively increasing consumption when food was available and decreasing running during the dark cycle, both of which led to weight stabilization. These findings provide a new framework for evaluating susceptibility to ABA, which not only takes into account behaviors driving changes in body weight, but captures individual differences in these responses. With our analysis of phenotypes, we now have the opportunity to compare the responses of males and females in a novel way, potentially providing new insights into sex differences in ABA vulnerability.

Here, we tested ABA in young adult male C57Bl/6N mice and compared their responses to those found in females of the same age and strain. After analyzing individual differences in food intake, water intake, and wheel running, we discovered that males exhibit the same vulnerable and resilient phenotypes as females. Notably, there was no difference between sexes in the proportion of mice exhibiting each phenotype, indicating that there are no sex differences in vulnerability to ABA in this mouse line. Analysis of wheel running during the period of food availability revealed that vulnerability is not characterized by a decision to run instead of eat in either sex. Instead, vulnerability is driven by a combination of low levels of voluntary consumption resembling self-starvation and excessive running throughout the light cycle, during which food is not available.

Materials and methods

Animals

Young adult C57Bl/6N mice of both sexes were purchased at 8 weeks of age (Taconic Biosciences, Germantown, NY) and group housed (4/cage) upon arrival. They were maintained on a 12-hour light-dark schedule with free access to standard laboratory chow (Prolab Isopro 3000 5P75, WF Fisher & Son Inc., Somerville, NY) and water in a temperature-controlled room dedicated to ABA testing. Estrous cycle was never monitored in females and none of the mice had breeding experience. Experiments were conducted in accordance with NIH guidelines and were approved by the Institutional Animal Care and Use Committee of Hunter College.

ABA procedure

ABA was tested as previously described (, ) with each sex tested separately. Males (n=19) and females (n=19) were tested under identical conditions by the same experimenter who used the exact same equipment, testing room, and procedures (). The room used for ABA only contained animals being tested in the experiment and only the experimenter was given access to the room. This limited unrelated noise from stimulating wheel running. Males were all tested at the same time and females were tested in two cohorts (n=8; n=11, similar proportion of vulnerable mice in each). One week after arrival, mice were individually housed with unlimited access to food, water, and a wireless running wheel (ENV-044, Med Associates, Inc., St. Albans, VT) that transmitted wheel rotations continuously to a computer. Mice were undisturbed the next day so they could acclimate to their new housing conditions. On the following 3 days, baseline bodyweight, food intake, water intake, and wheel running were recorded (baseline days 1-3). Food was removed from all mice on baseline day 3, two hours after the onset of the dark cycle. On the following 10 days (ABA days 1-10), mice were weighed and water intake was recorded immediately before the onset of the dark cycle. Once the lights turned off, large pre-weighed food pellets were immediately placed in the overhead food bin and mice were given 2 hours of unlimited access to food. At the end of this feeding window, the number of pellets retrieved was the same as what was provided, confirming that none fell into the cage. Mice were 68 days old on ABA day 1. If mice lost at least 25% of their baseline bodyweight (determined on baseline day 3), they were removed from the experiment and characterized as “vulnerable.” Those that did not require removal were characterized as “resilient.”

Statistical analyses

Data were analyzed with two-way ANOVA and Bonferroni’s multiple comparisons test, and Pearson correlation using GraphPad prism software (GraphPad, San Diego, CA). A repeated measures ANOVA (RM ANOVA) was used when the same mice were compared at 2 time-points (baseline vs. ABA). Survival curves were analyzed using the log-rank (Mantel-Cox) test. For analyses across days of food restriction, mice were removed at different time point resulting in data sets with missing values. These data sets were analyzed with the linear mixed-effects (LME) model, in which experimental day and hours were treated as continuous variables, effects of interest (e.g., sex and phenotype) were treated as fixed effects, and individual mice were treated as a random effect (see Supplementary Table 1 for a description of each model). Repeated measures correlations were calculated using the rmcorr package in R and significance was accepted for p<0.05. The female dataset was obtained from a previous ABA study and the female weight loss data has been published (). All other analyses of this female dataset, all male data, and all sex comparisons reported here are new.

Results

Males are as vulnerable to activity-based anorexia as females

At baseline, males (23.35 ± 0.36g) weighed significantly more than females (18.78 ± 0.30g) (t(36) = 9.68, p < 0.0001). To evaluate whether there are sex differences in weight loss during ABA, we compared the bodyweights of males and females (calculated as percent baseline bodyweight) across days of food restriction. We found that both sexes lost substantial weight during the first four days of restriction, but then exhibited weight stabilization and a modest increase in weight by day 10. Analysis of all 10 days indicated no sex difference in this response (Figure 1A, LME model, sex x day, F(1, 32) = 0.50, p=0.48; sex, F(1, 36) = 1.78, p = 0.19). In both sexes, we found that a subset of vulnerable animals required removal from the experiment, while a subset of resilient mice remained the entire 10 days (Figure 1B). Interestingly, there were no sex differences in survival (hazard ratio, 1.26; 95% CI, 0.54-3.18; p = 0.54) or the proportion of mice exhibiting each phenotype (Figure 1C, χ2(1) = 0.99, p = 0.32), indicating that ABA risk is similar in both sexes. Like females (), we found that vulnerability in males was marked by a progressive, persistent loss of bodyweight leading to removal from the experiment, while resilience was characterized by initial weight loss followed by weight stabilization or even weight gain (Figure 1D). These dynamic changes in bodyweight were further evaluated by analyzing daily change in bodyweight in both sexes. Regardless of sex, vulnerable and resilient mice initially lost weight at a similar rate, but resilient mice gradually stopped losing weight by the fourth day of food restriction, while weight loss was constant in the vulnerable group (Figure 1E, LME model, phenotype x day, F(1, 89) = 0.59, p = 0.44; phenotype, F(1, 39) = 10.45, p = 0.002; sex, F(1, 39) = 1.01, p = 0.32). Interestingly, the rate of weight loss began to differ between the two phenotypes as early as the third day of food restriction. A two-way ANOVA on average daily weight loss confirmed that there were no sex differences within each phenotype (Figure 1F, sex x phenotype, F(1, 34) = 3.45, p = 0.07; phenotype, F(1, 34) = 109.9, p<0.0001). Together, these results indicate that males exhibit the same two phenotypes that we previously reported in females and that there is no sex difference in ABA vulnerability or resilience.

Figure 1

Resilient males adapt food intake

We previously showed that ABA resilience in females is characterized by a progressive increase in food intake (). Here, we evaluated whether the same is true of males. After comparing vulnerable and resilient males in baseline consumption and consumption during ABA (Figure 2A, timepoint x phenotype, F(1, 17) = 24.69, p<0.001), we found that resilient mice did indeed eat more than vulnerable mice during ABA testing (ABA-R vs. ABA-V, Bonferroni, p = 0.0005). Interestingly, there was a trend for resilient males to eat less than vulnerable males at baseline, but this did not reach significance (Baseline-R vs. Baseline-V, Bonferroni, p = 0.058). Similarly, there was no correlation between food intake at baseline and the number of days males were tested in the experiment (Pearson r = -0.07, p = 0.76), indicating that consumption prior to ABA testing is not predictive of phenotype. Analysis of food intake across days of food restriction revealed the resilient males exhibited a steady increase in consumption over time (Figure 2B, paired t-test, day 1 vs. day 10, t(5) = 3.75, p = 0.01), an adaptive response that was not found in vulnerable males (Figure 2B, LME model, phenotype x day, F(1, 14) = 4.23, p = 0.058; paired t-test, day 1 vs. last ABA day, t(12) = 1.81, p = 0.10). Consistent with our previous reports in females (), only vulnerable males ate less on the day of removal than the previous days of food restriction (Figure 2C, time point x phenotype, F(1, 17) = 35.57, p<0.0001, vulnerable before removal vs. vulnerable at removal: Bonferroni, p<0.0001). Finally, changes in bodyweight correlated with consumption in resilient males only (Figure 2D, vulnerable: rrm=-0.23, p = 0.13; resilient: rrm=0.66, p<0.0001). Collectively, these results mirror our previous findings in females and indicate that resilience in both sexes is marked by the same adaptive changes in food intake.

Figure 2

Weight loss is not driven by reduced water intake in either sex

Food intake often correlates with water intake (), which also affects bodyweight (). Here we investigated the contribution of water intake to changes in bodyweight during ABA. In both males and females, water intake did not differ between phenotypes at baseline or during ABA (Supplementary Figure 1, males, phenotype x timepoint, F(1, 17) = 8.06, p = 0.01; vulnerable/baseline males vs. resilient/baseline males: Bonferroni, p>0.05; vulnerable/ABA males vs. resilient/ABA males, Bonferroni, p>0.05; Supplementary Figure 1, females, phenotype x timepoint, F(1, 17) = 2.42, p = 0.14; phenotype, F(1, 17) = 3.48, p = 0.08). Closer examination of ABA testing revealed that males of both phenotypes drank less on the day of removal than previous days of ABA testing (Supplementary Figure 1, phenotype x timepoint, F(1, 17) = 0.50, p = 0.49; timepoint, F(1, 17) = 5.25, p = 0.035). In contrast, in females, only the vulnerable phenotype exhibited this response (Supplementary Figure 1, phenotype x timepoint, F(1, 17) = 12.73, p = 0.0024; vulnerable/before vs. vulnerable/at removal: Bonferroni, p = 0.02). However, water intake did not correlate with change in bodyweight in any of the groups tested (Supplementary Figure 1, vulnerable male: rrm= 0.07, p = 0.65, resilient male: rrm=0.15, p = 0.32; Supplementary Figure 1, vulnerable female: rrm= -0.13, p = 0.50, resilient female: rrm=0.01, p = 0.94), indicating that weight loss during ABA is not caused by a reduction in water consumption.

Vulnerable males exhibit maladaptive light cycle running

We next investigated whether ABA vulnerability in males is associated with the same maladaptive changes in wheel running found in females (). At baseline, there were no significant differences between phenotypes in amount of running (Figure 3A, LME model: group, F(1, 32) = 3.77, p = 0.06) or its circadian distribution (Figure 3B1, LME model: phenotype x hour, F(1, 1347) = 1.36, p = 0.24). We also found no correlation between baseline running and the number of days males were tested in the experiment (baseline light cycle running: Pearson r = -0.23, p = 0.35; baseline dark cycle running: Pearson r = 0.33, p = 0.17), further indicating that baseline running is not predictive of phenotype. During ABA, food restriction elicited a similar increase in total running in both phenotypes (Figure 3C, timepoint, F(1, 17) = 49.17, p<0.0001; timepoint x phenotype, F(1, 17) = 3.64, p = 0.07). However, when light cycle running was analyzed separately, vulnerable males, like vulnerable females (), exhibited a more dramatic increase in running than resilient mice of the same sex (Figure 3B2, LME model: phenotype x hour, F(1, 3147) = 9.70, p = 0.0018; Figures 3D, H1, timepoint x phenotype, F(1, 17) = 5.47, p = 0.03, Bonferroni, ABA/resilient vs. ABA/vulnerable, p = 0.001). In contrast, dark cycle running increased similarly in both phenotypes during ABA testing (Figures 3E, H2, timepoint, F(1, 17) = 20.52, p = 0.0003; timepoint x phenotype, F(1, 17) = 1.12, p = 0.31). Like females, the increase in light cycle running was more abrupt in vulnerable than resilient males, as indicated by a bigger peak change in running within a 24-hour period (Figure 3F, phenotype x cycle, F(1, 34) = 5.55, p = 0.02, Bonferroni, light/resilient vs. light/vulnerable, p = 0.001), which preceded removal from the experiment by 1–2 days for most animals tested (Figure 3G). Light cycle running positively correlated with weight loss in vulnerable males (Figure 3I, vulnerable male: rrm= -0.29, p = 0.025), with effects that approached significance in resilient males (rrm=0.27, p = 0.05). Dark cycle running positively correlated with weight loss in both phenotypes (Figure 3J, vulnerable male: rrm= -0.56, p<0.0001; resilient male: rrm= -0.41, p = 0.002), effects that were also found in females. These results demonstrate that vulnerability to ABA is characterized by the same maladaptive changes in light cycle running in both sexes.

Figure 3

ABA vulnerability is not associated with more running during food availability

Previous studies report that animals choose to run instead of eat during ABA (, ). To evaluate whether there are differences between phenotypes in this response, we measured running during the two-hour window of food availability. We found that vulnerable females actually run less than resilient females and less than they did at baseline (Figure 4B, timepoint x phenotype, F(1, 17) = 11.19, p = 0.004, Bonferroni, ABA/resilient vs. ABA/vulnerable, p = 0.02, vulnerable/baseline vs. vulnerable/ABA, p<0.0001). Vulnerable males exhibited a similar pattern, although the phenotypes did not differ significantly from each other (Figure 4A, timepoint x phenotype, F(1, 17) = 1.88, p = 0.19). This occurred despite females exhibiting 2–3 times more baseline running in a 24-hour period than males (Figure 4C, sex, F(1, 34) = 54.33, p<0.0001). The reduction in wheel running during the 2-hour feeding window primarily occurred during the end stages of weight loss. Indeed, vulnerable mice of both sexes ran less than same-sex resilient mice on the day they were removed from the experiment, but not earlier in ABA testing (Figure 4D, timepoint x phenotype, F(1, 17) = 11.68, p = 0.003; Bonferroni, at removal/resilient male vs. at removal/vulnerable male, p = 0.0004; Figure 4E, timepoint x phenotype, F(1, 16) = 6.12, p = 0.03, Bonferroni, at removal/resilient female vs. at removal/vulnerable female, p = 0.0004). Finally, there was no correlation between running during the feeding window and daily weight loss in vulnerable females (Figure 4H, rrm= -0.13, p = 0.51), with effects in males that approached significance (Figure 4G, rrm= -0.29, p = 0.051). In resilient mice, weight gain did negatively correlate with running during this time window in both sexes (Figure 4G, resilient male: rrm= -0.44, p < 0.01; Figure 4H, resilient female: rm= -0.45, p < 0.0001), results that are in line with the adaptive decrease in dark cycle running previously described (). Interestingly, there were no differences between phenotypes during the 2-hour period prior to food availability (Figure 4F, sex x phenotype, F(1, 34) = 0.30, p = 0.59; sex, F(1, 34) = 3.80, p = 0.0595; phenotype, F(1, 34) = 0.37, p = 0.55) or the 2-hour period after feeding (Figure 4I, sex x phenotype, F(1, 34) = 0.03, p = 0.85, sex, F(1, 34) = 9.17, p = 0.005; phenotype, F(1, 34) = 0.03, p = 0.87) in males or females. Together, these findings reveal that vulnerable mice do not run more than resilient mice when food is available, indicating that ABA vulnerability is not characterized by a decision to run instead of eat in either sex.

Figure 4

Discussion

We tested ABA in C57Bl/6N male mice and found that they demonstrate the same distinct vulnerable and resilient phenotypes as age-matched C57Bl/6N females. We found no sex difference in the repertoire of behaviors characterizing each phenotype or the proportion of mice exhibiting vulnerability or resilience. Phenotypes were distinguishable based on daily changes in bodyweight that were associated with either excessive light cycle running (vulnerable phenotype) or increased food intake (resilient phenotype), none of which varied by sex. Together, our findings indicate that there are no sex differences in ABA, suggesting that ABA may be used to model adaptive and maladaptive responses of both men and women once they reach the same level of weight loss.

Our findings are consistent with ABA studies conducted in Wistar rats reporting that the percentage of baseline bodyweight lost across days of food restriction did not differ between males and females (, ). In studies that allowed rats to die over the course of ABA testing, there was also no sex difference in survival duration (, ) but see (). Interestingly, one of these studies found that female rats died earlier than males, but concluded that there was no sex effect after beginning bodyweight was included as a covariate in the analysis (). To control for sex differences in initial bodyweight, another group included older female Wistar rats with weights that were comparable to those of males (). Older females lost weight more slowly and met removal criterion later than younger mice of both sexes, consistent with the known protective effects of age on ABA vulnerability (, ). Like our study in mice, there were no differences between same-age male and female rats, despite males weighing more at baseline (). Another study used male Sprague-Dawley rats that were only slightly younger than females, in which case the initial weight difference was minimized, but still significant. Contrary to the findings described above, those younger males required removal from the experiment earlier than females (), which again may have reflected an effect of age. Furthermore, that study excluded rats that never required removal (i.e., resilient) from all analyses (), which may have affected experimental outcome. Our analyses of both vulnerable and resilient mice indicate that starting weight, which was higher in males of both phenotypes, is unrelated to mean daily weight loss or survival in the experiment.

Historically, ABA has been tested primarily in rats, resulting in far fewer mouse studies investigating sex differences. We are aware of only two sex difference studies that used C57Bl/6 mice, both of which progressively limited food access, which is different from the fixed feeding schedule we used. Interestingly, one study found that C57Bl/6 males lost more weight than females (), indicating higher ABA vulnerability in males, while the other concluded the opposite, with C57Bl/6 males surviving in the model longer than females (). Notably, both challenged mice in ways that were fundamentally different from our study. While they allowed mice to gradually adapt to a restricted feeding schedule, we evaluated responses to a sudden and consistent change in food access. They also gave mice more time to eat, with food access that was eventually limited to 3 hours compared to the 2-hour feeding window we provided. In addition, they removed mice from the experiment once they lost 20% of their baseline weight for 2–3 consecutive days (, ), which was less strict than our removal criterion (> 25%). If we had used their criterion, some of our resilient mice that lost 20% of their baseline weight would have been removed before they had a chance to show weight stabilization. While these methodological differences may account for our inability to replicate either of the previously reported sex differences in this strain, it remains unclear why those previous findings were the opposite of each other. Interestingly, when ABA was tested in Balb/cJ mice, which are more anxious than the C57Bl/6NCrl strain (), males were found to survive longer in the model than females (). Such findings highlight the role of mouse strain in ABA vulnerability (, –) and implicate a potential sex difference in how anxiety might affect ABA.

Consistent with numerous reports in mice (, ) and rats (, ), we found that females run significantly more than males before food is restricted (i.e., baseline) and during ABA testing. These baseline differences can affect the interpretation of sex differences during ABA, where higher running in females could reflect a stronger effect of food restriction and weight loss on activity or a sex difference that always existed. Given that more baseline running has been associated with stronger ABA susceptibility (, ), the expectation has been that females would be more vulnerable to ABA than males (). However, our findings demonstrate that there is no sex difference in ABA vulnerability or resilience and that mice of both phenotypes exhibit similar levels of running at baseline. In one mouse study, the association between baseline running and weight loss was based on 4 days of food restriction (), which is shorter than our 10-day protocol. It is possible that baseline running predicts early changes in bodyweight that are later corrected in resilient mice, potentially accounting for why baseline running did not differ between phenotypes. Indeed, we show that resilient mice stopped losing weight after the fourth day of food restriction, which is after data collection ended in that study.

Finally, we analyzed running during food availability to evaluate whether ABA vulnerability is characterized by a decision to run instead of eat (–). However, we found that, if anything, vulnerable mice run less than resilient mice during this time window, findings that are in line with a previous study reporting decreased running during feeding in both sexes (). Furthermore, the phenotypes did not differ in how much they ran 2 hours before or after food availability. Instead, we found that excessive running throughout the light cycle played a key role in driving weight loss in vulnerable mice, effects evident in both sexes.

Based on known sex differences in the prevalence of AN, the majority of AN studies have been conducted in women only. Similarly, ABA studies frequently use female rodents, with much less is known about the cellular and molecular basis of ABA in males (). However, the rates of eating disorders in males have been increasing () and are higher than previously thought (). A comparison of AN in men and women reveal that core symptoms, such as patterns of food restriction, are similar in both sexes (), but food restriction is often initiated for different reasons. While women are focused on being thin, men tend to be more concerned with muscle definition (, , 46) and men engage in more excessive exercise (47). Men are also less likely to seek treatment than women when they experience comparable levels of problematic eating behaviors (47), potentially skewing prevalence data. Our ABA findings suggest that once sufficient weight loss occurs, the likelihood of expressing an adaptive (i.e., resilient) or maladaptive (i.e., vulnerable) response is not sexually dimorphic. The findings from the Minnesota Starvation Study in which men developed symptomology similar to AN are consistent with this possibility (48). Future ABA studies identifying substrates underlying vulnerability and resilience may provide important insight into the etiology and maintenance of AN in both sexes. Notably, investigation of factors mediating ABA resilience, defined by maintenance of bodyweight rather than slower decline of bodyweight, may contribute importantly to our understanding of factors protecting individuals who diet and exercise from developing AN. Such information may not only lead to the development of novel treatment options, but the identification of biomarkers for early diagnosis in both sexes.

Statements

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

The animal study was approved by Institutional Animal Care and Use Committee of Hunter College. The study was conducted in accordance with the local legislation and institutional requirements.

Author contributions

JZ: Formal analysis, Writing – original draft, Writing – review & editing. JB: Data curation, Formal analysis, Funding acquisition, Writing – review & editing. NB: Data curation, Formal analysis, Funding acquisition, Writing – review & editing, Conceptualization, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by a Klarman Family Foundation Eating Disorders Grant (JB and NB) and PSC-CUNY Awards jointly funded by the Professional Staff Congress and The City University of New York (JB and NB).

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpsyt.2026.1897501/full#supplementary-material

References

Keywords

activity-based anorexia, anorexia nervosa, exercise, food restriction, mouse models, phenotypes, resilience, sex differences

Citation

Zhao J, Beeler JA and Burghardt NS (2026) Vulnerability and resilience to activity-based anorexia are not sex-dependent. Front. Psychiatry 17:1897501. doi: 10.3389/fpsyt.2026.1897501

Received

01 June 2026

Revised

19 August 2026

Accepted

07 September 2026

Published

02 October 2026

Volume

17 - 2026

Updates

Copyright

© 2026 Zhao, Beeler and Burghardt.

This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Nesha S. Burghardt, nb844@hunter.cuny.edu

Disclaimer

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

来源：Frontiers in Psychiatry · frontiersin.org

#研究发布#神经科学#焦虑抑郁#Frontiers

猜你喜欢