Growth Hormone Secretagogue Receptor (GHSR) Signaling in the Ventral Tegmental Area (VTA) Mediates Feeding Produced by Chronic Social Defeat Stress in Male Mice

Ghrelin, a hormone secreted by the stomach, binds to the growth hormone secretagogue receptor (GHSR) in various brain regions to produce a number of behavioral effects that include increased feeding motivation. During social defeat stress, ghrelin levels rise in correlation with increased feeding and potentially play a role in attenuating the anxiogenic effects of social defeat. One region implicated in the feeding effects of ghrelin is the ventral tegmental area (VTA), a region implicated in reward seeking behaviors, and linked to social defeat in mice. Here we examined the role of GHSR signaling in the VTA in feeding behavior in mice exposed to social defeat stress. Male C57BL/J6 mice that were socially defeated once daily for 3 weeks ate more, had higher plasma ghrelin level and increased GHSR expression in the VTA compared to non ‐ stressed mice. Socially defeated GHSR KO mice failed to increase their caloric intake in response to this stressor but rescue of GHSR expression in the VTA restored feeding responses. Finally, we pharmacologically blocked VTA GHSR signalling with JMV2959 infused via an indwelling VTA cannula connected to a minipump. Vehicle ‐ treated mice increased their caloric intake during social defeat, but JMV2959 ‐ infusions attenuated feeding responses and increased anxiety ‐ like behaviors. The data suggest that GHSR signalling in the VTA is critical for the increases in appetite observed during chronic social defeat stress. Furthermore, these data support the idea that GHSR signaling in the VTA may also have anxiolytic effects, and blocking GHSR in this region may result in an anxiety ‐ like phenotype.


Introduction
In mammals the stress response is associated with several physiological and behavioral changes that include alterations in food intake, changes in energy expenditure and overall metabolic adaptations that help the organism return to homeostasis.Generally, acute stressors are associated with decreases in food intake and rapid increases in energy expenditure driven by the rapid utilization of carbohydrates as a source of nutrients.Repeated social stressors, however, can lead to a chronic metabolic imbalance associated with increased food intake and altered metabolic regulation that can lead to obesity and insulin resistance (Patterson and Abizaid, 2013).Furthermore, availability of carbohydrates in the form of sucrose in drinking water, or as part of a high calorie diet, appear to be important in decreasing the activity of the hypothalamic pituitary adrenal (HPA) axis, potentially being a mechanism for coping with the stressors (Dallman et al., 2003).Over time, however, stress-induced overconsumption of foods rich in carbohydrates could ultimately lead to obesity and metabolic syndrome.One paradigm that reflect this is the chronic social defeat stress paradigm, a paradigm that takes advantage of male-male aggression, and one that has been successfully used to examine the effects of repeated social stress in male rodents including the effects of stress on ghrelin secretion (Krishnan et al., 2007;Lutter et al., 2008;Patterson et al., 2013).
Ghrelin, a hormone secreted by the stomach and associated with increased caloric intake, plays an important role in the metabolic adaptations required by the stress response.Indeed, the secretion of ghrelin is increased after acute psychogenic stressors potentially important for the mobilization of carbohydrates to meet the energetic demands posed by the stressor (Abizaid, 2019).Release of ghrelin under exposure to chronic unpredictable stressors or psychosocial stressors like social defeat, is also elevated and remains elevated for several days after animals are removed from the social defeat paradigm suggesting that ghrelin may be required as part of the metabolic processes facilitating carbohydrate use to meet the requirements of the stress response, and to increase caloric intake to replenish carbohydrate stores when the stressor is removed (Lutter et al., 2008;Chuang et al., 2011;Patterson et al., 2013).These effects are less pronounced in mice lacking ghrelin receptor (GHSR KO mice) suggesting that these metabolic effects are mediated in part by central ghrelin receptor signaling (Patterson et al., 2013).
One way in which stress induced ghrelin could increase feeding and adiposity could be through its action on the HPA axis.Indeed, ghrelin can stimulate the release of adrenocorticotropin hormone (ACTH) and corticosterone via central mechanisms that include increasing corticotropin releasing hormone (CRH) from the hypothalamic paraventricular nucleus (Asakawa et al., 2001;Stevanović et al., 2007).This relationship, however, is not as simple as suggested by the data from the studies cited above.For instance, chronic social defeat increases caloric intake in WT mice but not in GHSR KO mice, despite both showing stress induced increases in corticosterone (Patterson et al., 2013).In rats, the secretion of ghrelin is increased by adrenalectomy, and feeding responses to centrally administered ghrelin are not affected by this procedure (Proulx et al., 2005).In fact, data point to ghrelin as an important hormone in maintaining allostasis in the HPA axis.Male and female GHSR KO mice show exacerbated metabolic and behavioral responses to chronic corticosterone treatment (Hay et al., 2020;Silver et al., 2021).Moreover, ghrelin KO mice show increased ACTH secretion, and decreased adrenal sensitivity to ACTH at the level of the adrenal glands, in correlation with behavioral coping responses linked to anxiety (Spencer et al., 2012).Similarly, GHSR KO mice show increased depressive-like behaviours following chronic social defeat (Lutter et al., 2008).In all, these studies suggest that ghrelin is critical for the metabolic and hormonal adaptations required by the stress response.Thus, ghrelin ultimately attenuates some of the negative consequences associated with chronic social defeat stress, including changes in behavioral coping strategies that result from a stress induced dysregulated HPA axis.
Stress induced eating has been hypothesised to be partially driven by increased activation of the brain circuits regulating motivational and emotional states in the brain.Indeed, ghrelin increases feeding via activation of the GHSR, which is expressed in many regions of the brain that are recruited by the stress response, including the ventral tegmental area (VTA) (Cao et al., 2010;Holly et al., 2015).In the VTA, a central hub of the reward system, GHSR expression is found in about 50-60% of dopaminergic neurons (Abizaid et al., 2006;Skov et al., 2017).Importantly, ghrelin administration into the VTA stimulates dopaminergic neurons and their release of dopamine onto regions like the nucleus accumbens (Abizaid et al., 2006).Moreover, when ghrelin infusion into the VTA increases the number of lever presses rats are willing to do to obtain palatable food pellets, an effect that is blunted by the administration of the GHSR antagonist, D-Lys-GHRP6 (King et al., 2011).Male GHSR KO mice show deficits in forming conditioned place preferences to a context associated with palatable foods (Chuang et al., 2011), while mice generated to express the GHSR exclusively in dopaminergic neurons show increased intake of a palatable diet when satiated (Chuang et al., 2011).This suggests that, within dopamine cells throughout the brain, the GHSR plays a critical role in promoting ingestive behaviors towards palatable and rewarding diets.How GHSR-expressing cells exclusively in the VTA contributes to feeding behaviors in response to stress has not yet been examined.
In general, ghrelin secretion follows a circadian pattern where, in rodents, peak ghrelin secretion occurs just before the onset of the dark phase of the light/dark cycle (Blum et al., 2009).In humans, mice, and rats, ghrelin is also secreted in anticipation of scheduled meals and in anticipation of meaningful stimuli like an imminent stressor or in response to cues that predict a reinforcer like cocaine (Cummings et al., 2001;Drazen et al., 2006;Davis et al., 2007).One would expect that chronic social defeat would be a meaningful stimulus and would therefore also be associated with higher ghrelin secretion during the early phase of the light/dark cycle when ghrelin is secreted at low levels in non-stressed mice.
In the present study, we examined the role of GHSR in the VTA in mediating the feeding and behavioral coping responses of male mice exposed to chronic social defeat.Overall, our results show that chronic social defeat increases GHSR expression in the VTA, and that GHSR signalling in this region mediates appetite in response to chronic social defeat stress.

Animals
All procedures described in this project were approved by the Carleton University Animal Care Committee and followed the guidelines of the Canadian Council of Animal Care.Adult male C57BL/J6 mice from Charles Rivers farms (St.Constant, Quebec, Canada) were used as the experimental subjects in experiment 1 and 3.In experiment 2, adult male GHSR wildtype (WT) and transgenic GHSR null mice were bred in our facility from a stock that were a gift from Dr. Jeffrey Zigman at the University of Texas Southwestern medical center, in Dallas, Texas.The GHSR null mice were created via insertion of a loxP flanked transcriptional blocking cassette into the putative promoter region ultimately blocking Ghsr gene expression.This loxP flanked transcriptional blocking cassette can be removed by the enzyme CRErecombinase allowing for transcriptional rescue.The metabolic phenotype of these mice has been previously described (Zigman et al., 2005).Male CD-1 retired breeder mice purchased also from Charles Rivers farms, weighing 45-50 g.These mice served as the resident bully mice for the stressor.All mice were individually housed and allowed to acclimate for a minimum of 48 h prior to the onset of the study.All mice were provided with ad lib access to a standard chow (67% calories from carbohydrates, 13% calories from fat, 20% calories from protein, 2.9 kcal/g) diet and water for the duration of the experiment.In experiment 1 the mice were also provided with restricted access to a high fat diet (HFD) (60% calories from fat, 20% calories from protein, 20% calories from carbohydrates, 5.1 kcal/g), presented for 4-h daily.All mice were housed in a temperature and humidity-controlled vivarium, on a 12-h light-dark cycle, with lights on at 8:00am.Body weight and food intake of the experimental mice was recorded every 24 h.Prior to any experimental manipulations, mice underwent a 7-14day baseline period, during which body weight and food intake were measured daily.

Stereotaxic surgeries
In experiments 2 and 3 mice underwent stereotaxic surgeries for the delivery of adeno-associated viral vectors (AAV) or a GHSR antagonist.Mice were anesthetized with a mixture of isoflurane or oxygen (5% induction, 2% maintenance rate).After confirming the animal was anesthetized, the head was shaved and cleaned surrounding the surgical site.Meloxicam slow release was given as an analgesic at a dose of 2 mg/kg.Once the head was mounted onto the stereotaxic apparatus using ear bars an incision was made along the midline.Bregma was located and the coordinates of the VTA (AP: −2.95, ML: ± 0.7, DV: −4.55) were located.In experiment 2, mice received a bilateral infusion either the control AAV (pAAV.hSyn.eGFP.WPRE.bGH, 5 × 10 12 vg/mL, Addgene, catalogue #105539) or active CRE AAV (pENN.AAV.hSyn.HI.eGFP-Cre.WPRE.SV40, 1 × 10 13 vg/mL, Addgene catalogue #105540) at a rate of 75 nL/min, delivering 0.5 µL/side.The infuser was then removed, and the skin sutured.In experiment 3, a cannula was inserted into the VTA and either vehicle or ghrelin antagonist JMV2959 were delivered at a rate of 10 nM/day via osmotic minipumps implanted subcutaneously.Once the cannula and the minipumps were in place, the incisions were sutured.In both experiments mice recovered for 1 week prior to undergoing further experimentation.

Chronic social defeat paradigm
The animals assigned to stress groups were moved to a separate room from the non-stress controls.Here, the mice were placed into the cage of sexually experienced male CD-1 mouse and allowed to interact with this mouse until the CD-1 mouse became physically aggressive.At that time, mice were separated but experimental mice remained in the cage separated from the CD-1 mouse by a transparent mesh divider that prevented physical contact but allowed olfactory and visual contact between the mice.The experimental mice were housed opposite the CD-1 mouse for 24 h, at which time the experimental mouse would be exposed to a novel CD-1 mouse and the interaction repeated.This occurred daily at the same time for a period of 21 days.
Experiment 1.-Effects of chronic social defeat on ghrelin secretion and ghsr mRNA expression at the onset of the light or dark phase of the circadian cycle C57BL/J6 Mice (N = 40) were assigned to control or stress conditions where mice were exposed to 21-days of the chronic social defeat paradigm.Mice were then euthanized via rapid decapitation in the morning (8:00am) or in the evening (6:00 pm) of the day following their last stressor.Trunk blood was collected to measure plasm ghrelin concentrations and brains were extracted, flash frozen, and stored at −80 °C.
Reverse Transcriptase Quantitative Polymerase Chain Reaction (RT-qPCR).-Flash-frozenbrains were thawed and tissue micro punches were taken from 1 mm thick brain sections that contained the prefrontal cortex, medio-basal hypothalamus, ventral tegmental area, and hippocampus.Samples were processed for quantification ghsr, nr3c1 (glucocorticoid receptor), and β-actin mRNA expression.To do this, RNA was extracted using TRIzol (Life Technologies) as detailed in the manufacturers protocol.RNA integrity was confirmed using agarose gel electrophoresis (1% w/v agarose gel with ethidium bromide).Purity and concentration of mRNA were determined using a Nanodrop-spectrophotometer (Thermo Scientific).A A260/280 ratio between 1.8 and 2.2 was considered acceptable.RNA was reverse transcribed with iScript Supermix (Bio-Rad).cDNA samples were mixed with DEPC water, primers, and SYBR Green Super Mix (Bio-Rad) in duplicate on a 96-well plate.RT-qPCR was conducted on cDNA samples using a CFX Connect Real-Time PCR machine (Bio-Rad).Primer sequences are displayed in Table 1 (see Table 1).Annealing temperatures were tested using a temperature gradient and the temperature resulting in a reaction efficiency between 90% and 110% was selected.Relative mRNA expression was determined using the 2 −ΔΔCT method (Schmittgen and Livak, 2008).
Experiment 2.-Effects of GHSR rescue in the VTA on feeding in socially defeated GHSR KO mice Following the baseline period, WT and GHSR KO mice underwent stereotaxic surgery to receive bilateral infusions of an AAV vector to express GFP or CRE-GFP as described above.Following 7 days of recovery the mice were then exposed to the chronic social defeat paradigm.Mice were sacrificed by perfusion 24 h after the last bout of social defeat via intracardial perfusion of saline followed by 4% paraformaldehyde in 0.1 M phosphate buffered saline.Brains were kept in the same fixative for 48 h, then transferred to 30% sucrose solution (at 4 °C) until sectioned in a cryostat.Tissues were sliced 40 µm thick, using a cryostat, and stored in cryoprotectant at −20 °C.GFP expression from the AAV was used to examine accuracy of cannula placements into the VTA.Of the n = 92 mice included in this study, n = 50 contained accurate GFP expression in the VTA and were included in analyses.Mice that exhibited GFP expression in regions outside of the VTA (n = 42) were removed from analysis.
Fluorescent In-situ Hybridization.-To determine that GFP expression was indicative of true GHSR rescue, a RNAscope multiplex fluorescent V2 assay (ACD, Catalogue #323100 & 323270) was used to detect Ghsr (ACD #426141) and Th (ACD #431921) mRNA and was performed per manufacturer's instructions.Slides were stored in the dark at 4 °C until imaging.Fluorescent images were obtained using a Nikon C2Si confocal and Nikon Eclipse Ti2-E microscope at 20× magnification.Images were processed using NIS-Elements Imaging Software and ImageJ.
Experiment 3.-Effects of blocking GHSR in the VTA on feeding during social defeat stress Following a baseline period, C57BL/J6 mice underwent stereotaxic surgery to implant a cannula aimed unilaterally at the VTA.Cannulae were connected to osmotic with PE20 tubing to osmotic minipumps (ALZET, 1004) delivering saline or continuous dose of JMV2959 at a rate of 10 nM/day.Mice recovered for 7 days before being exposed to the chronic social defeat paradigm as described above.On the day after the last stressor, mice were killed by an overdose of sodium pentobarbitol, and perfused with saline followed by 4% paraformaldehyde.Brains were kept in the same fixative for 48 h and stored in 30% sucrose solution (at 4 °C) until sectioned in a cryostat.All brains were sliced to obtain 50 μm sections of the VTA that were mounted on gel coated slides and stained with Cresyl violet to confirm cannula placements.Of the n = 40 mice included in this study, n = 27 contained cannula placements that hit the VTA and were included in analyses.
Behavioral Testing.-During the last week of the stress paradigm mice were tested for differences in behavioral coping to novel situations.These included behavioral expression of anxiety-like behaviors in the open field and social approach tests.To conduct these tests mice were placed in an open field arena (50 × 40 cm) box with and allowed to explore freely for 5-minutes while the experimenter videorecorded behavior.Mice were returned to their home cage for 5-minutes.A novel male C57BL/J6 mouse was then placed in a perforated plexiglass cylinder at one end of the arena.The experimental mouse was then returned to the corner of the testing open field box and allowed to explore and interact with the novel con-specific for 5-minutes.Mice were removed from the cage and the cage was cleaned before the next mouse.Measures collected in the open field portion of the test included the time spent in the periphery of the testing box as a measure of anxiety like behavior and the number of lines crossed as a measure of locomotor activity.Measures collected in the social approach test included the latency to approach the stranger con-specific, the time spent investigating (sniffing) the con-specific, the number of sniffs, and the amount of time spent on the corner of the testing box observing the stranger con-specific as done in previous work (Park et al., 2021).All behaviors were recorded and later scored by a researcher blind to the treatment of each animal.

Statistical analyses
All data are expressed as mean +/− SEM.Normality of variance assumption was assessed using the Shapiro-Wilk test, with a p > 0.05 considered to be normally distributed and justified the use of parametric analyses.Experiment 1 data was analyzed using t-tests.Analyses of data that were not normally distributed were analyzed using non-parametric tests (Mann-Whitney test).Experiment 2 data were analyzed using 2-factor ANOVA, with group (non-stress vs stress) as one level, and genotype (GHSR +/+ , GHSR −/− , GHSR VTA+/+ ) as the second level.Experiment 3 data were analyzed using 2-factor ANOVA, with group (non-stress vs stress) as one level, and treatment (vehicle vs JMV-2959) as the second level.Analyses that failed to pass the assumptions were analyzed executing a log transformation.Statistical significance was set at p < 0.05 with *p < 0.05; **p < 0.01; ***p < 0.001.Statistical analyses were run using Prism GraphPad software.

Results
Experiment 1: Chronic social defeat stress increases caloric intake, plasma ghrelin and ghsr mRNA expression in the ventral tegmental area As shown here, mice exposed to chronic social defeat increased their total caloric intake throughout the stress period compared to control mice (t(38) = 2.182, p = 0.035, η 2 = 0.69; see Fig. 1(B)).This increase was related to an increase in caloric intake from the chow, but not HFD, as shown by a significant increase in the daily intake of chow (t(38) = 2.881, p = 0.006, η 2 = 0.91), and a significant decrease in the intake of the HFD in socially defeated mice (U = 117, p = 0.024, η 2 = 0.65; see Fig. 1(C-D)).In addition, mice that experienced social defeat stress showed an increase in weight gain that was borderline significant (t(38) = 2.018, p = 0.05, η 2 = 0.64; see Fig. 1(E)).Mice that were socially defeated also had higher plasma concentrations of ghrelin compared to control mice, an effect that was evident regardless of the time of the day in which the mice were sacrificed at (main effect of stress, F(1, 22) = 6.209, p = 0.021, η 2 = 0.21; see Fig. 1(F)).Finally, we analyzed mRNA expression of the ghsr and the glucocorticoid receptor, nr3c1, to determine genetic changes in response to chronic stress.Results from control and socially defeated mice revealed an increase in ghsr (F(1, 35) = 13.53,p < 0.001, η 2 = 0.28) and nr3c1 (F(1,36) = 8.906, p = 0.05, η 2 = 0.20) mRNA expression in the VTA of stressed mice, regardless of the time of day when these samples were collected (see Fig. 1(G-H)).Additionally, the expression of ghsr mRNA in the hypothalamus was higher just before the onset of the dark phase of the daily cycle (F(1, 36) = 6.859, p = 0.013, η 2 = 0.16) but this effect was not altered by exposure to stress (see Supplemental Fig. 1(A)).As expected, we found that the hippocampus of stressed mice expressed lower relative nr3c1 mRNA expression compared to control mice (F(1, 25) = 8.594, p = 0.007, η 2 = 0.24), but no changes in ghsr mRNA expression were detected in this region following chronic social defeat (see Supplemental Fig. 1 (B)).No changes in ghsr or Nr3c1 mRNA expression were observed in samples from the prefrontal cortex (p > 0.05; see Supplemental Fig. 1 (C)).
Experiment 2: Rescuing VTA GHSR expression restores stress-induced feeding In this study we wanted to investigate if restoring GHSR expression exclusively in the VTA of GHSR null mice would affect feeding behaviors in response to chronic social defeat stress.To do so, we examined feeding behavior in stressed and non-stressed WT (GHSR +/+ ) and GHSR null (GHSR −/− ) mice.We rescued GHSR expression in the VTA of a subset of GHSR −/− mice by delivering a neuron specific CRE-expressing AAV bilaterally into the VTA (GHSR VTA+/+ ) prior to chronic social defeat exposure (see Fig. 2(A)).To confirm that GHSR expression was successfully restored in the GHSR null mice, we used RNAscope In-Situ hybridization on sections of mice receiving the CRE-expressing AAV.As shown in Fig. 2(B), GHSR mRNA signal was observed in the midbrain of the GHSR VTA+/+ mice and it was confined to the VTA, where a large proportion of th mRNA expressing neurons were also observed (see Fig. 2(C)).Results revealed that, as in the previous experiment, stress caused a significant change in food intake based in GHSR +/+ but not GHSR −/− mice (F(1,43) = 19.31,p < 0.001, η 2 = 0.19) (See Fig. 2(F)).In the GHSR VTA+/+ mice, however, restoring expression of the GHSR induced an increase their daily caloric intake during the stress period (p = 0.003) (see Fig. 2(D)).Chronic social defeat induced a decrease in weight gain (F (1,43) = 13.51,p < 0.001, η 2 = 0.23) in mice regardless of genotype (See Fig. 2(E)).We investigated if restoring GHSR expression in the VTA was solely responsible for mediating these changes in intake by comparing the successful virus infusion, those that hit the target region, to unsuccessful infusions, that missed the target region and may have induced GHSR expression in regions other than the VTA in the GHSR null mice (see Supplemental Fig. 2 (A,B)).The infusions that hit the VTA, and therefore successfully rescued ghsr expression, resulted in a significant increase in caloric intake in the stressed mice, compared to mice in which the viral infusion missed the VTA (F(1, 30) = 6.519, p = 0.016, η 2 = 0.18; See Fig. 2(G)).Social defeat resulted in a significant reduction in weight gain in the mice where the viral infusion missed the VTA (F(1,30) = 7.02, p = 0.013, η 2 = 0.15; see Fig. 2 (H)), compared to mice in which the viral infusions hit the VTA target (p < 0.001).
Experiment 3: Unilateral infusions of JMV2959 into the VTA attenuate caloric intake but increase the expression of anxiety-like responses after chronic social defeat Fig. 3 shows data from mice in which we chronically infused the GHSR antagonist, JMV-2959, unilaterally into the VTA to determine if this would attenuate stress induced increases in caloric intake (see Fig. 3(A) for the experimental timeline).As shown in this figure, chronic infusions of JMV-2959 into the VTA eliminated the increase in caloric intake of socially stressed mice (F(1, 22) = 5.68, p = 0.026, η 2 = 0.44) (see Fig. 3(B)).Mice treated with JMV2959 also showed attenuated weight gain regardless of whether they were stressed or not (main effect of treatment, F(1,22) = 5.766, p = 0.025, η 2 = 0.11; Fig. 3(C)).While treatment with JMV2959 attenuated caloric intake and weight gain in socially defeated mice, it also resulted in altered behavioral coping responses, reflecting increased anxiety-like responses after chronic social defeat.Analyses of data from the open field test showed that stressed mice treated with JMV2959 spent more time in the periphery of the arena (F(1,22) = 6.295, p = 0.020; See Fig. 3(D)), compared to vehicle-treated stress counterparts (p = 0.005).Stressed mice, regardless of treatment, displayed a reduction in locomotor activity, and effect that was not enhanced by JMV2959 (main effect of stress, F(1,22) = 22.26, p < 0.001; See Fig. 3(E)).In the social approach test stressed mice displayed increased latency to approach the stranger mouse (main effect of stress, F(1,20) = 5.335, p = 0.032; See Fig. 3(F)).Similarly, the stressed, JMV2959 treated mice spent significantly less time interacting with the stranger mouse (F(1, 21) = 5.4, p = 0.03) compared to stressed, vehicle-treated mice (p = 0.024; See Fig. 3(G)).

Discussion
Results from this study show that chronic social defeat increases ghrelin secretion and potentially ghrelin sensitivity in the VTA, a region that plays a critical role in the regulation of reward seeking behaviors and in the adaptations to social defeat stress (Krishnan et al., 2007).Our data also show that, while a global deficiency in GHSR signalling prevents increased caloric intake in response to stress, rescuing GHSR signalling in the VTA restores this behavior.Conversely, chronic pharmacological reduction in GHSR signalling within the VTA leads to a reduction in stress-induced caloric intake.Overall, our data support the idea that stress induced feeding is mediated by increases in GHSR signalling in the VTA.
This data confirms previous work highlighting that chronic social defeat stress increases ghrelin secretion in male and female mice, and that this increase in ghrelin is linked to increases in caloric intake (Lutter et al., 2008;Patterson et al., 2013;Sanghez et al., 2013;Goto et al., 2014;Smith et al., 2023).Our results also confirm that the majority of calories consumed come from a diet in which the highest caloric content comes from carbohydrates, and not fat (Patterson et al., 2013).This suggests that social defeat results in increased ghrelin secretion at the onset of both the light and dark cycles, increasing orexigenic drive and carbohydrate utilization throughout the stressor and potentially beyond it, given that increased ghrelin secretion persists for weeks after the termination of the stressor (Lutter et al., 2008;Patterson et al., 2013;Smith et al., 2023).The mechanisms sustaining elevated ghrelin secretion during the chronic social defeat period are not yet clear, but could be related to depletion of carbohydrate stores produced by chronic activation of the sympathetic nervous system and the HPA axis (Gupta et al., 2019).The importance of increased ghrelin secretion, and ultimately increased caloric intake, is linked to the anxiolytic effects of this peptide.As shown by Lutter et al (2008), ghrelin treatment attenuates the anxiogenic effects of social defeat stress, particularly those involving social interactions (Lutter et al., 2008).In GHSR KO mice, ghrelin is ineffective in diminishing decreases in social approach following social defeat stress.In general, these mice show more symptoms associated with higher severity of behavioral symptoms produced by chronic social defeat exposure (Lutter et al., 2008).Our data show that GHSR expression in the VTA may play a critical role in the adaptive processes that help mice cope with this psychosocial stress paradigm.Indeed, social defeat increases ghsr mRNA expression in the VTA, and blocking these receptors makes animals consume less calories during the social defeat paradigm.Conversely, these mice also show reduced time investigating a novel conspecific compared to controls with functional GHSR signaling, indicating that they are more susceptible to the effects of the social defeat stress.Finally, while GHSR KO mice do not increase their caloric intake in response to chronic stress, increased feeding responses during social defeat were observed in mice in which the GHSR was rescued selectively in the VTA.While behavior was not measured in our rescue study, previous work in our lab shows that GHSR deficient mice of the same strain, show social approach deficits that can be restored by CRE-mediated VTA rescue of the GHSR (Park et al., 2021).Importantly, rescue of GHSR signalling in TH expressing neurons of GHSR KO mice also makes these mice less likely to show stress induced behavioral coping alterations, including social approach (Chuang et al., 2011).Overall, these data suggest that the VTA neurons become more sensitive to ghrelin in response to chronic social defeat, and this increase in expression may be required to defend against the stressor via responses that include increased caloric intake.The phenotype of the VTA neurons that increase ghsr mRNA expression has not been fully determined, but ghrelin does target dopamine cells to increase their activity, their release of dopamine at target sites and ultimately increase food intake and motivated behaviors (Abizaid et al., 2006;Jerlhag et al., 2007).Therefore, one could propose that the anxiolytic effects of ghrelin are mediated via the actions of ghrelin on dopamine cells in the VTA as supported by previous work (Perello et al., 2010;Chuang et al., 2011).We also demonstrated that GHSR rescue in the VTA was associated with a high degree of colocalization of GHSR within TH neurons.There are, however, other cell groups in the VTA that release glutamate and GABA and are important for the regulation of feeding and stress related behaviors.The role of these cells in mediating ghrelin's effects to social defeat needs to be further examined.
Chronic social defeat is a paradigm that takes advantage of natural behavioral responses associated with male territorial aggression.The resident mice in this paradigm are mature, sexually experienced CD-1 mice, that tend to display high levels of aggression and quickly become the dominant mouse following interaction with the experimental intruders.While there is some variability in the behavioral and endocrine responses of experimental intruders to the continuous social defeat, intruder mice tend to display not only submissive behav-iors but also several behavioral coping responses that have been associated with mood disorders like anxiety, depression, and posttraumatic stress disorder.Such mice typically display anhedonia as detected in the sucrose preference test, helplessness as detected in the forced swim test, anxiety-like behaviors as measured in the open field box or the plus maze tests, and social withdrawal as detected with the social interaction test (Krishnan et al., 2007;Golden et al., 2011).A drawback to this paradigm is that it is inherently male biased and provides information that may be specific to the male response to stress.In the past decade a number of paradigms adapting social defeat stress to study females have emerged, and in some cases have been used to study the effects of chronic social defeat on metabolic parameters, including some of those studied in this project (Lopez and Bagot, 2021).For instance, female mice coated with urine from male mice and then used as intruders in the cages of CD-1 males showed increased weight gain and displayed decreased exploratory behavior in the open field test (van Doeselaar et al., 2020).Recently, we used a model of social defeat described by Yohn et al (2019), where a female experimental intruder is placed in the cage of an aggressive CD-1 male resident at the same time as another experimental male intruder (Yohn et al., 2019).Using this paradigm we observed that females, like males in this study, also increased their caloric intake and this increase was accompanied by increases in plasma ghrelin (Smith et al., 2023).While the effects of the stressor on females in this paradigm were not as dramatic as those observed here or in other studies looking at chronic social defeat in males, they confirm that chronic social defeat has a similar endocrine impact on females.
Aside from changes in ghsr mRNA expression in the VTA, we also observed changes in the expression of this gene in the hypothalamus, that differed in terms of the time of day and regardless of whether the mice were stressed or not.As expected, mice sacrificed at the onset of the dark phase of the light/dark cycle showed higher hypothalamic ghsr mRNA expression than mice sacrificed at the onset of the light phase of the cycle, regardless of stress exposure (Supplemental Fig. 1).This pattern of ghsr mRNA expression is in line with circadian patterns of ghrelin secretion in mice and rats with ghrelin secretion peaking at the onset of the dark phase of the cycle (Drazen et al., 2006;Blum et al., 2009;LeSauter et al., 2009), and these do not seem altered by the stressor.
In addition, we also observed changes in the expression of the glucocorticoid receptor, nr3c1, in response to chronic social defeat stress.Chronic social defeat resulted in increased nr3c1 mRNA expression in the VTA, where glucocorticoids may drive reward seeking behaviors including those associated with food seeking (Ambroggi et al., 2009;Mizoguchi et al., 2021).These increases may be implicated in the social deficits that result chronic social defeat (Barik et al., 2013).Overall, the increase in ghsr and nr3c1 mRNA expression in the VTA suggests that stress increases sensitivity of the reward system to stress and metabolic hormones in response ultimately leading to changes in caloric consumption.As with previous work, nr3c1 mRNA expression was decreased in the hippocampus of socially defeated mice, an effect linked to excess in glucocorticoids (Hodes et al., 2012;Han et al., 2017).This effect however is not dependent on GHSR signalling as GHSR KO mice do not show further decreases in the expression of nr3c1 mRNA in response to chronic corticosterone treatment (Hay et al., 2020).Circulating ghrelin levels were elevated regardless of the time of day in which the samples were collected, supporting the idea that chronic social defeat induces chronic elevations in ghrelin secretion (Lutter et al., 2008;Patterson et al., 2013) and likely abolish the circadian release of ghrelin.
We also tested whether food intake or preference would change if mice were presented with the HFD closer to the onset of the dark phase, however, our results support previous findings that mice prefer a diet rich in carbohydrates and this does not change based on the time of day in which the HFD is provided.These results support the hypothesis that chronic stressors can alter diet preference, shifting them in favor of the diet that will provide for adequate energy to cope with the energetically demanding effects of repeated stress exposure (Patterson et al., 2013).While these results appear contradictory to data that has reported that chronic stress increases HFD intake (Chuang et al., 2011), it is likely that the preference for the chow, a diet rich in carbohydrates, reflects the potential for chronic social defeat to cause carbohydrate depletion.Alternatively, the increase in chow consumption may reflect a return to a comfort food, a food that the animal has experienced throughout its lifespan, while decreasing the more novel food that was presented at the onset of the experiment.Finally, it might be possible that the defeat episode itself produces signals that are anorectic and that are not cleared until the dark phase of the cycle.Once cleared, stress induced ghrelin may exacerbate feeding and this might occur when there is only chow available.Work from our lab suggests that chronic social defeat favours carbohydrate utilization as a source of fuel, and therefore the preference for the carbohydrate-rich chow might reflect carbohydrate depletion (Patterson et al., 2013;Abizaid, 2019).Additionally, chronic social defeat results in increased expression of feeding related peptides in the hypothalamus, including increased expression of npy and agrp mRNA (Patterson et al., 2013;Smith et al., 2023) which promote increased caloric intake.Nevertheless, more studies are required to determine if continuous access to the HFD would result less preference for the chow and more for the HFD.Alternatively, pair feeding studies where the chow is restricted so that animals cannot increase their chow intake during chronic social defeat, would provide an insight into the importance of this behavioral change for the metabolic responses to this challenge.
Overall, the current results point to increased ghrelin and GHSR signalling as key adaptations to meet the challenges imposed by chronic social defeat stress.Increased ghrelin secretion may mediate the metabolic challenges associated with chronic stress, by acting at brain regions like the arcuate nucleus and VTA to increase food intake and motivation, and ultimately increase resilience.It is, however, important to note that while ghrelin can access the arcuate nucleus with relative ease, peripheral ghrelin has limited access to the VTA and the ability of increased ghrelin to influence cells in the VTA of stressed mice may be questioned (see Edwards and Abizaid, 2017 for review).Notably, our results highlight that rescuing GHSR expression in the VTA is sufficient to restore stress-induced feeding, despite a lack of expression in ghrelin accessible regions, such as the arcuate nucleus.
It also remains possible that chronic social stressors alter brain permeability to circulating hormones like ghrelin, making these more able to enter the brain and influence the activity of cells in regions like the VTA, or some of its targets like the nucleus accumbens (Cornejo et al., 2020;Menard et al., 2017).While we did not investigate if rescuing GHSR expression in the VTA would restore the orexigenic response to peripheral ghrelin treatment in the GHSR KO mice, this would be a worthwhile investigation for future research.It has yet to be determined if chronic stress impacts ghrelin entry into the VTA, however, the results presented here do support a functional role of ghrelin in the VTA through activation of the GHSR.Alternatively, and given that the GHSR has high constitutive activity and that this receptor can influence the activity of other G-coupled protein receptors via dimerization (Wellman and Abizaid, 2015), one can speculate that increased GHSR expression following stress allows for increased or biased dopamine, serotonin or other types of receptors that facilitate dopamine neurotransmission from the VTA to its target regions and ultimately increased food intake.Additional research is required to determine how GHSR expression in the VTA is mediating the observed changes to food intake in response to chronic stress.
In summary, the results presented here highlight the importance of ghrelin and the GHSR within the VTA in regulating both the metabolic and behavioral responses to social defeat stress.Expression of the GHSR is upregulated in the VTA of mice undergoing repeated social defeat stress, and this is associated with an increase in caloric intake.These changes seem to be required to maintain an allostatic state and prevent stress induced pathological conditions that are displayed when the GHSR is blocked in the VTA, or that can be strengthened when the GHSR is restored in the VTA of GHSR KO mice.Ultimately, these data support the notion that the ghrelin system plays an important role in regulating responses to acute and chronic stressors and that, like glucocorticoids, may be critical for maintaining allostasis during metabolic challenges posed by psychosocial stressors.

Fig. 1
Fig.1depicts the timeline (panel A) and results from Experiment 1.As shown here, mice exposed to chronic social defeat increased their total caloric intake throughout the stress period compared to control mice (t(38) = 2.182, p = 0.035, η 2 = 0.69; see Fig.1(B)).This increase was related to an increase in caloric intake from the chow, but not HFD, as shown by a significant increase in the daily intake of chow (t(38) = 2.881, p = 0.006, η 2 = 0.91), and a significant decrease in the intake of the HFD in socially defeated mice (U = 117, p = 0.024, η 2 = 0.65; see Fig.1(C-D)).In addition, mice

Fig. 1 .
Fig. 1.Chronic social defeat stress increases caloric intake via increased ghrelin secretion and Ghsr expression in the ventral tegmental area.(A) Timeline of experiment 1. (B) Total caloric intake was increased in mice exposed to the chronic social defeat paradigm.(C) Stressed mice displayed an increase in caloric intake of the chow diet during the stress paradigm, compared to non-stressed counterparts.(D) Intake of the HFD, provided for 4-h daily, decreased in stressed mice, compared to non-stressed counterparts.(E) Weight gain increased in stressed mice, compared to non-stressed counterparts.(F) Circulating ghrelin levels and the relative expression of (G) Ghsr and (H) Nr3c1 was increased in the stressed mice at the end of the stress paradigm, regardless of diurnal or nocturnal collection.

Fig. 2 .
Fig. 2. Effects of rescuing GHSR expression on feeding behaviors in response to chronic social defeat stress.(A) Timeline of experiment 2, involving the rescue of GHSR expression in the VTA using AAV infusions.(B) Graphical schematic of transgenic GHSR mice used in experiment.(C) Image of successful GFPtagged AAV infusions into the VTA.(D) Images demonstrating successful GHSR expression following Cre-GFP AAV infusion into the VTA.Solid white arrows indicate neurons displaying overlap of GFP, GHSR, and TH expression in the VTA.Empty white arrows indicate Th+ neurons lacking GHSR expression.Majority of GHSR expressing cells were colocalized with Th signal.Scale bars measure 50 µm.(E) Chronic social defeat stress induced an increase in food intake in WT (GHSR +/+ ) mice and mice with VTA GHSR expression (GHSR VTA+/+ ), but not GHSR null (GHSR −/− ) mice.(F) Stress induced a reduction in weight gain in mice, regardless of genotype.(G) Mice with viral infusions that hit the VTA increased their food intake compared to miss whose viral infusions missed the VTA, which were no different than their non-stressed counterparts.(H) Missed AAV infusions resulted in reduced weight gain in stressed mice, compared to those with successful AAV infusions and non-stressed controls.

Fig. 3 .
Fig. 3. Effect of chronic JMV-2959 infusion into the VTA on the metabolic and behavioral response to chronic social defeat stress.(A) Timeline of experiment 3, involving the insertion of osmotic minipumps and VTA cannulas for the chronic infusion of the GHSR antagonist, JMV-2959.(B) Administration of JMV-2959 eliminated a stressed induced increase in caloric intake, as observed in vehicle-treated mice.(C) JMV-2959 treated mice exhibited a reduction in weight gain over the course of the experiment, compared to mice receiving the vehicle.Behavioral tests following the stress paradigm revealed that (D) stressed JMVtreated mice spent more time in the periphery of the open field arena, compared to vehicle-treated stressed mice.(E) Stressed mice exhibited greater locomotor activity, regardless of treatment.In the social approach test (F) stressed mice exhibited increased latency to approach the stranger mouse and (G) JMV-treated, stressed mice spent less time interacting with the stranger mouse, compared to vehicle-treated counterparts.

Table 1
Primer sequences used to measure ghsr and nr3c1 expression in the VTA of non-stress and stress mice