Impact of social isolation on corticosterone release and recovery after stroke in aged rats: A behavioral and biochemical analysis

Social isolation (SI) after stroke reduces recovery. The aim of this study was to evaluate the effects of SI on corticosterone release and recovery after stroke in aged rats. A total of 64 male Wistar rats (aged 24 months) were used in the present study. All rats were housed in pairs for two weeks. After two weeks, rats were randomly assigned to one of four groups: (1) rats underwent sham surgery and kept socially isolated (control/social isolated (CO/SI) group); (2) rats underwent sham surgery and kept pair housed (control/pair housed (CO/PH) group); (3) rats underwent middle cerebral artery occlusion (MCAO) surgery and kept socially isolated (stroke/ isolated (ST/SI) group); (4) rats underwent MCAO surgery and kept pair housed (stroke/pair housed (ST/PH)) group. Behaviors were assessed using the adhesive removal test, rotarod test and social interaction test at 1st, 7th, 14th and 21st days after stroke. Serum biochemical analysis was also performed on the behavioral testing days. Results showed THAT serum corticosterone and MDA levels in CO/PH group were significantly lower than CO/SI group. Serum BDNF levels in CO/PH group was significantly higher than CO/SI group. Serum cortico-sterone and MDA levels in ST/PH group were lower than ST/SI group. In ST/PH group, serum Total antioxidant capacity (TAC) and BDNF levels were significantly higher than ST/SI group. Biochemical analysis of certain regions of the brain (hippocampus, striatum and cerebral cortex) was performed on 21st day after stroke. In the hippocampus of CO/PH group, BDNF and TAC levels were significantly higher than CO/SI group. The hippocampal MDA level of CO/PH group were significantly lower than CO/SI group. BDNF and TAC levels in the hippocampus, striatum and cerebral cortex of ST/PH group were significantly higher and MDA level was significantly lower as compared with ST/SI group. Both ischemic groups showed sensorimotor recovery over a 21-day period, but recovery of ST/PH group was significantly greater than ST/SI group. Total social interaction time in ST/PH group was significantly longer than ST/SI group. Based on the results of this study, social interaction after stroke enhances histologic and sensorimotor recovery through reduction of HPA activity and corticosterone release, leading to increased TAC and BDNF levels.


Introduction
Stroke is a potent injury which leads to a variety of sensorimotor, behavioral, cognitive and emotional disorders.These events prone patients to social isolation (SI) and distanced relationships (Jellema et al., 2022).SI impairs functional and histological aspects of recovery after any types of injury (e.g.stroke, heart disease and cutaneous wounds) and increases mortality secondary to these causes (Hong et al., 2023).
Until recently, the possible harmful underlying mechanisms of SI after stroke were not known.Growing documents from human and animal studies have revealed that social interactions deeply influence health while increasing the likelihood of survival, whereas SI weakens physical and psychological health (Christiansen et al., 2021;Naito et al., 2023;Stuller et al., 2012).
Current evidence suggests an existing relationship between SI and the activity of hypothalamo-pituatary (HP) axis (Algamal et al., 2021;Biggio et al., 2014;Butler et al., 2014;Serra et al., 2005).Clearly, it has shown that SI acts as a strong psychological stressor via changing the hypothalamic-pituitary-adrenal (HPA) axis activity by increasing corticosterone secretions from adrenal glands (Boero et al., 2018).Therefore, it is suggested that SI both enhances the sensitivity of the anterior pituitary gland to corticotropin-releasing factor and weakens the negative feedback regulation of the HPA axis in rats and their offspring (Kamal et al., 2014;Pisu et al., 2013).After stroke, the HPA axis is activated for a longer duration of time due to several reasons such as stress from the stroke attack, destruction of the HPA inhibitory regions of the brain and the cytokine release following brain damage (Barugh et al., 2014;de la Tremblaye and Plamondon, 2016;Feng et al., 2014).Shao et al. in intact rats demonstrated that SI increases oxidative stress levels by reducing the activities of the antioxidant enzymes and increasing reactive oxygen species (ROS) production in several regions of brain (Shao et al., 2015).Studies in intact rats demonstrated that serum corticosterone elevation leads to oxidative stress in serum and brain tissue by both increasing ROS production and reducing antioxidant enzyme activities (Kim et al., 2021;Liu et al., 2014;Zafir andBanu, 2009a, 2009b).O' Keefe et al. (2014) displayed that SI after stroke can lead to depression-like behaviors in mice due to reduction in Brain-Derived Neurotrophic Factors (BDNF) levels (O'Keefe et al., 2014).BDNF is a neuroprotective agent, the concentrations of which is lower in the presence of serum corticosterone elevations in rats (Huang et al., 2010;Lin et al., 2022;Maghsoudi et al., 2014;Radahmadi et al., 2015).
Stroke is a leading cause of disability, often resulting in sensorimotor deficits and social withdrawal (Fanaei et al., 2014).While the negative impact of social isolation on recovery from various health conditions is recognized, the specific mechanisms by which it exacerbates post-stroke cerebral vulnerability remain elusive.This gap in knowledge hinders the development of targeted interventions to mitigate the adverse effects of social isolation on stroke outcomes.
Our study addresses this critical gap by exploring the relationship between social isolation and the hypothalamic-pituitary-adrenal (HPA) axis's activity post-stroke.We investigate the novel hypothesis that social interaction can modulate corticosterone release, oxidative stress, and brain-derived neurotrophic factor (BDNF) levels, thereby influencing both histological and behavioral recovery.This research is pioneering in its examination of the HPA axis's role in the context of social factors and stroke recovery, offering potential insights into new therapeutic strategies.

Material and methods
This study was approved by the Institutional Animal Research Ethics Committee at Zahedan University of Medical Sciences (ethical code: IR.ZAUMS.REC.1398.279).All chemicals were obtained from Sigma-Aldrich (St. Louis, Missouri, USA), unless stated otherwise.

Animals
A total of 64 male Wistar rats (aged 24 months) which were provided by the Laboratory Animal Research Center of Zahedan University of Medical Sciences, were used in the present study.The rats were maintained under a 12 h lighting cycle at room temperature and humidity controlled vivarium.The rats were permitted ad libitum access to water and standard lab chow.

Experimental design
As displayed in Fig. 1, all male rats were allowed to acclimate for two weeks with ovariectomized female rats prior to any housing manipulation.Subsequently, all rats were randomly housed in pairs per cage (one male with an ovariectomized female rat from their previous cage, with whom they had already acclimated) for another two weeks before being assigned to different groups (O'Keefe et al., 2014;Karelina et al., 2009;Venna et al., 2014).Then, male rats were randomly assigned into one of four groups (16 in each group): (1) Rats underwent sham surgery and kept socially isolated (control/social isolated (CO/SI) group); (2) Rats underwent sham surgery and kept pair housed (control/pair housed (CO/PH) group); (3) Rats underwent middle cerebral artery occlusion (MCAO) surgery and kept socially isolated (stroke/isolated (ST/SI) group); (4) Rats underwent MCAO surgery and kept pair housed (stroke/ pair housed (ST/PH) group).The male pair housed rats remained in the same cage as the familiar female rat until the end of the study.
All animals were euthanized on day 21 after transient MCAO surgery.Eight rats in each group were decapitated to obtain brain tissue samples for biochemical analysis and brain tissues of the remaining eight animals were used to measure infarct volume.

Surgical preparation and anesthesia
Prior to surgery, rats were subjected to a fasting period of 12 h, although they were allowed unrestricted access to water.Anesthesia was administered using a combination of ketamine (100 mg/kg, intraperitoneal, IP) sourced from Bremer Pharma GMBH, Germany, and Xylazine (10 mg/kg, IP) obtained from Alfasan, Netherlands, to ensure the animals were adequately sedated for the procedure.

Middle cerebral artery occlusion (MCAO) procedure
The experimental ischemic strokes were induced utilizing the wellestablished transient MCAO model, as detailed in prior studies Fig. 1.Experimental design overview.
H. Fanaei et al. (Bornavard et al., 2020;Fanaei et al., 2014Fanaei et al., , 2013;;Kumar and Gupta, 2016;Nazari et al., 2016;Yarahmadzehi et al., 2020).This involved the careful insertion of a silicone-coated nylon monofilament through the external carotid artery, navigating it further into the internal carotid artery and subsequently into the circle of Willis.The filament's advancement was meticulously monitored to occlude the right middle cerebral artery (MCA) effectively.

Monitoring and verification of ischemia
To confirm the successful induction of ischemia, we continuously monitored the blood flow in the MCA territory using laser-Doppler flowmetry (LDF) with the moorVMS-LDF1-HP device from Moor Instruments Ltd., Oxford, England.The accuracy of vascular occlusion was verified when a sustained reduction in the LDF signal dropped below 25 % of the pre-ischemic baseline, indicating a significant decrease in cerebral blood flow (Fanaei et al., 2014).

Reperfusion and post-surgical care
Following a 90-min occlusion period, the monofilament was carefully removed to initiate reperfusion and restore blood flow to the brain tissue.Throughout the surgical process, we maintained the core body temperatures of the rats at 37.0 • C ± 0.5 • C using a feedback-controlled heating pad that adjusted the heat output based on the rectal temperature readings of the animals.Post-surgery, the rats were placed in a temperature-controlled environment for two hours to ensure stable body temperature during recovery.

Sham procedure
For the sham-operated control groups (CO/SI and CO/PH groups), we performed the same surgical steps without inducing MCA occlusion to serve as a baseline for comparison (Fanaei et al., 2014).

Measurement of BDNF, MDA and TAC levels
The blood from animals were collected one day before stroke (as baseline) and on 1st, 7th, 14th and 21st days after stroke from rat's tail and centrifuged (4 ο C, 3000 g for 15 min) to separate the serum.Whole brain samples were obtained from the skull 21 days after stroke, and certain regions of the ischemic hemisphere (hippocampus (Hipp), striatum (STM), and cerebral cortex (CTX)) were dissected (Chiu et al., 2007;Spijker, 2011) and stored at − 80 • C until analysis.
For biochemical analysis, tissue samples were homogenized using a homogenizer (T10 basic ULTRA-TURRAX®, IKA, Germany).A 10 % tissue homogenate was prepared by homogenizing the samples in icecold phosphate-buffered saline (PBS) (0.1 M, pH 7.4) containing the protease inhibitor cocktail (Roche).Subsequently, the homogenized tissue samples were centrifuged at 4 • C, 10,000 g for 10 min, and the supernatant was collected.Serum and brain BDNF (ELISA kit, Promega, USA) and tissue Malondialdehyde levels (MDA, Zellbio, Germany) and Total Antioxidant Capacity (TAC) (Zellbio, Germany) were measured using commercial kits, per the manufacturer's instructions.Social stressors lead to changes in HPA axis activity.HPA axis activity biomarkers include glucocorticoids: free cortisol (or corticosterone in experimental animals) and adrenocorticotropic hormone (ACTH) (Oken et al., 2015;Stephens and Wand, 2012).In the present study, HPA axis activity was measured by changes in corticosterone concentrations.Serum corticosterone levels were measured using a commercial corticosterone rat ELISA kit (Zellbio, Germany) as per the manufacturer's instructions.

Behavioral tests
All behavioral evaluations were performed by an experimenter blinded to the animal's history.On 1st, 7th, 14th and 21st days after stroke, the animals underwent the adhesive removal test and rotarod test.On the last day of the experiment (21st day) social interaction test was performed.

Adhesive removal test
To assess somatosensory deficit, an adhesive removal test was performed (Bouet et al., 2009;Lin et al., 2021).One piece of adhesive tape (113 mm 2 ) was attached to distal-radial region of the wrist of each forelimb (Fanaei et al., 2014;Van Der Zijden et al., 2008).Then, the animal was placed in a clear plexiglass cage with stainless steel lid cover.Cage was without bedding for ease of recording (Gharbawie and Whishaw, 2003).Before ischemic stroke -to familiarize the animals with the test-the rats were trained for 3 days (five times in each day) (Chen et al., 2001).After stroke, five trials per testing day were performed.Each trial ended either after both pieces of the adhesive tape were removed from forelimbs or after 180 s (Fanaei et al., 2014;Van Der Zijden et al., 2008).

Rotarod test
Assessment of motor recovery after stroke was done by rotarod test (Fanaei et al., 2014;Ruan and Yao, 2020;Ulbrich et al., 2012).Before stroke, the animals were trained one session per day for 3 days.After stroke, four trials per testing day were performed with a 5 min cut-off time between trials (Ulbrich et al., 2012).Rotarod apparatus was set to accelerate slowly from 4 to 45 rpm speed over a 5 min period (Fanaei et al., 2014;Boussicault et al., 2016).For the rotarod test, rats were placed on a rotarod cylinder, and the time the rat was able to walk on the rod before falling was measured (Xu et al., 2015).The trial ended if the animal fell off the rotarod cylinder or reached criterion level (maximum value: 300 s).Mean duration (in seconds) of 4 trials per testing day was calculated for each animal.

Social interaction test
Social interaction test was performed to evaluate sociability and affective-like behavior of animals (Kaidanovich-Beilin et al., 2011;Nemeth et al., 2012).Each rat was placed in the center of a black Perspex box (75 m × 75 m), in which an ovariectomized unfamiliar female rat was placed as stimulus (Tao et al., 2019).The behavior of rats was recorded and scored by an investigator that was blind to the condition.The latency time of the experimental rats to their first interaction with the female rat was recorded.Then, the total interaction time, including sniffing, biting, grooming, following, mounting, attacking and jumping, was recorded.Behavior of each animal was recorded for 10 min (Tao et al., 2019).

Determination of infarct volume
Following 21 days of reperfusion, rats were deeply anesthetized by pentobarbital overdose and were transcardially perfused with cold phosphate-buffered saline followed by 4 % paraformaldehyde.After that, the brains were removed, sectioned into 50 μm thick coronal sections and stained with hematoxylin and eosin (H&E) (Fanaei et al., 2014).Sections were scanned using a flatbed scanner (Scanjet, Hewlett-Packard, USA) connected to a computer.The images were measured with Image J software (NIH).Final infarct volume is presented as a percentage using the following formula: Infarct volume (contralateral hemisphere volumenon-infarct volume in ipsilateral hemisphere) / contralateral hemisphere volume (Bornavard et al., 2020;Yarahmadzehi et al., 2020).

Statistical analysis
All data are presented as mean ± SD.Statistical analyses were performed using GraphPad Prism 7 (GraphPad Software, Inc.).To assess the effects of housing (single or paired) and intervention (sham or stroke), a 2-way ANOVA was conducted for infarct volume, biochemical analyses of the brain, and social interaction tests.For other assessments, including the Adhesive Removal and Rotarod tests, as well as serum H. Fanaei et al. biochemical data, a 3-way ANOVA was utilized to analyze the interactions between housing, intervention, and time.Bonferroni post hoc tests were applied following ANOVA to identify specific group differences.A p-value of <0.05 was considered statistically significant.

Results
Two of rats in the ST/SI group and one from the ST/PH group died during the study.Data of died animals were excluded from the analysis.However, mortality was not significantly different between the groups (P > 0.05).

Serum corticosterone level
As shown in Fig. 2, pair housing decreased serum corticosterone levels in the CO/PH group compared with the CO/SI group on the 1st, 7th, 14th and 21st days (P-Value <0.001 for 1st, 7th, 14th, and 21st days).Serum corticosterone levels in ST/PH group was significantly lower than ST/SI group on the 7th, 14th and 21st days after stroke (P-Values <0.001 for 7th, 14th and 21st days); however, there was no significant difference in serum corticosterone levels in the stroke groups on the first day after surgery (P-Value: 0.25).Corticosterone concentration in the MACO surgery groups were significantly higher than control groups on all sampling days (P-Values were <0.001 in all sampling days).The 2-way ANOVA revealed that there were significant interactions in corticosterone levels with stroke and housing in entire testing period, separately (F (4,220): 37.32, P-Value<0.001;F (4.220): 8.24, P-Value<0.001,respectively).Furthermore, a three-way ANOVA was run to examine the effect of stroke, housing and time on corticosterone levels.There was a significant three-way interaction (F (4,220): 5.11, P-Value: 0.0006).

Serum MDA, TAC and BDNF levels
Cerebral ischemia in ST/SI and ST/PH groups induced considerable increase in serum MDA levels than CO/SI and CO/PH groups (P-Val-ue<0.001)(Fig. 3a).Serum MDA levels of both ischemic groups decreased over times and the MDA levels in ST/PH group were significantly lower than ST/SI group on the 7th and 14th days after stroke (P-Value<0.001, and P-Value<0.01,respectively).The Serum MDA levels in the CO/PH group were consistently lower than those in the CO/SI group on the 1st, 7th, 14th, and 21st days post-surgery.This difference reached statistical significance only on the 21st day (P-value: 0.001).There was no statistically significant difference between the groups on the 1st, 7th, and 14th days (P-value >0.05 for all comparisons).The ST/ SI group had higher serum MDA concentrations than the CO/SI group on the 1st, 7th, and 14th days (P < 0.0001, P < 0.0001, P < 0.05.respectively).Serum MDA levels in the ST/PH group were significantly higher than those in CO/PH group on 1st and 7th days (P-Value <0.0001, and P-Value <0.0001).A significant interaction in MDA levels were seen between stroke by time and housing by time (F (4, 44): 110.3, P-Val-ue<0.001;F (4, 44): 6.58, P-Value<0.001,respectively).Furthermore, the results of 3-way ANOVA showed that there was a significant interaction (F (4, 44): 3.38, P-Value: 0.01) between stroke, housing and time.
Upon analysis, the mean serum TAC levels in the ST/PH group compared to the ST/SI group, and in the CO/PH group compared to the CO/SI group, did not demonstrate statistically significant differences at any of the time points measured (p > 0.05) (Fig. 3b).
As shown in Fig. 4, the BNDF levels of the ST/PH and ST/SI groups showed a decreasing trend until the 7th day after stroke, which was greater in ST/SI group.Thereafter, BDNF levels began to increase until the end of the experiment in both ischemic groups, and this increment was significantly greater in ST/PH group than ST/SI group on the 7th, 14th and 21st days after stroke (p˂0.05,p˂0.01, p˂0.01 respectively).Over the course of the study, there was a progressive decline in serum BDNF levels within the CO/SI group.Notably, these levels were significantly diminished when compared to the CO/PH group on both the 14th and 21st days, with p-values <0.05 for each comparison.ST/SI group had lower serum BDNF levels than CO/SI group at 7th, 14th and 21st days after ischemia (p˂0.0001,p˂0.0001, p˂0.0001 respectively).Serum BDNF levels in ST/PH group was significantly lower than CO/PH group at 7th, 14th and 21st days after stroke (p˂0.0001,p˂0.0001, p˂0.0001 respectively).The results of 2-way and 3-way ANOVA tests revealed that there were significant interactions between stroke by time and housing and time, separately (F (4,220): 18.54, P-Value<0.001,and F (4,220): 5.04, P-Value: 0.0007).However, there was no interaction between stroke, housing and time (F (4,220): 0.22, P-Value: 0.92).

MDA, TAC and BDNF levels in brain
Measurement of brain MDA levels at the 21st day after stroke showed improvement in stroke-induced oxidative stress due to pair housing (Fig. 5a).MDA levels in cortex (CTX) of ST/PH group were significantly lower than ST/SI group (P-Value ˂ 0.0001).The MDA levels in CTX of the ST/SI group were significantly higher than CO/SI and CO/PH groups (P-Value˂0.0001,P-Value˂0.0001respectively).MDA levels in CTX of ST/PH group were significantly higher than CO/PH and CO/SI groups (P-Value ˂0.0001, P-Value ˂0.0001 respectively).The results of 2-way ANOVA showed that there was a significant interaction between stroke and housing in CTX MDA levels (F (1, 40): 18.8, P-Value<0.001).
In the striatum (STM) region of the ST/PH group, MDA levels were significantly lower than the ST/SI group (P-Value ˂0.001).MDA levels in STM of ST/SI group were significantly higher than CO/PH and CO/SI groups (P-Value ˂0.0001, and P-Value ˂0.0001, respectively).MDA levels in STM region of ST/PH group were significantly higher than CO/ PH and CO/SI group (P-Value ˂0.01, P-Value ˂0.01 respectively).In addition, there was a significant stroke and housing interaction in STM MDA levels (F (1, 44): 12.8, P-Value = 0.0009).
MDA levels in hippocampus (Hipp) of the ST/PH group were significantly lower than the ST/SI group (P-Value ˂0.01).MDA levels in the Hipp of ST/SI group were significantly higher than both the CO/PH and the CO/SI groups (P-Value ˂0.0001, and P-Value ˂0.0001 respectively).The Hipp of CO/PH group had significantly lower levels of MDA than CO/SI group (P-Value ˂0.01).However, the 2-way ANOVA results revealed that there was no interaction between stroke by housing in Hipp MDA levels (F (1, 44): 0.03, P-Value: 0.86).
As shown in Fig. 5b, TAC levels in the CTX of the ST/PH group were significantly higher than the ST/SI group (P-Value ˂0.0001).TAC levels in the CTX of the ST/SI group were significantly lower than the CO/PH and CO/SI groups (P-Value ˂0.0001, and P-Value ˂0.0001, respectively).Furthermore, a significant interaction between stroke and housing condition demonstrated in CTX (F (1, 44): 10.07, P-Value: 0.003).
TAC levels in the STM of the CO/SI group were significantly higher than the ST/SI and ST/PH groups (P-Value ˂0.0001, and P-Value ˂0.0001, respectively).STM of the CO/PH group had significantly more TAC levels than the ST/SI and ST/PH groups (P-Value ˂0.0001, and P-Value ˂0.0001, respectively).However, the 2-way ANOVA analysis  H. Fanaei et al. revealed that stroke by housing had a significant interaction with STM TAC levels (F (1, 44): 0.17; P-Value: 0.68).
TAC levels in the Hipp of the CO/PH group were higher than the CO/ SI, ST/SI and ST/PH groups (P-Value ˂0.01, P-Value ˂0.0001, and P-Value ˂0.001, respectively).TAC levels in Hipp region of CO/SI group were higher in than ST/SI group (P-Value ˂0.05).Moreover, the 2-way ANOVA analysis showed an interaction between stroke and housing in Hipp TAC levels (F (1, 44): 1.54, P-Value: 0.22).
As displayed in Fig. 6, BDNF levels in the CTX of the ST/PH group were significantly higher than the ST/SI group (P-Value ˂0.0001).BDNF levels in the CTX of the ST/SI group were significantly lower than both the CO/SI and the CO/PH groups (P-Value ˂0.0001, and P-Value ˂0.0001, respectively).The 2-way ANOVA demonstrated that CTX BDNF levels interacted with stroke by housing (F (1, 40): 18.81, P-Value <0.0001).
In the STM of the ST/SI group, BDNF levels were significantly lower than the CO/SI and CO/PH groups (P-Value ˂0.001, P-Value ˂0.01 respectively).BDNF levels in the STM of the CO/SI group were significantly higher than the ST/PH group (P-Value ˂0.05).The 2-way ANOVA results revealed that there was no interaction between stroke by housing (F (1, 44): 2.39, P-Value: 0.13).In the Hipp of the ST/PH group, BDNF levels were significantly higher than the ST/SI group (P-Value ˂0.01).Hipp of the ST/SI group had significantly lower BDNF levels than the CO/SI and CO/PH groups (P-Value ˂0.0001, and P-Value ˂0.0001 respectively).BDNF levels in the Hipp of the CO/PH group were significantly higher than the CO/SI and ST/PH groups (P-Value ˂0.01 and P-Value ˂0.0001, respectively).However, the 2-way ANOVA revealed that BDNF in Hipp levels had no significant interaction between stroke and housing (F (1, 44): 0.08, P-Value: 0.77).

Behavioral recovery
Adhesive removal test was used to assess the recovery of the sensorimotor function after experimental stroke.This test revealed the extent of sensorimotor function of the ST/SI and the ST/PH groups after stroke.As shown in Fig. 8a, no statistical differences in adhesive removal times were observed between different groups at baseline.After stroke, both ischemic groups improved their adhesive removal times till the end of experiment.But recovery was significantly quicker in the ST/PH group on the 14th and 21st days after stroke as compared to the ST/SI group (P-Value ˂0.01, P-Value ˂0.001 respectively).There was no difference between the durations to adhesive removal between the CO/SI and the CO/PH groups.Furthermore, the findings showed a significant a stroke by time interaction (F (4,220): 45.8, P-Value<0.0001);however, there were not a significant interaction in housing by time and stroke, housing and time interaction (F (4, 220): 2.14, P-Value: 0.07; F (4, 220): 1.89, P-Value: 0.11).
The longitudinal assessment of sensorimotor function post-stroke revealed distinct recovery profiles within each experimental group.Using the adhesive removal test, we quantified the time taken for tape removal at baseline and at various intervals following stroke (1st, 7th, 14th, and 21st days).At baseline, the ST/SI group exhibited no significant differences in tape removal time compared to their respective controls.Post-stroke, a pronounced impairment was evident on the 1st day, with a significant increase in removal time (p < 0.0001) than baseline.This impairment persisted through the 7th day (p < 0.0001), with gradual improvements noted by the 14th (p < 0.01) and 21st days (p < 0.001), yet not returning to baseline levels.The ST/PH group's baseline performance was comparable to controls.Following stroke, a significant delay in tape removal was observed on the 1st day (p < 0.0001) than baseline.However, the group demonstrated a remarkable recovery trajectory, with significant improvements by the 7th day (p < 0.0001) and a return to near-baseline performance by the 21st day (p > 0.05), indicating no significant difference from baseline.Both control groups (CO/SI and CO/PH) maintained consistent sensorimotor function throughout the study duration, with no significant changes observed in tape removal times at any point, confirming the stability of the baseline performance.
These data illustrate the profound impact of social environment on recovery after stroke.The ST/SI group's slower recovery trajectory suggests that social isolation may impede the rehabilitation process, whereas the ST/PH group's rapid return to baseline function underscores the potential therapeutic benefits of social interaction poststroke.
Rotarod test results showed that on the 1st post-stroke day both ST/  SI and ST/PH groups had similar reduction in the duration to stay on the rotarod device (Fig. 8b).Rotarod time of both ischemic groups had improved by the 21st after stroke.Rotarod time of ST/PH group was significantly higher than ST/SI group at 14th (P-Value ˂0.01) and 21st (P-Value ˂0.01) days after stroke (Fig. 8b).The duration of the rotarod in CO/PH group was significantly higher than CO/SI group at 14th and 21st days (P-Value ˂0.05 and P-Value ˂0.001, respectively) (Figure 8b).Rotarod time in CO/SI group was significantly higher than ST/SI group at 1st, 7th, 14th and 21st days after stroke (P-Value ˂0.0001, P-Value ˂0.0001, P-Value ˂0.0001, P-Value ˂0.0001, respectively).Rotarod time of the CO/PH group was significantly higher than the ST/PH group on 1st, 7th, 14th and 21st days after stroke (P-Value ˂0.0001, P-Value ˂0.0001, P-Value ˂0.0001, P-Value ˂0.0001 respectively).The 2-way ANOVA results showed that there were significant interactions between stroke-by-time and housing-by-time in rotarod tests results, separately (F (4,220): 45.4,P-Value<0.0001;F (4,220): 7.0, P-Value <0.0001).Furthermore, the results of 3-way ANOVA showed that there was a significant interaction (F (4, 220): 0.20, P-Value: 0.93) between stroke, housing and time.
The longitudinal analysis of motor function post-stroke revealed distinct recovery profiles within each experimental group (Fig. 8b).The rotarod test was employed to quantify the endurance of rats at baseline and at various intervals following stroke (1st, 7th, 14th, and 21st days).In rotarod test, at baseline, the ST/SI group exhibited no significant differences in endurance compared to their respective controls.Poststroke, a pronounced impairment was evident on the 1st day, with a significant decrease in the time to fall (p < 0.0001) than base.This impairment persisted through the 7th day (p < 0.0001), with gradual improvements noted by the 14th (p < 0.01) and 21st days (p < 0.001), yet not returning to baseline levels.The ST/PH group's baseline performance was comparable to controls.Following stroke, a significant decrease in endurance was observed on the 1st day (p < 0.0001) than baseline.However, the group demonstrated a remarkable recovery trajectory, with significant improvements by the 7th day (p < 0.0001) and a return to near-baseline performance by the 21st day (p > 0.05), indicating no significant difference from baseline.Both control groups (CO/ SI and CO/PH) maintained consistent endurance throughout the study duration, with no significant changes observed in the time to fall at any point, confirming the stability of the baseline performance.These data illustrate the profound impact of social environment on recovery after stroke.The ST/SI group's slower recovery trajectory suggests that social isolation may impede the rehabilitation process, whereas the ST/PH group's rapid return to baseline function underscores the potential therapeutic benefits of social interaction post-stroke.
Social interaction with an unfamiliar ovariectomized female rat revealed that the latency to initiate social interaction in the ST/PH group was significantly shorter than the ST/SI group (Fig. 9a).Latency to initiate social interaction in the CO/SI and CO/PH groups was significantly shorter than both the ST/SI (both P-Value ˂0.0001) and the ST/ PH (both P-Value ˂0.0001) groups (Fig. 9a).Total interaction time in the ST/PH group was significantly longer than the ST/SI group (P-Value ˂0.001) (Fig. 8b).The total interaction times in the control groups (CO/ SI and CO/PH) were significantly longer than the ischemic groups ((ST/ SI and ST/PH)) (Fig. 9b).

Discussion
This study has elucidated several key findings regarding the impact of social interaction on post-stroke recovery.Notably, rats with active social interactions (ST/PH group) demonstrated a significantly enhanced recovery, as evidenced by laboratory analyses and behavioral tests.The improvements were observed across multiple domains: 1. Corticosterone Levels: The ST/PH group exhibited reduced corticosterone levels compared to their socially isolated counterparts (ST/SI group), suggesting that pair housing mitigates the stress response following stroke.2. Oxidative Stress Markers: A more pronounced decline in MDA levels was observed in the ST/PH group, both in serum and brain tissues, indicating that social interaction may play a role in reducing oxidative damage post-stroke.3. Neurotrophic Factors: There was a significant increase in BDNF levels in the ST/PH group when compared to the ST/SI group, highlighting the potential for social environments to promote neuroplasticity and brain recovery after a stroke.4. Sensorimotor Function: Behavioral assessments using the Adhesive Removal test and Rotarod test revealed improved sensorimotor function in the ST/PH group on days 14 and 21 post-stroke, underscoring the benefits of social interaction for physical rehabilitation.5. Social Interaction: The social interaction test conducted in this study offers profound insights into the rehabilitative potential of social engagement following a stroke.Our results indicate that rats in the ST/PH group, which were allowed to interact with a pair, displayed a significantly shorter latency to initiate social interaction compared to the ST/SI group, which experienced social isolation.This suggests that social interaction inherently motivates proactive engagement, which could be beneficial for cognitive and emotional recovery post-stroke.Furthermore, the total interaction time was notably longer in the ST/PH group, underscoring the value of sustained social contact.In contrast, the ST/SI group's reluctance and reduced interaction duration may reflect the detrimental effects of isolation on social motivation and engagement.These behavioral differences are critical as they highlight the potential of social interaction to not only improve physical and neurochemical outcomes but also to enhance the social behavior of stroke survivors, which is essential for their overall well-being and integration into their pre-stroke social environments.While social interaction appears to be beneficial, it is important to note that it does not completely reverse the detrimental effects of stroke.This is evident from the still lower performance of post-stroke rats in the socially active groups compared to the control groups who underwent sham surgery.
The most critical insight from this research is the role of social interaction (SI) in modulating the physiological responses to stroke.SI is known to be a potent psychological stressor that can alter hypothalamicpituitary-adrenal (HPA) axis activity, leading to increased corticosterone secretion from the adrenal glands (Cené et al., 2012;Du Preez et al., 2021;Reis et al., 2012).Stroke, in turn, increases the activity of the HPA axis for a prolonged duration due to several reasons, such as strokeinduced stress, destruction of the HPA inhibitory areas, and cytokine release secondary to brain damage (Barugh et al., 2014;Feng et al., 2014).Human studies have reported that elevated cortisol levels have been associated with worse outcomes post-stroke, including increased dependency, morbidity, and mortality (Barugh et al., 2014).
Our findings indicate that corticosterone levels in paired-housed animals were lower than those in socially isolated animals.Following stroke, corticosterone levels spiked in both ischemic groups but subsequently decreased.The decline was more pronounced in the pairedhoused animals, suggesting that social interaction may facilitate the normalization of HPA axis activity after stroke.
These results align with previous research, such as Boero et al. which linked increased corticosterone to psychological stressors and altered HPA activity (Boero et al., 2018).Moreover, the presence of pre-existing stressors, such as social isolation, has been associated with poorer stroke outcomes in both human patients and animal models (Stuller et al., 2012).These results further verify our finding of poor outcomes in poststroke SI animals.
Our findings indicate that there was no significant difference in serum corticosterone levels between baseline and one day post-surgery in the CO/PH group.This suggests that the presence of a familiar female may have provided a buffer against the stress typically associated with surgical procedures, as evidenced by the stable corticosterone levels.Notably, this effect was not observed in the CO/SI group, which exhibited elevated corticosterone levels post-surgery, underscoring the potential impact of social support on stress responses.The absence of a significant increase in corticosterone levels in the CO/PH group aligns with the broader literature on the benefits of social support in postsurgical recovery (Brembo et al., 2017;Ellingsen et al., 2023;Kulik and Mahler, 1989).Studies have demonstrated that environmental enrichment, including social housing, can mitigate stress and promote recovery in rodent models (Westenbroek et al., 2004;Atrooz et al., 2021).Conversely, isolation or a lack of social support has been associated with delayed recovery and increased stress markers (Brembo et al., 2017;Ellingsen et al., 2023;Kulik and Mahler, 1989;Westenbroek et al., 2004;Atrooz et al., 2021).Our study contributes to this body of evidence, suggesting that social factors play a crucial role in modulating physiological stress responses following surgical interventions.
Our study revealed that control animals subjected to social isolation exhibited increased oxidative stress, as evidenced by elevated MDA levels in the brain, particularly within the striatum.This suggests that SI alone may act as a detrimental factor for brain recovery.When coupled with a significant insult such as a stroke, social isolation could further hinder the brain's recuperative processes.The oxidative stress observed post-stroke is likely attributable to the accumulation of reactive oxygen species (ROS) and the concomitant degradation of antioxidant enzymes (Fanaei et al., 2014;Crack and Wong, 2008;Rodrigo et al., 2013;Bornavard et al., 2020).These changes can severely compromise neuronal function by disrupting the plasma membrane's integrity, leading to altered ion fluxes, neuronal depolarization, and excessive glutamate release (Li et al., 2018;Manzanero et al., 2013;Orellana-Urzúa et al., 2020).Corroborating our findings, multiple studies have reported similar outcomes, noting reduced activities of key antioxidant enzymes-catalase, superoxide dismutase, and glutathione peroxidase-in various brain regions of social isolation rats, along with increased hydrogen peroxide levels (Shao et al., 2015) (Kim et al., 2021;Liu et al., 2014;Zafir andBanu, 2009a, 2009b).These observations underscore social isola as a potential stressor on a healthy brain, one that could negatively influence recovery following a traumatic event.Conversely, pair-housing appears to offer a protective effect, potentially mitigating the adverse impacts of social isolation on brain recovery.
Serum MDA levels in the ST groups were significantly higher than those in the CO groups, which indeed suggests that the effects of stroke were more pronounced.This pronounced effect likely amplified the body's response to environmental factors, such as housing conditions, making the impact of PH more noticeable in the ST groups.
The early and significant reduction of MDA levels in the ST/PH group compared to the ST/SI group on the 7th and 14th days post-stroke indicates that PH may have a therapeutic potential as an environmental H. Fanaei et al. intervention after stroke.It seems that PH can enhance the resilience of the brain to ischemic damage, thereby mitigating oxidative stress more effectively than in non-stroke conditions.
In this study, we observed a difference in serum MDA levels between the control groups subjected to pair housing (CO/PH) and social isolation (CO/SI).While both groups underwent the same surgical procedure, the CO/SI group exhibited higher MDA levels as early as the first day post-surgery, compared to the CO/PH group.This difference was found to be statistically significant on the 21st day post-surgery.The increase in MDA levels in the CO/SI group suggests that even short-term social isolation can induce a stress response that accelerates oxidative stress.This is consistent with the literature indicating that social isolation can lead to physiological changes that may increase the production of reactive oxygen species (ROS) and, consequently, MDA levels (Huang et al., 2015;Drinkwater et al., 2022;Vitale and Smith, 2022;Cacioppo et al., 2014).The activation of the hypothalamic-pituitary-adrenal (HPA) axis and subsequent release of corticosterone in response to social isolation is a well-documented phenomenon that could explain the elevated MDA levels observed (Cacioppo et al., 2014).
Furthermore, the significant difference in MDA levels between the CO/PH and CO/SI groups at the 21st day post-surgery underscores the potential long-term effects of social housing conditions on oxidative stress.The protective effect of social interaction against oxidative damage, as indicated by lower MDA levels in the CO/PH group, aligns with previous findings that highlight the role of social support in mitigating stress-related biomarkers (Huang et al., 2015;Drinkwater et al., 2022;Vitale and Smith, 2022;Cacioppo et al., 2014).
Our findings contribute to the understanding of how social factors can influence physiological processes and recovery outcomes.The impact of social isolation on oxidative stress markers such as MDA emphasizes the need for considering social environmental factors in postoperative care and recovery protocols.Future research should aim to further elucidate the mechanisms by which social environments modulate oxidative stress and investigate the long-term implications of these findings for post-surgical recovery and overall health.
Research has consistently demonstrated a decline in BDNF across various brain regions post-stroke, correlating with the emergence of depressive-like behaviors in animal models (O'Keefe et al., 2014;Verma et al., 2014).These behaviors, characterized by despair, anhedonia, and diminished sociability, have been linked to BDNF reduction in studies (O'Keefe et al., 2014;Ieraci et al., 2016;Scaccianoce et al., 2006).Furthermore, a decrease in BDNF has been associated with increased serum corticosterone levels, suggesting a relationship between stress response and neurotrophic support (Huang et al., 2010;Lin et al., 2022;Maghsoudi et al., 2014;Radahmadi et al., 2015).In our investigation, we observed elevated BDNF levels in the pair-housed groups, irrespective of stroke status.These groups also exhibited enhanced social behaviors-a finding consistent with previous research connecting increased BDNF levels to improved social interaction.Notably, the pairhoused groups demonstrated a sustained decrease in corticosterone release post-stroke, which coincided with heightened BDNF expression in critical brain areas such as the hippocampus, striatum, and orbitofrontal cortex, thereby promoting sociability (Fuentealba et al., 2019;Li et al., 2021;Manfré et al., 2018).From a physiological standpoint, BDNF plays a crucial role in attenuating neuronal apoptosis, stimulating neurogenesis, and ameliorating sensorimotor deficits caused by stroke (Bornavard et al., 2020;Fanaei et al., 2013;Li et al., 2018;Liu et al., 2010;Mang et al., 2013;Ploughman et al., 2009;Zhang et al., 2012;Fereidooni et al., 2020).These functions position BDNF as a key target for research aimed at developing interventions to foster functional recovery after stroke (Mang et al., 2013;Ploughman et al., 2009;Li et al., 2014;Nagahara and Tuszynski, 2011;Pezet and Malcangio, 2004).Our findings suggest that increased BDNF levels contribute to sensorimotor, histological, and behavioral enhancements, alongside reductions in serum corticosterone and oxidative stress, further underscoring the therapeutic potential of BDNF modulation in stroke recovery.
During the course of the study, male rat in the ST/PH and CO/PH groups were housed with familiar female rat, while male rat in the ST/SI and CO/SI groups were housed alone.This housing arrangement was maintained to investigate the effects of social housing versus social isolation on the recovery and behavior of male rat post-stroke.In the social interaction test, all male rat, regardless of their prior housing conditions, were introduced to an unfamiliar female rat.This was done to ensure that any observed differences in social interaction could be attributed to the effects of social housing or isolation, rather than familiarity with the specific female rat used in the test.The introduction of an unfamiliar female rat aimed to simulate a novel social encounter, which is a standard approach in behavioral studies to assess social curiosity and interaction in rodents (Schulingkamp et al., 2023;Schneider et al., 2016;Beery and Shambaugh, 2021).The decision to use an unfamiliar female rat during the social interaction test is supported by literature indicating that male rodents typically show robust investigatory behavior towards novel females, which is not significantly altered by prior familiarity with other females (Beery and Shambaugh, 2021;Paré, 2000).This behavior is thought to be driven by the innate reproductive and social behaviors of male rodents, which are heightened in the presence of a potential mating partner, regardless of familiarity (Beery and Shambaugh, 2021;Paré, 2000).
Recent literature suggests that the hippocampus and striatum are functionally linked via the prefrontal cortex, playing a crucial role in spatial navigation and decision-making processes (Goodroe et al., 2018;Graham et al., 2009).The hippocampus is traditionally associated with episodic and spatial memory, while the dorsal striatum is important for stimulus-response and action-outcome associations (Goodroe et al., 2018;Graham et al., 2009).These regions interact dynamically to support a range of navigational and cognitive functions (Goodroe et al., 2018;Graham et al., 2009).Moreover, the striatum receives projections from multiple regions of the hippocampal formation, emphasizing the flow of spatial and contextual information to motivational control systems (Goodroe et al., 2018;Graham et al., 2009;Liljeholm and O'Doherty, 2012).This interaction is critical for the learning and expression of place-reward associations, which could be relevant to our findings on social interaction and motor function recovery (Goodroe et al., 2018;Graham et al., 2009;Liljeholm and O'Doherty, 2012).Regarding the types of neurons involved, mirror neurons in the premotor and parietal cortices are known to be activated both when performing and observing actions, suggesting their importance in social cognition and empathy (Acharya and Shukla, 2012;Woodruff, 2018).These neurons could potentially contribute to the observed behaviors in our study, as they are essential for understanding the actions, intentions, and emotions of others (Acharya and Shukla, 2012;Woodruff, 2018).Furthermore, the cortex and hippocampus work together during social interactions through a network involving the medial prefrontal cortex, hippocampus, and amygdala (Sigurdsson and Duvarci, 2016;Qi et al., 2018;Sun et al., 2020).This network is responsible for the perception and integration of social cues, which is essential for complex social behaviors (Sigurdsson and Duvarci, 2016;Qi et al., 2018;Sun et al., 2020).For instance, the lateral orbitofrontal cortex and hippocampus have been implicated in encoding, maintaining, and retrieving social cues, particularly when multiple interactions need to be disambiguated in a rapidly changing social context (Ross et al., 2013).In light of these insights, we propose that the slower recovery trajectory and reduced social engagement observed in the ST/SI group may be due to impaired interactions among these neural regions and circuits.The ST/PH group's better recovery and return to baseline performance suggest that social housing may facilitate the proper functioning of these neural pathways, thereby enhancing both motor and social recovery post-stroke.
This study provides compelling evidence for the influence of social environment on motor function recovery and social behavior following a stroke.The results from the rotarod test and social interaction assessments paint a nuanced picture of recovery, highlighting the interdependence of physical and social rehabilitation.The rotarod test results H. Fanaei et al. revealed that the ST/SI group, which experienced post-stroke recovery under conditions of social isolation, demonstrated a significantly slower recovery trajectory compared to the ST/PH group, which was housed with a partner.This difference was particularly pronounced at the 14th and 21st days post-stroke, suggesting that social isolation may have a lasting impact on motor function recovery.In parallel, the social interaction test results indicated that the ST/SI group engaged less with an unfamiliar female rat, when compared to the ST/PH group.This reduced social engagement could be reflective of a broader decline in activity levels and motivation, which may also contribute to the observed deficits in motor function.The findings underscore the potential therapeutic benefits of social interaction in post-stroke rehabilitation.The ST/PH group's recovery and return to near-baseline performance by the 21st day post-stroke emphasize the positive impact of a socially enriched environment.In contrast, the ST/SI group's persistent deficits highlight the detrimental effects of social isolation on both motor and social recovery.These observations are consistent with existing literature that associates social support with better health outcomes in stroke survivors (Kruithof et al., 2013;Butsing et al., 2019;Zhou and Kulick, 2023).Social interaction has been shown to enhance neuroplasticity, reduce stress, and improve overall well-being, which can all contribute to more effective recovery processes (Kruithof et al., 2013;Butsing et al., 2019;Zhou and Kulick, 2023;Davidson and McEwen, 2012).
Finally, our study suggests that incorporating social interaction into post-stroke care may be beneficial for enhancing recovery.Future research should explore the mechanisms underlying these effects and determine the optimal conditions and timing for social interventions to maximize rehabilitation outcomes.

Limitations and future directions
Our study has several limitations that warrant consideration.Firstly, the influence of pre-stroke social housing conditions, specifically living with only another rat, on post-stroke recovery has not been explored in our research.Prior social interactions may prime neuroplasticity and resilience, potentially affecting recovery outcomes (Neves et al., 2024).Future studies should investigate the pre-stroke social environment's impact on post-stroke rehabilitation.Secondly, the percentage of blood flow interruption during surgery is a critical variable that can influence stroke outcomes (Ström et al., 2013).Variability in the extent of ischemia may lead to differences in infarct volume and severity (Ström et al., 2013), which could affect the generalizability of our results.It is essential for subsequent research to consider the consistency of surgical procedures to ensure comparable infarct volumes across studies.Lastly, the olfactory bulb plays a crucial role in social interaction and recognition (Bakker et al., 2020), and its potential effects on our findings have not been evaluated.Given the olfactory system's significance in rodent social behaviors (Bakker et al., 2020), the exclusion of this evaluation is a limitation of our study.We did not assess the olfactory bulb due to the focus on motor function and social interaction as measured by physical proximity and contact.However, future research should include an evaluation of the olfactory bulb to fully understand its contribution to social behavior post-stroke.

Conclusion
In conclusion, our study highlights the multifaceted role of social interaction in stroke recovery.We have demonstrated that social engagement post-stroke not only reduces infarct volume but also significantly improves sensorimotor and behavioral outcomes.These benefits are further illuminated by the observed modulation of the HPA axis and corticosterone levels, suggesting that social interaction may exert its therapeutic effects by attenuating the stress response.Elevated BDNF levels, associated with better recovery metrics, indicate that BDNF may mediate these positive effects.The interplay between reduced corticosterone release, HPA axis regulation, and increased BDNF levels presents a promising therapeutic triad for enhancing recovery in stroke patients.Our findings advocate for the integration of social interaction into therapeutic strategies and underscore the potential of targeting the HPA axis and BDNF pathways to optimize stroke rehabilitation.