Enriched housing promotes post-stroke functional recovery through astrocytic HMGB1-IL-6-mediated angiogenesis

Enriched environment (EE) is shown to promote angiogenesis, neurogenesis and functional recovery after ischemic stroke. However, the underlying mechanisms remain unclear. C57BL/6 mice underwent middle cerebral artery occlusion (60 min) followed by reperfusion, after which mice were housed in either standard environment (SE) or EE. Here we found that post-ischemic EE exhibited decreased depression and anxiety-like behavior, and promoted angiogenesis and functional recovery compared to SE mice. EE mice treated with high-mobility group box-1 (HMGB1) inhibitor glycyrrhizin had an increased post-stroke depression and anxiety-like behavior, and the angiogenesis and functional recovery were decreased. HMGB1 and interleukin-6 (IL-6) expression in astrocyte were increased in EE mice. EE mice treated with glycyrrhizin decreased, whereas EE mice treated with recombinant HMGB1 (rHMGB1) increased the levels of IL-6 and p-AKT. Blockade of IL-6 with anti-IL-6-neutralizing antibody in EE mice attenuated EE-mediated angiogenesis and functional recovery. Furthermore, our in vitro data revealed that in primary astrocyte cultures rHMGB1 promoted the expression of IL-6 in activated astrocytes. PI3K/AKT signaling pathway was involved in HMGB1-mediated expression of astrocytic IL-6. Thus, our results reveal a previously uncharacterized property of HMGB1/IL-6 signaling pathway in EE-mediated angiogenesis and functional recovery after ischemic stroke.


INTRODUCTION
Stroke is the major cause of permanent disability in adults worldwide because of the brain's limited capacity for neural repair. 1,2 An enriched environment (EE) has been a classic paradigm for studying the effects of a complex combination of physical, cognitive and social stimulation in rodents. EE (social interactions, voluntary and varied physical activity, and introduction of novel objects) is shown to have important roles in a normal or injured brain, ultimately beneficially influencing brain function and recovery after injury. [3][4][5][6] EE increases neurogenesis in the adult subventricular zone (SVZ) and angiogenesis during stroke recovery, [6][7][8] promotes spontaneous recovery after ischemic stroke. [7][8][9] However, the underlying molecular mechanisms remain unclear.
High-mobility group box-1 protein (HMGB1) is a member of the damage-associated-molecular-pattern family of proteins that is rapidly released from necrotic neurons that amplify neuronal death in the penumbra in the acute stages of ischemic stroke. 10,11 However, recent study suggests that astrocytic HMGB1 could promote peri-infarct angiogenesis and functional recovery in the delayed phases of stroke recovery. 12 Previous studies have shown that EE could enhanced angiogenesis after cerebral ischemic injury. 13,14 However, whether EE could promote angiogenesis and functional recovery through astrocytic HMGB1 during stroke recovery is unclear. One aim of our study is to investigate whether HMGB1 is an important mediator of EE on angiogenesis and long-term functional recovery after ischemic stroke.
Interleukin-6 (IL-6) belongs to the family of glycoprotein 130activating cytokines. IL-6 has been shown to promote neuronal differentiation of neural progenitor cells (NPCs) dissociated from normal adult mice. 15 Under physiological conditions, adult IL-6 knockout mice exhibit significantly lower NPCs survival and proliferation in the dentate gyrus and SVZ. 16 Our previous study has found that IL-6 is essential for the promoting effects of social support on the neurogenesis and long-term outcome after ischemic stroke. 17 A previous study demonstrates that IL-6 produced locally by resident brain cells promotes angiogenesis and affords long-term histological and functional protection after ischemic stroke. 18 HMGB1 can bind to its receptors (glycation end products (RAGE), Toll-like receptor 2 (TLR2) and TLR4) and then trigger inflammatory cytokine expression. 19 Evidence has shown that astrocytic HMGB1 promotes neurovascular remodeling via RAGE receptors. 12 RAGE expression at the cell surface membrane of astrocytes mediates the expression of IL-6 in astrocytes. 20 Thus, we speculate that astrocytic HMGB1 could promote the production and secretion of IL-6 from astrocyte in the delayed phases of stroke recovery.
In this study, we examined the hypothesis that EE could promote astrocytic HMGB1-induced production and secretion of IL-6 from astrocyte, which promoted angiogenesis and functional recovery following focal cerebral ischemia. Our study could provide a possible mechanism for explaining how EE promotes neurovascular remodeling and functional recovery after brain injury.

RESULTS
EE increases the number of HMGB1-positive astrocytes in ischemic hemisphere during stroke recovery As previously reported, astrocytes are the major cellular source of HMGB1 during stroke recovery. 12  EE mice show decreased depression and anxiety during stroke recovery Depression-like behavior was assessed with forced swim task (FST) and sucrose consumption test (SCT). There was an observed significant effect of EE on FST after ischemic stroke, as measured by spending less time floating in the FST in enriched mice, relative to mice housed in SE ( HMGB1 is involved in EE-mediated angiogenesis in the peri-infarct region after ischemic stroke To study whether EE-induced inhibition of depression and anxietylike behavior was involved in the promoting effects of HMGB1 in post-stroke angiogenesis, the microvascular density (MVD) in the peri-infarct area was assessed at 21 d.p.i. using CD34, which is expressed on early and vascular-associated tissue. Notably, EE significantly increased the CD34-positive MVD in the peri-infarct area (Figures 2c and d  We found that deficits in the pole test (Figures 2g and h), rotarod test ( Figure 2i) and elevated body swing test (EBST; Figure 2j) were better in the EE 3 and 4 weeks after stroke. However, the deficits in the pole test (Figures 2g and h), rotarod test ( Figure 2i) and EBST ( Figure 2j) were worse in the enriched mice after treatment with glycyrrhizin. HMGB1 promoted the production and secretion of IL-6 from astrocytes in EE during stroke recovery IL-6 expression was increased in all stroke mice regardless of housing conditions compared to their respective sham-operated group at 21 d.p.i. (Figure 3a; Po0.01 and Po0.05, respectively). However, EE exhibited significantly increased IL-6 level in ischemic hemisphere at 21 d.p.i., compared to mice housed in SE (Figure 3a; Po0.05). Glycyrrhizin treatment significantly decreased, whereas rHMGB1 treatment significantly increased the protein levels of IL-6 in EE mice, relative to mice housing in SE (Figures 3b and c; Po0.05).
Furthermore, in primary astrocytes cultures from the ischemic mice at 21 d.p.i., astrocytes were stimulated with lipopolysaccharide (LPS) to mimic a reactive phenotype. We observed a significant increase of IL-6 in both the cell lysate and culture supernatant of astrocytes treated with LPS (Figures 3d-f; P o0.01 and P o 0.05, respectively). Administration of rHMGB1 further significantly increased the levels of IL-6 in both the cell lysate and culture supernatant of LPS-treated astrocytes (Figures 3d, g and h; Po 0.01 and P o0.05, respectively). PI 3 K/AKT signaling pathway is involved in HMGB1-mediated production and secretion of IL-6 from astrocytes in EE after stroke In primary astrocyte cultures, PI 3 K/AKT pathway inhibitor LY294002 treatment significantly attenuated the promoting effects of rHMGB1 on the expression of IL-6 in both the cell lysate and culture supernatant of LPS-treated astrocytes in vitro (Figures 3d, i and j; P o0.05).
In vivo, we found that the expression of phospho-AKT (p-AKT) was increased in all stroke mice regardless of housing conditions compared to their respective sham-operated group at 21 d.p.i.

DISCUSSION
Our study confirmed that post-ischemic environmental enrichment could increase angiogenesis in the peri-infarct area and functional recovery in experimental animals during stroke recovery. PI 3 K/AKT signaling pathway was involved in HMGB1-induced IL-6 production from astrocytes in enriched mice during stroke recovery. HMGB1-AKT-IL-6 signaling pathway was involved in EE-mediated promotion of post-stroke angiogenesis and functional recovery ( Figure 6).
An EE has been a classic paradigm for studying the effects of a complex combination of physical, cognitive and social stimulation in rodents, which includes running wheels, novel objects and social interactions. In many studies, an EE has exhibited therapeutic effects on stroke such as enhancement of neurogenesis in the SVZ 6 and angiogenesis around the peri-infarction region. 7,8 These effects have been hypothesized to lead to the promoting effects of EE on post-stroke functional recovery. [7][8][9] Our current study is the first to investigate the promoting effects of EE on post-stroke angiogenesis in the peri-infarct area in mice. In our study, EE increased the CD34-positive MVD in the peri-infarct area by 60%. Evidence has shown that brain angiogenesis may provide the critical neurovascular substrates for neuronal remodeling during stroke recovery. 21 Furthermore, one clinical study suggests that stroke patients with a higher density of blood vessels have reduced morbidity and longer survival. 22 Therefore, EE has an important role in the neurorepair and functional recovery via promoting angiogenesis after ischemic stroke. However, the underlying molecular mechanisms are unknown. Across 51 clinical studies, approximately one-third of stroke survivors are diagnosed with PSD. 23 About a quarter of stroke survivors are diagnosed with PSD and PSA at the same time. 24 PSD and PSA are associated with higher morbidity and mortality, greater disability and poorer recovery after stroke. 24,25 In this work, we found significant inhibitory effects of EE on depressive-like phenotypes in the FST and SCT, and anxiety-like behavior in the OFT after ischemic stroke. Increasing evidences have revealed that depressive-and anxiety-like behaviors have been linked to impairments of all major aspects of plasticity in the adult mammalian brain such as neurogenesis, 26,27 gliogenesis 28,29 and angiogenesis. 30 Our present study found that EE significantly increased angiogenesis around the peri-infarction region. We also showed that EE significantly increased the expression of astrocytic HMGB1 in the ischemic hemisphere, which has been shown to be able to promote peri-infarct angiogenesis and functional recovery in the delayed phases of stroke recovery. 12 Our further results showed that administration of HMGB1 inhibitor glycyrrhizin increased PSD and PSA in EE mice, and attenuated the promoting effects of EE in angiogenesis and functional recovery in the delayed phases of stroke recovery. Thus, our present data suggested that EE could promote angiogenesis and functional recovery through increasing astrocytic HMGB1 expression and subsequent inhibition of PSD and PSA during stroke recovery. However, how PSD and PSA could directly inhibit post-stroke angiogenesis and functional recovery is unclear in our present study. The intrinsic mechanisms involved in PSD-and PSA-mediated inhibition of post-stroke neurorepair should be investigated further in a future study. A previous study has shown that IL-6 produced locally by resident brain cells could promote angiogenesis and long-term functional recovery after ischemic stroke. 18 In our previous study, we reported that IL-6 was essential for the promoting effects of social support on the neurogenesis and long-term outcome after ischemic stroke. 17 Evidence has shown that HMGB1 can bind to its receptors (RAGE, TLR2 and TLR4) and then trigger inflammatory cytokine expression. 19 Astrocytic HMGB1 promotes neurovascular remodeling via RAGE receptors. 12 RAGE expression at the cell surface membrane of astrocytes is shown to mediates the expression of IL-6 in astrocytes. 20 On the basis of these findings, we explored the intrinsic link between astrocytic HMGB1 and IL-6 in the promoting effects of EE in angiogenesis and functional recovery after ischemic stroke. In this study, our in vivo results showed that astrocytic HMGB1 played an essential role in the promoting effects of EE in the expression of IL-6 in the ischemic hemisphere during stroke recovery. Our in vitro study confirmed that HMGB1 could promote the production and secretion of IL-6 from activated astrocytes. We further demonstrated that IL-6 promoted angiogenesis in enriched animals after ischemic stroke, indicating that the promoting effects of astrocytic HMGB1 in the post-stroke angiogenesis and functional recovery were mediated by the secretion of IL-6 from astrocytes.
Evidence has shown that IL-6 secretion by astrocytes is regulated by PI 3 K/AKT signaling in the sub-acute phases of central nervous system injury. 31 Our and other previous studies have shown that PI 3 K/AKT signaling is the major signaling pathway implicated in NPC proliferation and differentiation in the delayed phases of stroke recovery. 32,33 In the present study, astrocytic HMGB1-induced angiogenesis in enriched mice was dependent on PI 3 K/AKT signaling-mediated production and secretion of IL-6 from astrocytes during stroke recovery. We also found that astrocytic HMGB1-mediated inhibition of PSD and PSA could also promote angiogenesis and functional recovery. However, the intrinsic link between PSD (PSA) and AKT-IL-6 signaling in astrocytes after ischemic stroke in enriched mice is unclear and needed to be further studied.
Angiogenesis and neurogenesis holds promise for brain repair and long-term functional recovery after ischemic stroke. Angiogenesis has been shown to be coupled with neurogenesis in brain tissue repair and remodeling after ischemic stroke. 34 Thus, proangiogenesis and recovery-modulating strategy are needed. In summary, our data here suggest that post-stroke EE improves stroke outcomes in an apparently pro-angiogenesis manner through astrocytic HMGB1-mediated inhibition of PSD and PSA, and also through astrocytic HMGB1-induced production and secretion of IL-6 from activated astrocytes. PI 3 K/AKT signaling is involved in astrocytic HMGB1-induced production and secretion of IL-6 from activated astrocytes in post-ischemic EE. Together, these data provide further evidence of the powerful effect that a rehabilitative strategy with EE has on the treatment of ischemic stroke.  the Animal Care and Use Committee of Tongji Medical College, Huazhong University of Science and Technology. Mice were anesthetized i.p. with ketamine (100 mg/kg) and xylazine (8 mg/kg). Focal cerebral ischemia was induced by middle cerebral artery occlusion (MCAO) with a 6-0 siliconecoated nylon monofilament for 1 h, to block the origin of the MCA as previously described. 35 Occlusion was confirmed by laser-Doppler flowmeter (Periflux System 5000, PERIMED, Stockholm, Sweden) with a probe placed on thinned skull over the lateral parietal cortex before, during and after MCAO, as well as before death. 35 Abrupt reduction in rCBF by ≈75-90% indicated a successful occlusion of the MCA. Mice in which  . Main mechanisms of EE in post-stroke neurogenesis and functional recovery. Post-ischemic EE increased the production and secretion of activated astrocytes after ischemic stroke. Astrocytic HMGB1 then enhances the production and secretion of activated astrocytes through PI 3 K/AKT signaling pathway. Astrocytic IL-6 promotes post-stroke neurogenesis and functional recovery.

MATERIALS AND METHODS
ipsilateral blood flow was not reduced to o 20% of the baseline after placement of the intraluminal filament and whose ipsilateral blood flow was not rapidly restored during reperfusion were excluded from subsequent experiments (≈10% in each group). Sham-operated mice were manipulated in the same way, but the MCA was not occluded. Body temperature was maintained at 37 ± 0.5°C with a feedback temperature control unit until the mice had recovered from surgery. SE-housed controls were housed in a standard cage (27 × 22.5 × 14 cm 3 ; 3-4 mice/cage). The EE mice were introduced in EE 2 days after MCAO or sham operation. The EE mice were housed in a spacious cage (86 × 76 × 31 cm 3 ) containing novel objects such as tunnels, shelters, toys and running wheels for voluntary exercise (10 mice/cage).

Experimental groups and drug administration
All treatments were administered in a blinded manner. The mice were randomly divided into nine groups (Figure 7): (1) sham operation mice housed in SE (sham+SE; n = 12/group/time point for each functional assay, and 5/group for western blotting and 5/group for immunofluorescence (IF)); (2) sham operation mice housed in EE (sham+EE; n = 12/group/time point for each functional assay, and 5/group for western blotting and  either anti-IL-6 mAbs (10 ng in 2 μl aCSF; R&D Systems) or 2 μl aCSF. This dose has been used successfully to neutralize IL-6 signaling in mice. 36 Functional assays EBST was performed to evaluate the symmetry of motor behavior 3, 4, 6 and 10 weeks after stroke. 37 Mice (n = 12/group/time point) were examined for lateral movements/turning when their bodies were suspended 10 cm above the testing table by lifting their tails. A swing was recorded when mice moved their head away from the vertical axis (angle 410°) in three sets of 10 trials, performed over 5 min. Results are expressed as the ratio of total number of contralateral swings. The rotarod test provided an index of forelimb and hindlimb motor coordination and balance. 37 Mice (n = 12/group/time point) were trained daily on an accelerating (5-40 r.p.m.) rotating rod for 3 days before MCAO; only those mice able to remain on the rod for 20 s at 40 r.p.m. were subjected to MCAO. Test sessions consisting of three trials at 40 r.p.m. were carried out 3, 4, 6 and 10 weeks after stroke, by an investigator who was blinded to the experimental groups. The final score was expressed as the mean time that a mouse could remain on the rod over three trials.
The pole test was used to assess forelimb strength, ability to grasp and balance performed in a blinded fashion 3, 4, 6 and 10 weeks after stroke. 18 Mice (n = 12/group/time point) were placed head upward near the top of a vertical steel pole (60 cm high with rough surface). Thereafter, both time taken to orientate the body completely downwards and to reach the floor with all four paws were recorded.
FST: To assess depression-like behavior at 3 weeks after stroke, mice (n = 12/group) were placed into an opaque cylinder tank (24 cm in diameter and 53 cm high) filled to a depth of 30 cm with water (25 ± 1°C). Swimming behavior was recorded for 5 min and scored for time spent actively swimming versus floating. Quantification of float versus swim time was performed with Observer software (Version 5; Exeter Software, New York, NY, USA).
SCT was used to assess depression-like behavior at 3 weeks after stroke. Mice (n = 12/group) are presented with the option of consuming either water or a 3% sucrose solution. Two identical 10 ml vials were placed on a custom-made wire cage top. At 3 days before MCAO, each individual mouse was provided two 10 ml vials of water for 12 h. The following morning, all mice were returned to their original housing condition and provided their normal drinking water and cage tops. That evening mice were again provided with two 10 ml vials of 3% sucrose overnight. After completion of habituation, all mice were again returned to their original housing condition and provided their normal drinking water and cage tops. On the day of testing, mice were water-deprived in their original housing condition for 6 h, then kept in a new cage with the customized cage top and one 10 ml vial of water and one 10 ml vial of 3% sucrose solution overnight. The volume of consumption of both solutions was recorded the following day by an observer blinded to housing condition.
OFT: Anxiety-like behavior of mice (n = 12/group) were assessed during a 60 min session in an open field apparatus (40 × 40 × 37.5 cm) using Flex Field photobeam activity (San Diego, CA, USA) at 3 weeks after stroke. Data were analyzed to determine the relative amount of activity occurring in the periphery versus the center of the apparatus (anxiety-like behavior).
In vitro LPS stimulation and treatments in the astrocyte culture Cells were dissociated from ischemic hemispheric brains at 21 d.p.i. or from sham-operated brains as described in our previous study. 38 Purified astrocytes were treated with LPS (100 ng/ml; Sigma, St. Louis, MO, USA) in the presence or absence of pre-treatment with LY294002 (200 nM, 30 min before LPS treatment). rHMGB1 (500 ng/ml) was added to the media 30 min after LY294002 treatment. At 16 h after treatment, the cells were fixed with 4% paraformaldehyde and subjected to histological analysis. The supernatant of cell culture was collected for western blotting analysis.

Immunofluorescence
Mice were transcardially perfused with 4% paraformaldehyde, and the brains were paraffin-embedded. Brains were cut into 4 μm-thick coronal sections. Cell cultures were fixed using 4% paraformaldehyde for 15 min. For both sections and cells, nonspecific binding was blocked using normal goat serum. Immunoassays were performed using the following antibodies: rabbit polyclonal HMGB1 (1 : 20, Abclonal); mouse monoclonal GFAP (1 : 300, Cell Signaling Technology); rabbit polyclonal IL-6 (1 : 20, Abclonal); or rabbit polyclonal CD34 (1 : 20, Abclonal). Primary antibodies were detected with dylight 549-conjugated goat anti-rabbit, dylight 488conjugated goat anti-mouse secondary antibody. For the quantitative analysis, eight non-overlapping fields in the peri-infarct area per slide at a magnification of × 400 were recorded by an observer who was blinded to the experimental groups. The mean volume of MVD from five sequential brain sections of individual mouse was calculated and expressed as numbers/mm 2 .

Data analysis
Multiple comparisons were performed by one-way ANOVA followed by Newman-Keuls multiple comparison tests for multiple comparisons (GraphPad Prism statistics software version 5.0, La Jolla, CA, USA). Two groups were compared by two-tailed Student's t-test. Behavioral data were analyzed by two-way ANOVA with repeated measures, followed by post hoc multiple comparison tests. All data are presented as mean ± S.E.M. The P-values o0.05 were considered statistically significant.