Effects of environmental enrichment on white matter glial responses in a mouse model of chronic cerebral hypoperfusion

Abstract

Background: This study was designed to explore the beneficial effects of environmental enrichment (EE) on white matter glial changes in a mouse model of chronic cerebral hypoperfusion induced by bilateral common carotid artery stenosis (BCAS).

Methods: A total of 74 wild-type male C57BL/6J mice underwent BCAS or sham surgery. One week after surgery, the mice were randomly assigned into three different groups having varied amounts of EE-standard housing with no EE conditions (std), limited exposure with 3 h EE a day (3 h) and full-time exposure to EE (full) for 12 weeks. At 16 weeks after BCAS surgery, behavioural and cognitive function were assessed prior to euthanasia. Brain tissues were analysed for the degree of gliosis including morphological changes in astrocytes and microglia.

Results: Chronic cerebral hypoperfusion (or BCAS) increased clasmatodendrocytes (damaged astrocytes) with disruption of aquaporin-4 immunoreactivity and an increased degree of microglial activation/proliferation. BCAS also impaired behavioural and cognitive function. These changes were significantly attenuated, by limited exposure compared to full-time exposure to EE.

Conclusions: Our results suggest that moderate or limited exposure to EE substantially reduced glial damage/activation. Our findings also suggest moderate rather than continuous exposure to EE is beneficial for patients with subcortical ischaemic vascular dementia characterised by white matter disease-related inflammation.

Keywords: Animal model; Chronic cerebral hypoperfusion; Clasmatodendrosis; Environmental enrichment; Glial activation/proliferation.

Fig. 1

Experimental protocol and animal groups. a Experimental protocol − BCAS, bilateral common carotid artery stenosis; EE, environmental enrichment. b Animal groups and the number of animals in each group. Sham-std, n = 11; Sham-3 h, n = 11; Sham-full, n = 11; BCAS-std, n = 13, BCAS-3 h, n = 10; BCAS-full, n = 12. c(A and B) Images of standard housing cage (A) and EE cage (B)

Fig. 2

Assessment of astrogliosis and clasmatodendrosis in the corpus callosum (CC). a(A–F) Representative images of GFAP-stained corpus callosum (CC) in each group. a(A) Sham-std; a(B) Sham-3 h; a(C) Sham-full; a(D) BCAS-std; a(E) BCAS-3 h; a(F) BCAS-full. Scale bar represents 25 μm. a(A–C) Arrow heads indicate normal astrocytes. a(D–F) Arrows indicate clasmatodendrocytes. b–c, Histograms showing GFAP-positively stained astrocytes per area of CC (% GFAP-stained area) (b) as well as numerical density of GFAP-positive astrocytes in the entire CC (c). BCAS-3 h showed decreased GFAP-stained area and decreased numerical density of GFAP-positive astrocytes in the entire CC compared with BCAS-std (**P < 0.01). BCAS-full showed decreased GFAP-stained area in the entire CC compared with BCAS-std (**P < 0.01) (b, c). BCAS-3 h also showed decreased numerical density of GFAP-positive astrocytes in the entire CC compared with BCAS-full (**P < 0.01) (c). d, e Histograms showing numerical density of GFAP-positive clasmatodendrocytes (damaged astrocytes) in the entire CC (d) and percentage of clasmatodendrocytes per total number of GFAP-positive astrocytes in the entire CC (e). BCAS-3 h and BCAS-full showed reduced numerical density of GFAP-positive clasmatodendrocytes in the entire CC and % of clasmatodendrocytes per all GFAP-positive astrocytes compared with BCAS-std (**P < 0.01) (d and e). BCAS-3 h also showed reduced numerical density of GFAP-positive clasmatodendrocytes in the entire CC compared with BCAS-full (*P < 0.05) (d) and % of clasmatodendrocytes per all GFAP-positive astrocytes compared with BCAS-std (**P < 0.01) (e). f, g Histograms showing negative correlation between numerical density of clasmatodendrocytes and CC volume (f) as well as correlation between % of clasmatodendrocytes and CC volume (g) in the CC of BCAS cohort. Pearson’s correlation analysis revealed that degree of clasmatodendrosis in the CC of BCAS cohort exhibited strong negative correlation with CC atrophy (r = 0.615, P < 0.01, between numerical density of clasmatodendrocytes and CC volume, (f); r = 0.583, P < 0.01, between % of clasmatodendrocytes in the CC and CC volume, (g)). Sham-std, n = 11; Sham-3 h, n = 11; Sham-full, n = 11; BCAS-std, n = 13, BCAS-3 h, n = 10; BCAS-full, n = 12

Fig. 3

GFAP and AQP4 distribution in the corpus callosum (CC). a(A–F) Representative images of double immunofluorescent staining for GFAP and AQP4 in the corpus callosum (CC). a(A) Sham-std; a(B) Sham-3 h; a(C) Sham-full; a(D) BCAS-std; a(E) BCAS-3 h; a(F) BCAS-full. Scale bar represents 10 μm. a(A-C) In Sham subgroups, AQP4 was normally distributed within the astrocytic end-feet around the vessels (arrow heads). a(D) and a(F) In BCAS subgroups, AQP4 was abnormally aggregated at the periphery of GFAP-positive astrocytes/clasmatodendrocytes, especially in BCAS-std and BCAS-full subgroups (arrows). a(E) In limited exposure to EE (BCAS-3 h), less abnormal distribution of AQP4 was seen in the CC. b Histogram showing % AQP4 dislocation in each group. EE, especially by limited exposure to EE (BCAS-3 h) attenuated AQP4 dislocation in the CC (**P < 0.01). Sham-std, n = 11; Sham-3 h, n = 11; Sham-full, n = 11; BCAS-std, n = 13, BCAS-3 h, n = 10; BCAS-full, n = 12

Fig. 4

Assessment of microglial proliferation and activation in the corpus callosum (CC). a (A–F) Representative images of Iba-1 stained corpus callosum (CC) in each group. a(A) Sham-std; a(B) Sham-3 h; a(C) Sham-full; a(D) BCAS-std; a(E) BCAS-3 h; a(F) BCAS-full. Scale bar represents 50 μm. a(A) Inset showing a normal microglial cell. a(D) Inset showing activated microglial cells. b, c Histograms showing degree of microglial proliferation in each subgroup. EE, especially by limited exposure to EE (BCAS-3 h) suppressed microglial proliferation in the CC induced by BCAS (**P < 0.01). d–f Histograms showing degree of microglial activation in each subgroup. Limited exposure to EE suppressed degree of microglial activation in both moderate (microglial cell body >7 μm) (**P < 0.01) (d) and highly (microglial cell body >10 μm) activated (*P < 0.05) (e) levels, as well as suppressed the number of non-activated microglial cells (**P < 0.01) (f). However, BCAS-full did not show suppressed highly activated microglia in the CC (e). Sham-std, n = 11; Sham-3 h, n = 11; Sham-full, n = 11; BCAS-std, n = 13, BCAS-3 h, n = 10; BCAS-full, n = 12

Fig. 5

Nesting ability and cognitive function assessed in the nine-arm radial maze [72, 73]. a(A) Representative images of nest created from the Nestlet in Sham and BCAS group. a(B) Boxplot showing Nestlet score at baseline level and post operation in each group. a(C, D) Histograms showing height of the nest (C) and % used Nestlet (D) at baseline level and post operation in each group. BCAS-std subgroup had impaired nesting ability compared to baseline level (*P < 0.05). EE reversed impaired nesting ability induced by BCAS, especially by limited EE (BCAS-3 h) (a(B)). b(A, B), Histograms showing number of arm entries before first repeat in all sham and all BCAS (A) as well as number of arm entries before first repeat of each subgroup (B) assessed in the nine-arm radial maze. Every consecutive 10 sessions and overall 20 sessions were averaged. b(A) All sham completed more arm entries before first repeat, compared with all BCAS in sessions 1–10, sessions 11–20 and sessions 1–20 (*P < 0.01 vs sham; †P < 0.01 vs sham; ‡P < 0.01 vs sham). b(B) Between BCAS subgroup, BCAS-3 h completed more arm entries compared with BCAS-std and BCAS-full in sessions 1–10, sessions 11–20 and sessions 1–20. Sham-std, n = 11; Sham-3 h, n = 11; Sham-full, n = 11; BCAS-std, n = 13, BCAS-3 h, n = 10; BCAS-full, n = 12