Blocking of microglia-astrocyte proinflammatory signaling is beneficial following stroke

Abstract

Microglia and astrocytes play an important role in the neuroinflammatory response and contribute to both the destruction of neighboring tissue as well as the resolution of inflammation following stroke. These reactive glial cells are highly heterogeneous at both the transcriptomic and functional level. Depending upon the stimulus, microglia and astrocytes mount a complex, and specific response composed of distinct microglial and astrocyte substates. These substates ultimately drive the landscape of the initiation and recovery from the adverse stimulus. In one state, inflammation- and damage-induced microglia release tumor necrosis factor (TNF), interleukin 1α (IL1α), and complement component 1q (C1q), together "TIC." This cocktail of cytokines drives astrocytes into a neurotoxic reactive astrocyte (nRA) substate. This nRA substate is associated with loss of many physiological astrocyte functions (e.g., synapse formation and maturation, phagocytosis, among others), as well as a gain-of-function release of neurotoxic long-chain fatty acids which kill neighboring cells. Here we report that transgenic removal of TIC led to reduction of gliosis, infarct expansion, and worsened functional deficits in the acute and delayed stages following stroke. Our results suggest that TIC cytokines, and likely nRAs play an important role that may maintain neuroinflammation and inhibit functional motor recovery after ischemic stroke. This is the first report that this paradigm is relevant in stroke and that therapies against nRAs may be a novel means to treat patients. Since nRAs are evolutionarily conserved from rodents to humans and present in multiple neurodegenerative diseases and injuries, further identification of mechanistic role of nRAs will lead to a better understanding of the neuroinflammatory response and the development of new therapies.

Keywords: astrocyte; infarct volume reduction; macrophage; recovery of function; stroke.

Figure 1

Experimental timeline and survival rate of animals 30 days after photothrombotic stroke. (A) Timeline showing 1- and 7-day sacrifice after dMCAO and 30-day sacrifice following photothrombotic stroke. Immunohistochemistry was performed to examine infarct volume (NeuN/cresyl violet), astrogliosis (GFAP), and microgliosis (CD68). Functional recovery assessment was performed up to 28 days after photothrombotic stroke include the rotating beam, tapered beam, and grid walk tasks. Survival rates were not significantly different between wildtype (WT) and Tnf^−/−Il1a^−/−C1q^−/− triple knockout (TKO) mice 1 (B) or 7 days after dMCAO (C), or 30 days (D) after photothrombotic stroke (X²= 1.85, p = 0.17). IHC; immunohistochemistry, GFAP; glial fibrillary acidic protein, CD68; Cluster of Differentiation 68, d, day; dMCAO, distal middle cerebral artery occlusion; TKO, Tnf^−/−Il1a^−/−C1q^−/− triple knockout mouse; WT, wild type.

Figure 2

TNF, IL1α, and C1q knockout reduces GFAP+ cell number in the peri-infarct area 7- and 30-days post ischemic stroke. Low- (5x) and high-magnification (40x) representative images of GFAP+ cells in the infarct and peri-infarct area for wildtype (WT) and Tnf^−/−Il1a^−/−C1q^−/− triple knockout (TKO) mice 1- (A), 7-days post-dMCAO (C) and 30-days post-photothrombotic stroke (E). (B) No difference was found in GFAP+ cells in the peri-infarct area 1 day after ischemia between the WT and TKO mice, t₍₂₉₎ = 1.29, p = 0.104 post-dMCAO. (D) TKO mice had significantly reduced GFAP+ cells in the peri-infarct area than WT mice 7 days post-dMCAO, t₍₁₂₎ = 2.64, p = 0.011. (F) TKO mice had significantly reduced GFAP+ cells in the peri-infarct area than WT mice 30 days post-photothrombotic stroke, t₍₁₄₎ = 5.03, p < 0.0001. * p < 0.05, ** p < 0.001, *** p < 0.0001. GFAP; glial fibrillary acidic protein, TKO, Tnf^−/−Il1a^−/−C1q^−/− triple knockout mouse; WT, wild type; DV, dorsoventral; ML, mediolateral.

Figure 3

TNF, IL1α, and C1q knockout reduces CD68+ cells in the peri-infarct area 7- and 30-days post ischemic stroke. Low- (5x) and high-magnification (40x) representative images of CD68+ cells in the infarct and peri-infarct area for wildtype (WT) and Tnf^−/−Il1a^−/−C1q^−/− triple knockout (TKO) mice 1- (A), 7-days post-dMCAO (C), and 30-days post-photothrombotic stroke (E). (B) No difference was found in CD68+ cells in the peri-infarct area 1 day after ischemia between the WT and TKO mice, t₍₂₉₎ = 0.29, p = 0.285, post-dMCAO. (D) TKO mice had significantly reduced CD68+ cells in the peri-infarct area than WT mice 7 days post-dMCAO, t₍₁₅₎ = 3.51, p = 0.002. (F) TKO mice had significantly reduced CD68+ cells in the peri-infarct area than WT mice 30 days post-photothrombotic stroke, t₍₁₉₎ = 4.55, p = 0.0001. * p < 0.05, ** p < 0.001, *** p < 0.0001. CD68; Cluster of Differentiation 68, TKO, Tnf^−/−Il1a^−/−C1q^−/− triple knockout mouse; WT, wild type; DV, dorsoventral; ML, mediolateral.

Figure 4

TNF, IL1α, and C1q knockout reduces infarct volume 1-, 7-, and 30-days post stroke. (A) Representative images of stroke volume using NeuN/cresyl violet stain of wildtype (WT) and Tnf^−/−Il1a^−/−C1a^−/− triple knockout (TKO) mice 1-, 7-days post-dMCAO, and 30-days post-photothrombotic stroke. (B) TKO mice had significantly reduced infarct volume compared to WT mice 1-day post-dMCAO, t₍₂₉₎ = 2.09, p = 0.022. (C) TKO mice had significantly reduced infarct volume than WT mice 7 days post-dMCAO, t₍₁₂₎ = 4.64, p < 0.0001. (D) TKO mice had a significantly reduced infarct volume compared to the WT mice 30 days post-photothrombotic stroke, t₍₁₅₎ = 2.712, p = 0.008. * p < 0.05, ** p < 0.001, *** p < 0.0001. dMCAO, distal middle cerebral artery occlusion; TKO, Tnf^−/−Il1a^−/−C1q^−/− triple knockout mouse; WT, wild type.

Figure 5

TNF, IL1α, and C1q deletion reduces behavioral deficits after ischemic stroke. (A) Tnf^−/−Il1a^−/−C1q^−/− triple knockout (TKO) mice a significantly further distance on the rotating beam as the wildtype (WT) mice, F_{(1, 24)} = 0.3969, p = 0.026. A post-hoc analysis revealed that TKO mice made fewer errors 1- (p = 0.0358) and 3 days (p = 0.0244) after stroke. (B) TKO mice mad significantly fewer errors on the tapered beam task than WT mice, F_{(1, 24)} = 6.342, p = 0.0189. A post-hoc analysis revealed that TKO mice made fewer errors 1- (p < 0.0124), 3- (p < 0.023), 7- (p < 0.0329), and 14 (p < 0.0453) days after stroke. (C) TKO mice and WT mice made a similar number of contralateral forelimb errors on the grid walk task, F_{(1, 32.008)} = 0.015, p = 0.91. * p < 0.05. TKO, Tnf^−/−Il1a^−/−C1q^−/− triple knockout mouse; WT, wild type.