Dose-dependent modulation of microglia activation in rats after penetrating traumatic brain injury (pTBI) by transplanted human neural stem cells

Abstract

Traumatic brain injury (TBI) often results in long-lasting patterns of neurological deficits including motor, sensory, and cognitive abnormalities. Cranial gunshot survivors are among the most disabled TBI patients and face a lifetime of disability with no approved strategies to protect or repair the brain after injury. Recent studies using a model of penetrating TBI (pTBI) have reported that human neural stem cells (hNSCs) transplantation can lead to dose and location-dependent neuroprotection. Evidence for regional patterns of microglial activation has also been reported after pTBI with evidence for microglial cell death by pyroptosis. Because of the importance of injury-induced microglial activation in the pathogenesis of TBI, we tested the hypothesis that dose-dependent hNSC mediated neuroprotection after pTBI was associated with reduced microglial activation in pericontusional cortical areas. To test this hypothesis, quantitative microglial/macrophage Iba1 immunohistochemistry and Sholl analysis was conducted to investigate the arborization patterns using four experimental groups including, (i) Sham operated (no injury) + low dose (0.16 million cells/rat), (ii) pTBI + vehicle (no cells), (iii) pTBI + low dose hNSCs (0.16 million/rat), and (iv) pTBI + high dose hNSCs (1.6 million cells/rat). At 3 months post-transplantation (transplants at one week after pTBI), the total number of intersections was significantly reduced in vehicle treated pTBI animals versus sham operated controls indicating increased microglia/macrophage activation. In contrast, hNSC transplantation led to a dose-dependent increase in the number of intersections compared to pTBI vehicle indicating less microglia/macrophage activation. The peak of Sholl intersections at 1 μm from the center of the microglia/macrophages ranged from ~6,500-14,000 intersections for sham operated, ~250-500 intersections for pTBI vehicle, ~550-1,000 intersections for pTBI low dose, and ~2,500-7,500 intersections for pTBI high dose. Plotting data along the rostrocaudal axis also showed that pericontusional cortical areas protected by hNSC transplantation had increased intersections compared to nontreated pTBI animals. These studies using a non-biased Sholl analysis demonstrated a dose-dependent reduction in inflammatory cell activation that may be associated with a neuroprotective effect driven by the cellular transplant in perilesional regions after pTBI.

Fig 1. Schematic diagram of a saggital brain section to illustrate the bregma range of the lesion and sites examined.

The schematic shows a sagittal view of the 11mm brain block (top row) to provide the context for the lesion, transplant locations with respect to bregma. Next set of rows lists sites examined for microglial morphology. The first column lists names of the rows and subsequent 11 columns represent 11 sections 1 mm apart along rostro caudal axis with features such as lesion and transplant. The rostro caudal axis with respect to bregma (first row) controls for location. The pTBI lesion (filled red squares) in next row is present only on one side of the brain and across 6mm (+3.0mm to -2mm). The transplants (green bars) in row below show locations of the hNSC deposits. In the next set the top row show brain sections stained with IbaI (filled black squares), the sites at which the microglia morphology was examined which includes penumbra regions with spared cortex (blue filled squares) and hippocampus (blue filled squares).

Fig 2. Filament reconstruction.

Filament reconstruction was utilized to render a microglia structure. The blue sphere indicated the soma of the microglia. The seed points were connected to create red filaments of varying length. The sphere and filaments were connected to form entire microglia (A). This procedure was performed utilizing Filament Tracer Module in Imaris Software. Magnified representations of the microglia are shown on the right panel with (B) and without (C) fluorescence from other channels. These images correspond to the ipsilateral cortex.

Fig 3. Cell dose dependent mitigation of pTBI lesion progression.

(A) The mean lesion volume (±SEM) as a % of the left hemisphere shows the injury effect and hNSC neuroprotection. Statistical significance determined using a one-way ANOVA followed by a Dunnett’s post hoc test. (B) The GFP volume as a % of the left hemisphere captures dose response with the high dose group higher than the low dose group and sham group. Statistical significance was similarly determined using a one-way ANOVA followed by a Dunnett’s post hoc test. ****p < 0.0001; **p < 0.01.

Fig 4. High power images for morphological analysis.

High power images of (A) ipsilateral cortex and (B) ipsilateral dentate gyrus from four experimental groups: Sham (first), Vehicle (second), low dose (third) and high dose (fourth rows) in channels (labeled above the image) shows activated morphology (rounded microglia without processes), ramified microglia (highly branched). The merge column indicates evidence for the degree of microglia activation. Scale bar is 50 μm.

Fig 5. Microglia morphological transition.

The microglia activation transition can be represented from lowest to highest microglia activation corresponding to the analyzed microglia morphology: sham (resting) < high dose (ramified)< low dose (ramified with partial ameboid) < vehicle (ameboid).

Fig 6. Rostrocaudal axis.

(A) A scatter plot of individuals from experimental groups along the rostrocaudal axis shows the extent of injury and location of therapeutic effect in penumbra. The number of intersections trends higher in Sham and lower in Vehicle along the rostrocaudal axis, with low dose and high dose between them. Peak radius intersections (low microglia activation) was similar to Sham only in the penumbra (-2.5 mm to -3.5 mm bregma) of high dose while at the lesion core (-0.5 mm to -2.0 mm bregma) values were still low (activated microglia). (B) Along the rostrocaudal axis, statistical significance was determined for the mean number of intersections (+SEM) using a repeated measures ANOVA followed by Tukey’s post hoc test.

Fig 7. Sholl analysis of distance from soma graphs.

(A) Ipsilateral Cortex Peak Sham Intersections: (~12,000 intersections), Vehicle (~500 intersections), Low dose (~1,000 intersections), High dose (~7,500 intersections). (B) Contralateral Cortex Peak Intersections: Sham (~14,000 intersections), Vehicle (~500 intersections), Low dose (~750 intersections), High dose (~6,500 intersections). (C) Ipsilateral Dentate Gyrus Peak Intersections: Sham (~6,500 intersections), Vehicle (~250 intersections), Low dose (~550 intersections), High dose (~3,000 intersections). (D) Contralateral Dentate Gyrus Peak Intersections: Sham (~6,500 intersections), Vehicle (~250 intersections), Low dose (~550 intersections), High dose (~2,500 intersections).