Early life stress increases vulnerability to the sequelae of pediatric mild traumatic brain injury

Abstract

Early life stress (ELS) is a risk factor for many psychopathologies that happen later in life. Although stress can occur in cases of child abuse, studies on non-accidental brain injuries in pediatric populations do not consider the possible increase in vulnerability caused by ELS. Hence, we sought to determine whether ELS increases the effects of pediatric mild traumatic brain injury (mTBI) on cognition, hippocampal inflammation, and plasticity. Male rats were subjected to maternal separation for 180 min per day (MS180) or used as controls (CONT) during the first 21 post-natal (P) days. At P21 the rats were anesthetized with isoflurane and subjected to a mild controlled cortical impact or sham injury. At P32 the rats were injected with the cell proliferation marker bromodeoxyuridine (BrdU, 500 mg/kg), then evaluated for spatial learning and memory in a water maze (P35–40) and sacrificed for quantification of Ki67⁺, BrdU⁺ and Iba1⁺ (P42). Neither MS180 nor mTBI impacted cognitive outcome when provided alone but their combination (MS180 + mTBI) decreased spatial learning and memory relative to Sham controls (p< 0.01). mTBI increased microglial activation and affected BrdU⁺ cell survival in the ipsilateral hippocampus without affecting proliferation rates. However, only MS180 + mTBI increased microglial activation in the area adjacent to the injury and the contralateral CA1 hippocampal subfield and decreased cell proliferation in the ipsilateral neurogenic niche. Overall, the data show that ELS increases the vulnerability to the sequelae of pediatric mTBI and may be mediated by increased neuroinflammation.

Introduction

Abusive head trauma (AHT) is one of the leading causes of traumatic brain injury (TBI) in young children. In the United States, AHT represents the most common cause of death in physically abused children under age 5 (Klevens and Leeb, 2010). The cognitive and behavioral deficits caused by AHT are more severe than non-abusive head trauma (Niederkrotenth et al., 2013) and most AHT survivors show long-term disabilities (Chevignard and Lind, 2014). Child abuse is a significant health issue that deserves especial attention. International data from the World Health Organization reported that 25% of adults were physically abused as children and approximately 20% of women and 5–10% of men reported being victims of sexual abuse during childhood (Krug et al., 2002). In Mexico, which is first in indices of child abuse within the countries of the Organization for Economic Cooperation and Development (Secretaria de Salud, 2006), one third of children aged 6–9 years reported that they experienced violence at school and within their family. Early life stress (ELS) can occur in cases of abuse and neglect and has long-term consequences for the individual (Carr et al., 2013; Pechtel et al., 2011). Chronic exposure to stress during early life increases the likelihood of experiencing psychopathologies, such as anxiety, depression, and cognitive impairments during adolescence (Burton, 2008; Calvete, 2014; Chapman et al., 2004) and adulthood (Heim et al., 2008, 1997; Pesonen et al., 2007). Moreover, ELS causes a dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis that could increase vulnerability later in life (Heim et al., 2008; McEwen, 2012; Teicher et al., 2003). Albeit situations leading to AHT may be accompanied by chronic stress exposure, studies on pediatric[populations have not fully considered the possible interaction of ELS combined with TBI.

The inflammatory response is a significant component of the pathophysiology of TBI (Kochanek et al., 2015; Semple et al., 2016). During TBI, microglia rapidly become activated and undergo morphological changes that turn them from a branched resting phenotype into an active hypertrophic, “bushy” morphology (for review see Chiu et al., 2016; Simon et al., 2017). Activated microglia release pro-inflammatory cytokines and decrease anti-inflammatory cytokines (Loane and Byrnes, 2010; Hernandez-Ontiveros et al., 2013; Karve et al., 2016). Increased neuroinflammation during the acute TBI phase exerts beneficial roles when the response is controlled and regulated but a failure in the resolution of the inflammatory response and increased microglial activation persists for many years after TBI in humans (Johnson et al., 2013; Ramlackhansingh et al., 2011). It has been suggested that microglial activity may be involved in the ongoing pathogenesis after TBI in the immature brain (Raghupathi and Huh, 2017). Similarly, it is known that the release of pro-inflammatory cytokines and the activation of microglia are induced by chronic stressful events (Blandino et al., 2006, 2009; Johnson et al., 2005). Moreover, victims of AHT showed increased levels of macrophage-microglia derived neurotoxin in the cerebrospinal fluid (Berger et al., 2004) suggesting a shared role of chronic microglial activation in cognitive and functional impairments after both TBI and stress. ELS and TBI can also have detrimental effects on hippocampal plasticity. The subgranular zone (SGZ) of the hippocampus dentate gyrus (DG) is a neurogenic niche that generates neurons constantly (Altman and Das, 1965). Hippocampal neurogenesis is related to spatial learning and memory (Gould et al., 1999) and it has been shown that newly generated hippocampal DG cells are necessary for appropriate HPA axis negative feedback to occur (Schloesser et al., 2009). Although it is well documented that there is an increase in hippocampal neurogenesis during the immediate period after TBI (Dash et al., 2001), it has also been observed that TBI affects hippocampal neurogenesis long-term by depleting the proliferative capacity of the neurogenic niche (Neuberger et al., 2017). Moreover, ELS also affects the hippocampus neurogenic niche (Aisa et al., 2009; Lajud et al., 2012; Mirescu et al., 2004; Ruiz et al., 2018; Suri et al., 2014). Furthermore, a common consequence of ELS and TBI are deficits in neurogenesis related functions such as learning and memory (Teicher and Samson, 2016; Hamm et al., 1992), which suggest that the alterations on this type of hippocampal plasticity may be a plausible explanation for the cognitive impairments observed after TBI or chronic stress.

The most common rodent model to study the effects of ELS is maternal separation for 180 min per day (MS180) (Plotsky and Meaney, 1993; Wigger and Neumann, 1999; Lajud et al., 2012; Ruiz et al., 2018). MS180 consists of the separation of the pups from their mother during the first weeks of life (Levine, 1967). As adults, MS180 rats show an increase in basal HPA axis activity that is related to depressive-like behavior and cognitive impairments (Aisa et al., 2007; Lajud et al., 2012; Reshetnikov et al., 2018). Moreover, MS180 increases activated microglia in various parts of the hippocampus and promotes neuroinflammation (Roque et al., 2015; Gracia-Rubio et al., 2016). Also, MS180 can cause a decrease in hippocampal neurogenesis that is related to behavioral and neuroendocrine alterations (Mirescu et al., 2004; Lajud et al., 2012; Suri et al., 2014; Ruiz et al., 2018). Hence, the aim of this study was to test the hypothesis that ELS induced by MS180 increases the effects of pediatric mTBI on cognition, hippocampal inflammation, and plasticity relative to un-injured Sham controls.

Materials and Methods

Rats

Fourteen time-pregnant female Sprague-Dawley rats (Envigo RMS, Inc., Indianapolis, IN) were obtained on embryonic day-14. The rats were maintained in temperature-controlled rooms with a 12-h light/dark cycle and ad libitum food and water. Day of birth was considered as postnatal day zero (P0). On P1, litters were cross-fostered, adjusted to eight pups (2–4 females), and randomly assigned to the MS180 protocol or left undisturbed with their mothers (CONT). Pups were weaned at P21 and received a mild CCI or sham injury (Sham). To avoid sex-and litter-dependent effects, only males were evaluated, and each experimental group consisted of pups from at least six different dams (CONT + Sham, n = 15; MS180 + Sham, n = 15; CONT + mTBI, n = 15; and MS180 + mTBI, n = 15). Females were reserved for a different study. All procedures were approved by the University of Pittsburgh’s Institutional Animal Care and Use Committee and the National Research Committee of the Mexican Social Security System (R-2016-785-054). All experiments were carried out in accordance with the Institute of Laboratory Animal Resources Guide for the Care and Use of Laboratory Animals (National Research Council, revised 1996) and the official regulations for use and care of laboratory animals of Mexico (NOM-062-ZOO-1999). Every effort was made to reduce the number of rats used and to minimize suffering.

Maternal separation

On P1, pups of the MS180 group were separated from their dam and placed in a box filled with clean sawdust and placed in a room separated from the colony, with controlled temperature (30–32°C) for 3 h as previously described (Lajud et al., 2012; Ruiz et al., 2018) and then returned to the dam. This procedure was repeated every day at the same time (09:00–12:00) until weaning (P21). The control group remained undisturbed except for routine cage cleaning.

Controlled cortical injury (CCI)

Rats weighing 45–65 g were subjected to a CCI or sham injury on P21. Briefly, anesthesia was induced and maintained with inspired concentrations of 4% and 2% isoflurane, respectively, in 2:1 N₂O:O₂. The rats were secured in a stereotaxic frame, and a midline scalp incision was made utilizing aseptic conditions. The skin and fascia were reflected to expose the skull, then a craniectomy encompassing bregma and lambda and the sagittal and coronal sutures was made in the right hemisphere with a dental drill. The bone flap was removed, the impacting rod was extended and the impact tip (6 mm, flat) was centered and lowered through the craniectomy until it touched the dura mater, then the rod was retracted, and the impact tip was advanced 2.2 mm farther to produce a mTBI (2.2 mm tissue deformation at 4 m/s). Body temperature was monitored with a rectal probe and maintained at 37 ± 0.5°C with a heating blanket. Immediately after the CCI, anesthesia was discontinued, and the incision was promptly sutured. The rats were assessed for acute neurological outcome. Sham-operated rats underwent similar surgical procedures but were not subjected to the impact.

Acute neurological evaluation

Hind limb reflexive ability was assessed immediately after the cessation of anesthesia and removal from the stereotaxic apparatus by gently squeezing the rats’ paw every 5 s and recording the time to elicit a withdrawal response. Return of the righting reflex was determined by the time required to turn from the supine to prone position. These neurological indices are used to determine the level of injury severity (Kline et al, 2002, 2010, 2012; Bondi et al., 2014a,b; Radabaugh et al., 2017; Lajud et al., 2019; Njoku et al., 2019).

BrdU injections

To evaluate learning-induced neurogenesis, the rats were injected twice at ten days before sacrifice (12-hr apart, P32) with bromodeoxyuridine (BrdU, Sigma-Aldrich 500 mg/kg) and trained in a spatial learning task as previously described (Lajud et al., 2019). A different cohort of rats were injected on the same dates but were evaluated under baseline conditions (naïve).

Spatial learning evaluation

Cognitive performance was assessed in a Morris water maze (MWM) task (Morris, 1984) on P35 to P39 (post injury days 14–18). The maze consisted of a circular pool (180 cm diameter × 60 cm height) that was filled with water (26°C ± 1°C) to a height of 28 cm. The pool was divided into four quadrants and located in the center of a room with salient visual cues that remained constant throughout the training. The platform (26 cm in height) was placed 2 cm below the water′s surface in the southwest quadrant. Each rat was provided four trials per day (4-min inter-trial interval) for five consecutive days of training. In each trial, the rat was placed in the maze facing the wall from a different quadrant and given a maximum time of 120 s to find the escape platform. If the rat was unable to find the platform, it was manually guided to it where it remained for 30 s before being returned to a heated incubator between trials. The distance traveled to reach the platform was recorded in all trials. On P40, a probe trial was performed with the platform removed to assess memory retention. Each rat was placed in the pool and allowed to swim freely for 30 s. The percent of distance traveled within the target quadrant relative to their respective totals were calculated. On the same day, but after the probe trial, a visible platform test was performed with the platform elevated 2 cm above the water level to determine the contribution of factors such as visual acuity and sensory-motor performance (Kline et al, 2002, 2010, 2012; Cheng et al., 2008, 2012; Radabaugh et al., 2017; Lajud et al., 2019).

Histology

On P42, rats received an overdose of sodium pentobarbital (i.p.) and perfused intracardially with 200 mL of saline solution (0.9% NaCl) and 200 mL of 4% paraformaldehyde [PFA] in 0.1 M phosphate buffer [PB]). The brains were extracted, placed in 4% PFA for 24 hr, transferred to 30% sucrose PB until sunken, then frozen. Random systematic sampling of 40 μm thick slices along the rostro-caudal axis of the hippocampus was performed using a cryostat. The sections were placed in tubes with cryoprotective solution (25% glycerol, 25% ethylene glycol, 50% PB) until processing.

Immunohistochemistry

A series of 40 μm-thick coronal sections were cut at 240 μm intervals and randomly selected for immunohistochemistry. For BrdU staining, brain slices were rinsed in PB and incubated in PBT (PB + Triton X-100 0.3%) containing 10% hydrogen peroxide. Subsequently, sections were incubated in absolute methanol, rinsed in PB and incubated in formamide (50% SSC, Sigma-Aldrich) at 74°C. After SSC wash, DNA denaturation was performed in HCl (1N) at 37°C followed by incubation in a borate buffer solution (pH 8.4). After pre-treatments, slices were incubated overnight in the primary anti-BrdU antibody solution (mouse anti-BrdU, 1:1000, Roche Molecular). Brains were incubated in the biotinylated secondary antibody solution (goat anti-mouse, 1:750, Vector Laboratories,) and visualized with avidin-biotin complex (Elite ABC kit, Vector Laboratories), and Ni-DAB solution (DAB staining kit, Vector Laboratories). Brains were mounted on gelatinized slides, counter stained with hematoxylin-eosin (Vector laboratories), and cover-slipped. For quantification of proliferation and microglia labeling, the brain sections were pretreated with hydrogen peroxide and methanol and further incubated in primary antibody solution for detecting the proliferation associated protein Ki-67 (rabbit anti Ki-67, 1:3000 ABCAM) or ionized calcium-binding adaptor molecule 1 (mouse anti-Iba1 / IAF1, 1:3000, Millipore). Sections were visualized as described for BrdU.

Immunofluorescence

To characterize the cellular phenotype of the BrdU-labeled nuclei, triple immunofluorescence was performed for BrdU, the mature neuron nuclear antigen NeuN, the marker of immature neurons doublecortin (DCX), or the glial cell marker GFAP as previously described (Lajud et al., 2019). For BrdU/GFAP/NeuN labeling, sections were incubated overnight with a mixture of mouse anti-NeuN antibody (1:300, Millipore), rat anti BrdU antibody (1:200, ABD Serotec), and rabbit anti GFAP (1:300, Millipore). Slices were further visualized with corresponding goat anti mouse Alexa Fluor 647 (1:200, Invitrogene), donkey anti rat DyLight 405 (1:200, Jackson Immuno Research), and goat anti rabbit Alexa Fluor 488 (1:200, Jackson Immuno Research), mounted and cover-slipped with DAKO (DAKO, Glostrup, Denmark). For BrdU/DCX/NeuN labeling, slices were first incubated overnight in a mixture of two primary antibodies that detect different DCX epitopes, guinea pig anti DCX (1:400, Millipore) and goat anti DCX (1:200, Santa Cruz). Sections were further placed in a mixture of a biotinylated goat anti guinea pig IgG and donkey anti goat IgG (1:300, Vector laboratories) antibodies and visualized with the fluorescent avidin-FITC substrate kit (Vector laboratories). After DCX staining, slices were further incubated in a mixture of primary rat anti BrdU IgG (1:200, ABD Serotec, Oxford, UK) and mouse anti NeuN IgG (1:300, Millipore). Sections were then visualized with secondary donkey anti rat IgG (1:300, Alexa 405 Jackson Immuno) and goat anti mouse IgG (1:300, Alexa 643, Jackson Immuno) antibodies, mounted and cover-slipped with DAKO mounting media (DAKO, Glostrup, Denmark). This 2-phase immunostaining procedure was done to ensure that no cross reactivity was observed between Alexa 643 goat anti mouse fluorescent antibody used to visualize NeuN and donkey anti goat biotinylated secondary used to visualize DCX.

BrdU+ and Ki67+ Cell counting

For stereology, the numbers of BrdU⁺ and Ki67⁺ cells along the entire hippocampus were evaluated by an experimenter blinded to group conditions. The boundaries of the granular and SGZ of the DG were digitally outlined using the software Axio Vision 4.6, Karl Zeiss. Total cell numbers were calculated in the granular and SGZ. DG volume was determined according to the Cavalieri principle as previously described (Lajud et al., 2012, 2019; Ruiz et al., 2018). Evaluation of co-localization of the total BrdU⁺ nuclei was made with an Olympus Fluoview FV1000 confocal microscope. A series of microphotographs 1 μm apart along the z-axis of the slice were obtained to generate an orthogonal view. The images were processed with the program “FluoView FV10-ASW 2.0”, and only BrdU⁺ nuclei that showed co-localization on the z-axis were considered for quantification.

Microglia evaluation

For Iba1⁺ microglial cell evaluation, four sections of the dorsal hippocampus were randomly selected for evaluation. Iba1⁺ cells were quantified in four different optical frames per section, captured at a 20x magnification by an observer blinded to experimental conditions and assessed in the areas of interest of each section. Iba1⁺ microglia were classified as resting glia when they presented a small soma with long and thin ramifications, and as an activated glia when a large soma with short and thick ramifications (bushy) were present. The proportion of activated Iba1⁺ cells was estimated as previously described (Roque et al., 2015).

Cortical lesion volume

40 μm-thick coronal sections were cut at 480 μm intervals through the lesion on a cryostat and mounted on Superfrost®/Plus glass microscope slides. After drying at room temperature, the sections were rehydrated and stained with cresyl violet. Cortical lesion volume (mm³) was assessed by an observer blinded to experimental conditions using a Nikon Eclipse 90i microscope (Nikon Corporation, Tokyo, Japan). The area of the lesion (mm²) was first calculated by outlining the inferred area of missing cortical tissue for each section (Nikon NIS-Elements AR 3.22.14 software Nikon Corp, Tokyo, Japan), and then by summing the lesions obtained, as previously reported (Olsen et al., 2012; Monaco et al., 2013, 2014).

Data analyses

Statistical analyses were performed on data generated by experimenters blinded to conditions using GraphPad Prism 7.0 (GraphPad software, US.). Cognitive data were analyzed by three-way repeated-measures analysis of variance (ANOVA) with ELS, injury, and training day as factors. The acute neurological assessment, probe trial, visible platform, swim speed, Ki67, BrdU, and Iba1 quantifications were analyzed by two-way ANOVAs with ELS and injury as factors. When the ANOVA showed a significant effect, the Tukey post-hoc test was utilized to determine specific group differences. The results are expressed as the mean ± standard error of the mean (S.E.M.) and are considered significant when p≤ 0.05.

Results

Acute neurological evaluation

Two-way ANOVAs showed significant effects of mTBI (vs. Sham) for return of hind limb reflex ability after a brief paw pinch (right, F_1,85 = 1654, p< 0.0001 and left, F_1,85 = 1700, p< 0.0001) and righting reflex latency (F_1,85 = 428, p< 0.0001) following the cessation of anesthesia. No significant differences were observed between CONT + mTBI and MS180 + mTBI in any of the parameters (p>0.05). (see Table 1 for data).

Table 1.

Acute Neurological Assessments

Groups	Right limb withdrawal (s)	Left limb withdrawal (s)	Righting reflex (s)
CONT + Sham	11.5 ± 0.8	15.2 ± .9	76.9 ± 2.7
MS + Sham	9.5 ± 0.6	13.1 ± .5	88.5 ± 4.5
CONT + TBI	123.4 ± 4.6**	127.2 ± 4.6**	212.6 ± 6.6**
MS + TBI	122.8 ± 4.3**	126.7 ± 4.2**	202.5 ± 8.3**

Mean (± S.E.M.) acute neurological assessments. No significant differences were revealed between the CONT + Sham and MS + Sham groups or between the CONT + TBI and MS + TBI groups in time (s) to elicit a right and left hind paw reflexive withdrawal (after a brief paw pinch) or righting reflex after the cessation of anesthesia. There was, however, a difference in reflexive ability between the Sham and TBI groups.

^**(p< 0.01 vs. all Sham groups. CONT + Sham, n = 17; MS + Sham, n = 17; CONT + TBI, n = 17; MS + TBI, n = 17).

MS180 causes cognitive deficiencies after mTBI

To evaluate the effects of ELS on cognitive performance after mTBI, we used a well-established MWM task as a measure of hippocampal-dependent spatial learning and memory (Fig. 1A). During the acquisition phase, significant main effects of injury (F_1,44 = 4.7, p= 0.03) and training day (F4,176 = 9.7, p< 0.0001), but not ELS, were observed in the distance traveled to locate the escape platform. Multiple comparisons revealed that only the MS180 + mTBI group exhibited cognitive deficits as evidenced by a significantly greater distance traveled to find the platform compared to the CONT + Sham group (p< 0.01), suggesting that the combination of MS180 and mTBI affects spatial learning in this task (Fig. 1B). On P40, the platform was removed, and the percentage of distance traveled in the target quadrant was calculated. There was a significant injury effect (F_1,44 = 4.0, p< 0.05) and interaction (F_1,44 = 4.1, p= 0.04) but not ELS. Multiple comparisons indicated that only the MS180 + mTBI group swam significantly shorter distances in the target quadrant compared to the CONT + Sham group (Fig. 1C). Analysis of the routes in the MWM showed that the MS180 + mTBI group swam predominantly at the edges of the tank (i.e., thigmotaxis) from days 1 to 5 and during the probe trial, which was markedly different from the CONT + Sham group that was more likely to swim in the center of the tank (Fig. 1D).

Fig. 1.

Early life stress causes cognitive impairments in a Morris water maze task after pediatric mild traumatic brain injury (mTBI). (A) Control (CONT) and maternally separated (180 min/day, MS180) rats were subjected to either a mTBI or sham surgery on postnatal day (P) 21. (B) Distance traveled to locate the submerged platform was evaluated on P35- 39 (Mean ± S.E.M. *p<0.05 and **p<0.01 vs. CONT + Sham. ^$p<0.05 vs. MS180 + Sham). No differences were revealed for distance to reach the visible platform on P40. (C) A probe trial was provided on P40 and the percent of distance traveled in the target quadrant was evaluated (Mean ± S.E.M. *p<0.05 vs. CONT + Sham) (D) Representative heat maps demonstrating swim trajectories on days 1 and 5 of spatial learning (P35 and P39) as well as the probe trial (i.e., memory retention). (CONT + Sham, n=12; MS180 + Sham, n=12; CONT + mTBI, n=12; MS180 + mTBI, n=12).

Effects of mTBI and MS180 on cortical lesion and subgranular cell layer volume

Cortical lesion and hippocampal neurogenic niche volumes were evaluated by stereology, as well as Iba1⁺ cell density in the area adjacent to the lesion. No statistically significant differences were observed between the CONT + mTBI and MS180 + mTBI groups on lesion volume (p> 0.05, Fig. 2A). Two-way ANOVA for the proportion of Iba1⁺ cells that presented an activated phenotype indicated a significant effect of injury (F_1,28 = 7.21, p= 0.012) but not MS180 or interaction (p> 0.05). Multiple comparisons revealed that only the MS180 + mTBI group showed a significant 81% increase in the percentage of activated Iba1+ cells in the area adjacent to the lesion compared to the CONT + Sham group (p< 0.05, Fig. 2B). We also evaluated the effect of MS180 and mTBI on hippocampal neurogenic niche volume (Fig. 2C and D). A significant main effect of mTBI (F_1,36 = 41.9, p< 0.0001) was observed in the ipsilateral subgranular cell layer volume, but not of MS180 or interaction (p> 0.05). Multiple comparisons indicated that the CONT + mTBI group showed a 26 ± 6% (p< 0.01) and MS180 + mTBI a 36 ± 4% decrease (p< 0.0001) in the ipsilateral SGZ volume compared to CONT + Sham, and a 26 ± 6% (p< 0.001) and 35 ± 4% (p< 0.0001) decrease compared to MS180 + Sham group (Fig. 2C). There was no effect of mTBI or MS180 on the contralateral SGZ volume (p> 0.05, Fig. 2D).

Fig. 2.

Pediatric mild traumatic brain injury (mTBI) increases microglial activation in the lesion area and affects hippocampal dentate gyrus volume. (A) Cortical lesion volume (Mean ± S.E.M., CONT + mTBI, n = 5 and MS180 + mTBI, n = 5). (B) Percent of activated microglial (Iba1⁺) cells in the lesion area. (Mean ± S.E.M., ANOVA *p<0.05 vs. CONT + Sham. CONT + Sham, n = 8; MS180 + Sham, n = 8; CONT + mTBI, n = 8; MS180 + mTBI, n = 8). Stereological estimations of (C) ipsilateral and (D) ontralateral hippocampus dentate gyrus volume (Mean ± S.E.M. ANOVA **p<0.01 and ***p<0.001 vs. CONT + Sham. ^##p<0.01 and ^###p<0.001 vs. MS180 + Sham. CONT + Sham, n=10; MS180 + Sham, n=10; CONT + mTBI, n=10; MS180 + mTBI, n=10).

ELS increases microglial activation in the contralateral CA₁ hippocampus subfield

To test whether an increase in hippocampal inflammatory tone could explain the cognitive deficiency observed in the MWM in the MS180 + mTBI group, we estimated the proportion of Iba1⁺ cells that presented an activated phenotype in the ipsilateral CA₁ hippocampal subfield (Fig. 3). A significant effect of mTBI (F_1,28 = 37.63, p< 0.0001) but not of MS180 or interaction (p> 0.05) was revealed in the ipsilateral CA₁. Multiple comparisons indicated that the CONT + mTBI and MS180 + mTBI groups showed a 102 ± 18 (p< 0.01) and 140 ± 21 (p< 0.0001) increase in the percentage of activated Iba1⁺ cells in the ipsilateral CA₁ subfield compared to CONT + Sham (Fig. 3C) and a 63 ± 15% (p< 0.05) and 94± 17% (p< 0.001) increase compared to the MS180 + Sham group (Fig. 3C). In the contralateral CA₁ hippocampus subfield, statistical analysis showed that there was a significant effect of mTBI (F_1,28 = 12.3, p< 0.001) and MS180 (F_1,28 = 5.0, p< 0.05) but no interaction (p> 0.05). Multiple comparisons revealed that only the MS180 + mTBI group caused a 97 ± 19% increase (p< 0.01) in the proportion of activated Iba1⁺ cells in the contralateral CA₁ hippocampal subfield compared to the CONT + Sham group (Fig. 3D).

Fig. 3.

Early life stress increases microglial activation in the contralateral CA hippocampal subfield after pediatric mild traumatic brain injury (mTBI). (A) Control (CONT) and maternally separated (180 min/day, MS180) rats were subjected to either a mTBI or sham surgery on postnatal day (P) 21. Rats were sacrificed on P42 and microglial cell activation was quantified in the CA. (B) Representative images of Iba1 immunostaining. Inserts illustrate higher magnifications of activated Iba1⁺ cells depicted by arrows (scale bar = 200 μm). (C) Percent of activated Iba1⁺cells in the ipsilateral and (D) contralateral CA hippocampus subfield. (Mean ± S.E.M. ANOVA **p<0.01 and ***p<0.001 vs CONT + Sham. ^#p<0.05, and ^###p<0.001 vs MS180 + Sham. CONT + Sham, n=8; MS180 + Sham, n=8; CONT + mTBI, n=8; MS180 + mTBI, n=8).

ELS does not further increase microglial activation in the hilus

We estimated the proportion of Iba1⁺ cells that presented an activated phenotype and the statistical analysis showed there was a significant effect of mTBI (F_1,28 = 74.5, p> 0.0001) but not of MS180 or interaction (p> 0.05). Multiple comparisons showed that CONT + mTBI caused a 121 ± 7% (p< 0.0001) increase in the proportion of activated Iba1⁺ cells compared to CONT + Sham, and a 95 ± 6% increase (p< 0.0001) compared to MS180 + Sham, while MS180 + mTBI caused a 124 ± 9% (p< 0.0001) increase compared to CONT + Sham and a 97 ± 8% (p< 0.0001) increase compared to the MS180 + Sham group (Fig. 4C). In the contralateral hilus we observed a significant effect of mTBI (F_1,28 = 9.5, p = 0.003) but not of MS180 or interaction (p> 0.05). Multiple comparisons showed that CONT + mTBI caused a 58 ± 13% (p< 0.05) increase in the proportion of activated Iba1⁺ cells compared to CONT + Sham but no difference compared to MS180 + Sham (p> 0.05), while MS180 + mTBI caused a 73 ± 15% (p< 0.01) increase only compared to CONT + Sham (Fig. 4D).

Fig. 4.

Pediatric mild traumatic brain injury (mTBI) increases microglial activation in the hilus. (A) Control (CONT) and maternally separated (180 min/day, MS180) rats were subjected to either a mTBI or sham surgery on postnatal day (P) 21. Rats were sacrificed on P42 and microglial cell activation was quantified in the hilus region of the hippocampus dentate gyrus. (B) Representative images of Iba1 immunostaining. Inserts illustrate higher magnifications of activated Iba1⁺ cells depicted by arrows (scale bar = 200 μm). (C) Percent of activated Iba1⁺ cells in the ipsilateral and (D) contralateral hilus. (Mean ± S.E.M. ANOVA *p<0.05, **p<0.01 and ***p<0.001 vs CONT + Sham. ^###p<0.001 vs. MS180 + Sham. CONT + Sham, n=8; MS180 + Sham, n=8; CONT + mTBI, n=8; MS180 + mTBI, n=8).

MS180 + mTBI decreases SGZ proliferation in the ipsilateral hippocampus.

To test the hypothesis that MS180 + mTBI-induced behavioral deficits could be related to alterations in the proliferation rates of the hippocampus neurogenic niche we performed immunohistochemistry against the cell proliferation marker Ki-67 and assessed the total positive nuclei number in the SGZ by stereology (Fig. 5B). Because training in learning tasks, such as the MWM, may have effects on hippocampal neurogenesis, a naïve group that was not tested in the MWM was included as a reference for the analysis. Three-way ANOVA indicated a significant effect of mTBI (F_1,36 = 6.006, p = 0.02) but not of MS180, training, or interactions and hence, groups were pooled for all subsequent analysis (Supplemental Fig. 1). In the ipsilateral hippocampus (Fig. 5C) we observed a main effect of mTBI (F_1,36 = 4.619, p = 0.03) and MS180 (F_1,36 =5.890, p = 0.02) but no interaction. Multiple comparisons indicated that only the MS180 + mTBI group caused a 43 ± 5% reduction of Ki-67⁺ nuclei number compared to the CONT + Sham group (p< 0.01). Cell proliferation in the contralateral hippocampus was not affected (Fig. 5D).

Fig. 5.

Early life stress decreases cell proliferation in the ipsilateral subgranular cell layer neurogenic niche after mild pediatric traumatic brain injury (mTBI). (A) Control (CONT) and maternally separated (180 min/day, MS180) rats were subjected to either a mTBI or sham surgery on postnatal day (P) 21. Rats were sacrificed on P42 and the total number of Ki67⁺ nuclei number was estimated by stereology. (B) Representative images of Ki67 immunostaining (hematoxylin and eosin counterstain). Inserts illustrate higher magnifications of cells depicted by arrows (scale bar = 200 μm). Stereological estimations of Ki67⁺ total nuclei number in the (C) ipsilateral and (D) contralateral hippocampus dentate gyrus subgranular zone. (Mean ± S.E.M. ANOVA **p< 0.01 vs. CONT + Sham. CONT + Sham, n=10; MS180 + Sham n=10; CONT + mTBI, n=10; MS180 + mTBI, n=10).

MS180 does not further increase the effects of mTBI on the survival of BrdU⁺ nuclei in the hippocampus neurogenic niche.

To test whether MS180 affects the survival of newly generated cells in the hippocampal neurogenic niche, we injected P32 rats with BrdU and quantified nuclei number by stereology ten days after (P42) in naïve rats or after training in a spatial learning task. Three-way ANOVA indicated a significant effect of mTBI (F_1,35 = 25.869, p< 0.001) on BrdU⁺ nuclei number, but not of MS180, training, or interactions and hence, groups were pooled for all subsequent analysis (Supplemental Fig. 1). Two-way ANOVA for the ipsilateral SGZ indicated a main effect of mTBI (F_1,35 = 12.589; p = 0.001) but not of MS180 or interaction on BrdU⁺ nuclei number (Fig. 6C). Multiple comparisons showed that CONT + mTBI caused a 59 ± 8% decrease (p< 0.001) compared to CONT + Sham and 60 ± 8% decrease (p< 0.001) compared to MS180 + Sham, while MS180 + mTBI caused a 52 ± 5% decrease (p< 0.01) compared to CONT + Sham and a 53 ± 5% decrease (p< 0.01) in the number of BrdU+ nuclei that survived in the SGZ 10 days after injection compared to the MS180 + Sham. In the contralateral hippocampus neurogenic niche, a two-way ANOVA showed a main effect of mTBI (F_1,35 = 10.8, p= 0.002) but not of MS180 or interaction; however multiple comparisons failed to show significance between groups (Fig 6D).

Fig. 6.

Pediatric mild traumatic brain injury (mTBI) causes long-term decreases in the survival of newly generated cells in the ipsilateral subgranular cell layer neurogenic niche. (A) Control (CONT) and maternally separated (180 min/day, MS180) rats were subjected to either a mTBI or sham surgery on postnatal day (P) 21. On P32 the rats were injected twice (12 hours apart) with BrdU (500 mg/kg). Rats were sacrificed 10 days later (P42) and total number of BrdU⁺ nuclei number was estimated by stereology. (B) Representative images of BrdU immunostaining (hematoxylin and eosin counterstain). Inserts illustrate higher magnifications of cells depicted by arrows (scale bar = 200 mm). Stereological estimations of BrdU⁺ total nuclei number in the (C) ipsilateral and (D) contralateral hippocampus dentate gyrus granular cell layer and subgranular zone. (Mean ± S.E.M. ANOVA **p<0.01 and ***p<0.001 vs. CONT + Sham; ^##p 0.01 and ^###p<0.001 vs. MS180 + Sham. CONT + Sham, n=10; MS180 + Sham, n=9; CONT + mTBI, n=10; MS180 + mTBI, n=10).

MS180 and mTBI do not affect the differentiation rate of BrdU+ nuclei

To determine if MS180 or mTBI affected BrdU⁺ nuclei differentiation we performed triple immunofluorescence staining for BrdU, the mature neuronal marker NeuN, and either the immature neuron marker doublecortin (DCX) (Fig. 7A) or the astrocyte marker GFAP (Fig. 7B). By estimating the proportion of BrdU⁺ cells that differentiate into neural (BrdU⁺/DCX⁺ and BrdU⁺/NeuN⁺) and glial phenotype (BrdU⁺/GFAP⁺) we observed that mTBI decreased the estimated number of immature (F_1,35 = 84.8, p< 0.0001. Fig. 7C) and mature (F_{1, 35} = 42.2, p< 0.0001) newly generated neurons in the ipsilateral hippocampus (Fig. 7D), as well as the estimated number of BrdU⁺/GFAP⁺ (F_1,35 = 12, p= 0.001. Fig. 7E) newly generated cells. Multiple comparisons indicated that both CONT + mTBI and MS180 + mTBI groups showed a significant decrease in the number of BrdU⁺/DCX⁺ and BrdU⁺/NeuN⁺ cells compared to CONT + Sham and MS180 + Sham groups (p< 0.01), while only MS180 + mTBI showed a significant decrease in BrdU⁺/GFAP⁺ nuclei compared to CONT + Sham (p< 0.05) and MS180 + Sham (p< 0.01). In the contralateral hippocampus neurogenic niche, a two-way ANOVA indicated a significant effect of mTBI on immature (F_1,35 = 47.8, p< 0.0001) and mature (F_1,35 = 13.3 p= 0.0008) newly generated neurons, but not on BrdU⁺/GFAP⁺ nuclei number estimations (p> 0.05). No effect of MS180 or interaction was observed (p> 0.05).

Fig. 7.

Early life stress decreases the survival of newly generated astrocytes after pediatric mild traumatic brain injury (mTBI). Control (CONT) and maternally separated (180 min/day, MS180) rats were subjected to either a mTBI or sham surgery on postnatal day (P) 21. On P32 the rats were injected twice (12 hr apart) with BrdU (500 mg/kg). Rats were sacrificed 10 days later (P42) and the phenotype of BrdU⁺ cells was evaluated by confocal microscopy. (A) Representative images of confocal imaging z-stack orthogonal reconstruction of triple immunofluorescent staining for BrdU (red), doublecortin (DCX, green) and NeuN (blue), or (B) of a triple immunofluorescent staining for BrdU (red), GFAP (green) and NeuN (blue). Inserts show higher magnification of cells depicted by white arrows (scale bar = 50 μm). The percentage of co-localization was calculated for six rats in each group and the number of (C) immature neurons (BrdU⁺/DCX⁺), (D) mature neurons (BrdU⁺/NeuN⁺) and (E) astrocytes (BrdU⁺/GFAP⁺) was extrapolated from stereology counting for BrdU. (Mean ± S.E.M. ANOVA *p<0.05 and **p<0.01 vs. CONT + Sham, ^##p<0.01 vs. MS180 + Sham. CONT + Sham, n=10; MS180 + Sham, n=10; CONT + mTBI, n=10; MS180 + mTBI, n=10).

Discussion

Child abuse represents one of the main causes of brain injury in children. However, most of the studies on non-accidental TBI do not consider the effect of chronic stress exposure associated with abuse. The few studies that exist on stress and TBI use adult models (Kwon et al., 2011; Acosta et al., 2013; Xing et al., 2013; Ojo et al., 2014; Davies et al., 2016; Ogier et al., 2017; Algamal et al., 2019). Because of the vast differences between adults and pediatrics the findings cannot be extrapolated accurately. Hence, we evaluated the effect of ELS exposure in a rodent model of pediatric mTBI. To determine the consequences of MS180 on the injured brain, we opted to utilize a mild model of TBI that would not produce a salient cognitive deficit so that we could detect differences in vulnerability of ELS and TBI and more importantly the effect of their combination on spatial learning and memory. A moderate-to-severe TBI would produce marked cognitive deficits (Kochanek et al., 2017) that would mask the contribution of MS180.

Although many models of early life stress have been developed through the years, MS180 remains the most widely used. Several variations on separation procedures have been analyzed in the literature, ranging from short daily 15 min separations (Levine 1957; Plotsky and Meaney 1993), two separate 3-h long episodes in one day (Riveros-Barrera and Dueñas 2016), or even total maternal deprivation for 24 h (Oomen et al. 2009). During development, rodent offspring feed approximately every 3 h and hence longer separations can interfere with normal feeding patterns of pups and induce undernutrition and metabolic programing. Additionally, it has been shown that MS180 activates the HPA axis response even during the stress hyporesponsive period, while short separations (MS15 or handling) caused a form of stimulation that programed HPA axis in an opposite direction from chronic stress (Meaney et al., 1989; Plotsky and Meaney, 1993). 24-h maternal deprivation, unlike MS180, is an acute traumatic event whose effects are mediated mainly by nutritional factors and lacks the chronicity component of child abuse (Rosenfeld et al., 1993; Suchecki et al., 1995; Suchecki et al., 1993). Lately, another postnatal stress model that involves housing dams in a cage with limited bedding material has been gaining popularity (Rice et al., 2008). However, although offspring of limited nesting stress model showed behavioral alterations as adults (Rice et al., 2008), they did not exhibit depressive-like behaviors in the forced swim test (Molet et al., 2016). Hence, in the present study we aimed to evaluate whether ELS increases the effects of pediatric mTBI utilizing the well-established MS180 model that efficiently reproduces the long-term consequences of child abuse without other confounding variables, such as undernutrition or sub-chronic exposure to stress (Lehmann and Feldon, 2000; Tractenberg et al., 2016; van Bodegom et al., 2017).

As predicted, the behavioral data revealed that the mTBI did not affect cognitive performance in the spatial learning task, which was determined by the distance traveled during the acquisition phase (i.e., learning) and the percentage of distance traveled in the target quadrant during the probe trial (i.e., memory). These findings are in accord with previous reports on mTBI in adolescent rats (Prins et al., 1998; Kochanek et al., 2017). On the other hand, it has been shown that chronic stress during early life significantly affects performance in the MWM in adulthood (Huot et al., 2002; Aisa et al., 2007; Hui et al., 2011; Cao et al., 2014; Sousa et al., 2014; Guo et al., 2018) but studies addressing the effects of ELS on cognitive performance in adolescent rats are scarce. Contrary to our data, which revealed that MS180 did not significantly impact cognitive performance, a study using adolescents did show that daily separation from both the mother and siblings for 4 h from P1 to P21 affected spatial learning (Wang et al., 2015), suggesting that a more severe stressor could elicit cognitive impairments in adolescents. While MS180 did not cause a deficit in the acquisition of spatial learning in adolescent rats in the current study, the combination of ELS and mTBI did lead to an impairment relative to Sham controls. Moreover, MS180 + mTBI also affected the consolidation of memory as evidenced by significantly shorter distances swum in the target quadrant compared to the CONT + Sham group (Fig. 1C). The cognitive deficiencies revealed by the combination of ELS and mTBI indicates that ELS increases the vulnerability of pediatric mTBI.

Cortical lesion and hippocampal volumes are morphological parameters that correlate with injury severity and functional outcome after pediatric TBI (DeMaster et al., 2017). However, we found no statistical differences between CONT + mTBI and MS180 + mTBI groups on lesion volume. Moreover, mTBI affected the ipsilateral hippocampus dentate gyrus volume, but this effect was not further aggravated by prior exposure to ELS. Hence, these morphological parameters do not explain the cognitive deficits observed in the MS180 + mTBI rats. Indeed, the data suggest that the cognitive deficiency could be due to either an increase in inflammation caused by the activation of microglia or to a reduction in hippocampal neurogenesis after mTBI.

Although the mechanisms underlying the inflammatory response in the acute phase of pediatric TBI have been widely studied (for review see Nasr et al., 2019), microglial activation during the chronic phase of TBI has received limited attention. Acute microglial inflammatory responses have been shown to be necessary for recovery after repeated closed head pediatric TBI, but chronic activation has been related to cognitive deficiencies (Hanlon et al., 2019). Moreover, microglia, are important modulators of many stages of brain development and neurogenesis (Paolicelli et al., 2011; Cunningham et al., 2013) and are highly responsive cells that react to stress (Frank et al., 2011), making them suitable candidates to mediate the effects of stress exposure and pediatric mTBI. Hence, we quantified the proportion of activated microglial cells in the area adjacent to the lesion, the hippocampal CA₁ and hilus, and observed that mTBI caused an increase in the proportion of activated Iba1⁺ cells in the ipsilateral hippocampus without affecting the area adjacent to the lesion. Additionally, mTBI increased microglial activation in the contralateral hilus but not in the CA₁. Our results also show that although no significant differences were observed between the CONT + mTBI and MS180 + mTBI groups, stressed rats consistently exhibited a higher increase of microglial activation 21 days post injury. Moreover, only MS180 + mTBI rats showed a significant increase in microglial activation in the area adjacent to the lesion and the contralateral CA₁ hippocampal subfield compared to the CONT + Sham group.

In agreement with our results, it has been shown that after mild CCI in adults, Iba1⁺ cells in the area adjacent to the lesion are morphologically round (activated) at 7 days post injury (Jin et al., 2012), whereas activated microglia were no longer present at 21 days (Jin et al., 2012) or 28 days post-injury (Aungst et al., 2014). Furthermore, no effects on cortical Iba1 immunoreactivity was observed 21 days after repeated closed head injury in pediatric rats (Hanlon et al., 2016). This change in microglia reactivity could be indicative of a switch between the pro-inflammatory M1-like profile to the anti-inflammatory M2-like microglia profile (Ziebell et al., 2015). Taken together, the data suggest that ELS promotes a proinflammatory profile in the hippocampus after mTBI, which is likely related to the reduction of the DG volume and cognitive deficits observed in the MS180 + mTBI group. That microglial cells initially maintained a reactive M1-phenotype during the mTBI chronic phase in MS180 rats suggests that tissue repair processes are delayed, and that alteration in the resolution of inflammation in MS180 + mTBI rats may be related to cognitive deficits. However, further molecular characterizations utilizing other markers for M1 and M2 phenotypes might be necessary to appropriately classify microglia after stress and mTBI.

The remote effect of brain injury on the microglia of distant regions that are anatomically and functionally connected to the lesion site, such as the contralateral hippocampus, has been previously documented for ischemic injuries (for review see Bisicchia et al. 2019) and to a lesser extent in TBI (Niesman et al., 2014; Taib et al., 2017). It has been proposed that axonal damage in the lesion site causes remote degeneration in the areas to which it is connected (Donat et al., 2017; Bisicchia et al., 2019). Because microglia is involved in synaptic pruning, as well as the removal of damaged processes and degenerating axons after an injury, it is plausible that this remote axonal damage could be triggering inflammatory processes in the contralateral hemisphere.

The hippocampus and cortex showed differential microglial activation after mTBI in ELS rats. The cortex and hippocampal region had a resting M2-microglial phenotype related to an anti-inflammatory profile indicated by a lack of significance in the percentage of activated Iba1⁺ cells, while the hippocampal hilus showed a predominantly M1-profile. These results suggest a higher vulnerability to the insult in these regions compared to the cortex. In support of this idea, recent transcriptomic analysis has shown that the expression of immune and inflammatory genes is over-represented in the hippocampal cluster compared to other brain regions, indicating region-specific microglial functions (Grabert et al., 2016).

Recently, it has been proposed that abnormal microglial function caused by ELS during the critical period affects brain development and leads to long-term behavioral consequences (Johnson and Kaffman, 2018). However, studies evaluating the effect of ELS on microglia and inflammation are scarce. It has been shown that MS180 causes an increase in hippocampal microglial activation of neonatal male rats (Roque et al., 2015) and in the prefrontal cortex, CA₁, CA₃ and the dentate gyrus of adolescent females (Gracia-Rubio et al., 2016). However, the effect of MS180 on hippocampal and cortical microglial cells of adolescent males had not been studied until now. In agreement with our results, recent evidence showed that while 4 h of maternal separation from P1 to P21 increases hippocampal Iba1 immunoreactivity in adult rats, no effects were observed in the cortical areas (Banqueri et al., 2019). It could be that Iba1 immunostaining may not be sufficient to detect more subtle changes in microglial function as it has been shown that ELS in the form of limited nesting with the mother in combination with brief daily separations (15 min) has no effect on hippocampal microglia morphology at P28; however, microglia harvested from the hippocampus at the same age showed an increase in phagocytic activity and reduced expression of genes that normally increase across development (Delpech et al., 2016).

Because the effects of TBI on neurogenesis vary depending on the post-injury period in which they are measured, we evaluated neurogenesis at the beginning of the chronic phase of mTBI. It has been reported that after injury, there is an increase in the rate of cell proliferation in the subgranular zone of the dentate gyrus (Dash et al., 2001). However, this increase is only temporary as evidenced by a marked reduction in survival relative to controls during the chronic phase of TBI (Neuberger et al., 2017). In agreement with these observations we observed that mTBI caused a decrease in the survival BrdU⁺ cells in the ipsilateral hippocampus, but mTBI was not sufficient to affect cell proliferation. MS180 + mTBI rats showed a decrease in cell proliferation at the beginning of the chronic phase. Furthermore, by estimating the proportion of BrdU⁺ nuclei that differentiate into neural or glial phenotype we observed that this effect is accompanied by a decrease in gliogenesis in the MS180 + mTBI rats.

While it has been proposed that the developing brain is more vulnerable to TBI- and ELS-induced neuroinflammation than the adult brain (Johnson and Kaffman, 2018; Nasr et al., 2019), the effects of ELS on hippocampal neurogenesis and the neuroendocrine system are divergent. There is evidence that ELS causes an activation of the stress response after each separation episode that is accompanied by a decrease in the hippocampal neurogenesis of neonate rats (Lajud et al., 2012). However, MS180 has no effect on hippocampal neurogenesis of adolescent rats (Suri et al., 2014). Subsequently, upon reaching adulthood, MS180 rats show a dysregulation of the stress response that is accompanied by a decrease in hippocampal neurogenesis (Suri et al., 2014; Ruiz et al., 2018), suggesting that ELS-induced alterations appear later in development. Accordingly, we did not observe a significant effect of MS180 alone in any of the parameters evaluated, therefore further studies should be conducted to evaluate whether TBI could have a differential effect on adult MS180 rats that do show neuroendocrine deficiencies and decreased neurogenesis.

Conclusion

The major conclusion that emerges from this study is that ELS increases the vulnerability to pediatric mTBI. This novel finding suggests that combining ELS with pediatric mTBI in rats could be a relevant model to further evaluate AHT that reproduces a wider spectrum of the factors involved in the phenomena and could increase translatability. Second, the data indicate that the ELS-induced vulnerability to pediatric mTBI could be related to increased neuroinflammation. More studies should evaluate resolution of the inflammatory response in MS180 + mTBI rats.

Supplementary Material

1

Supp. Fig 1. Training in a spatial learning task has no effect on the proliferation and survival rates of the newly generated cells of the subgranular cell layer neurogenic niche. Control (CONT) and maternally separated (180 min/day, MS180) rats were subjected to either a mTBI or sham surgery on postnatal day (P) 21. On P32 the rats were injected twice (12 hr apart) with BrdU (500 mg/kg). (A) A group of rats were sacrificed 10 days later (P42) without further manipulations (Naïve) or (B) trained in a spatial learning task from P35-P40 (Behavior). Stereological estimations of Ki67⁺ and BrdU⁺ nuclei number in the (C and E) ipsilateral and (D and F) contralateral hippocampus. (Mean ± S.E.M. CONT + Sham + Naïve, n=5; MS180 + Sham + Naïve, n=5; CONT + mTBI + Naïve, n=5; MS180 + mTBI + Naïve n=5; CONT + Sham + Behavior, n=5; MS180 + Sham + Behavior, n=5; CONT + mTBI + Behavior, n=5; MS180 + mTBI + Behavior, n=5).

Highlights.

• Early life stress (ELS) increases cognitive impairments after mild pediatric TBI

• ELS increases hippocampal microglial activation after mild pediatric TBI

• ELS + TBI decreases proliferation in the hippocampal neurogenic niche

• ELS + TBI did not further affect the survival or differentiation of newly generated cells