Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Mouse closed head traumatic brain injury replicates the histological tau pathology pattern of human disease: characterization of a novel model and systematic review of the literature

Mouse closed head traumatic brain injury replicates the histological tau pathology pattern of... Traumatic brain injury ( TBI) constitutes one of the strongest environmental risk factors for several progressive neu- rodegenerative disorders of cognitive impairment and dementia that are characterized by the pathological accu- mulation of hyperphosphorylated tau (p-Tau). It has been questioned whether mouse closed-head TBI models can replicate human TBI-associated tauopathy. We conducted longitudinal histopathological characterization of a mouse closed head TBI model, with a focus on pathological features reported in human TBI-associated tauopathy. Male C57BL/6 J mice were subjected to once daily TBI for 5 consecutive days using a weight drop paradigm. Histologi- cal analyses (AT8, TDP-43, pTDP-43, NeuN, GFAP, Iba-1, MBP, SMI-312, Prussian blue, IgG, βAPP, alpha-synuclein) were conducted at 1 week, 4 weeks, and 24 weeks after rTBI and compared to sham operated controls. We conducted a systematic review of the literature for mouse models of closed-head injury focusing on studies referencing tau protein assessment. At 1-week post rTBI, p-Tau accumulation was restricted to the corpus callosum and perivascular spaces adjacent to the superior longitudinal fissure. Progressive p-Tau accumulation was observed in the superficial layers of the cerebral cortex, as well as in mammillary bodies and cortical perivascular, subpial, and periventricular locations at 4 to 24 weeks after rTBI. Associated cortical histopathologies included microvascular injury, neuroaxonal rarefaction, astroglial and microglial activation, and cytoplasmatic localization of TDP-43 and pTDP-43. In our system- atic review, less than 1% of mouse studies (25/3756) reported p-Tau using immunostaining, of which only 3 (0.08%) reported perivascular p-Tau, which is considered a defining feature of chronic traumatic encephalopathy. Commonly reported associated pathologies included neuronal loss (23%), axonal loss (43%), microglial activation and astrogliosis (50%, each), and beta amyloid deposition (29%). Our novel model, supported by systematic review of the literature, indicates progressive tau pathology after closed head murine TBI, highlighting the suitability of mouse models to replicate pertinent human histopathology. Keywords: Animal model, Chronic traumatic encephalopathy, Concussion, Systematic review, Tauopathy, Traumatic brain injury Introduction Traumatic brain injury (TBI) represents a major public *Correspondence: nils.henninger@umassmed.edu health problem affecting more than 10 million people Department of Neurology, Medical School, University of Massachusetts, worldwide each year [36]. TBI is a leading cause of adult 55 Lake Ave, Worcester, USA death and disability worldwide. It has been estimated Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Kahriman et al. acta neuropathol commun (2021) 9:118 Page 2 of 17 that annually 150 to 200/1,000,000 people become disa- histopathological features of human TBI-associated bled as a result of brain trauma, and TBI was declared a tauopathy according to reported consensus criteria for major public health problem by the National Institutes of human CTE [8, 46]. In addition, we conducted a system- Health in 1999 [4, 20, 53]. atic literature review of reported tau pathology in mouse There is a long history of epidemiological evidence closed-head TBI and its relationship to pertinent histo- that TBI represents one of the strongest environmental logical features of human CTE to provide a contempo- risk factors for several progressive neurodegenerative rary overview of the state of the field. disorders of cognitive impairment and dementia that are characterized by the pathological accumulation of hyper- Material and methods phosphorylated tau (p-Tau). In particular, TBI has been Mouse study linked to Alzheimer’s disease (AD) and a unique clinico- Mice were randomly allocated to sham surgery versus pathological entity termed chronic traumatic encepha- rTBI except for 4 mice allowed to survive for 24-weeks lopathy (CTE) [8, 17, 29, 46, 66, 67]. Yet, the exact after rTBI, which were added to gain insight into the mechanism(s) driving pathological tau accumulation and longer-term evolution of p-Tau and associated pathology. spread, cognitive impairment, and dementia after TBI Histological analyses were conducted by an investigator are poorly understood. Defining the mechanisms that masked to the experimental groups. explain the link between TBI and dementia at the cellular level is a public health priority [61]. Animals Mouse models of TBI-associated pathology are of great Male C57BL/6  J mice (Jackson Laboratories) were value because of the ability to conduct detailed, longi- socially housed with same-sex mice (n = 4 per cage) tudinal histopathological characterization in a tempo- on 12-h light/dark cycle with food and water ad  libi- rally accelerated fashion, as well as the unique possibility tum. Spontaneously breathing mice (n = 27) weighing to explore underlying molecular mechanisms through 26.8 ± 2.6  g (age 8–12  weeks) were subjected to closed genetic manipulation without confounding by comorbid head injury (n = 20) or sham surgery (n = 11). Brains conditions. This appears particularly important in light were removed 1  week after rTBI (n = 8), 4  weeks (sham of a current lack of a uniformly accepted definition for n = 8, rTBI n = 8), and 24 weeks (rTBI n = 4) for histolog- TBI-specific tauopathy. Although pathological consen - ical analyses. Additionally, we conducted AT8-staining in sus criteria have been developed for CTE [8, 46], which 3 sham operated mice that survived for 12 months to rule is considered the prototypical TBI-associated tauopathy, out any potential age-related effects on p-Tau pathology. challenges in applying these criteria relate to the fact that many pathologies are also present in the normally Anesthesia, analgesia, and traumatic brain injury induction aging brain and other neurodegenerative conditions [27, Animals were anesthetized with isoflurane (5% for induc - 38]. Hence, distinguishing the direct effects of TBI from tion, 2% for surgery, 1.5% for maintenance) in room a sporadic progressive neurodegenerative process in air. Anesthesia was discontinued immediately prior to humans is difficult. Moreover, animal models are needed TBI and sham injury. Body temperature was monitored to dissect the specific neuropathology of TBI and how continuously with a rectal probe and maintained at pathological tau deposition relates to progressive neuro- 37.0 ± 0.5 °C. To alleviate pain, animals received 0.05 mg/ degeneration. However, concerns have been raised that kg subcutaneous buprenorphine (Med-Vet International, murine closed head injury models may not be suitable to Mettawa, Il, USA) 30 min before anesthesia and every 6 h replicate pertinent histopathological features of human afterwards for 24  h. Additionally, each animal received TBI-associated tauopathy [18]. Depending on the model 5  mg/kg subcutaneous carprofen (Patterson Veterinary, used, variability in post-TBI histopathology may be con- Devens, MA, USA) at the end of the anesthesia. siderable [9, 10]. Further, there are distinct differences TBI was produced using a weight drop device as pre- in the cerebral anatomy as well as expression of tau iso- viously described in detail [11, 31] with the modifica - forms in the adult brains of mice and humans [32], which tion that animals were subjected to TBI or sham injury could conceivably contribute to disparate pathology [41]. once daily for 5 consecutive days. Briefly, a 50  g weight Therefore, it is critical to understand to what degree was freely dropped 15  cm to strike a cylindrical polyac- murine closed-head TBI can replicate human TBI-associ- etal transducer rod (Delrin , tip-diameter 2  mm, 17.4  g) ated tau pathology and related histopathological features. that was placed with its tip directly on the exposed In this study, we characterized the histopathological skull (target 2.5  mm posterior and 2.5  mm lateral from features of a mouse closed head repetitive TBI (rTBI) Bregma). Following TBI, the wound closed with inter- model with specific focus on the presence and evolution rupted sutures. Sham animals were anesthetized, surgi- of p-Tau accumulation and its association with pertinent cally prepared (including skin incision), and placed under K ahriman et al. acta neuropathol commun (2021) 9:118 Page 3 of 17 the impact device with the impactor touching the skull, Hematoxylin counterstaining was used for AT8, but were not subjected to head impact. One mouse with AT180, βAPP, and α-synuclein labeled tissues. For a skull fracture was excluded. immunofluorescence staining tissue sections labeled with the primary antibodies (TDP-43, pTDP-43, NeuN, Immunohistochemistry staining GFAP, Iba-1, MBP, SMI-312, βAPP, alpha-synuclein) For histology, animals received an overdose of pentobar- were incubated in appropriate secondary antibodies bital (150  mg/kg Fatal-Plus, Vortech Pharmaceuticals). conjugated with Alexa Fluor 488 (1:250, Abcam, Cat# Then animals were perfused under isoflurane anesthesia ab150113, RRID: AB_2576208 and Cat# ab150077, through the ascending aorta with 50 mL saline and then RRID: AB_2630356), Alexa Fluor 555 (1:250, Abcam, with ice cold phosphate-buffered 4% paraformaldehyde Cat# ab150106, RRID: AB_2857373), and Alexa Fluor (PFA) for 10  min. Brains were removed from the cra- 647 (1:250, Abcam, Cat# ab150075, RRID: AB_2752244 nium, postfixed overnight in the same fixative, and then and Cat# ab150115, RRID: AB_2687948). Omitting the stored in 0.4% PFA at 4 °C until further processing. Prior primary antibody in a subset of slides served as nega- to paraffin embedding brains were pre-sectioned using a tive controls. brain matrix. Immunohistochemistry was performed against Prussian blue staining Ser−202/Thr205 p-Tau , (AT8, 1:250, Thermo Fisher Sci- To assess for microhemorrhages, sections were stained entific, Cat# MN1020, RRID: AB_223647), p-Tau- for Prussian blue reaction using an Iron Stain Kit (# Thr231 (AT180, 1:250, Thermo Fisher Scientific, Cat# HT20, Sigma-Aldrich), following the manufacturer’s MN1040, RRID: AB_223649), TAR DNA-bind- instructions. ing protein 43 (TDP-43, 1:250, Proteintech, Cat# Ser−409/410 10,782–2-AP, RRID: AB_615042), pTDP-43 Microscopy (1:250, Proteintech, Cat# 22,309–1-AP, RRID: Paraffin sections, 10-µm thick coronal, were obtained AB_11182943), neuronal nuclei (NeuN, 1:200, Pro- at approximately Bregma -3.07  mm (impact center; teintech, Cat# 26,975–1-AP, RRID: AB_2880708), s3), −1.67  mm (adjacent to the impact center; s2), glial fibrillary acidic protein (GFAP, 1:250, Agilent, and + 1.21 mm (remote to the impact center; s1) for his- Cat# Z0334, RRID: AB_10013382), ionized calcium tological assessment (Fig.  1a, b). For quantitative and binding adaptor molecule 1 (1:250, Iba-1, Wako, Cat# qualitative analyses of AT8, NeuN, GFAP, Iba-1, MBP, 019–19,741, RRID: AB_839504), myelin basic pro- and SMI-312 staining one coronal section each from tein (MBP, 1:200, Santa Cruz Biotechnology, Cat# these coordinates (s1-s3) was used. For qualitative anal- M3821, RRID: AB_1841021), SMI-312 (1:200, BioLe- yses of βAPP, alpha-synuclein, IgG, and Prussian blue gend Cat# 837,904, RRID: AB_2566782), beta amyloid staining we reviewed one coronal section each from precursor protein (βAPP, 1:200, Zymed, CT695, Cat# s1-s3. For quantitative and qualitative analyses of TDP-43 51–2700, RRID: AB_2533902), α-synuclein (1:250, and pTDP-43 one coronal section from s2 was used. All Biolegend, Cat# 824,301, RRID: AB_2564879), and histological analyses were performed by an investigator immunoglobulin G (IgG, 1:100, Abcam, Cat# ab6708, masked to the animal groups. RRID: AB_956005). For chromogenic staining, tissue sections labeled with the primary antibodies (AT8, AT180, NeuN, βAPP, α-synuclein, IgG) were incu- Image acquisition and quantification bated with appropriate biotin-conjugated second- To acquire images of all stained sections for subsequent ary antibodies followed by avidin–biotin complex offline analysis we used a Leica DM6 B microscopy sys - (Vector Laboratories) incubation and treatment with tem equipped with a brightfield DMC5400 color CMOS diaminobenzidine as directed by the manufacturer. (See figure on next page.) Fig. 1 Patterns and evolution of p-Tau accumulation after closed head repetitive traumatic brain injury (rTBI). (a) Approximate location of the impact center over the intact mouse skull (blue circle) relative to the brain sections sampled for histological analysis (s1-s3; dashed lines). (b) Approximate location of p-Tau positive cells at 4 weeks after rTBI (composite of 8 mice; each red dot represents 8 p-Tau positive cells, blue ellipses indicate the spatial relation between the impactor and brain surface). Black boxes indicate the location of photomicrographs shown in panel c-e (box 1) and f–h (box 2). (c) Intact cerebral cortex without p-Tau accumulation at 1 week after rTBI. Progressive accumulation of p-Tau in the superficial layers of the cerebral cortex at (d) 4 weeks and (e) 24 weeks after rTBI. P-Tau accumulation in the corpus callosum at (f) 1 week, (g) 4 weeks, and (h) 24 weeks after rTBI (red asterisks). (f ) In contrast to later time points, AT8-immunoreactivity at 1 week was restricted to dot-like staining in a subset of cells (arrowheads). Examples of p-Tau accumulation in (i) perivascular, (j) subpial, (k, l) periventricular, and (l) mammillary body locations as well as (m) at the depth of the superficial longitudinal fissure (white arrowheads) in perivascular (black arrowheads) and subpial (red arrowheads) locations at 4 weeks after rTBI. Scale bars are 30 µm (in c–j), 1 mm (k), and 300 µm (in l–m) Kahriman et al. acta neuropathol commun (2021) 9:118 Page 4 of 17 Fig. 1 (See legend on previous page.) K ahriman et al. acta neuropathol commun (2021) 9:118 Page 5 of 17 camera and an immunofluorescent DFC9000 sCMOS section were taken at 63 × magnification and analyzed camera. as described for GFAP. To determine the spatial distribution of p-Tau accumulation after rTBI, sections were imaged at Assessment of chronic traumatic encephalopathy (CTE) 63 × magnification and the approximate location of all related pathology AT8-positive profiles systematically recorded within With respect to assessing CTE-like pathology in our each of the sampled sections and transferred to a model, we defined histopathological features according standard atlas [52] to provide a visual representation to recently published consensus group criteria developed of the observed tau pathology relative to the impact by an NINDS/NIBIB panel [8, 46] with modifications for location. Given reported p-Tau accumulation in both use in mice (Table 1). Specifically, a defining criterion for neurons and astrocytes in human and mouse TBI, we human CTE includes the presence of perivascular foci quantified the number of p-Tau positive cells across of p-Tau immunoreactive neurofibrillary tangles (NFTs) time points as stratified by neuronal versus astrocytic and abnormal neurites, with or without p-Tau immu- p-Tau. Moreover, because we observed the earliest noreactive astrocytes, in an irregular pattern in the cer- p-Tau accumulation in the corpus callosum with only ebral cortex, with a tendency to involve the sulcal depths. later involvement of the cerebral cortex we addition- Because the lissencephalic brains of mice lack sulci, we ally stratified these analyses according to the location considered the presence of perivascular p-Tau in the cer- in the superficial cortex (approximately cortical lay- ebral cortex a pathognomonic criterion. With respect ers I-III), deep cortex (approximately cortical layers to supportive features, we assessed for several support- IV-VI), and the corpus callosum. ive tau- and nontau-related histopathological features To determine the extent of neuronal loss, chromo- (Table  1). We did not assess for macroscopic pathology gen stained NeuN-positive cells were assessed in each such as dilation of ventricles, septal pathology, atrophy, coronal section. Images of 16 nonoverlapping fields contusions, or other signs of previous trauma [8, 46]. of view (FOV; 8 per hemisphere; 659 × 439  µm, each) Finally, for additional context we also assessed several covering the dorsal cerebral cortex [11] were taken at pathological features that have been repeatedly described 20 × magnification. NeuN positive cells were semiau- in human TBI but are not specific to CTE and may be tomatically quantified using ImageJ [60] as previously shared with TBI and other neurodegenerative conditions described [15]. First, 16-bit color images were con- including the presence of reactive microglia, astrogliosis, verted to 8-bit grayscale followed by automatic thresh- neuronal and axonal loss, presence of beta amyloid and olding. The Analyze Particle tool was then used to alpha-synuclein depositions as well as evidence of prior count all particles with a circularity of 0.1–1.0 and a blood brain barrier (BBB) disruption and microvascular size of 25 to 250 µm . Results were adjusted by man- injury/cerebral microbleeds [45, 49]. ually counting all overlapping cells (ie, particle size greater than 250 µm ). Neurologic evaluation To assess microglia and astroglia in the dorsal cor- Return of the righting reflex was measured as the time (s) tex we used fluorescence staining. For GFAP, images from TBI/sham injury to righting from a supine to prone of 16 nonoverlapping FOVs (8 per hemisphere; position after discontinuation of anesthesia. The neuro - 667 × 667  µm, each) covering the dorsal cerebral cor- logical severity score (NSS) was assessed on a scale from tex were taken at 20 × magnification. First, 16-bit color 0 (no deficit) to 10 (maximal deficit) prior to TBI as well images were split into the individual color channels as serially until sacrifice as previously described in detail (red, green, blue) followed by automatic threshold- [31]. ing of the 8-bit green channel and black-white color inversion. The Analyze Particle tool was then used to Statistical analysis quantify the total thresholded area (µm ). For Iba-1, Unless otherwise stated, continuous variables are TDP-43, and pTDP-43 images of 16 FOVs (8 per hemi- reported as mean ± standard error of the mean. Normal- sphere; 211 × 211 µm, each) centered within the corre- ity of data was examined using the Shapiro–Wilk test. sponding FOV used for the GFAP analyses were taken One-way analysis of variance (ANOVA) on Ranks with at 63 × magnification and analyzed as described for post-hoc Dunn’s test was used to assess for between- GFAP. group differences in histopathology (p-Tau, TDP-43, To assess the impact of rTBI on axonal integrity in pTDP-43, NeuN, Iba-1, GFAP, MBP, SMI-312). Between- the injured cortex, we used fluorescence staining for group comparisons of continuous variables over time the myelin marker MBP and pan-axonal neurofilament (body weight, return of the righting reflex, NSS) were marker SMI312. Images of one FOV (206 × 206 µm) per conducted using longitudinal mixed models. Time was Kahriman et al. acta neuropathol commun (2021) 9:118 Page 6 of 17 Table 1 Chronic traumatic encephalopathy (CTE) related pathology as adapted from consensus criteria [8, 46] and including frequently associated pathology [38, 45, 48, 49] for use in mice in the systematic review and characterization of our mouse model 1 week 4 weeks 24 weeks Neuropathology considered Pathognomonic for CTE Perivascular p-Tau accumulation P1: p-Tau immunoreactive neurons – x x P2: p-Tau immunoreactive astrocytes – x x Neuropathological features Supportive of CTE P-Tau related pathologies S1: Cortical p-Tau (preferentially in superficial layers) – x x S2: Hippocampal p-Tau – – – S3: p-Tau present in subcortical nuclei Mamillary bodies – x x Amygdala – – – Thalamus – – – S4: Astroglial p-Tau in subpial and periventricular regions p-Tau immunoreactive astrocytes in the subpial regions – x x p-Tau immunoreactive astrocytes in the periventricular regions – – – Non-p-Tau related histological pathologies S5: TDP-43 immunoreactive neuronal cytoplasmic inclusions Cortex – x x Hippocampus – – x Amygdala – – – Select nonspecific neuropathological features A ssociated with CTE* A1: β-amyloid precursor protein depositions – – – A2: α-synuclein depositions – – – A3: Hemosiderin laden macrophages x x x A4: Reactive microglia – x x A5: Astrogliosis x x x A6: Neuronal loss – x x A7: Axonal loss – x x A8: Blood brain barrier disruption X x not done *These selected histopathological features have been commonly reported to accompany CTE pathology but are not used to in the consensus criteria to define CTE. P, S, and A refer to pathognomonic, supportive, and associated pathology. “x “ and “–” denote histopathological feature present and absent in our model, respectively treated as a categorical variable. The models included outcomes that have been associated with CTE, includ- group (Sham versus rTBI) and time as fixed covariates, as ing assessment of axonal and neuronal injury, astroglio- well as the group × time interactions. Two-sided signifi - sis, microglial activation, TDP-43 pathology, as well as cance tests were used throughout and unless stated oth- presence of amyloid pathology, α-synuclein, cerebral erwise a two-sided P < 0.05 was considered statistically microbleeds, and evidence of BBB disruption. Overall, significant. All statistical analyses were performed using we identified 1940 articles in PubMed and 2057 articles SigmaPlot 12.5 (Systat Software, Inc., Germany) or IBM in Scopus. After removal of duplicates (n = 241), 3756 ® ® SPSS Statistics Version 26 (IBM -Armonk, NY). papers were included for screening. Details of the review methodology including the search strategy, inclusion and Systematic review exclusion criteria, and retrieval of information from the We conducted a systematic review of the literature by full-text articles is summarized in the Additional file  1: searching PubMed and Scopus using search criteria Supplementary information. that were established to be specific for mouse models of closed head injury (including blast injuries) and that con- tained reference to tau protein assessment in wild type mouse strains. In addition to characterizing model char- acteristics of included studies, we collected histological K ahriman et al. acta neuropathol commun (2021) 9:118 Page 7 of 17 Fig. 2 Temporal and spatial distribution of p-Tau pathology in neurons and astroglia. (a,b) Double staining indicates colocalization (white arrows) of hyperphosphorylated tau (p-Tau; AT8) with (a) neurons (NeuN) and (b) astrocytes (GFAP) in the cerebral cortex at 4 weeks after repetitive traumatic brain injury (rTBI). White scale bars correspond to 50 µm in low power and 17 µm in high power magnified panels. DAPI (blue) channel omitted from high power magnifications. (c) Dotted lines delineate cortical layers I-III (superficial cortex) from cortical layers IV-VI (deep cortex) and corpus callosum used to assess the presence of p-Tau stained cells in the ipsilateral hemisphere. Asterisk denotes the approximate location of the images taken for a‑b. (d) Proportion of AT8 positive neurons and astrocytes at the investigated time points after rTBI. (e) Distribution of AT8 stained cells in the cerebral cortex (superficial + deep combined) relative to the impact center (black bars). There was no difference in the number and distribution of p-Tau positive cells at 4 weeks and 24 weeks after rTBI (P > 0.05). Because sham and 1-week rTBI animals had no cortical AT8-positive cells they were omitted from this analysis. Each bar corresponds to one cortical field of view (FOV ) arranged from left, contralateral (FOV 1) to right, ipsilateral (FOV 16), whereby corresponding FOVs in the three investigated sections s1–s3 were summed. Data are mean ± SD. (f ) Number of p-Tau positive neurons (green shades) and astrocytes (blue shades) in the traumatized hemisphere stratified by location in the superficial cortex, deep cortex, and corpus callosum over time (total number of cells counted in the three investigated sections s1–s3). *P < 0.05. Data are mean ± SEM. n = 8 mice for sham, 1 week, and 4 weeks, n = 4 for 24 weeks (there were no p-Tau positive cells in sham operated mice). All analyses were done using one-way ANOVA on Ranks with post-hoc Dunn’s Results BBB disruption as assessed by IgG staining at 1 week and Mouse rTBI model 4 weeks post rTBI (Additional file 1: Figure S2). Similar to sham operated mice, brains of rTBI mice appeared macroscopically intact and without evidence pT ‑ au spreads from the corpus callosum to superficial for macroscopic cerebral hemorrhages at all time points cortical layers between 1 to 24 weeks after rTBI post injury. Microscopically, the cerebral cortex appeared The primary goal of this study was to determine the normal at 1 week after rTBI. However, at 4 and 24 weeks spatial and temporal evolution of p-Tau pathology over after rTBI we observed focal tissue disruption within the 24 weeks after murine closed head-injury. We used both superficial cerebral cortex approximately corresponding AT8 and AT180 to detect p-Tau. AT180 staining yielded to cortical layer I (Fig. 1c-e). similar results as AT8 staining (Additional file  1: Figure Consistent with human pathology [28], Prussian blue S3) and we exclusively refer to AT8 staining to describe staining showed the presence of hemosiderin laden p-Tau pathology below. Importantly, we found no AT8 macrophages indicative of microvascular injury in the positive cells in sham operated mice that survived for superficial layers of the ipsilateral cortex, at the depth 4 weeks and 12 months after surgery, respectively (Addi- of the superior longitudinal fissure, and grey-white mat - tional file 1: Figure S4). ter junction between cortex and corpus callosum (Addi- We found that 1  week after rTBI, 4 mice (n = 50%) tional file  1: Figure S1). There was evidence of subtle showed faint AT8 staining restricted to the corpus cal- losum without any p-Tau accumulation in the overlying Kahriman et al. acta neuropathol commun (2021) 9:118 Page 8 of 17 Fig. 3 Patterns and evolution of TDP43 (a–h) and pTDP43 ▸ (j–q) pathology in the cerebral cortex after repetitive closed head traumatic brain injury (rTBI) in the mouse. (f) Increased linear pattern of TDP-43 reactivity suggesting neuritic distribution at 1 week after rTBI. Nuclear loss (red arrowheads) and cytoplasmatic localization (white arrowheads) of TDP-43 (g‑h) and pTDP-43 (p‑ q) at 4 and 24 weeks. Aggregation of cytoplasmatic localized TDP-43 (g‑h) and pTDP-43 (q). (r) Apparent reduction in signal intensity of pTDP-43 in the cortex next to the impact center (§). Scale bars correspond to 240 µm for low power and 20 µm for high power magnified images. In each group, bars correspond to one cortical field of view (FOV ) arranged from left, contralateral (contra), to right, ipsilateral (ipsi). Data are shown as mean (+ SD). n = 8 for sham, 1 week, and 4 week post rTBI and n = 4 for 24 week post rTBI. Analyses were done at Bregma − 1.67 mm (corresponding to s2 in Fig. 1) *P < 0.05. All analyses were done using one-way ANOVA on Ranks with post-hoc Dunn’s cerebral cortex (Figs.  1c, f, 2f ). P-Tau pathology did not substantially progress in the corpus callosum between 1 and 24 weeks and AT8 positive cells were present in only a subset of mice (n = 3 [38%] at 4  weeks, n = 2 [50%] at 24  weeks), typically within the section adjacent to the impact center (Figs.  1b, f–h, 2f). Conversely, by 4 and 24  weeks we observed significant accumulation of AT8- stained cells in all investigated mice, particularly in sub- pial locations and the superficial cortex (approximately cortical layers I-III) in the traumatized hemisphere (Figs.  1d–e, 2e–f). Figure  1b depicts the approximate distribution of AT8-positive cells within the traumatized hemisphere at 4  weeks after rTBI. Interestingly, stain- ing was most abundant in the cerebral cortex adjacent to the impact site (Bregma − 1.67 mm) rather than directly below the impact center (Bregma −3.07). pT ‑ au accumulation after murine rTBI occurs in cerebral locations reported for human TBI‑associated tauopathy At 4  weeks post rTBI we observed perivascular p-Tau accumulation (Fig.  1i, m), a pathognomonic histological feature of CTE. Additional, though less common, sites of p-Tau accumulation included the mamillary body (Fig. 1l) and periventricular tissues (Fig.  1k, l), which are consid- ered histological features supporting a CTE diagnosis. Notably, in all examined rTBI mice p-Tau accumulation was present in the subpial cortex adjoining the superficial longitudinal fissure (Fig.  1b, m). We did not find any AT8 stained cells in the hippocampus, thalamus, and amyg- dala (Table 1). the superficial cerebral cortex without overt co-staining pT ‑ au accumulates in cortical neurons and astrocytes in deeper cortical layers or the corpus callosum (Fig.  2d, We found distinct co-localization of AT8 with both NeuN f ). We found that perivascular p-Tau was present in both and GFAP in the traumatized hemisphere consistent with neurons and astrocytes (Additional file  1: Figure S5). The neuronal and astrocytic tau accumulation (Fig.  2a, b). majority of AT8 positive cells co-stained with NeuN, and Interestingly, colocalization of AT8 and GFAP was only observed at 4 and 24  weeks after rTBI and restricted to K ahriman et al. acta neuropathol commun (2021) 9:118 Page 9 of 17 Fig. 4 Progression of cortical neuronal loss and glial activation after repetitive traumatic brain injury (rTBI). (a) Cortical neuronal loss after rTBI as shown by NeuN staining. (b) Astroglial activation at 4 weeks as indicated by (a) the presence of numerous hypertrophied GFAP stained astrocytes (inset) and (c) a significant increase in GFAP staining signal. Decrease in astroglial activation at 24 weeks after rTBI as shown by reduced cell hypertrophy (inset in (a)) and attenuated GFAP staining (c). Microglial activation was observed at 4 weeks as indicated by the presence of numerous bipolar/rod shaped Iba-1 stained microglia (inset in (a)) and significantly increased in Iba-1 stained area (d). Normalization of microglial activation at 24 weeks after rTBI as noted by increased ramification of Iba-1 stained cells (inset in (a)) and similar Iba-1 staining intensity as compared to sham operated mice (d). Each bar corresponds to one cortical field of view (FOV ) arranged from left, contralateral (contra), to right, ipsilateral (ipsi). Data are shown as mean (+ SD). n = 8 (sham, 1 week and 4 week post rTBI, each) and n = 4 (24 week post rTBI) mice. Analyses are based on three coronal sections (corresponding to s1–s3 in Fig. 1). *P < 0.05. All analyses were done using one-way ANOVA on Ranks with post-hoc Dunn’s less than 10% of cells were positive for both AT8 and points after rTBI (not shown). Nonetheless, by 24 weeks GFAP (Fig. 2d, f ). after rTBI, nuclear loss and cytoplasmic localization of TDP-43 (Additional file  1: Figure S6) and pTDP-43 (not shown) was present in the contralateral cerebral cortex rTBI causes persistent nuclear loss and cytoplasmatic and bilateral CA1 of the hippocampus. localization of TDP‑43 and pTDP‑43 Compared to sham and 1 week post rTBI mice, we noted rTBI causes astroglial and microglial activation an overall increase in TDP-43 and pTDP-43 in the ipsi- and progressive neurodegeneration lateral cerebral cortex, particularly adjacent to the impact In addition to assessing pertinent p-Tau and TDP-43 center, at 4  weeks and 24  weeks after rTBI (Fig.  3). On related histological features of human CTE (Table  1), a cellular level, there was nuclear loss and cytoplasmatic we sought to determine pathology that occurs in human localization of TDP-43 and pTDP-43 at the same time TBI-associated neurodegenerative disease, including points (Fig.  3). However, while conspicuous aggregation the presence of reactive migroglia and astroglia, neu- of cytoplasmatic TDP-43 was present at 4 weeks (Fig. 3g), ronal and axonal loss, and accumulation of βAPP and this was only observed at 24 weeks for pTDP-43 (Fig. 3q). α-synuclein. In the contralateral (non-traumatized) hemisphere Sham and 1-week rTBI animals had few GFAP posi- staining intensity for TDP-43 and pTDP-43 in the cor- tive cells in the cerebral cortex and corpus callosum (< 1% pus callosum, subcortical nuclei, and hippocampus staining signal coverage) without difference between was overall similar to sham operated animals at all time hemispheres (Fig. 4c). We observed substantial astroglial Kahriman et al. acta neuropathol commun (2021) 9:118 Page 10 of 17 Fig. 5 Axonal loss in cortex following repetitive traumatic brain injury (rTBI). (a) Cartoon depicting the 3 sections s1–s3 used for analysis. Black boxes indicate the location of the three fields of view (FOVs) used to quantify the immunofluorescent imaging signal for the myelin marker myelin basic protein (MBP) and the pan-axonal neurofilament marker SMI-312. (b) Representative photomicrographs showing progressive loss of the MBP and SMI-312 staining signal from 1 to 24 weeks after rTBI. Quantified staining signal for (c) MBP and (d) SMI-312. Data are mean ± SEM. n = 8 mice for sham, 1 week, and 4 weeks, n = 4 for 24 weeks. *P < 0.05. n.s. indicates not significant. All analyses were done using one-way ANOVA on Ranks with post-hoc Dunn’s. Scale bar is 50 µm activation at 4 weeks after rTBI as indicated by the pres- b). However, by 4 weeks after rTBI, we observed a signifi - ence of hypertrophied GFAP stained astrocytes and over- cant loss of NeuN profiles, which worsened by 24 weeks all significantly increased GFAP staining in the cerebral indicative of progressive neurodegeneration (Fig.  4a, b). cortex (Fig.  4a, c). Although GFAP staining remained Loss of NeuN positive cells was most pronounced within significantly elevated by 24  weeks as compared to sham the ipsilateral cerebral cortex adjacent to the impact site operated animals, overall staining intensity appeared (rather than beneath the impact center) corresponding to attenuated, and cells appeared less hypertrophied when the area of maximal p-Tau accumulation (compare with compared to 4 weeks (Fig. 4a, c). Fig. 1a). There was significant microglial activation in the cer - At 1  week after rTBI, MBP and SMI-312 staining in ebral cortex at 4  weeks after rTBI as indicated by the the traumatized cerebral cortex appeared morphologi- presence of bipolar/rod shaped Iba-1-stained microglia cally similar without significant difference in the quanti - and overall increased Iba-1 staining (Fig.  4a, d). Activa- fied signal intensity as compared to sham operated mice tion subsided by 24  weeks as indicated by a more rami- (Fig.  5b). Concurrent with neuronal injury, we observed fied appearance of Iba-1-stained cells and return of Iba-1 a significant loss of MBP and SMI-312 stained profiles staining intensity to sham levels (Fig. 4a, d). indicating axonal degeneration by 4 weeks and 24 weeks At 1  week after rTBI, neuronal density in the ipsilat- after rTBI (Fig. 5c, d). eral cortex was similar to sham operated mice (Fig.  4a, Lastly, we found no βAPP and α-synuclein depositions in any of the operated mice (not shown). K ahriman et al. acta neuropathol commun (2021) 9:118 Page 11 of 17 rTBI causes long‑term functional deficits Sham operated mice regained their righting reflex within a median of 24  s (interquartile range 14–32  s) after dis- continuation of anesthesia. In contrast, righting reflex was significantly suppressed in rTBI mice and only returned after a median of 135  s (interquartile range 79–229 s) (Fig. 6a). We used the NSS, which is a composite of ratings measuring a combination of overall inquisitiveness, pos- tural stability, and motor function, to examine the tem- poral evolution of functional deficits in rTBI animals versus controls. Whereas sham operated animals had no change in the NSS over the observation period, rTBI mice had significant neurological deficits at 2 h after the last impact (day 5) when compared to baseline and con- trols. Although deficits partially improved over time, persistent residual neurological deficits were present by 24 weeks (Fig. 6b). Although both sham and rTBI mice lost weight after surgery, sham injured animals regained their baseline weight by 1  week whereas rTBI mice had a persistent weight loss up to 4 weeks after rTBI (Fig. 6c). Systematic literature review pT ‑ au pathology is infrequently reported in murine closed head models From searches on both PubMed and Scopus, a total of 3997 articles were initially included. After removal of 241 duplicates 3756 articles were screened for inclu- sion and exclusion criteria (Additional file  1: Figure S7a). From this screening, we excluded 2394 articles because they did not include the predetermined key- words and 411 studies that did not use mice, leaving 951 articles included in the full text search. After full text review, we excluded studies that were not pub- lished in a peer-reviewed journal (n = 5); review articles (n = 235); conference proceeding without primary data (n = 30); studies using transgenic animals without wild- Fig. 6 Persistent neurological deficits in mice subjected to repetitive type mice (n = 106); lacking assessment of cerebral tau traumatic brain injury (rTBI). (a) Significantly prolonged time to the return (n = 507); and penetrating head injury (n = 12). L astly, of the righting reflex (RR) after discontinuation of anesthesia in mice two studies identified from references were added, subjected to repetitive TBI (rTBI) versus sham operated animals. There was resulting in a total of 58 studies (1.5% of screened pub- as a significant group (P < 0.001) effect but no significant time (P = 0.958) lications) that investigated the presence of pathologi- effect or a group x time (P = 0.574) interaction (*P < 0.05, **P < 0.01, ***P < 0.001 versus sham). (b). There were significant group (P = 0.003) cal tau in the brains of mice subjected to closed-head and time (P < 0.001) effects as well as presence of a significant group injury. x time interaction (P < 0.001) for the composite neurological severity score (**P < 0.01, ***P < 0.001 versus baseline). (c) There were significant group and time effects as well as presence of a significant group x time Characteristics of included studies reporting tau pathology interaction (P < 0.001, each) for the change in body weight during the first assessment 4 weeks after surgery (*P < 0.05, **P < 0.01, ***P < 0.001 versus sham). Data Among the 58 included studies, 114 different injury are mean ± SEM. All statistical comparisons were made using mixed effects paradigms were used (42 single TBI and 72 repetitive models (n = 8 for sham; n = 20 for baseline to 1 week after rTBI, n = 12 for 2 to 4 weeks after rTBI, and n = 4 for 24 weeks after rTBI). For clarity in TBI models). Most models utilized male mice (112/114 the figure, post-hoc pairwise comparisons are only shown for significant [98%]), the C57BL/6 strain (111/114 [97%]), and anes- differences between groups (a, c) and versus baseline (b ) thesia (111/114 [97%]). The median mouse age at the Kahriman et al. acta neuropathol commun (2021) 9:118 Page 12 of 17 Table 2 Mouse traumatic brain injury ( TBI) models reporting p-Tau histopathology in the brain Reference Model Impacts ITI EDPI p‑ Tau Ab CTE‑like pathology* Repetitive TBI Pathognomonic Supportive Associated This study WD 5 24 7 AT8 P1, P2 S1,S3,S4, S5 A3 to A8 Petraglia et al. [54]** PD 6 24 30 AT8 P1, P2 S1,S2,S3,S4 A4, A5 Shin et al. [64] WD 20 24 42 P( T205) P1 S1, S2, S4 A5 Zhang et al. [78] PD 3 24 8 AT8 – S1, S2, S5 A1, A4, A5, A6, A7 Luo et al. [44] PD 3 24 180 AT8 – S1, S2, S3 A5 Briggs et al. [13] WD 30 24 93 AT8 – S1, S5 A1, A4, A5, A7 Albayram et al. [1]** RIA 7 24 14 Cisptau – S1, S5 A1, A4,A5, A7 Selvaraj et al. [62] PD 3 24 8 P(S404) – S1, S2 A1, A4, A5, A6 Kondo et al. [40]** RIA 7 24 10 Cisptau – S1, S2 A6, A7 Albayram et al. [2] RIA 5 24 240 AT8 – S1, S2 A4 Yang et al. [76] PD 4 72 10 CP13 – S1, S2 A5 Sacramento et al. [58] PD 10 2 21 AT8 – S1,S2 – Tagge et al. [69] RIA 2 0.25 1 Cisptau – S1 A4, A5, A7, A8 Rehman et al. [56] WD 3 24 7 P(S413) – S1 A1, A6, A8 Seo et al. [63] WD 3 72 7 PHF – S2 – Niziolek et al. [51] WD 4 48 30 P(S262) – S2 – Single TBI Liu et al. [42] RIA 1 n/a 30 T231 P1 S1, S2 – Petraglia et al. [54]** PD 1 n/a 30 AT8 – S1, S2, S3 A5 Goldstein et al. [26] PD 1 n/a 14 CP13 – S1,S2 A4, A5, A6, A7 Kondo et al. [40]** RIA 1 n/a 1 Cisptau – S1, S2 A6, A7 Huber et al. [35] Blast 1 n/a 30 CP13 – S1,S2 A6 Sabbagh et al. [57] Blast 1 n/a 56 T231 – S1, S2 – Iliff et al. [37] PD 1 n/a 28 AT8 – S1 A4, A5, A7 Albayram et al. [1]** RIA 1 n/a 14 Cisptau – S1 A1, A7 Logsdon et al. [43] Blast 1 n/a 3 T22 – S1 A5,A8 Niziolek et al. [50] WD 1 n/a 30 P(S262) – S2 A4 *See Table 1 for definitions. **Used both single and repetitive TBI. EDPI indicates the earliest day post injury at which p-Tau was found after TBI. If more than one p-Tau Ab was used only the antibody used to depict the main results is indicated. ITI, inter-injury-interval (hours), n/a, not applicable; PD, piston driven; RIA, rotational impact acceleration; WD, weight drop. Four studies reporting p-Tau exclusively in the optic tract and cerebellum are excluded from this table. P, S, and A refer to pathognomonic, supportive, and associated pathology (see Table 1 for details). Dashes indicate histopathological feature not reported or not found. Studies not showing p-Tau in the cerebral cortex or hippocampus are omitted from this table time of injury was 12  weeks (range 5 to 48  weeks). summarizes key characteristics of the 25 TBI models The most commonly used models were weight drop (15 repetitive TBI and 10 single TBI) that were associ- (n = 51), piston driven (n = 42), blast injury (n = 12), ated with cortical or hippocampal p-Tau, as assessed by and rotational impact acceleration (n = 9) models . immunohistochemistry. Additional file  1: Figure S7b shows the primary meth- odologies used for p-Tau detection and the frequency Presence of CTE‑like pathology in mouse models reporting of p-Tau detection. Of the included studies, 40 (68%) tau pathology reported p-Tau pathology and 25 (44%) described With respect to the presence of CTE-like pathol- p-Tau pathology by immunostaining (Additional file  1: ogy, only 3 studies (5.2% of included and 0.1% of all Figure S7c-d). The most commonly used antibody to screened papers) reported perivascular p-Tau, a defin - detect p-Tau was AT8 (8/25 [32%] of studies). Of the 25 ing feature of human CTE. One of these studies spe- studies that observed tau pathology, 22 found p-Tau in cifically mentioned the presence of both neuronal and the cerebral cortex, hippocampus, or subcortical struc- astroglial p-Tau (Additional file  1: Figure S7c). One tures and 3 studies reported the presence of p-Tau in additional study reported concurrent p-Tau pathology the optic tract (n = 2) and cerebellum (n = 1). Table  2 in both astroglia and neurons (located in the cortex and K ahriman et al. acta neuropathol commun (2021) 9:118 Page 13 of 17 hippocampus) but did not report on potential perivas- as well as in the adjacent corpus callosum. This is con - cular location. sistent with computational models of the biomechani- In terms of non-p-Tau related supportive histopatho- cal forces predicting that cerebral areas with a change in logical CTE-like features, 3 studies reported the presence morphology (such as at the site of sulci and the superior of cytoplasmatic localization of TDP-43 pathology (none longitudinal fissure) represent locations of high stress specifically commented on pTDP-43) (Additional file  1: and strain and thus greatest susceptibility to axonal and Figure S7c). vascular injury [12, 16, 23]. Forty-six (79%) of the included studies reported non- Given the perivascular location of many neurodegen- specific CTE-associated pathologies (Additional file  1: erative features it has been suggested that TBI triggers Figure S7d). Most commonly, these included neuronal the neurodegenerative cascade by damaging the neuro- loss (23%), axonal loss (43%), micro- and astrogliosis vascular unit. Indeed, consistent with human data [28, (50%, each), and beta amyloid deposition (29%) (Addi- 39], we observed microvascular injury in our model, par- tional file  1: Figure S7c). Interestingly, no study reported ticularly in locations with p-Tau accumulation. Moreover, cerebral microbleeds. using IgG staining, we found evidence for BBB-disrup- tion adding to the notion that BBB hyperpermeability is Discussion an important aspect of murine TBI-associated tauopa- Pathological hyperphosphorylation and aggregation thy. Nevertheless, IgG staining was faint and restricted of Tau protein is observed in a wide range of neurode- to the optic tract. Injury to the visual pathway including generative disorders and is the key defining feature of BBB-disruption and p-Tau accumulation in the optic a heterogeneous class of diseases called tauopathies. tract has been shown after murine closed head injury [3, TBI has been identified as a strong risk factor for many 13, 14, 19]. Yet, we did not observe previously reported of these tauopathies, highlighting the potential lifelong perivascular IgG staining in the cerebral cortex [69]. This consequences of TBI exposure. Recently, CTE has been is inconsistent with the observed microvascular injury described as the prototypical TBI-associated tauopathy. as assessed by Prussian blue staining in our study. In Its diagnosis presently rests with the unique distribution this respect it is noteworthy that we used the cyclooxy- of tau pathologies on a macroscopic and cellular level, yet genase-2 (COX-2) inhibitor carprofen as a post-opera- little is known to what extent these pathologies are rep- tive analgesic. Carprofen has been shown to attenuate licated by mouse closed-head TBI. To close this knowl- microglial activation, inflammation, and brain edema edge gap we described the pertinent histopathological formation at therapeutic doses in mice, which may have features of human TBI-associated tauopathy in our rTBI attenuated BBB-disruption in our model [72]. Thus, while model as well as by conducting a systematic review of the IgG-staining provided proof-of-principle that BBB-integ- literature. rity was impaired in our model, further studies using Our concussive mouse rTBI model showed p-Tau different BBB-integrity markers as well as avoidance of accumulation within the corpus callosum of the trau- anti-inflammatory drugs is required to better character - matized hemisphere as early as 1  week after injury with ize the extent of BBB permeability and its relationship to subsequent involvement of the cerebral cortex, mimick- p-Tau and other observed pathologies. ing the pathological tau progression described in human In addition to pathological p-Tau accumulation, sev- disease. Importantly, we found p-Tau in multiple cer- eral other proteinopathies have been described following ebral locations that are frequently involved in human human TBI. In particular, widespread TDP-43 inclusions disease and are included in the CTE consensus crite- in the neocortex of patients with CTE have been reported ria. These included, superficial cortical layers, the hip - and the presence of TDP-43 pathology has been used pocampus, periventricular tissues, mamillary bodies, and to support the diagnosis of CTE [47, 49]. Nevertheless, perivascular locations [8, 46]. Nevertheless, we defined TDP-43-related pathology is not specific to CTE. It has vessels based on morphology only and future studies been described in a range of conditions, and colocaliza- should include staining for vascular markers to deter- tion of tau and TDP-43 is often limited. For this reason, it mine the specific association of p-Tau with the vascular has yet to be determined exactly how TDP-43 aggregates compartment. coincide and interact with pathological p-Tau accumula- In this regard, it should be noted that human CTE crite- tion. Overall, few mouse studies have sought to evaluate ria include the location of perivascular p-Tau at the depth TDP-43 pathology after closed-head TBI [1, 13, 73, 78]. of cerebral sulci [8, 46]. Because mice are lisencephalic, Here, we found widespread TDP-43 expression following this criterion cannot be fulfilled in a murine model. How - rTBI with a persistent presence of cytoplasmatic mislo- ever, we consistently observed p-Tau within the subpial calization by 6 months post rTBI. This is consistent with cerebral cortex lining the superficial longitudinal fissure, prior studies reporting persistent TDP-43 pathology in Kahriman et al. acta neuropathol commun (2021) 9:118 Page 14 of 17 murine TBI associated with tauopathy [1, 13, 78]. We screened studies (58/3,756) sought to assess pathologi- found long-term accumulation of pTDP-43 expression cal tau accumulation. Nevertheless, of these studies, with associated nuclear loss and cytoplasmatic aggre- approximately 40% found evidence of p-Tau pathology gation up to 24  weeks after rTBI. This is an important by immunohistochemistry. Despite the overall scarcity extension of previous studies that were limited to shorter of investigations, these and our observations provide (1  week) observation periods [55, 70, 73] and in light of cumulative evidence that murine closed head TBI rep- reported transient alterations [55, 73]. Our observation licates the critical histopathological aspects of human that pathological TDP-43, but not tau, accumulation was TBI-associated tau pathology with specific features of present in the hippocampus adds to the notion that while CTE. Reflecting human disease, pathology was seen these proteinopathies share a common pathophysiology, across a wide variety of TBI-paradigms, highlight- there may exist cell-specific susceptibilities [1, 68]. ing that tau pathology after murine closed head TBI is In contrast to pathological tau and TDP-43 accumu- reproducible and robust. This indicates that model dif - lation after TBI, persistent α-synuclein and β-amyloid ferences could be leveraged to study the association of proteinopathy represent less consistent histopathologi- pathophysiological mechanisms with varying mechan- cal features after TBI [7, 8, 22, 38, 67]. Although several ics of injury [9] as well as the ability to control for pos- prior animal studies reported increased β-amyloid and sible confounders. For example, similar to other studies α-synuclein after TBI, we did not find this present in our reporting p-Tau accumulation after murine closed-head rTBI model [1, 13, 56, 62, 75, 78]. rTBI [56, 62, 75, 78] we exposed the skull for precise Mixed neuronal and astroglial tau pathology is consid- impact delivery to the same coordinates across animals. ered a hallmark of CTE; as such, it is important to evalu- Yet, this is inconsistent with the clinical situation, and ate the precise cell types involved with TBI-associated our systematic review showed that this approach is not tauopathy. For example, astrocyte activation is common critical for inducing pertinent CTE-like neuropatholog- after TBI, accompanies virtually all neurodegenerative ical features. tauopathies, and astrocytes may serve as a source for tau Finally, though accumulation of p-Tau is a critical and thus could conceivably contribute to pathological early event in the cascade leading to tauopathy, forma- tau accumulation [34, 41, 65, 69]. Yet, whether astroglial tion of small, soluble oligomeric tau species as well as tau expression serves as a driver for injury-associated the aggregation into larger insoluble filaments known as tauopathy remains uncertain [41]. Many groups hypoth- neurofibrillary tangles (NFTs), represent the hallmark esize that astrocytes may promote neurodegeneration of tauopathies including CTE [5, 30, 33]. There is strong because astroglial tau pathology has been observed in the evidence that pathogenic tau, sometimes referred to as absence of neuronal tau inclusions, possibly through pro- a tau prion or prion-like tau, self-templates to progres- inflammatory mechanisms [41]. We observed significant sively spread disease in tauopathy patients. In a sub-set neuronal loss by 4  weeks after rTBI, which coincided of rTBI patients, the repeated injury gives rise to the with significant astroglial and microglial activation. How - formation of tau prions, which include a toxic spe- ever, we noted that neuronal p-Tau expression occurred cies responsible for neuronal death. However, further prior to significant microglial and astroglial activation. studies are needed to determine if murine closed-head Moreover, neuronal p-Tau expression was present earlier models are able to replicate the formation of tau prions and progressed more widespread than astroglial p-Tau in mice [24, 25, 33]. expression. Lastly, p-Tau-stained astrocytes were located In addition to ongoing efforts to better understand the in the superficial cortex, which had the greatest burden effects of TBI in wild-type mice, several groups have used of neuronal p-Tau, but there were no p-Tau expressing transgenic mouse models expressing human tau to inves- astrocytes in deeper cortical layers and the corpus callo- tigate the link between traumatic injury, tau misfolding, sum. Together, these observations are consistent with the and p-Tau pathology. For example, the rTg4510 mouse hypothesis that astroglial p-Tau expression after closed- model, which expresses a doxycycline-repressible isoform head TBI is a secondary event rather than primary driver of tau containing the P301L mutation [59] developed of tau pathology, possibly related to internalization of elevated p-Tau levels in the cortex 7  days after either a p-Tau into astrocytes from neighboring neurons and syn- single or double closed-head injury [6]. Investigating the apses [41]. longer-term effects of TBI on tau aggregation in Tg mice, To put our results in context with ongoing work in others have used the PS19 mouse model [77], which the field, we conducted a systematic review of the lit - expresses human tau with the P301S mutation, and per- erature. Our analysis showed a striking paucity of formed a semi-quantitative analysis of p-Tau pathology studies that sought to determine tau pathology in up to 7 months post-injury [71]. While these studies have experimental closed-head mouse TBI. Fewer than 2% of helped address several questions about the link between K ahriman et al. acta neuropathol commun (2021) 9:118 Page 15 of 17 References TBI and tau spreading, there are also important caveats 1. Albayram O, Kondo A, Mannix R, Smith C, Tsai CY, Li C, Herbert MK, Qiu to consider with regard to these mouse models. A care- J, Monuteaux M, Driver J et al (2017) Cis P-tau is induced in clinical and ful analysis of the transgene insertion sites in the rTg4510 preclinical brain injury and contributes to post-injury sequelae. Nat Com- mun 8:1000. https:// doi. org/ 10. 1038/ s41467- 017- 01068-4 model revealed that disruption of essential genes is at 2. Albayram O, MacIver B, Mathai J, Verstegen A, Baxley S, Qiu C, Bell C, least partially responsible for the p-Tau pathology that Caldarone BJ, Zhou XZ, Lu KP et al (2019) Traumatic Brain Injury-related results from overexpressing the human protein [21]. And voiding dysfunction in mice is caused by damage to rostral pathways, altering inputs to the reflex pathways. Sci Rep 9:8646. https:// doi. org/ 10. extensive variability in the PS19 model, with regard to 1038/ s41598- 019- 45234-8 both the rate of tau spreading and p-Tau formation [74], 3. Angoa-Perez M, Zagorac B, Anneken JH, Briggs DI, Winters AD, Greenberg limits the interpretations that can be drawn from studies JM, Ahmad M, Theis KR, Kuhn DM (2020) Repetitive, mild traumatic brain injury results in a progressive white matter pathology, cognitive deterio- using the mouse line. ration, and a transient gut microbiota dysbiosis. Sci Rep 10:8949. https:// In conclusion, our study highlights the translational doi. org/ 10. 1038/ s41598- 020- 65972-4 value of murine closed skull TBI to replicate the perti- 4. Anonymous (1999) Consensus conference. Rehabilitation of persons with traumatic brain injury. NIH consensus development panel on rehabilita- nent aspects of human TBI-associated tauopathy with a tion of persons with traumatic brain injury. JAMA 282: 974-983 wide range of related histopathological features. 5. Arena JD, Smith DH, Lee EB, Gibbons GS, Irwin DJ, Robinson JL, Lee VM, Trojanowski JQ, Stewart W, Johnson VE (2020) Tau immunophenotypes in chronic traumatic encephalopathy recapitulate those of ageing and Supplementary Information Alzheimer’s disease. Brain 143:1572–1587. https:// doi. org/ 10. 1093/ brain/ The online version contains supplementary material available at https:// doi. awaa0 71 org/ 10. 1186/ s40478- 021- 01220-8. 6. Bachstetter AD, Morganti JM, Bodnar CN, Webster SJ, Higgins EK, Roberts KN, Snider H, Meier SE, Nation GK, Goulding DS et al (2020) The effects of mild closed head injuries on tauopathy and cognitive deficits in rodents: Additional file 1. Supplementary information. Primary results in wild type and rTg4510 mice, and a systematic review. Exp Neurol 326:113180. https:// doi. org/ 10. 1016/j. expne urol. 2020. 113180 7. Baugh CM, Stamm JM, Riley DO, Gavett BE, Shenton ME, Lin A, Authors’ contributions Nowinski CJ, Cantu RC, McKee AC, Stern RA (2012) Chronic traumatic All authors contributed to the study conception and design. A.K. conducted encephalopathy: neurodegeneration following repetitive concussive histological preparations, histological analysis, data analyses. J.B. conducted and subconcussive brain trauma. Brain Imaging Behav 6:244–254. animal surgery, behavioral testing, histological preparations, and genotyping. https:// doi. org/ 10. 1007/ s11682- 012- 9164-5 N.H. conceived and designed the study, conducted data analyses, animal 8. Bieniek KF, Cairns NJ, Crary JF, Dickson DW, Folkerth RD, Keene CD, surgery, behavioral testing, histological preparations, and wrote the paper. All Litvan I, Perl DP, Stein TD, Vonsattel JP et al (2021) The second NINDS/ authors discussed the results, commented on the manuscript for important NIBIB consensus meeting to define neuropathological criteria for the intellectual content, read, and approved the final manuscript. diagnosis of chronic traumatic encephalopathy. J Neuropathol Exp Neurol 80:210–219. https:// doi. org/ 10. 1093/ jnen/ nlab0 01 Funding 9. Bodnar CN, Roberts KN, Higgins EK, Bachstetter AD (2019) A systematic This study was supported by institutional Grants, National Institutes of Health review of closed head injury models of mild traumatic brain injury in Grant K08NS091499 to N.H. mice and rats. J Neurotrauma 36:1683–1706. https:// doi. org/ 10. 1089/ neu. 2018. 6127 Data availability 10. Bolton-Hall AN, Hubbard WB, Saatman KE (2019) Experimental designs Data supporting the findings of this study are available from the correspond- for repeated mild traumatic brain injury: challenges and considera- ing author upon reasonable request. tions. J Neurotrauma 36:1203–1221. https:// doi. org/ 10. 1089/ neu. 2018. Declarations 11. Bouley J, Chung DY, Ayata C, Brown RH Jr, Henninger N (2019) Corti- cal spreading depression denotes concussion injury. J Neurotrauma Ethics approval 36:1008–1017. https:// doi. org/ 10. 1089/ neu. 2018. 5844 All animal procedures were approved by the University of Massachusetts 12. Bradshaw DR, Ivarsson J, Morfey CL, Viano DC (2001) Simulation of Medical School Institutional Animal Care and Use Committee. This manuscript acute subdural hematoma and diffuse axonal injury in coronal head was prepared in adherence to the ARRIVE and PRISMA guidelines (Additional impact. J Biomech 34:85–94. https:// doi. org/ 10. 1016/ s0021- 9290(00) file 1: Supplementary information). 00135-4 13. Briggs DI, Angoa-Perez M, Kuhn DM (2016) Prolonged repetitive head Competing interests trauma induces a singular chronic traumatic encephalopathy-like pathol- The authors declare no competing financial interests. ogy in white matter despite transient behavioral abnormalities. Am J Pathol 186:2869–2886. https:// doi. org/ 10. 1016/j. ajpath. 2016. 07. 013 Author details 14. Cansler SM, Guilhaume-Correa F, Day D, Bedolla A, Evanson NK Indirect Department of Neurology, Medical School, University of Massachusetts, 55 traumatic optic neuropathy after head trauma in adolescent male mice Lake Ave, Worcester, USA. Department of Pathology, Medical School, Univer- is associated with behavioral visual impairment, neurodegeneration, and sity of Massachusetts, 55 Lake Ave, Worcester, USA. Department of Biology, elevated endoplasmic reticulum stress markers at acute and subacute University of Massachusetts Amherst, Amherst, MA 01003, USA. Depar t- times. BioRxiv [Preprint] June 12, 2020 (Cited 06/01/2021). https:// doi. org/ ment of Psychiatry, Medical School, University of Massachusetts, 55 Lake Ave, 10. 1101/ 2020. 06. 11. 144766 Worcester, USA. 15. Cheng WH, Martens KM, Bashir A, Cheung H, Stukas S, Gibbs E, Namjoshi DR, Button EB, Wilkinson A, Barron CJ et al (2019) CHIMERA repetitive mild Received: 11 May 2021 Accepted: 18 June 2021 traumatic brain injury induces chronic behavioural and neuropathologi- cal phenotypes in wild-type and APP/PS1 mice. Alzheimers Res Ther 11:6. https:// doi. org/ 10. 1186/ s13195- 018- 0461-0 16. Cloots RJ, Gervaise HM, van Dommelen JA, Geers MG (2008) Bio- mechanics of traumatic brain injury: influences of the morphologic Kahriman et al. acta neuropathol commun (2021) 9:118 Page 16 of 17 heterogeneities of the cerebral cortex. Ann Biomed Eng 36:1203–1215. 35. Huber BR, Meabon JS, Martin TJ, Mourad PD, Bennett R, Kraemer BC, Cer- https:// doi. org/ 10. 1007/ s10439- 008- 9510-3 nak I, Petrie EC, Emery MJ, Swenson ER et al (2013) Blast exposure causes 17. Crane PK, Gibbons LE, Dams-O’Connor K, Trittschuh E, Leverenz JB, Keene early and persistent aberrant phospho- and cleaved-tau expression in a CD, Sonnen J, Montine TJ, Bennett DA, Leurgans S et al (2016) Association murine model of mild blast-induced traumatic brain injury. J Alzheimers of traumatic brain injury with late-life neurodegenerative conditions and Dis 37:309–323. https:// doi. org/ 10. 3233/ JAD- 130182 neuropathologic findings. JAMA Neurol 73:1062–1069. https:// doi. org/ 10. 36. Hyder AA, Wunderlich CA, Puvanachandra P, Gururaj G, Kobusingye 1001/ jaman eurol. 2016. 1948 OC (2007) The impact of traumatic brain injuries: a global perspective. 18. DeWitt DS, Hawkins BE, Dixon CE, Kochanek PM, Armstead W, Bass CR, NeuroRehabilitation 22:341–353 Bramlett HM, Buki A, Dietrich WD, Ferguson AR et al (2018) Pre-clinical 37. Iliff JJ, Chen MJ, Plog BA, Zeppenfeld DM, Soltero M, Yang L, Singh I, testing of therapies for traumatic brain injury. J Neurotrauma 35:2737– Deane R, Nedergaard M (2014) Impairment of glymphatic pathway 2754. https:// doi. org/ 10. 1089/ neu. 2018. 5778 function promotes tau pathology after traumatic brain injury. J Neurosci 19. Evanson NK, Guilhaume-Correa F, Herman JP, Goodman MD (2018) Optic 34:16180–16193. https:// doi. org/ 10. 1523/ JNEUR OSCI. 3020- 14. 2014 tract injury after closed head traumatic brain injury in mice: A model of 38. Iverson GL, Gardner AJ, McCrory P, Zafonte R, Castellani RJ (2015) A critical indirect traumatic optic neuropathy. PLoS ONE 13:e0197346. https:// doi. review of chronic traumatic encephalopathy. Neurosci Biobehav Rev org/ 10. 1371/ journ al. pone. 01973 46 56:276–293. https:// doi. org/ 10. 1016/j. neubi orev. 2015. 05. 008 20. Fleminger S, Ponsford J (2005) Long term outcome after traumatic brain 39. Kenney K, Amyot F, Haber M, Pronger A, Bogoslovsky T, Moore C, Diaz- injury. BMJ 331:1419–1420. https:// doi. org/ 10. 1136/ bmj. 331. 7530. 1419 Arrastia R (2016) Cerebral vascular injury in traumatic brain injury. Exp 21. Gamache J, Benzow K, Forster C, Kemper L, Hlynialuk C, Furrow E, Ashe Neurol 275(Pt 3):353–366. https:// doi. org/ 10. 1016/j. expne urol. 2015. 05. KH, Koob MD (2019) Factors other than hTau overexpression that contrib- 019 ute to tauopathy-like phenotype in rTg4510 mice. Nat Commun 10:2479. 40. Kondo A, Shahpasand K, Mannix R, Qiu J, Moncaster J, Chen CH, Yao https:// doi. org/ 10. 1038/ s41467- 019- 10428-1 Y, Lin YM, Driver JA, Sun Y et al (2015) Antibody against early driver of 22. Gavett BE, Stern RA, McKee AC (2011) Chronic traumatic encephalopathy: neurodegeneration cis P-tau blocks brain injury and tauopathy. Nature a potential late effect of sport-related concussive and subconcussive 523:431–436. https:// doi. org/ 10. 1038/ natur e14658 head trauma. Clin Sports Med 30:179–188. https:// doi. org/ 10. 1016/j. csm. 41. Kovacs GG (2020) Astroglia and Tau: new perspectives. Front Aging 2010. 09. 007 Neurosci 12:96. https:// doi. org/ 10. 3389/ fnagi. 2020. 00096 23. Ghajari M, Hellyer PJ, Sharp DJ (2017) Computational modelling of trau- 42. Liu B, Cao Y, Shi F, Wang L, Li N, Cheng X, Du J, Tian Q, Zhou X (2020) The matic brain injury predicts the location of chronic traumatic encepha- overexpression of RBM3 alleviates TBI-induced behaviour impairment lopathy pathology. Brain 140:333–343. https:// doi. org/ 10. 1093/ brain/ and AD-like tauopathy in mice. J Cell Mol Med 24:9176–9188. https:// doi. aww317org/ 10. 1111/ jcmm. 15555 24. Goedert M, Eisenberg DS, Crowther RA (2017) Propagation of tau aggre- 43. Logsdon AF, Lucke-Wold BP, Turner RC, Li X, Adkins CE, Mohammad AS, gates and neurodegeneration. Annu Rev Neurosci 40:189–210. https:// Huber JD, Rosen CL, Lockman PR (2017) A mouse model of focal vascular doi. org/ 10. 1146/ annur ev- neuro- 072116- 031153 injury induces astrocyte reactivity, tau oligomers, and aberrant behavior. 25. Goedert M, Spillantini MG (2017) Propagation of tau aggregates. Mol. Arch Neurosci. https:// doi. org/ 10. 5812/ archn euros ci. 44254 Brain 10:18. https:// doi. org/ 10. 1186/ s13041- 017- 0298-7 44. Luo J, Nguyen A, Villeda S, Zhang H, Ding Z, Lindsey D, Bieri G, Castellano 26. Goldstein LE, Fisher AM, Tagge CA, Zhang XL, Velisek L, Sullivan JA, Upreti JM, Beaupre GS, Wyss-Coray T (2014) Long-term cognitive impairments C, Kracht JM, Ericsson M, Wojnarowicz MW et al (2012) Chronic traumatic and pathological alterations in a mouse model of repetitive mild trau- encephalopathy in blast-exposed military veterans and a blast neuro- matic brain injury. Front Neurol 5:12. https:// doi. org/ 10. 3389/ fneur. 2014. trauma mouse model. Sci Transl Med 4:134–160. https:// doi. org/ 10. 1126/ 00012 scitr anslm ed. 30037 16 45. McKee AC, Abdolmohammadi B, Stein TD (2018) The neuropathology 27. Graham NS, Sharp DJ (2019) Understanding neurodegeneration after of chronic traumatic encephalopathy. Handb Clin Neurol 158:297–307. traumatic brain injury: from mechanisms to clinical trials in dementia. https:// doi. org/ 10. 1016/ B978-0- 444- 63954-7. 00028-8 J Neurol Neurosurg Psychiatry 90:1221–1233. https:// doi. org/ 10. 1136/ 46. McKee AC, Cairns NJ, Dickson DW, Folkerth RD, Keene CD, Litvan I, Perl DP, jnnp- 2017- 317557 Stein TD, Vonsattel JP, Stewart W et al (2016) The first NINDS/NIBIB con- 28. Griffin AD, Turtzo LC, Parikh GY, Tolpygo A, Lodato Z, Moses AD, Nair G, sensus meeting to define neuropathological criteria for the diagnosis of Perl DP, Edwards NA, Dardzinski BJ et al (2019) Traumatic microbleeds chronic traumatic encephalopathy. Acta Neuropathol 131:75–86. https:// suggest vascular injury and predict disability in traumatic brain injury. doi. org/ 10. 1007/ s00401- 015- 1515-z Brain 142:3550–3564. https:// doi. org/ 10. 1093/ brain/ awz290 47. McKee AC, Gavett BE, Stern RA, Nowinski CJ, Cantu RC, Kowall NW, Perl 29. Guo Z, Cupples LA, Kurz A, Auerbach SH, Volicer L, Chui H, Green RC, DP, Hedley-Whyte ET, Price B, Sullivan C et al (2010) TDP-43 proteinopa- Sadovnick AD, Duara R, DeCarli C et al (2000) Head injury and the risk of thy and motor neuron disease in chronic traumatic encephalopathy. J AD in the MIRAGE study. Neurology 54:1316–1323. https:// doi. org/ 10. Neuropathol Exp Neurol 69:918–929. https:// doi. org/ 10. 1097/ NEN. 0b013 1212/ wnl. 54.6. 1316e3181 ee7d85 30. Gyparaki MT, Arab A, Sorokina EM, Santiago-Ruiz AN, Bohrer CH, Xiao J, 48. McKee AC, Robinson ME (2014) Military-related traumatic brain injury and Lakadamyali M (2021) Tau forms oligomeric complexes on microtubules neurodegeneration. Alzheimers Dement 10:S242-253. https:// doi. org/ 10. that are distinct from tau aggregates. Proc Natl Acad Sci U S A. https:// 1016/j. jalz. 2014. 04. 003 doi. org/ 10. 1073/ pnas. 20214 61118 49. Mez J, Daneshvar DH, Kiernan PT, Abdolmohammadi B, Alvarez VE, Huber 31. Henninger N, Bouley J, Sikoglu EM, An J, Moore CM, King JA, Bowser R, BR, Alosco ML, Solomon TM, Nowinski CJ, McHale L et al (2017) Clinico- Freeman MR, Brown RH Jr (2016) Attenuated traumatic axonal injury and pathological evaluation of chronic traumatic encephalopathy in players improved functional outcome after traumatic brain injury in mice lacking of american football. JAMA 318:360–370. https:// doi. org/ 10. 1001/ jama. Sarm1. Brain 139:1094–1105. https:// doi. org/ 10. 1093/ brain/ aww0012017. 8334 32. Hernandez F, Merchan-Rubira J, Valles-Saiz L, Rodriguez-Matellan A, Avila 50. Niziolek GM, Boudreau RM, Baker J, Friend LA, Makley AT, Edwards MJ, J (2020) Differences between human and murine tau at the N-terminal Gulbins E, Goodman MD (2020) Acid sphingomyelinase inhibition end. Front Aging Neurosci 12:11. https:// doi. org/ 10. 3389/ fnagi. 2020. mitigates histopathological and behavioral changes in a murine model of 00011 traumatic brain injury. J Neurotrauma 37:1902–1909. https:// doi. org/ 10. 33. Holtzman DM, Carrillo MC, Hendrix JA, Bain LJ, Catafau AM, Gault LM, 1089/ neu. 2019. 6436 Goedert M, Mandelkow E, Mandelkow EM, Miller DS et al (2016) Tau: From 51. Niziolek GM, Hoehn RS, Seitz AP, Jernigan PL, Makley AT, Gulbins E, research to clinical development. Alzheimers Dement 12:1033–1039. Edwards MJ, Goodman MD (2021) The role of acid sphingomyelinase https:// doi. org/ 10. 1016/j. jalz. 2016. 03. 018 inhibition in repetitive mild traumatic brain injury. J Surg Res 259:296– 34. Hsu ET, Gangolli M, Su S, Holleran L, Stein TD, Alvarez VE, McKee AC, 304. https:// doi. org/ 10. 1016/j. jss. 2020. 09. 034 Schmidt RE, Brody DL (2018) Astrocytic degeneration in chronic trau- 52. Paxinos G, Franklin KBJ (2013) The mouse brain in stereotaxic coordinates. matic encephalopathy. Acta Neuropathol 136:955–972. https:// doi. org/ Elsevier Academic Press, Cambridge 10. 1007/ s00401- 018- 1902-3 K ahriman et al. acta neuropathol commun (2021) 9:118 Page 17 of 17 53. Peeters W, van den Brande R, Polinder S, Brazinova A, Steyerberg EW, 67. Smith DH, Johnson VE, Stewart W (2013) Chronic neuropathologies Lingsma HF, Maas AI (2015) Epidemiology of traumatic brain injury in of single and repetitive TBI: substrates of dementia? Nat Rev Neurol Europe. Acta Neurochir ( Wien) 157:1683–1696. https:// doi. org/ 10. 1007/ 9:211–221. https:// doi. org/ 10. 1038/ nrneu rol. 2013. 29 s00701- 015- 2512-7 68. Smith VD, Bachstetter AD, Ighodaro E, Roberts K, Abner EL, Fardo DW, 54. Petraglia AL, Plog BA, Dayawansa S, Dashnaw ML, Czerniecka K, Walker Nelson PT (2018) Overlapping but distinct TDP-43 and tau pathologic CT, Chen M, Hyrien O, Iliff JJ, Deane R et al (2014) The pathophysiology patterns in aged hippocampi. Brain Pathol 28:264–273. https:// doi. org/ underlying repetitive mild traumatic brain injury in a novel mouse model 10. 1111/ bpa. 12505 of chronic traumatic encephalopathy. Surg Neurol Int 5:184. https:// doi. 69. Tagge CA, Fisher AM, Minaeva OV, Gaudreau-Balderrama A, Moncaster org/ 10. 4103/ 2152- 7806. 147566 JA, Zhang XL, Wojnarowicz MW, Casey N, Lu H, Kokiko-Cochran ON et al 55. Rajic Bumber J, Pilipovic K, Jankovic T, Dolenec P, Grzeta N, Kriz J, Zupan G (2018) Concussion, microvascular injury, and early tauopathy in young (2021) Repetitive traumatic brain injury is associated with TDP-43 altera- athletes after impact head injury and an impact concussion mouse tions, neurodegeneration, and glial activation in mice. J Neuropathol Exp model. Brain 141:422–458. https:// doi. org/ 10. 1093/ brain/ awx350 Neurol 80:2–14. https:// doi. org/ 10. 1093/ jnen/ nlaa1 30 70. Tan XL, Sun M, Brady RD, Liu S, Llanos R, Cheung S, Wright DK, Casillas- 56. Rehman SU, Ahmad A, Yoon GH, Khan M, Abid MN, Kim MO (2018) Espinosa PM, Sashindranath M, O’Brien TJ et al (2019) Transactive Inhibition of c-Jun N-terminal kinase protects against brain damage and Response DNA-Binding Protein 43 Abnormalities after Traumatic Brain improves learning and memory after traumatic brain injury in adult mice. Injury. J Neurotrauma. 36:87–99. https:// doi. org/ 10. 1089/ neu. 2017. 5491 Cereb Cortex 28:2854–2872. https:// doi. org/ 10. 1093/ cercor/ bhx164 71. Tang SJ, Fesharaki-Zadeh A, Takahashi H, Nies SH, Smith LM, Luo A, 57. Sabbagh JJ, Fontaine SN, Shelton LB, Blair LJ, Hunt JB Jr, Zhang B, Gut- Chyung A, Chiasseu M, Strittmatter SM (2020) Fyn kinase inhibition mann JM, Lee DC, Lloyd JD, Dickey CA (2016) Noncontact rotational head reduces protein aggregation, increases synapse density and improves injury produces transient cognitive deficits but lasting neuropathological memory in transgenic and traumatic Tauopathy. Acta Neuropathol Com- changes. J Neurotrauma 33:1751–1760. https:// doi. org/ 10. 1089/ neu. mun 8:96. https:// doi. org/ 10. 1186/ s40478- 020- 00976-9 2015. 4288 72. Thau-Zuchman O, Shohami E, Alexandrovich AG, Trembovler V, Leker RR 58. Sacramento CB, Sondhi D, Rosenberg JB, Chen A, Giordano S, Pey E, Lee (2012) The anti-inflammatory drug carprofen improves long-term out - V, Stiles KM, Havlicek DF, Leopold PL et al (2020) Anti-phospho-tau gene come and induces gliogenesis after traumatic brain injury. J Neurotrauma therapy for chronic traumatic encephalopathy. Hum Gene Ther 31:57–69. 29:375–384. https:// doi. org/ 10. 1089/ neu. 2010. 1673 https:// doi. org/ 10. 1089/ hum. 2019. 174 73. Wiesner D, Tar L, Linkus B, Chandrasekar A, Olde Heuvel F, Dupuis L, Tsao 59. Santacruz K, Lewis J, Spires T, Paulson J, Kotilinek L, Ingelsson M, Guima- W, Wong PC, Ludolph A, Roselli F (2018) Reversible induction of TDP-43 raes A, DeTure M, Ramsden M, McGowan E et al (2005) Tau suppression in granules in cortical neurons after traumatic injury. Exp Neurol 299:15–25. a neurodegenerative mouse model improves memory function. Science https:// doi. org/ 10. 1016/j. expne urol. 2017. 09. 011 309:476–481. https:// doi. org/ 10. 1126/ scien ce. 11136 94 74. Woerman AL, Patel S, Kazmi SA, Oehler A, Freyman Y, Espiritu L, Cotter R, 60. Schneider CA, Rasband WS, Eliceiri KW (2012) NIH Image to ImageJ: 25 Castaneda JA, Olson SH, Prusiner SB (2017) Kinetics of human mutant tau years of image analysis. Nat Methods 9:671–675. https:// doi. org/ 10. 1038/ prion formation in the brains of 2 transgenic mouse lines. JAMA Neurol nmeth. 2089 74:1464–1472. https:// doi. org/ 10. 1001/ jaman eurol. 2017. 2822 61. Schneider J, Jeon S, Gladman JT, Corriveau RA ADRD Summit 2019 75. Xu X, Cowan M, Beraldo F, Schranz A, McCunn P, Geremia N, Brown Z, Report to the National Advisory Neurological Disorders and Stroke Patel M, Nygard KL, Khazaee R et al (2021) Repetitive mild traumatic brain Council. https:// www. ninds. nih. gov/ News- Events/ Events- Proce edings/ injury in mice triggers a slowly developing cascade of long-term and per- Events/ Alzhe imers- Disea se- Relat ed- Demen tias- Summit- 2019 accessed sistent behavioral deficits and pathological changes. Acta Neuropathol 05/13/2020 Commun 9:60. https:// doi. org/ 10. 1186/ s40478- 021- 01161-2 62. Selvaraj P, Wen J, Tanaka M, Zhang Y (2019) Therapeutic effect of a novel 76. Yang Z, Wang P, Morgan D, Lin D, Pan J, Lin F, Strang KH, Selig TM, Perez fatty acid amide hydrolase inhibitor PF04457845 in the repetitive closed PD, Febo M et al (2015) Temporal MRI characterization, neurobiochemi- head injury mouse model. J Neurotrauma 36:1655–1669. https:// doi. org/ cal and neurobehavioral changes in a mouse repetitive concussive head 10. 1089/ neu. 2018. 6226 injury model. Sci Rep 5:11178. https:// doi. org/ 10. 1038/ srep1 1178 63. Seo JS, Lee S, Shin JY, Hwang YJ, Cho H, Yoo SK, Kim Y, Lim S, Kim YK, 77. Yoshiyama Y, Higuchi M, Zhang B, Huang SM, Iwata N, Saido TC, Maeda Hwang EM et al (2017) Transcriptome analyses of chronic traumatic J, Suhara T, Trojanowski JQ, Lee VM (2007) Synapse loss and microglial encephalopathy show alterations in protein phosphatase expression activation precede tangles in a P301S tauopathy mouse model. Neuron associated with tauopathy. Exp Mol Med 49:e333. https:// doi. org/ 10. 53:337–351. https:// doi. org/ 10. 1016/j. neuron. 2007. 01. 010 1038/ emm. 2017. 56 78. Zhang J, Teng Z, Song Y, Hu M, Chen C (2015) Inhibition of monoacylglyc- 64. Shin N, Kim HG, Shin HJ, Kim S, Kwon HH, Baek H, Yi MH, Zhang E, Kim erol lipase prevents chronic traumatic encephalopathy-like neuropathol- JJ, Hong J et al (2019) Uncoupled endothelial nitric oxide synthase ogy in a mouse model of repetitive mild closed head injury. J Cereb enhances p-Tau in chronic traumatic encephalopathy mouse model. Blood Flow Metab 35:706. https:// doi. org/ 10. 1038/ jcbfm. 2014. 247 Antioxid Redox Signal 30:1601–1620. https:// doi. org/ 10. 1089/ ars. 2017. Publisher’s Note 65. Sidoryk-Wegrzynowicz M, Struzynska L (2019) Astroglial contribution to Springer Nature remains neutral with regard to jurisdictional claims in pub- tau-dependent neurodegeneration. Biochem J 476:3493–3504. https:// lished maps and institutional affiliations. doi. org/ 10. 1042/ BCJ20 190506 66. Smith DH, Dolle JP, Ameen-Ali KE, Bretzin A, Cortes E, Crary JF, Dams- O’Connor K, Diaz-Arrastia R, Edlow BL, Folkerth R et al (2021) Collaborative neuropathology network characterizing outcomes of TBI (CONNECT- TBI). Acta Neuropathol Commun 9:32. https:// doi. org/ 10. 1186/ s40478- 021- 01122-9 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acta Neuropathologica Communications Springer Journals

Mouse closed head traumatic brain injury replicates the histological tau pathology pattern of human disease: characterization of a novel model and systematic review of the literature

Loading next page...
 
/lp/springer-journals/mouse-closed-head-traumatic-brain-injury-replicates-the-histological-BE1YKjSa7K

References (87)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
2051-5960
DOI
10.1186/s40478-021-01220-8
Publisher site
See Article on Publisher Site

Abstract

Traumatic brain injury ( TBI) constitutes one of the strongest environmental risk factors for several progressive neu- rodegenerative disorders of cognitive impairment and dementia that are characterized by the pathological accu- mulation of hyperphosphorylated tau (p-Tau). It has been questioned whether mouse closed-head TBI models can replicate human TBI-associated tauopathy. We conducted longitudinal histopathological characterization of a mouse closed head TBI model, with a focus on pathological features reported in human TBI-associated tauopathy. Male C57BL/6 J mice were subjected to once daily TBI for 5 consecutive days using a weight drop paradigm. Histologi- cal analyses (AT8, TDP-43, pTDP-43, NeuN, GFAP, Iba-1, MBP, SMI-312, Prussian blue, IgG, βAPP, alpha-synuclein) were conducted at 1 week, 4 weeks, and 24 weeks after rTBI and compared to sham operated controls. We conducted a systematic review of the literature for mouse models of closed-head injury focusing on studies referencing tau protein assessment. At 1-week post rTBI, p-Tau accumulation was restricted to the corpus callosum and perivascular spaces adjacent to the superior longitudinal fissure. Progressive p-Tau accumulation was observed in the superficial layers of the cerebral cortex, as well as in mammillary bodies and cortical perivascular, subpial, and periventricular locations at 4 to 24 weeks after rTBI. Associated cortical histopathologies included microvascular injury, neuroaxonal rarefaction, astroglial and microglial activation, and cytoplasmatic localization of TDP-43 and pTDP-43. In our system- atic review, less than 1% of mouse studies (25/3756) reported p-Tau using immunostaining, of which only 3 (0.08%) reported perivascular p-Tau, which is considered a defining feature of chronic traumatic encephalopathy. Commonly reported associated pathologies included neuronal loss (23%), axonal loss (43%), microglial activation and astrogliosis (50%, each), and beta amyloid deposition (29%). Our novel model, supported by systematic review of the literature, indicates progressive tau pathology after closed head murine TBI, highlighting the suitability of mouse models to replicate pertinent human histopathology. Keywords: Animal model, Chronic traumatic encephalopathy, Concussion, Systematic review, Tauopathy, Traumatic brain injury Introduction Traumatic brain injury (TBI) represents a major public *Correspondence: nils.henninger@umassmed.edu health problem affecting more than 10 million people Department of Neurology, Medical School, University of Massachusetts, worldwide each year [36]. TBI is a leading cause of adult 55 Lake Ave, Worcester, USA death and disability worldwide. It has been estimated Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Kahriman et al. acta neuropathol commun (2021) 9:118 Page 2 of 17 that annually 150 to 200/1,000,000 people become disa- histopathological features of human TBI-associated bled as a result of brain trauma, and TBI was declared a tauopathy according to reported consensus criteria for major public health problem by the National Institutes of human CTE [8, 46]. In addition, we conducted a system- Health in 1999 [4, 20, 53]. atic literature review of reported tau pathology in mouse There is a long history of epidemiological evidence closed-head TBI and its relationship to pertinent histo- that TBI represents one of the strongest environmental logical features of human CTE to provide a contempo- risk factors for several progressive neurodegenerative rary overview of the state of the field. disorders of cognitive impairment and dementia that are characterized by the pathological accumulation of hyper- Material and methods phosphorylated tau (p-Tau). In particular, TBI has been Mouse study linked to Alzheimer’s disease (AD) and a unique clinico- Mice were randomly allocated to sham surgery versus pathological entity termed chronic traumatic encepha- rTBI except for 4 mice allowed to survive for 24-weeks lopathy (CTE) [8, 17, 29, 46, 66, 67]. Yet, the exact after rTBI, which were added to gain insight into the mechanism(s) driving pathological tau accumulation and longer-term evolution of p-Tau and associated pathology. spread, cognitive impairment, and dementia after TBI Histological analyses were conducted by an investigator are poorly understood. Defining the mechanisms that masked to the experimental groups. explain the link between TBI and dementia at the cellular level is a public health priority [61]. Animals Mouse models of TBI-associated pathology are of great Male C57BL/6  J mice (Jackson Laboratories) were value because of the ability to conduct detailed, longi- socially housed with same-sex mice (n = 4 per cage) tudinal histopathological characterization in a tempo- on 12-h light/dark cycle with food and water ad  libi- rally accelerated fashion, as well as the unique possibility tum. Spontaneously breathing mice (n = 27) weighing to explore underlying molecular mechanisms through 26.8 ± 2.6  g (age 8–12  weeks) were subjected to closed genetic manipulation without confounding by comorbid head injury (n = 20) or sham surgery (n = 11). Brains conditions. This appears particularly important in light were removed 1  week after rTBI (n = 8), 4  weeks (sham of a current lack of a uniformly accepted definition for n = 8, rTBI n = 8), and 24 weeks (rTBI n = 4) for histolog- TBI-specific tauopathy. Although pathological consen - ical analyses. Additionally, we conducted AT8-staining in sus criteria have been developed for CTE [8, 46], which 3 sham operated mice that survived for 12 months to rule is considered the prototypical TBI-associated tauopathy, out any potential age-related effects on p-Tau pathology. challenges in applying these criteria relate to the fact that many pathologies are also present in the normally Anesthesia, analgesia, and traumatic brain injury induction aging brain and other neurodegenerative conditions [27, Animals were anesthetized with isoflurane (5% for induc - 38]. Hence, distinguishing the direct effects of TBI from tion, 2% for surgery, 1.5% for maintenance) in room a sporadic progressive neurodegenerative process in air. Anesthesia was discontinued immediately prior to humans is difficult. Moreover, animal models are needed TBI and sham injury. Body temperature was monitored to dissect the specific neuropathology of TBI and how continuously with a rectal probe and maintained at pathological tau deposition relates to progressive neuro- 37.0 ± 0.5 °C. To alleviate pain, animals received 0.05 mg/ degeneration. However, concerns have been raised that kg subcutaneous buprenorphine (Med-Vet International, murine closed head injury models may not be suitable to Mettawa, Il, USA) 30 min before anesthesia and every 6 h replicate pertinent histopathological features of human afterwards for 24  h. Additionally, each animal received TBI-associated tauopathy [18]. Depending on the model 5  mg/kg subcutaneous carprofen (Patterson Veterinary, used, variability in post-TBI histopathology may be con- Devens, MA, USA) at the end of the anesthesia. siderable [9, 10]. Further, there are distinct differences TBI was produced using a weight drop device as pre- in the cerebral anatomy as well as expression of tau iso- viously described in detail [11, 31] with the modifica - forms in the adult brains of mice and humans [32], which tion that animals were subjected to TBI or sham injury could conceivably contribute to disparate pathology [41]. once daily for 5 consecutive days. Briefly, a 50  g weight Therefore, it is critical to understand to what degree was freely dropped 15  cm to strike a cylindrical polyac- murine closed-head TBI can replicate human TBI-associ- etal transducer rod (Delrin , tip-diameter 2  mm, 17.4  g) ated tau pathology and related histopathological features. that was placed with its tip directly on the exposed In this study, we characterized the histopathological skull (target 2.5  mm posterior and 2.5  mm lateral from features of a mouse closed head repetitive TBI (rTBI) Bregma). Following TBI, the wound closed with inter- model with specific focus on the presence and evolution rupted sutures. Sham animals were anesthetized, surgi- of p-Tau accumulation and its association with pertinent cally prepared (including skin incision), and placed under K ahriman et al. acta neuropathol commun (2021) 9:118 Page 3 of 17 the impact device with the impactor touching the skull, Hematoxylin counterstaining was used for AT8, but were not subjected to head impact. One mouse with AT180, βAPP, and α-synuclein labeled tissues. For a skull fracture was excluded. immunofluorescence staining tissue sections labeled with the primary antibodies (TDP-43, pTDP-43, NeuN, Immunohistochemistry staining GFAP, Iba-1, MBP, SMI-312, βAPP, alpha-synuclein) For histology, animals received an overdose of pentobar- were incubated in appropriate secondary antibodies bital (150  mg/kg Fatal-Plus, Vortech Pharmaceuticals). conjugated with Alexa Fluor 488 (1:250, Abcam, Cat# Then animals were perfused under isoflurane anesthesia ab150113, RRID: AB_2576208 and Cat# ab150077, through the ascending aorta with 50 mL saline and then RRID: AB_2630356), Alexa Fluor 555 (1:250, Abcam, with ice cold phosphate-buffered 4% paraformaldehyde Cat# ab150106, RRID: AB_2857373), and Alexa Fluor (PFA) for 10  min. Brains were removed from the cra- 647 (1:250, Abcam, Cat# ab150075, RRID: AB_2752244 nium, postfixed overnight in the same fixative, and then and Cat# ab150115, RRID: AB_2687948). Omitting the stored in 0.4% PFA at 4 °C until further processing. Prior primary antibody in a subset of slides served as nega- to paraffin embedding brains were pre-sectioned using a tive controls. brain matrix. Immunohistochemistry was performed against Prussian blue staining Ser−202/Thr205 p-Tau , (AT8, 1:250, Thermo Fisher Sci- To assess for microhemorrhages, sections were stained entific, Cat# MN1020, RRID: AB_223647), p-Tau- for Prussian blue reaction using an Iron Stain Kit (# Thr231 (AT180, 1:250, Thermo Fisher Scientific, Cat# HT20, Sigma-Aldrich), following the manufacturer’s MN1040, RRID: AB_223649), TAR DNA-bind- instructions. ing protein 43 (TDP-43, 1:250, Proteintech, Cat# Ser−409/410 10,782–2-AP, RRID: AB_615042), pTDP-43 Microscopy (1:250, Proteintech, Cat# 22,309–1-AP, RRID: Paraffin sections, 10-µm thick coronal, were obtained AB_11182943), neuronal nuclei (NeuN, 1:200, Pro- at approximately Bregma -3.07  mm (impact center; teintech, Cat# 26,975–1-AP, RRID: AB_2880708), s3), −1.67  mm (adjacent to the impact center; s2), glial fibrillary acidic protein (GFAP, 1:250, Agilent, and + 1.21 mm (remote to the impact center; s1) for his- Cat# Z0334, RRID: AB_10013382), ionized calcium tological assessment (Fig.  1a, b). For quantitative and binding adaptor molecule 1 (1:250, Iba-1, Wako, Cat# qualitative analyses of AT8, NeuN, GFAP, Iba-1, MBP, 019–19,741, RRID: AB_839504), myelin basic pro- and SMI-312 staining one coronal section each from tein (MBP, 1:200, Santa Cruz Biotechnology, Cat# these coordinates (s1-s3) was used. For qualitative anal- M3821, RRID: AB_1841021), SMI-312 (1:200, BioLe- yses of βAPP, alpha-synuclein, IgG, and Prussian blue gend Cat# 837,904, RRID: AB_2566782), beta amyloid staining we reviewed one coronal section each from precursor protein (βAPP, 1:200, Zymed, CT695, Cat# s1-s3. For quantitative and qualitative analyses of TDP-43 51–2700, RRID: AB_2533902), α-synuclein (1:250, and pTDP-43 one coronal section from s2 was used. All Biolegend, Cat# 824,301, RRID: AB_2564879), and histological analyses were performed by an investigator immunoglobulin G (IgG, 1:100, Abcam, Cat# ab6708, masked to the animal groups. RRID: AB_956005). For chromogenic staining, tissue sections labeled with the primary antibodies (AT8, AT180, NeuN, βAPP, α-synuclein, IgG) were incu- Image acquisition and quantification bated with appropriate biotin-conjugated second- To acquire images of all stained sections for subsequent ary antibodies followed by avidin–biotin complex offline analysis we used a Leica DM6 B microscopy sys - (Vector Laboratories) incubation and treatment with tem equipped with a brightfield DMC5400 color CMOS diaminobenzidine as directed by the manufacturer. (See figure on next page.) Fig. 1 Patterns and evolution of p-Tau accumulation after closed head repetitive traumatic brain injury (rTBI). (a) Approximate location of the impact center over the intact mouse skull (blue circle) relative to the brain sections sampled for histological analysis (s1-s3; dashed lines). (b) Approximate location of p-Tau positive cells at 4 weeks after rTBI (composite of 8 mice; each red dot represents 8 p-Tau positive cells, blue ellipses indicate the spatial relation between the impactor and brain surface). Black boxes indicate the location of photomicrographs shown in panel c-e (box 1) and f–h (box 2). (c) Intact cerebral cortex without p-Tau accumulation at 1 week after rTBI. Progressive accumulation of p-Tau in the superficial layers of the cerebral cortex at (d) 4 weeks and (e) 24 weeks after rTBI. P-Tau accumulation in the corpus callosum at (f) 1 week, (g) 4 weeks, and (h) 24 weeks after rTBI (red asterisks). (f ) In contrast to later time points, AT8-immunoreactivity at 1 week was restricted to dot-like staining in a subset of cells (arrowheads). Examples of p-Tau accumulation in (i) perivascular, (j) subpial, (k, l) periventricular, and (l) mammillary body locations as well as (m) at the depth of the superficial longitudinal fissure (white arrowheads) in perivascular (black arrowheads) and subpial (red arrowheads) locations at 4 weeks after rTBI. Scale bars are 30 µm (in c–j), 1 mm (k), and 300 µm (in l–m) Kahriman et al. acta neuropathol commun (2021) 9:118 Page 4 of 17 Fig. 1 (See legend on previous page.) K ahriman et al. acta neuropathol commun (2021) 9:118 Page 5 of 17 camera and an immunofluorescent DFC9000 sCMOS section were taken at 63 × magnification and analyzed camera. as described for GFAP. To determine the spatial distribution of p-Tau accumulation after rTBI, sections were imaged at Assessment of chronic traumatic encephalopathy (CTE) 63 × magnification and the approximate location of all related pathology AT8-positive profiles systematically recorded within With respect to assessing CTE-like pathology in our each of the sampled sections and transferred to a model, we defined histopathological features according standard atlas [52] to provide a visual representation to recently published consensus group criteria developed of the observed tau pathology relative to the impact by an NINDS/NIBIB panel [8, 46] with modifications for location. Given reported p-Tau accumulation in both use in mice (Table 1). Specifically, a defining criterion for neurons and astrocytes in human and mouse TBI, we human CTE includes the presence of perivascular foci quantified the number of p-Tau positive cells across of p-Tau immunoreactive neurofibrillary tangles (NFTs) time points as stratified by neuronal versus astrocytic and abnormal neurites, with or without p-Tau immu- p-Tau. Moreover, because we observed the earliest noreactive astrocytes, in an irregular pattern in the cer- p-Tau accumulation in the corpus callosum with only ebral cortex, with a tendency to involve the sulcal depths. later involvement of the cerebral cortex we addition- Because the lissencephalic brains of mice lack sulci, we ally stratified these analyses according to the location considered the presence of perivascular p-Tau in the cer- in the superficial cortex (approximately cortical lay- ebral cortex a pathognomonic criterion. With respect ers I-III), deep cortex (approximately cortical layers to supportive features, we assessed for several support- IV-VI), and the corpus callosum. ive tau- and nontau-related histopathological features To determine the extent of neuronal loss, chromo- (Table  1). We did not assess for macroscopic pathology gen stained NeuN-positive cells were assessed in each such as dilation of ventricles, septal pathology, atrophy, coronal section. Images of 16 nonoverlapping fields contusions, or other signs of previous trauma [8, 46]. of view (FOV; 8 per hemisphere; 659 × 439  µm, each) Finally, for additional context we also assessed several covering the dorsal cerebral cortex [11] were taken at pathological features that have been repeatedly described 20 × magnification. NeuN positive cells were semiau- in human TBI but are not specific to CTE and may be tomatically quantified using ImageJ [60] as previously shared with TBI and other neurodegenerative conditions described [15]. First, 16-bit color images were con- including the presence of reactive microglia, astrogliosis, verted to 8-bit grayscale followed by automatic thresh- neuronal and axonal loss, presence of beta amyloid and olding. The Analyze Particle tool was then used to alpha-synuclein depositions as well as evidence of prior count all particles with a circularity of 0.1–1.0 and a blood brain barrier (BBB) disruption and microvascular size of 25 to 250 µm . Results were adjusted by man- injury/cerebral microbleeds [45, 49]. ually counting all overlapping cells (ie, particle size greater than 250 µm ). Neurologic evaluation To assess microglia and astroglia in the dorsal cor- Return of the righting reflex was measured as the time (s) tex we used fluorescence staining. For GFAP, images from TBI/sham injury to righting from a supine to prone of 16 nonoverlapping FOVs (8 per hemisphere; position after discontinuation of anesthesia. The neuro - 667 × 667  µm, each) covering the dorsal cerebral cor- logical severity score (NSS) was assessed on a scale from tex were taken at 20 × magnification. First, 16-bit color 0 (no deficit) to 10 (maximal deficit) prior to TBI as well images were split into the individual color channels as serially until sacrifice as previously described in detail (red, green, blue) followed by automatic threshold- [31]. ing of the 8-bit green channel and black-white color inversion. The Analyze Particle tool was then used to Statistical analysis quantify the total thresholded area (µm ). For Iba-1, Unless otherwise stated, continuous variables are TDP-43, and pTDP-43 images of 16 FOVs (8 per hemi- reported as mean ± standard error of the mean. Normal- sphere; 211 × 211 µm, each) centered within the corre- ity of data was examined using the Shapiro–Wilk test. sponding FOV used for the GFAP analyses were taken One-way analysis of variance (ANOVA) on Ranks with at 63 × magnification and analyzed as described for post-hoc Dunn’s test was used to assess for between- GFAP. group differences in histopathology (p-Tau, TDP-43, To assess the impact of rTBI on axonal integrity in pTDP-43, NeuN, Iba-1, GFAP, MBP, SMI-312). Between- the injured cortex, we used fluorescence staining for group comparisons of continuous variables over time the myelin marker MBP and pan-axonal neurofilament (body weight, return of the righting reflex, NSS) were marker SMI312. Images of one FOV (206 × 206 µm) per conducted using longitudinal mixed models. Time was Kahriman et al. acta neuropathol commun (2021) 9:118 Page 6 of 17 Table 1 Chronic traumatic encephalopathy (CTE) related pathology as adapted from consensus criteria [8, 46] and including frequently associated pathology [38, 45, 48, 49] for use in mice in the systematic review and characterization of our mouse model 1 week 4 weeks 24 weeks Neuropathology considered Pathognomonic for CTE Perivascular p-Tau accumulation P1: p-Tau immunoreactive neurons – x x P2: p-Tau immunoreactive astrocytes – x x Neuropathological features Supportive of CTE P-Tau related pathologies S1: Cortical p-Tau (preferentially in superficial layers) – x x S2: Hippocampal p-Tau – – – S3: p-Tau present in subcortical nuclei Mamillary bodies – x x Amygdala – – – Thalamus – – – S4: Astroglial p-Tau in subpial and periventricular regions p-Tau immunoreactive astrocytes in the subpial regions – x x p-Tau immunoreactive astrocytes in the periventricular regions – – – Non-p-Tau related histological pathologies S5: TDP-43 immunoreactive neuronal cytoplasmic inclusions Cortex – x x Hippocampus – – x Amygdala – – – Select nonspecific neuropathological features A ssociated with CTE* A1: β-amyloid precursor protein depositions – – – A2: α-synuclein depositions – – – A3: Hemosiderin laden macrophages x x x A4: Reactive microglia – x x A5: Astrogliosis x x x A6: Neuronal loss – x x A7: Axonal loss – x x A8: Blood brain barrier disruption X x not done *These selected histopathological features have been commonly reported to accompany CTE pathology but are not used to in the consensus criteria to define CTE. P, S, and A refer to pathognomonic, supportive, and associated pathology. “x “ and “–” denote histopathological feature present and absent in our model, respectively treated as a categorical variable. The models included outcomes that have been associated with CTE, includ- group (Sham versus rTBI) and time as fixed covariates, as ing assessment of axonal and neuronal injury, astroglio- well as the group × time interactions. Two-sided signifi - sis, microglial activation, TDP-43 pathology, as well as cance tests were used throughout and unless stated oth- presence of amyloid pathology, α-synuclein, cerebral erwise a two-sided P < 0.05 was considered statistically microbleeds, and evidence of BBB disruption. Overall, significant. All statistical analyses were performed using we identified 1940 articles in PubMed and 2057 articles SigmaPlot 12.5 (Systat Software, Inc., Germany) or IBM in Scopus. After removal of duplicates (n = 241), 3756 ® ® SPSS Statistics Version 26 (IBM -Armonk, NY). papers were included for screening. Details of the review methodology including the search strategy, inclusion and Systematic review exclusion criteria, and retrieval of information from the We conducted a systematic review of the literature by full-text articles is summarized in the Additional file  1: searching PubMed and Scopus using search criteria Supplementary information. that were established to be specific for mouse models of closed head injury (including blast injuries) and that con- tained reference to tau protein assessment in wild type mouse strains. In addition to characterizing model char- acteristics of included studies, we collected histological K ahriman et al. acta neuropathol commun (2021) 9:118 Page 7 of 17 Fig. 2 Temporal and spatial distribution of p-Tau pathology in neurons and astroglia. (a,b) Double staining indicates colocalization (white arrows) of hyperphosphorylated tau (p-Tau; AT8) with (a) neurons (NeuN) and (b) astrocytes (GFAP) in the cerebral cortex at 4 weeks after repetitive traumatic brain injury (rTBI). White scale bars correspond to 50 µm in low power and 17 µm in high power magnified panels. DAPI (blue) channel omitted from high power magnifications. (c) Dotted lines delineate cortical layers I-III (superficial cortex) from cortical layers IV-VI (deep cortex) and corpus callosum used to assess the presence of p-Tau stained cells in the ipsilateral hemisphere. Asterisk denotes the approximate location of the images taken for a‑b. (d) Proportion of AT8 positive neurons and astrocytes at the investigated time points after rTBI. (e) Distribution of AT8 stained cells in the cerebral cortex (superficial + deep combined) relative to the impact center (black bars). There was no difference in the number and distribution of p-Tau positive cells at 4 weeks and 24 weeks after rTBI (P > 0.05). Because sham and 1-week rTBI animals had no cortical AT8-positive cells they were omitted from this analysis. Each bar corresponds to one cortical field of view (FOV ) arranged from left, contralateral (FOV 1) to right, ipsilateral (FOV 16), whereby corresponding FOVs in the three investigated sections s1–s3 were summed. Data are mean ± SD. (f ) Number of p-Tau positive neurons (green shades) and astrocytes (blue shades) in the traumatized hemisphere stratified by location in the superficial cortex, deep cortex, and corpus callosum over time (total number of cells counted in the three investigated sections s1–s3). *P < 0.05. Data are mean ± SEM. n = 8 mice for sham, 1 week, and 4 weeks, n = 4 for 24 weeks (there were no p-Tau positive cells in sham operated mice). All analyses were done using one-way ANOVA on Ranks with post-hoc Dunn’s Results BBB disruption as assessed by IgG staining at 1 week and Mouse rTBI model 4 weeks post rTBI (Additional file 1: Figure S2). Similar to sham operated mice, brains of rTBI mice appeared macroscopically intact and without evidence pT ‑ au spreads from the corpus callosum to superficial for macroscopic cerebral hemorrhages at all time points cortical layers between 1 to 24 weeks after rTBI post injury. Microscopically, the cerebral cortex appeared The primary goal of this study was to determine the normal at 1 week after rTBI. However, at 4 and 24 weeks spatial and temporal evolution of p-Tau pathology over after rTBI we observed focal tissue disruption within the 24 weeks after murine closed head-injury. We used both superficial cerebral cortex approximately corresponding AT8 and AT180 to detect p-Tau. AT180 staining yielded to cortical layer I (Fig. 1c-e). similar results as AT8 staining (Additional file  1: Figure Consistent with human pathology [28], Prussian blue S3) and we exclusively refer to AT8 staining to describe staining showed the presence of hemosiderin laden p-Tau pathology below. Importantly, we found no AT8 macrophages indicative of microvascular injury in the positive cells in sham operated mice that survived for superficial layers of the ipsilateral cortex, at the depth 4 weeks and 12 months after surgery, respectively (Addi- of the superior longitudinal fissure, and grey-white mat - tional file 1: Figure S4). ter junction between cortex and corpus callosum (Addi- We found that 1  week after rTBI, 4 mice (n = 50%) tional file  1: Figure S1). There was evidence of subtle showed faint AT8 staining restricted to the corpus cal- losum without any p-Tau accumulation in the overlying Kahriman et al. acta neuropathol commun (2021) 9:118 Page 8 of 17 Fig. 3 Patterns and evolution of TDP43 (a–h) and pTDP43 ▸ (j–q) pathology in the cerebral cortex after repetitive closed head traumatic brain injury (rTBI) in the mouse. (f) Increased linear pattern of TDP-43 reactivity suggesting neuritic distribution at 1 week after rTBI. Nuclear loss (red arrowheads) and cytoplasmatic localization (white arrowheads) of TDP-43 (g‑h) and pTDP-43 (p‑ q) at 4 and 24 weeks. Aggregation of cytoplasmatic localized TDP-43 (g‑h) and pTDP-43 (q). (r) Apparent reduction in signal intensity of pTDP-43 in the cortex next to the impact center (§). Scale bars correspond to 240 µm for low power and 20 µm for high power magnified images. In each group, bars correspond to one cortical field of view (FOV ) arranged from left, contralateral (contra), to right, ipsilateral (ipsi). Data are shown as mean (+ SD). n = 8 for sham, 1 week, and 4 week post rTBI and n = 4 for 24 week post rTBI. Analyses were done at Bregma − 1.67 mm (corresponding to s2 in Fig. 1) *P < 0.05. All analyses were done using one-way ANOVA on Ranks with post-hoc Dunn’s cerebral cortex (Figs.  1c, f, 2f ). P-Tau pathology did not substantially progress in the corpus callosum between 1 and 24 weeks and AT8 positive cells were present in only a subset of mice (n = 3 [38%] at 4  weeks, n = 2 [50%] at 24  weeks), typically within the section adjacent to the impact center (Figs.  1b, f–h, 2f). Conversely, by 4 and 24  weeks we observed significant accumulation of AT8- stained cells in all investigated mice, particularly in sub- pial locations and the superficial cortex (approximately cortical layers I-III) in the traumatized hemisphere (Figs.  1d–e, 2e–f). Figure  1b depicts the approximate distribution of AT8-positive cells within the traumatized hemisphere at 4  weeks after rTBI. Interestingly, stain- ing was most abundant in the cerebral cortex adjacent to the impact site (Bregma − 1.67 mm) rather than directly below the impact center (Bregma −3.07). pT ‑ au accumulation after murine rTBI occurs in cerebral locations reported for human TBI‑associated tauopathy At 4  weeks post rTBI we observed perivascular p-Tau accumulation (Fig.  1i, m), a pathognomonic histological feature of CTE. Additional, though less common, sites of p-Tau accumulation included the mamillary body (Fig. 1l) and periventricular tissues (Fig.  1k, l), which are consid- ered histological features supporting a CTE diagnosis. Notably, in all examined rTBI mice p-Tau accumulation was present in the subpial cortex adjoining the superficial longitudinal fissure (Fig.  1b, m). We did not find any AT8 stained cells in the hippocampus, thalamus, and amyg- dala (Table 1). the superficial cerebral cortex without overt co-staining pT ‑ au accumulates in cortical neurons and astrocytes in deeper cortical layers or the corpus callosum (Fig.  2d, We found distinct co-localization of AT8 with both NeuN f ). We found that perivascular p-Tau was present in both and GFAP in the traumatized hemisphere consistent with neurons and astrocytes (Additional file  1: Figure S5). The neuronal and astrocytic tau accumulation (Fig.  2a, b). majority of AT8 positive cells co-stained with NeuN, and Interestingly, colocalization of AT8 and GFAP was only observed at 4 and 24  weeks after rTBI and restricted to K ahriman et al. acta neuropathol commun (2021) 9:118 Page 9 of 17 Fig. 4 Progression of cortical neuronal loss and glial activation after repetitive traumatic brain injury (rTBI). (a) Cortical neuronal loss after rTBI as shown by NeuN staining. (b) Astroglial activation at 4 weeks as indicated by (a) the presence of numerous hypertrophied GFAP stained astrocytes (inset) and (c) a significant increase in GFAP staining signal. Decrease in astroglial activation at 24 weeks after rTBI as shown by reduced cell hypertrophy (inset in (a)) and attenuated GFAP staining (c). Microglial activation was observed at 4 weeks as indicated by the presence of numerous bipolar/rod shaped Iba-1 stained microglia (inset in (a)) and significantly increased in Iba-1 stained area (d). Normalization of microglial activation at 24 weeks after rTBI as noted by increased ramification of Iba-1 stained cells (inset in (a)) and similar Iba-1 staining intensity as compared to sham operated mice (d). Each bar corresponds to one cortical field of view (FOV ) arranged from left, contralateral (contra), to right, ipsilateral (ipsi). Data are shown as mean (+ SD). n = 8 (sham, 1 week and 4 week post rTBI, each) and n = 4 (24 week post rTBI) mice. Analyses are based on three coronal sections (corresponding to s1–s3 in Fig. 1). *P < 0.05. All analyses were done using one-way ANOVA on Ranks with post-hoc Dunn’s less than 10% of cells were positive for both AT8 and points after rTBI (not shown). Nonetheless, by 24 weeks GFAP (Fig. 2d, f ). after rTBI, nuclear loss and cytoplasmic localization of TDP-43 (Additional file  1: Figure S6) and pTDP-43 (not shown) was present in the contralateral cerebral cortex rTBI causes persistent nuclear loss and cytoplasmatic and bilateral CA1 of the hippocampus. localization of TDP‑43 and pTDP‑43 Compared to sham and 1 week post rTBI mice, we noted rTBI causes astroglial and microglial activation an overall increase in TDP-43 and pTDP-43 in the ipsi- and progressive neurodegeneration lateral cerebral cortex, particularly adjacent to the impact In addition to assessing pertinent p-Tau and TDP-43 center, at 4  weeks and 24  weeks after rTBI (Fig.  3). On related histological features of human CTE (Table  1), a cellular level, there was nuclear loss and cytoplasmatic we sought to determine pathology that occurs in human localization of TDP-43 and pTDP-43 at the same time TBI-associated neurodegenerative disease, including points (Fig.  3). However, while conspicuous aggregation the presence of reactive migroglia and astroglia, neu- of cytoplasmatic TDP-43 was present at 4 weeks (Fig. 3g), ronal and axonal loss, and accumulation of βAPP and this was only observed at 24 weeks for pTDP-43 (Fig. 3q). α-synuclein. In the contralateral (non-traumatized) hemisphere Sham and 1-week rTBI animals had few GFAP posi- staining intensity for TDP-43 and pTDP-43 in the cor- tive cells in the cerebral cortex and corpus callosum (< 1% pus callosum, subcortical nuclei, and hippocampus staining signal coverage) without difference between was overall similar to sham operated animals at all time hemispheres (Fig. 4c). We observed substantial astroglial Kahriman et al. acta neuropathol commun (2021) 9:118 Page 10 of 17 Fig. 5 Axonal loss in cortex following repetitive traumatic brain injury (rTBI). (a) Cartoon depicting the 3 sections s1–s3 used for analysis. Black boxes indicate the location of the three fields of view (FOVs) used to quantify the immunofluorescent imaging signal for the myelin marker myelin basic protein (MBP) and the pan-axonal neurofilament marker SMI-312. (b) Representative photomicrographs showing progressive loss of the MBP and SMI-312 staining signal from 1 to 24 weeks after rTBI. Quantified staining signal for (c) MBP and (d) SMI-312. Data are mean ± SEM. n = 8 mice for sham, 1 week, and 4 weeks, n = 4 for 24 weeks. *P < 0.05. n.s. indicates not significant. All analyses were done using one-way ANOVA on Ranks with post-hoc Dunn’s. Scale bar is 50 µm activation at 4 weeks after rTBI as indicated by the pres- b). However, by 4 weeks after rTBI, we observed a signifi - ence of hypertrophied GFAP stained astrocytes and over- cant loss of NeuN profiles, which worsened by 24 weeks all significantly increased GFAP staining in the cerebral indicative of progressive neurodegeneration (Fig.  4a, b). cortex (Fig.  4a, c). Although GFAP staining remained Loss of NeuN positive cells was most pronounced within significantly elevated by 24  weeks as compared to sham the ipsilateral cerebral cortex adjacent to the impact site operated animals, overall staining intensity appeared (rather than beneath the impact center) corresponding to attenuated, and cells appeared less hypertrophied when the area of maximal p-Tau accumulation (compare with compared to 4 weeks (Fig. 4a, c). Fig. 1a). There was significant microglial activation in the cer - At 1  week after rTBI, MBP and SMI-312 staining in ebral cortex at 4  weeks after rTBI as indicated by the the traumatized cerebral cortex appeared morphologi- presence of bipolar/rod shaped Iba-1-stained microglia cally similar without significant difference in the quanti - and overall increased Iba-1 staining (Fig.  4a, d). Activa- fied signal intensity as compared to sham operated mice tion subsided by 24  weeks as indicated by a more rami- (Fig.  5b). Concurrent with neuronal injury, we observed fied appearance of Iba-1-stained cells and return of Iba-1 a significant loss of MBP and SMI-312 stained profiles staining intensity to sham levels (Fig. 4a, d). indicating axonal degeneration by 4 weeks and 24 weeks At 1  week after rTBI, neuronal density in the ipsilat- after rTBI (Fig. 5c, d). eral cortex was similar to sham operated mice (Fig.  4a, Lastly, we found no βAPP and α-synuclein depositions in any of the operated mice (not shown). K ahriman et al. acta neuropathol commun (2021) 9:118 Page 11 of 17 rTBI causes long‑term functional deficits Sham operated mice regained their righting reflex within a median of 24  s (interquartile range 14–32  s) after dis- continuation of anesthesia. In contrast, righting reflex was significantly suppressed in rTBI mice and only returned after a median of 135  s (interquartile range 79–229 s) (Fig. 6a). We used the NSS, which is a composite of ratings measuring a combination of overall inquisitiveness, pos- tural stability, and motor function, to examine the tem- poral evolution of functional deficits in rTBI animals versus controls. Whereas sham operated animals had no change in the NSS over the observation period, rTBI mice had significant neurological deficits at 2 h after the last impact (day 5) when compared to baseline and con- trols. Although deficits partially improved over time, persistent residual neurological deficits were present by 24 weeks (Fig. 6b). Although both sham and rTBI mice lost weight after surgery, sham injured animals regained their baseline weight by 1  week whereas rTBI mice had a persistent weight loss up to 4 weeks after rTBI (Fig. 6c). Systematic literature review pT ‑ au pathology is infrequently reported in murine closed head models From searches on both PubMed and Scopus, a total of 3997 articles were initially included. After removal of 241 duplicates 3756 articles were screened for inclu- sion and exclusion criteria (Additional file  1: Figure S7a). From this screening, we excluded 2394 articles because they did not include the predetermined key- words and 411 studies that did not use mice, leaving 951 articles included in the full text search. After full text review, we excluded studies that were not pub- lished in a peer-reviewed journal (n = 5); review articles (n = 235); conference proceeding without primary data (n = 30); studies using transgenic animals without wild- Fig. 6 Persistent neurological deficits in mice subjected to repetitive type mice (n = 106); lacking assessment of cerebral tau traumatic brain injury (rTBI). (a) Significantly prolonged time to the return (n = 507); and penetrating head injury (n = 12). L astly, of the righting reflex (RR) after discontinuation of anesthesia in mice two studies identified from references were added, subjected to repetitive TBI (rTBI) versus sham operated animals. There was resulting in a total of 58 studies (1.5% of screened pub- as a significant group (P < 0.001) effect but no significant time (P = 0.958) lications) that investigated the presence of pathologi- effect or a group x time (P = 0.574) interaction (*P < 0.05, **P < 0.01, ***P < 0.001 versus sham). (b). There were significant group (P = 0.003) cal tau in the brains of mice subjected to closed-head and time (P < 0.001) effects as well as presence of a significant group injury. x time interaction (P < 0.001) for the composite neurological severity score (**P < 0.01, ***P < 0.001 versus baseline). (c) There were significant group and time effects as well as presence of a significant group x time Characteristics of included studies reporting tau pathology interaction (P < 0.001, each) for the change in body weight during the first assessment 4 weeks after surgery (*P < 0.05, **P < 0.01, ***P < 0.001 versus sham). Data Among the 58 included studies, 114 different injury are mean ± SEM. All statistical comparisons were made using mixed effects paradigms were used (42 single TBI and 72 repetitive models (n = 8 for sham; n = 20 for baseline to 1 week after rTBI, n = 12 for 2 to 4 weeks after rTBI, and n = 4 for 24 weeks after rTBI). For clarity in TBI models). Most models utilized male mice (112/114 the figure, post-hoc pairwise comparisons are only shown for significant [98%]), the C57BL/6 strain (111/114 [97%]), and anes- differences between groups (a, c) and versus baseline (b ) thesia (111/114 [97%]). The median mouse age at the Kahriman et al. acta neuropathol commun (2021) 9:118 Page 12 of 17 Table 2 Mouse traumatic brain injury ( TBI) models reporting p-Tau histopathology in the brain Reference Model Impacts ITI EDPI p‑ Tau Ab CTE‑like pathology* Repetitive TBI Pathognomonic Supportive Associated This study WD 5 24 7 AT8 P1, P2 S1,S3,S4, S5 A3 to A8 Petraglia et al. [54]** PD 6 24 30 AT8 P1, P2 S1,S2,S3,S4 A4, A5 Shin et al. [64] WD 20 24 42 P( T205) P1 S1, S2, S4 A5 Zhang et al. [78] PD 3 24 8 AT8 – S1, S2, S5 A1, A4, A5, A6, A7 Luo et al. [44] PD 3 24 180 AT8 – S1, S2, S3 A5 Briggs et al. [13] WD 30 24 93 AT8 – S1, S5 A1, A4, A5, A7 Albayram et al. [1]** RIA 7 24 14 Cisptau – S1, S5 A1, A4,A5, A7 Selvaraj et al. [62] PD 3 24 8 P(S404) – S1, S2 A1, A4, A5, A6 Kondo et al. [40]** RIA 7 24 10 Cisptau – S1, S2 A6, A7 Albayram et al. [2] RIA 5 24 240 AT8 – S1, S2 A4 Yang et al. [76] PD 4 72 10 CP13 – S1, S2 A5 Sacramento et al. [58] PD 10 2 21 AT8 – S1,S2 – Tagge et al. [69] RIA 2 0.25 1 Cisptau – S1 A4, A5, A7, A8 Rehman et al. [56] WD 3 24 7 P(S413) – S1 A1, A6, A8 Seo et al. [63] WD 3 72 7 PHF – S2 – Niziolek et al. [51] WD 4 48 30 P(S262) – S2 – Single TBI Liu et al. [42] RIA 1 n/a 30 T231 P1 S1, S2 – Petraglia et al. [54]** PD 1 n/a 30 AT8 – S1, S2, S3 A5 Goldstein et al. [26] PD 1 n/a 14 CP13 – S1,S2 A4, A5, A6, A7 Kondo et al. [40]** RIA 1 n/a 1 Cisptau – S1, S2 A6, A7 Huber et al. [35] Blast 1 n/a 30 CP13 – S1,S2 A6 Sabbagh et al. [57] Blast 1 n/a 56 T231 – S1, S2 – Iliff et al. [37] PD 1 n/a 28 AT8 – S1 A4, A5, A7 Albayram et al. [1]** RIA 1 n/a 14 Cisptau – S1 A1, A7 Logsdon et al. [43] Blast 1 n/a 3 T22 – S1 A5,A8 Niziolek et al. [50] WD 1 n/a 30 P(S262) – S2 A4 *See Table 1 for definitions. **Used both single and repetitive TBI. EDPI indicates the earliest day post injury at which p-Tau was found after TBI. If more than one p-Tau Ab was used only the antibody used to depict the main results is indicated. ITI, inter-injury-interval (hours), n/a, not applicable; PD, piston driven; RIA, rotational impact acceleration; WD, weight drop. Four studies reporting p-Tau exclusively in the optic tract and cerebellum are excluded from this table. P, S, and A refer to pathognomonic, supportive, and associated pathology (see Table 1 for details). Dashes indicate histopathological feature not reported or not found. Studies not showing p-Tau in the cerebral cortex or hippocampus are omitted from this table time of injury was 12  weeks (range 5 to 48  weeks). summarizes key characteristics of the 25 TBI models The most commonly used models were weight drop (15 repetitive TBI and 10 single TBI) that were associ- (n = 51), piston driven (n = 42), blast injury (n = 12), ated with cortical or hippocampal p-Tau, as assessed by and rotational impact acceleration (n = 9) models . immunohistochemistry. Additional file  1: Figure S7b shows the primary meth- odologies used for p-Tau detection and the frequency Presence of CTE‑like pathology in mouse models reporting of p-Tau detection. Of the included studies, 40 (68%) tau pathology reported p-Tau pathology and 25 (44%) described With respect to the presence of CTE-like pathol- p-Tau pathology by immunostaining (Additional file  1: ogy, only 3 studies (5.2% of included and 0.1% of all Figure S7c-d). The most commonly used antibody to screened papers) reported perivascular p-Tau, a defin - detect p-Tau was AT8 (8/25 [32%] of studies). Of the 25 ing feature of human CTE. One of these studies spe- studies that observed tau pathology, 22 found p-Tau in cifically mentioned the presence of both neuronal and the cerebral cortex, hippocampus, or subcortical struc- astroglial p-Tau (Additional file  1: Figure S7c). One tures and 3 studies reported the presence of p-Tau in additional study reported concurrent p-Tau pathology the optic tract (n = 2) and cerebellum (n = 1). Table  2 in both astroglia and neurons (located in the cortex and K ahriman et al. acta neuropathol commun (2021) 9:118 Page 13 of 17 hippocampus) but did not report on potential perivas- as well as in the adjacent corpus callosum. This is con - cular location. sistent with computational models of the biomechani- In terms of non-p-Tau related supportive histopatho- cal forces predicting that cerebral areas with a change in logical CTE-like features, 3 studies reported the presence morphology (such as at the site of sulci and the superior of cytoplasmatic localization of TDP-43 pathology (none longitudinal fissure) represent locations of high stress specifically commented on pTDP-43) (Additional file  1: and strain and thus greatest susceptibility to axonal and Figure S7c). vascular injury [12, 16, 23]. Forty-six (79%) of the included studies reported non- Given the perivascular location of many neurodegen- specific CTE-associated pathologies (Additional file  1: erative features it has been suggested that TBI triggers Figure S7d). Most commonly, these included neuronal the neurodegenerative cascade by damaging the neuro- loss (23%), axonal loss (43%), micro- and astrogliosis vascular unit. Indeed, consistent with human data [28, (50%, each), and beta amyloid deposition (29%) (Addi- 39], we observed microvascular injury in our model, par- tional file  1: Figure S7c). Interestingly, no study reported ticularly in locations with p-Tau accumulation. Moreover, cerebral microbleeds. using IgG staining, we found evidence for BBB-disrup- tion adding to the notion that BBB hyperpermeability is Discussion an important aspect of murine TBI-associated tauopa- Pathological hyperphosphorylation and aggregation thy. Nevertheless, IgG staining was faint and restricted of Tau protein is observed in a wide range of neurode- to the optic tract. Injury to the visual pathway including generative disorders and is the key defining feature of BBB-disruption and p-Tau accumulation in the optic a heterogeneous class of diseases called tauopathies. tract has been shown after murine closed head injury [3, TBI has been identified as a strong risk factor for many 13, 14, 19]. Yet, we did not observe previously reported of these tauopathies, highlighting the potential lifelong perivascular IgG staining in the cerebral cortex [69]. This consequences of TBI exposure. Recently, CTE has been is inconsistent with the observed microvascular injury described as the prototypical TBI-associated tauopathy. as assessed by Prussian blue staining in our study. In Its diagnosis presently rests with the unique distribution this respect it is noteworthy that we used the cyclooxy- of tau pathologies on a macroscopic and cellular level, yet genase-2 (COX-2) inhibitor carprofen as a post-opera- little is known to what extent these pathologies are rep- tive analgesic. Carprofen has been shown to attenuate licated by mouse closed-head TBI. To close this knowl- microglial activation, inflammation, and brain edema edge gap we described the pertinent histopathological formation at therapeutic doses in mice, which may have features of human TBI-associated tauopathy in our rTBI attenuated BBB-disruption in our model [72]. Thus, while model as well as by conducting a systematic review of the IgG-staining provided proof-of-principle that BBB-integ- literature. rity was impaired in our model, further studies using Our concussive mouse rTBI model showed p-Tau different BBB-integrity markers as well as avoidance of accumulation within the corpus callosum of the trau- anti-inflammatory drugs is required to better character - matized hemisphere as early as 1  week after injury with ize the extent of BBB permeability and its relationship to subsequent involvement of the cerebral cortex, mimick- p-Tau and other observed pathologies. ing the pathological tau progression described in human In addition to pathological p-Tau accumulation, sev- disease. Importantly, we found p-Tau in multiple cer- eral other proteinopathies have been described following ebral locations that are frequently involved in human human TBI. In particular, widespread TDP-43 inclusions disease and are included in the CTE consensus crite- in the neocortex of patients with CTE have been reported ria. These included, superficial cortical layers, the hip - and the presence of TDP-43 pathology has been used pocampus, periventricular tissues, mamillary bodies, and to support the diagnosis of CTE [47, 49]. Nevertheless, perivascular locations [8, 46]. Nevertheless, we defined TDP-43-related pathology is not specific to CTE. It has vessels based on morphology only and future studies been described in a range of conditions, and colocaliza- should include staining for vascular markers to deter- tion of tau and TDP-43 is often limited. For this reason, it mine the specific association of p-Tau with the vascular has yet to be determined exactly how TDP-43 aggregates compartment. coincide and interact with pathological p-Tau accumula- In this regard, it should be noted that human CTE crite- tion. Overall, few mouse studies have sought to evaluate ria include the location of perivascular p-Tau at the depth TDP-43 pathology after closed-head TBI [1, 13, 73, 78]. of cerebral sulci [8, 46]. Because mice are lisencephalic, Here, we found widespread TDP-43 expression following this criterion cannot be fulfilled in a murine model. How - rTBI with a persistent presence of cytoplasmatic mislo- ever, we consistently observed p-Tau within the subpial calization by 6 months post rTBI. This is consistent with cerebral cortex lining the superficial longitudinal fissure, prior studies reporting persistent TDP-43 pathology in Kahriman et al. acta neuropathol commun (2021) 9:118 Page 14 of 17 murine TBI associated with tauopathy [1, 13, 78]. We screened studies (58/3,756) sought to assess pathologi- found long-term accumulation of pTDP-43 expression cal tau accumulation. Nevertheless, of these studies, with associated nuclear loss and cytoplasmatic aggre- approximately 40% found evidence of p-Tau pathology gation up to 24  weeks after rTBI. This is an important by immunohistochemistry. Despite the overall scarcity extension of previous studies that were limited to shorter of investigations, these and our observations provide (1  week) observation periods [55, 70, 73] and in light of cumulative evidence that murine closed head TBI rep- reported transient alterations [55, 73]. Our observation licates the critical histopathological aspects of human that pathological TDP-43, but not tau, accumulation was TBI-associated tau pathology with specific features of present in the hippocampus adds to the notion that while CTE. Reflecting human disease, pathology was seen these proteinopathies share a common pathophysiology, across a wide variety of TBI-paradigms, highlight- there may exist cell-specific susceptibilities [1, 68]. ing that tau pathology after murine closed head TBI is In contrast to pathological tau and TDP-43 accumu- reproducible and robust. This indicates that model dif - lation after TBI, persistent α-synuclein and β-amyloid ferences could be leveraged to study the association of proteinopathy represent less consistent histopathologi- pathophysiological mechanisms with varying mechan- cal features after TBI [7, 8, 22, 38, 67]. Although several ics of injury [9] as well as the ability to control for pos- prior animal studies reported increased β-amyloid and sible confounders. For example, similar to other studies α-synuclein after TBI, we did not find this present in our reporting p-Tau accumulation after murine closed-head rTBI model [1, 13, 56, 62, 75, 78]. rTBI [56, 62, 75, 78] we exposed the skull for precise Mixed neuronal and astroglial tau pathology is consid- impact delivery to the same coordinates across animals. ered a hallmark of CTE; as such, it is important to evalu- Yet, this is inconsistent with the clinical situation, and ate the precise cell types involved with TBI-associated our systematic review showed that this approach is not tauopathy. For example, astrocyte activation is common critical for inducing pertinent CTE-like neuropatholog- after TBI, accompanies virtually all neurodegenerative ical features. tauopathies, and astrocytes may serve as a source for tau Finally, though accumulation of p-Tau is a critical and thus could conceivably contribute to pathological early event in the cascade leading to tauopathy, forma- tau accumulation [34, 41, 65, 69]. Yet, whether astroglial tion of small, soluble oligomeric tau species as well as tau expression serves as a driver for injury-associated the aggregation into larger insoluble filaments known as tauopathy remains uncertain [41]. Many groups hypoth- neurofibrillary tangles (NFTs), represent the hallmark esize that astrocytes may promote neurodegeneration of tauopathies including CTE [5, 30, 33]. There is strong because astroglial tau pathology has been observed in the evidence that pathogenic tau, sometimes referred to as absence of neuronal tau inclusions, possibly through pro- a tau prion or prion-like tau, self-templates to progres- inflammatory mechanisms [41]. We observed significant sively spread disease in tauopathy patients. In a sub-set neuronal loss by 4  weeks after rTBI, which coincided of rTBI patients, the repeated injury gives rise to the with significant astroglial and microglial activation. How - formation of tau prions, which include a toxic spe- ever, we noted that neuronal p-Tau expression occurred cies responsible for neuronal death. However, further prior to significant microglial and astroglial activation. studies are needed to determine if murine closed-head Moreover, neuronal p-Tau expression was present earlier models are able to replicate the formation of tau prions and progressed more widespread than astroglial p-Tau in mice [24, 25, 33]. expression. Lastly, p-Tau-stained astrocytes were located In addition to ongoing efforts to better understand the in the superficial cortex, which had the greatest burden effects of TBI in wild-type mice, several groups have used of neuronal p-Tau, but there were no p-Tau expressing transgenic mouse models expressing human tau to inves- astrocytes in deeper cortical layers and the corpus callo- tigate the link between traumatic injury, tau misfolding, sum. Together, these observations are consistent with the and p-Tau pathology. For example, the rTg4510 mouse hypothesis that astroglial p-Tau expression after closed- model, which expresses a doxycycline-repressible isoform head TBI is a secondary event rather than primary driver of tau containing the P301L mutation [59] developed of tau pathology, possibly related to internalization of elevated p-Tau levels in the cortex 7  days after either a p-Tau into astrocytes from neighboring neurons and syn- single or double closed-head injury [6]. Investigating the apses [41]. longer-term effects of TBI on tau aggregation in Tg mice, To put our results in context with ongoing work in others have used the PS19 mouse model [77], which the field, we conducted a systematic review of the lit - expresses human tau with the P301S mutation, and per- erature. Our analysis showed a striking paucity of formed a semi-quantitative analysis of p-Tau pathology studies that sought to determine tau pathology in up to 7 months post-injury [71]. While these studies have experimental closed-head mouse TBI. Fewer than 2% of helped address several questions about the link between K ahriman et al. acta neuropathol commun (2021) 9:118 Page 15 of 17 References TBI and tau spreading, there are also important caveats 1. Albayram O, Kondo A, Mannix R, Smith C, Tsai CY, Li C, Herbert MK, Qiu to consider with regard to these mouse models. A care- J, Monuteaux M, Driver J et al (2017) Cis P-tau is induced in clinical and ful analysis of the transgene insertion sites in the rTg4510 preclinical brain injury and contributes to post-injury sequelae. Nat Com- mun 8:1000. https:// doi. org/ 10. 1038/ s41467- 017- 01068-4 model revealed that disruption of essential genes is at 2. Albayram O, MacIver B, Mathai J, Verstegen A, Baxley S, Qiu C, Bell C, least partially responsible for the p-Tau pathology that Caldarone BJ, Zhou XZ, Lu KP et al (2019) Traumatic Brain Injury-related results from overexpressing the human protein [21]. And voiding dysfunction in mice is caused by damage to rostral pathways, altering inputs to the reflex pathways. Sci Rep 9:8646. https:// doi. org/ 10. extensive variability in the PS19 model, with regard to 1038/ s41598- 019- 45234-8 both the rate of tau spreading and p-Tau formation [74], 3. Angoa-Perez M, Zagorac B, Anneken JH, Briggs DI, Winters AD, Greenberg limits the interpretations that can be drawn from studies JM, Ahmad M, Theis KR, Kuhn DM (2020) Repetitive, mild traumatic brain injury results in a progressive white matter pathology, cognitive deterio- using the mouse line. ration, and a transient gut microbiota dysbiosis. Sci Rep 10:8949. https:// In conclusion, our study highlights the translational doi. org/ 10. 1038/ s41598- 020- 65972-4 value of murine closed skull TBI to replicate the perti- 4. Anonymous (1999) Consensus conference. Rehabilitation of persons with traumatic brain injury. NIH consensus development panel on rehabilita- nent aspects of human TBI-associated tauopathy with a tion of persons with traumatic brain injury. JAMA 282: 974-983 wide range of related histopathological features. 5. Arena JD, Smith DH, Lee EB, Gibbons GS, Irwin DJ, Robinson JL, Lee VM, Trojanowski JQ, Stewart W, Johnson VE (2020) Tau immunophenotypes in chronic traumatic encephalopathy recapitulate those of ageing and Supplementary Information Alzheimer’s disease. Brain 143:1572–1587. https:// doi. org/ 10. 1093/ brain/ The online version contains supplementary material available at https:// doi. awaa0 71 org/ 10. 1186/ s40478- 021- 01220-8. 6. Bachstetter AD, Morganti JM, Bodnar CN, Webster SJ, Higgins EK, Roberts KN, Snider H, Meier SE, Nation GK, Goulding DS et al (2020) The effects of mild closed head injuries on tauopathy and cognitive deficits in rodents: Additional file 1. Supplementary information. Primary results in wild type and rTg4510 mice, and a systematic review. Exp Neurol 326:113180. https:// doi. org/ 10. 1016/j. expne urol. 2020. 113180 7. Baugh CM, Stamm JM, Riley DO, Gavett BE, Shenton ME, Lin A, Authors’ contributions Nowinski CJ, Cantu RC, McKee AC, Stern RA (2012) Chronic traumatic All authors contributed to the study conception and design. A.K. conducted encephalopathy: neurodegeneration following repetitive concussive histological preparations, histological analysis, data analyses. J.B. conducted and subconcussive brain trauma. Brain Imaging Behav 6:244–254. animal surgery, behavioral testing, histological preparations, and genotyping. https:// doi. org/ 10. 1007/ s11682- 012- 9164-5 N.H. conceived and designed the study, conducted data analyses, animal 8. Bieniek KF, Cairns NJ, Crary JF, Dickson DW, Folkerth RD, Keene CD, surgery, behavioral testing, histological preparations, and wrote the paper. All Litvan I, Perl DP, Stein TD, Vonsattel JP et al (2021) The second NINDS/ authors discussed the results, commented on the manuscript for important NIBIB consensus meeting to define neuropathological criteria for the intellectual content, read, and approved the final manuscript. diagnosis of chronic traumatic encephalopathy. J Neuropathol Exp Neurol 80:210–219. https:// doi. org/ 10. 1093/ jnen/ nlab0 01 Funding 9. Bodnar CN, Roberts KN, Higgins EK, Bachstetter AD (2019) A systematic This study was supported by institutional Grants, National Institutes of Health review of closed head injury models of mild traumatic brain injury in Grant K08NS091499 to N.H. mice and rats. J Neurotrauma 36:1683–1706. https:// doi. org/ 10. 1089/ neu. 2018. 6127 Data availability 10. Bolton-Hall AN, Hubbard WB, Saatman KE (2019) Experimental designs Data supporting the findings of this study are available from the correspond- for repeated mild traumatic brain injury: challenges and considera- ing author upon reasonable request. tions. J Neurotrauma 36:1203–1221. https:// doi. org/ 10. 1089/ neu. 2018. Declarations 11. Bouley J, Chung DY, Ayata C, Brown RH Jr, Henninger N (2019) Corti- cal spreading depression denotes concussion injury. J Neurotrauma Ethics approval 36:1008–1017. https:// doi. org/ 10. 1089/ neu. 2018. 5844 All animal procedures were approved by the University of Massachusetts 12. Bradshaw DR, Ivarsson J, Morfey CL, Viano DC (2001) Simulation of Medical School Institutional Animal Care and Use Committee. This manuscript acute subdural hematoma and diffuse axonal injury in coronal head was prepared in adherence to the ARRIVE and PRISMA guidelines (Additional impact. J Biomech 34:85–94. https:// doi. org/ 10. 1016/ s0021- 9290(00) file 1: Supplementary information). 00135-4 13. Briggs DI, Angoa-Perez M, Kuhn DM (2016) Prolonged repetitive head Competing interests trauma induces a singular chronic traumatic encephalopathy-like pathol- The authors declare no competing financial interests. ogy in white matter despite transient behavioral abnormalities. Am J Pathol 186:2869–2886. https:// doi. org/ 10. 1016/j. ajpath. 2016. 07. 013 Author details 14. Cansler SM, Guilhaume-Correa F, Day D, Bedolla A, Evanson NK Indirect Department of Neurology, Medical School, University of Massachusetts, 55 traumatic optic neuropathy after head trauma in adolescent male mice Lake Ave, Worcester, USA. Department of Pathology, Medical School, Univer- is associated with behavioral visual impairment, neurodegeneration, and sity of Massachusetts, 55 Lake Ave, Worcester, USA. Department of Biology, elevated endoplasmic reticulum stress markers at acute and subacute University of Massachusetts Amherst, Amherst, MA 01003, USA. Depar t- times. BioRxiv [Preprint] June 12, 2020 (Cited 06/01/2021). https:// doi. org/ ment of Psychiatry, Medical School, University of Massachusetts, 55 Lake Ave, 10. 1101/ 2020. 06. 11. 144766 Worcester, USA. 15. Cheng WH, Martens KM, Bashir A, Cheung H, Stukas S, Gibbs E, Namjoshi DR, Button EB, Wilkinson A, Barron CJ et al (2019) CHIMERA repetitive mild Received: 11 May 2021 Accepted: 18 June 2021 traumatic brain injury induces chronic behavioural and neuropathologi- cal phenotypes in wild-type and APP/PS1 mice. Alzheimers Res Ther 11:6. https:// doi. org/ 10. 1186/ s13195- 018- 0461-0 16. Cloots RJ, Gervaise HM, van Dommelen JA, Geers MG (2008) Bio- mechanics of traumatic brain injury: influences of the morphologic Kahriman et al. acta neuropathol commun (2021) 9:118 Page 16 of 17 heterogeneities of the cerebral cortex. Ann Biomed Eng 36:1203–1215. 35. Huber BR, Meabon JS, Martin TJ, Mourad PD, Bennett R, Kraemer BC, Cer- https:// doi. org/ 10. 1007/ s10439- 008- 9510-3 nak I, Petrie EC, Emery MJ, Swenson ER et al (2013) Blast exposure causes 17. Crane PK, Gibbons LE, Dams-O’Connor K, Trittschuh E, Leverenz JB, Keene early and persistent aberrant phospho- and cleaved-tau expression in a CD, Sonnen J, Montine TJ, Bennett DA, Leurgans S et al (2016) Association murine model of mild blast-induced traumatic brain injury. J Alzheimers of traumatic brain injury with late-life neurodegenerative conditions and Dis 37:309–323. https:// doi. org/ 10. 3233/ JAD- 130182 neuropathologic findings. JAMA Neurol 73:1062–1069. https:// doi. org/ 10. 36. Hyder AA, Wunderlich CA, Puvanachandra P, Gururaj G, Kobusingye 1001/ jaman eurol. 2016. 1948 OC (2007) The impact of traumatic brain injuries: a global perspective. 18. DeWitt DS, Hawkins BE, Dixon CE, Kochanek PM, Armstead W, Bass CR, NeuroRehabilitation 22:341–353 Bramlett HM, Buki A, Dietrich WD, Ferguson AR et al (2018) Pre-clinical 37. Iliff JJ, Chen MJ, Plog BA, Zeppenfeld DM, Soltero M, Yang L, Singh I, testing of therapies for traumatic brain injury. J Neurotrauma 35:2737– Deane R, Nedergaard M (2014) Impairment of glymphatic pathway 2754. https:// doi. org/ 10. 1089/ neu. 2018. 5778 function promotes tau pathology after traumatic brain injury. J Neurosci 19. Evanson NK, Guilhaume-Correa F, Herman JP, Goodman MD (2018) Optic 34:16180–16193. https:// doi. org/ 10. 1523/ JNEUR OSCI. 3020- 14. 2014 tract injury after closed head traumatic brain injury in mice: A model of 38. Iverson GL, Gardner AJ, McCrory P, Zafonte R, Castellani RJ (2015) A critical indirect traumatic optic neuropathy. PLoS ONE 13:e0197346. https:// doi. review of chronic traumatic encephalopathy. Neurosci Biobehav Rev org/ 10. 1371/ journ al. pone. 01973 46 56:276–293. https:// doi. org/ 10. 1016/j. neubi orev. 2015. 05. 008 20. Fleminger S, Ponsford J (2005) Long term outcome after traumatic brain 39. Kenney K, Amyot F, Haber M, Pronger A, Bogoslovsky T, Moore C, Diaz- injury. BMJ 331:1419–1420. https:// doi. org/ 10. 1136/ bmj. 331. 7530. 1419 Arrastia R (2016) Cerebral vascular injury in traumatic brain injury. Exp 21. Gamache J, Benzow K, Forster C, Kemper L, Hlynialuk C, Furrow E, Ashe Neurol 275(Pt 3):353–366. https:// doi. org/ 10. 1016/j. expne urol. 2015. 05. KH, Koob MD (2019) Factors other than hTau overexpression that contrib- 019 ute to tauopathy-like phenotype in rTg4510 mice. Nat Commun 10:2479. 40. Kondo A, Shahpasand K, Mannix R, Qiu J, Moncaster J, Chen CH, Yao https:// doi. org/ 10. 1038/ s41467- 019- 10428-1 Y, Lin YM, Driver JA, Sun Y et al (2015) Antibody against early driver of 22. Gavett BE, Stern RA, McKee AC (2011) Chronic traumatic encephalopathy: neurodegeneration cis P-tau blocks brain injury and tauopathy. Nature a potential late effect of sport-related concussive and subconcussive 523:431–436. https:// doi. org/ 10. 1038/ natur e14658 head trauma. Clin Sports Med 30:179–188. https:// doi. org/ 10. 1016/j. csm. 41. Kovacs GG (2020) Astroglia and Tau: new perspectives. Front Aging 2010. 09. 007 Neurosci 12:96. https:// doi. org/ 10. 3389/ fnagi. 2020. 00096 23. Ghajari M, Hellyer PJ, Sharp DJ (2017) Computational modelling of trau- 42. Liu B, Cao Y, Shi F, Wang L, Li N, Cheng X, Du J, Tian Q, Zhou X (2020) The matic brain injury predicts the location of chronic traumatic encepha- overexpression of RBM3 alleviates TBI-induced behaviour impairment lopathy pathology. Brain 140:333–343. https:// doi. org/ 10. 1093/ brain/ and AD-like tauopathy in mice. J Cell Mol Med 24:9176–9188. https:// doi. aww317org/ 10. 1111/ jcmm. 15555 24. Goedert M, Eisenberg DS, Crowther RA (2017) Propagation of tau aggre- 43. Logsdon AF, Lucke-Wold BP, Turner RC, Li X, Adkins CE, Mohammad AS, gates and neurodegeneration. Annu Rev Neurosci 40:189–210. https:// Huber JD, Rosen CL, Lockman PR (2017) A mouse model of focal vascular doi. org/ 10. 1146/ annur ev- neuro- 072116- 031153 injury induces astrocyte reactivity, tau oligomers, and aberrant behavior. 25. Goedert M, Spillantini MG (2017) Propagation of tau aggregates. Mol. Arch Neurosci. https:// doi. org/ 10. 5812/ archn euros ci. 44254 Brain 10:18. https:// doi. org/ 10. 1186/ s13041- 017- 0298-7 44. Luo J, Nguyen A, Villeda S, Zhang H, Ding Z, Lindsey D, Bieri G, Castellano 26. Goldstein LE, Fisher AM, Tagge CA, Zhang XL, Velisek L, Sullivan JA, Upreti JM, Beaupre GS, Wyss-Coray T (2014) Long-term cognitive impairments C, Kracht JM, Ericsson M, Wojnarowicz MW et al (2012) Chronic traumatic and pathological alterations in a mouse model of repetitive mild trau- encephalopathy in blast-exposed military veterans and a blast neuro- matic brain injury. Front Neurol 5:12. https:// doi. org/ 10. 3389/ fneur. 2014. trauma mouse model. Sci Transl Med 4:134–160. https:// doi. org/ 10. 1126/ 00012 scitr anslm ed. 30037 16 45. McKee AC, Abdolmohammadi B, Stein TD (2018) The neuropathology 27. Graham NS, Sharp DJ (2019) Understanding neurodegeneration after of chronic traumatic encephalopathy. Handb Clin Neurol 158:297–307. traumatic brain injury: from mechanisms to clinical trials in dementia. https:// doi. org/ 10. 1016/ B978-0- 444- 63954-7. 00028-8 J Neurol Neurosurg Psychiatry 90:1221–1233. https:// doi. org/ 10. 1136/ 46. McKee AC, Cairns NJ, Dickson DW, Folkerth RD, Keene CD, Litvan I, Perl DP, jnnp- 2017- 317557 Stein TD, Vonsattel JP, Stewart W et al (2016) The first NINDS/NIBIB con- 28. Griffin AD, Turtzo LC, Parikh GY, Tolpygo A, Lodato Z, Moses AD, Nair G, sensus meeting to define neuropathological criteria for the diagnosis of Perl DP, Edwards NA, Dardzinski BJ et al (2019) Traumatic microbleeds chronic traumatic encephalopathy. Acta Neuropathol 131:75–86. https:// suggest vascular injury and predict disability in traumatic brain injury. doi. org/ 10. 1007/ s00401- 015- 1515-z Brain 142:3550–3564. https:// doi. org/ 10. 1093/ brain/ awz290 47. McKee AC, Gavett BE, Stern RA, Nowinski CJ, Cantu RC, Kowall NW, Perl 29. Guo Z, Cupples LA, Kurz A, Auerbach SH, Volicer L, Chui H, Green RC, DP, Hedley-Whyte ET, Price B, Sullivan C et al (2010) TDP-43 proteinopa- Sadovnick AD, Duara R, DeCarli C et al (2000) Head injury and the risk of thy and motor neuron disease in chronic traumatic encephalopathy. J AD in the MIRAGE study. Neurology 54:1316–1323. https:// doi. org/ 10. Neuropathol Exp Neurol 69:918–929. https:// doi. org/ 10. 1097/ NEN. 0b013 1212/ wnl. 54.6. 1316e3181 ee7d85 30. Gyparaki MT, Arab A, Sorokina EM, Santiago-Ruiz AN, Bohrer CH, Xiao J, 48. McKee AC, Robinson ME (2014) Military-related traumatic brain injury and Lakadamyali M (2021) Tau forms oligomeric complexes on microtubules neurodegeneration. Alzheimers Dement 10:S242-253. https:// doi. org/ 10. that are distinct from tau aggregates. Proc Natl Acad Sci U S A. https:// 1016/j. jalz. 2014. 04. 003 doi. org/ 10. 1073/ pnas. 20214 61118 49. Mez J, Daneshvar DH, Kiernan PT, Abdolmohammadi B, Alvarez VE, Huber 31. Henninger N, Bouley J, Sikoglu EM, An J, Moore CM, King JA, Bowser R, BR, Alosco ML, Solomon TM, Nowinski CJ, McHale L et al (2017) Clinico- Freeman MR, Brown RH Jr (2016) Attenuated traumatic axonal injury and pathological evaluation of chronic traumatic encephalopathy in players improved functional outcome after traumatic brain injury in mice lacking of american football. JAMA 318:360–370. https:// doi. org/ 10. 1001/ jama. Sarm1. Brain 139:1094–1105. https:// doi. org/ 10. 1093/ brain/ aww0012017. 8334 32. Hernandez F, Merchan-Rubira J, Valles-Saiz L, Rodriguez-Matellan A, Avila 50. Niziolek GM, Boudreau RM, Baker J, Friend LA, Makley AT, Edwards MJ, J (2020) Differences between human and murine tau at the N-terminal Gulbins E, Goodman MD (2020) Acid sphingomyelinase inhibition end. Front Aging Neurosci 12:11. https:// doi. org/ 10. 3389/ fnagi. 2020. mitigates histopathological and behavioral changes in a murine model of 00011 traumatic brain injury. J Neurotrauma 37:1902–1909. https:// doi. org/ 10. 33. Holtzman DM, Carrillo MC, Hendrix JA, Bain LJ, Catafau AM, Gault LM, 1089/ neu. 2019. 6436 Goedert M, Mandelkow E, Mandelkow EM, Miller DS et al (2016) Tau: From 51. Niziolek GM, Hoehn RS, Seitz AP, Jernigan PL, Makley AT, Gulbins E, research to clinical development. Alzheimers Dement 12:1033–1039. Edwards MJ, Goodman MD (2021) The role of acid sphingomyelinase https:// doi. org/ 10. 1016/j. jalz. 2016. 03. 018 inhibition in repetitive mild traumatic brain injury. J Surg Res 259:296– 34. Hsu ET, Gangolli M, Su S, Holleran L, Stein TD, Alvarez VE, McKee AC, 304. https:// doi. org/ 10. 1016/j. jss. 2020. 09. 034 Schmidt RE, Brody DL (2018) Astrocytic degeneration in chronic trau- 52. Paxinos G, Franklin KBJ (2013) The mouse brain in stereotaxic coordinates. matic encephalopathy. Acta Neuropathol 136:955–972. https:// doi. org/ Elsevier Academic Press, Cambridge 10. 1007/ s00401- 018- 1902-3 K ahriman et al. acta neuropathol commun (2021) 9:118 Page 17 of 17 53. Peeters W, van den Brande R, Polinder S, Brazinova A, Steyerberg EW, 67. Smith DH, Johnson VE, Stewart W (2013) Chronic neuropathologies Lingsma HF, Maas AI (2015) Epidemiology of traumatic brain injury in of single and repetitive TBI: substrates of dementia? Nat Rev Neurol Europe. Acta Neurochir ( Wien) 157:1683–1696. https:// doi. org/ 10. 1007/ 9:211–221. https:// doi. org/ 10. 1038/ nrneu rol. 2013. 29 s00701- 015- 2512-7 68. Smith VD, Bachstetter AD, Ighodaro E, Roberts K, Abner EL, Fardo DW, 54. Petraglia AL, Plog BA, Dayawansa S, Dashnaw ML, Czerniecka K, Walker Nelson PT (2018) Overlapping but distinct TDP-43 and tau pathologic CT, Chen M, Hyrien O, Iliff JJ, Deane R et al (2014) The pathophysiology patterns in aged hippocampi. Brain Pathol 28:264–273. https:// doi. org/ underlying repetitive mild traumatic brain injury in a novel mouse model 10. 1111/ bpa. 12505 of chronic traumatic encephalopathy. Surg Neurol Int 5:184. https:// doi. 69. Tagge CA, Fisher AM, Minaeva OV, Gaudreau-Balderrama A, Moncaster org/ 10. 4103/ 2152- 7806. 147566 JA, Zhang XL, Wojnarowicz MW, Casey N, Lu H, Kokiko-Cochran ON et al 55. Rajic Bumber J, Pilipovic K, Jankovic T, Dolenec P, Grzeta N, Kriz J, Zupan G (2018) Concussion, microvascular injury, and early tauopathy in young (2021) Repetitive traumatic brain injury is associated with TDP-43 altera- athletes after impact head injury and an impact concussion mouse tions, neurodegeneration, and glial activation in mice. J Neuropathol Exp model. Brain 141:422–458. https:// doi. org/ 10. 1093/ brain/ awx350 Neurol 80:2–14. https:// doi. org/ 10. 1093/ jnen/ nlaa1 30 70. Tan XL, Sun M, Brady RD, Liu S, Llanos R, Cheung S, Wright DK, Casillas- 56. Rehman SU, Ahmad A, Yoon GH, Khan M, Abid MN, Kim MO (2018) Espinosa PM, Sashindranath M, O’Brien TJ et al (2019) Transactive Inhibition of c-Jun N-terminal kinase protects against brain damage and Response DNA-Binding Protein 43 Abnormalities after Traumatic Brain improves learning and memory after traumatic brain injury in adult mice. Injury. J Neurotrauma. 36:87–99. https:// doi. org/ 10. 1089/ neu. 2017. 5491 Cereb Cortex 28:2854–2872. https:// doi. org/ 10. 1093/ cercor/ bhx164 71. Tang SJ, Fesharaki-Zadeh A, Takahashi H, Nies SH, Smith LM, Luo A, 57. Sabbagh JJ, Fontaine SN, Shelton LB, Blair LJ, Hunt JB Jr, Zhang B, Gut- Chyung A, Chiasseu M, Strittmatter SM (2020) Fyn kinase inhibition mann JM, Lee DC, Lloyd JD, Dickey CA (2016) Noncontact rotational head reduces protein aggregation, increases synapse density and improves injury produces transient cognitive deficits but lasting neuropathological memory in transgenic and traumatic Tauopathy. Acta Neuropathol Com- changes. J Neurotrauma 33:1751–1760. https:// doi. org/ 10. 1089/ neu. mun 8:96. https:// doi. org/ 10. 1186/ s40478- 020- 00976-9 2015. 4288 72. Thau-Zuchman O, Shohami E, Alexandrovich AG, Trembovler V, Leker RR 58. Sacramento CB, Sondhi D, Rosenberg JB, Chen A, Giordano S, Pey E, Lee (2012) The anti-inflammatory drug carprofen improves long-term out - V, Stiles KM, Havlicek DF, Leopold PL et al (2020) Anti-phospho-tau gene come and induces gliogenesis after traumatic brain injury. J Neurotrauma therapy for chronic traumatic encephalopathy. Hum Gene Ther 31:57–69. 29:375–384. https:// doi. org/ 10. 1089/ neu. 2010. 1673 https:// doi. org/ 10. 1089/ hum. 2019. 174 73. Wiesner D, Tar L, Linkus B, Chandrasekar A, Olde Heuvel F, Dupuis L, Tsao 59. Santacruz K, Lewis J, Spires T, Paulson J, Kotilinek L, Ingelsson M, Guima- W, Wong PC, Ludolph A, Roselli F (2018) Reversible induction of TDP-43 raes A, DeTure M, Ramsden M, McGowan E et al (2005) Tau suppression in granules in cortical neurons after traumatic injury. Exp Neurol 299:15–25. a neurodegenerative mouse model improves memory function. Science https:// doi. org/ 10. 1016/j. expne urol. 2017. 09. 011 309:476–481. https:// doi. org/ 10. 1126/ scien ce. 11136 94 74. Woerman AL, Patel S, Kazmi SA, Oehler A, Freyman Y, Espiritu L, Cotter R, 60. Schneider CA, Rasband WS, Eliceiri KW (2012) NIH Image to ImageJ: 25 Castaneda JA, Olson SH, Prusiner SB (2017) Kinetics of human mutant tau years of image analysis. Nat Methods 9:671–675. https:// doi. org/ 10. 1038/ prion formation in the brains of 2 transgenic mouse lines. JAMA Neurol nmeth. 2089 74:1464–1472. https:// doi. org/ 10. 1001/ jaman eurol. 2017. 2822 61. Schneider J, Jeon S, Gladman JT, Corriveau RA ADRD Summit 2019 75. Xu X, Cowan M, Beraldo F, Schranz A, McCunn P, Geremia N, Brown Z, Report to the National Advisory Neurological Disorders and Stroke Patel M, Nygard KL, Khazaee R et al (2021) Repetitive mild traumatic brain Council. https:// www. ninds. nih. gov/ News- Events/ Events- Proce edings/ injury in mice triggers a slowly developing cascade of long-term and per- Events/ Alzhe imers- Disea se- Relat ed- Demen tias- Summit- 2019 accessed sistent behavioral deficits and pathological changes. Acta Neuropathol 05/13/2020 Commun 9:60. https:// doi. org/ 10. 1186/ s40478- 021- 01161-2 62. Selvaraj P, Wen J, Tanaka M, Zhang Y (2019) Therapeutic effect of a novel 76. Yang Z, Wang P, Morgan D, Lin D, Pan J, Lin F, Strang KH, Selig TM, Perez fatty acid amide hydrolase inhibitor PF04457845 in the repetitive closed PD, Febo M et al (2015) Temporal MRI characterization, neurobiochemi- head injury mouse model. J Neurotrauma 36:1655–1669. https:// doi. org/ cal and neurobehavioral changes in a mouse repetitive concussive head 10. 1089/ neu. 2018. 6226 injury model. Sci Rep 5:11178. https:// doi. org/ 10. 1038/ srep1 1178 63. Seo JS, Lee S, Shin JY, Hwang YJ, Cho H, Yoo SK, Kim Y, Lim S, Kim YK, 77. Yoshiyama Y, Higuchi M, Zhang B, Huang SM, Iwata N, Saido TC, Maeda Hwang EM et al (2017) Transcriptome analyses of chronic traumatic J, Suhara T, Trojanowski JQ, Lee VM (2007) Synapse loss and microglial encephalopathy show alterations in protein phosphatase expression activation precede tangles in a P301S tauopathy mouse model. Neuron associated with tauopathy. Exp Mol Med 49:e333. https:// doi. org/ 10. 53:337–351. https:// doi. org/ 10. 1016/j. neuron. 2007. 01. 010 1038/ emm. 2017. 56 78. Zhang J, Teng Z, Song Y, Hu M, Chen C (2015) Inhibition of monoacylglyc- 64. Shin N, Kim HG, Shin HJ, Kim S, Kwon HH, Baek H, Yi MH, Zhang E, Kim erol lipase prevents chronic traumatic encephalopathy-like neuropathol- JJ, Hong J et al (2019) Uncoupled endothelial nitric oxide synthase ogy in a mouse model of repetitive mild closed head injury. J Cereb enhances p-Tau in chronic traumatic encephalopathy mouse model. Blood Flow Metab 35:706. https:// doi. org/ 10. 1038/ jcbfm. 2014. 247 Antioxid Redox Signal 30:1601–1620. https:// doi. org/ 10. 1089/ ars. 2017. Publisher’s Note 65. Sidoryk-Wegrzynowicz M, Struzynska L (2019) Astroglial contribution to Springer Nature remains neutral with regard to jurisdictional claims in pub- tau-dependent neurodegeneration. Biochem J 476:3493–3504. https:// lished maps and institutional affiliations. doi. org/ 10. 1042/ BCJ20 190506 66. Smith DH, Dolle JP, Ameen-Ali KE, Bretzin A, Cortes E, Crary JF, Dams- O’Connor K, Diaz-Arrastia R, Edlow BL, Folkerth R et al (2021) Collaborative neuropathology network characterizing outcomes of TBI (CONNECT- TBI). Acta Neuropathol Commun 9:32. https:// doi. org/ 10. 1186/ s40478- 021- 01122-9

Journal

Acta Neuropathologica CommunicationsSpringer Journals

Published: Jun 29, 2021

Keywords: Animal model; Chronic traumatic encephalopathy; Concussion; Systematic review; Tauopathy; Traumatic brain injury

There are no references for this article.