Pomalidomide Improves Motor Behavioral Deficits and Protects Cerebral Cortex and Striatum Against Neurodegeneration Through a Reduction of Oxidative/Nitrosative Damages and Neuroinflammation After Traumatic Brain Injury

Abstract

Neuronal damage resulting from traumatic brain injury (TBI) causes disruption of neuronal projections and neurotransmission that contribute to behavioral deficits. Cellular generation of reactive oxygen species (ROS) and reactive nitrogen species (RNS) is an early event following TBI. ROS often damage DNA, lipids, proteins, and carbohydrates while RNS attack proteins. The products of lipid peroxidation 4-hydroxynonenal (4-HNE) and protein nitration 3-nitrotyrosine (3-NT) are often used as indicators of oxidative and nitrosative damages, respectively. Increasing evidence has shown that striatum is vulnerable to damage from TBI with a disturbed dopamine neurotransmission. TBI results in neurodegeneration, oxidative stress, neuroinflammation, neuronal apoptosis, and autophagy in the striatum and contribute to motor or behavioral deficits. Pomalidomide (Pom) is a Food and Drug Administration (FDA)-approved immunomodulatory drug clinically used in treating multiple myeloma. We previously showed that Pom reduces neuroinflammation and neuronal death induced by TBI in rat cerebral cortex. Here, we further compared the effects of Pom in cortex and striatum focusing on neurodegeneration, oxidative and nitrosative damages, as well as neuroinflammation following TBI. Sprague–Dawley rats subjected to a controlled cortical impact were used as the animal model of TBI. Systemic administration of Pom (0.5 mg/kg, intravenous [i.v.]) at 5 h post-injury alleviated motor behavioral deficits, contusion volume at 24 h after TBI. Pom alleviated TBI-induced neurodegeneration stained by Fluoro-Jade C in both cortex and striatum. Notably, Pom treatment reduces oxidative and nitrosative damages in cortex and striatum and is more efficacious in striatum (93% reduction in 4-HNE-positive and 84% reduction in 3-NT-positive neurons) than in cerebral cortex (42% reduction in 4-HNE-positive and 55% reduction in 3-NT-positive neurons). In addition, Pom attenuated microgliosis, astrogliosis, and elevations of proinflammatory cytokines in cortical and striatal tissue. We conclude that Pom may contribute to improved motor behavioral outcomes after TBI through targeting oxidative/nitrosative damages and neuroinflammation.

Introduction

Traumatic brain injury (TBI) resulting from an impact to the head may temporarily or permanently impair brain functions. TBI-associated brain damage includes two phases. Primary injury involves immediate diffuse axonal injury, intracranial bleeding, and immediate cell death ¹ , which is largely considered irreversible. Secondary injury involves pathophysiological cascades that are associated with neuronal degeneration leading to microglial and astrocyte reactivity which may amplify and prolong the primary injury and eventually lead to delayed neuronal death^2–5. Membrane disruption resulted from the mechanical injury during TBI causes widespread neuronal depolarization and massive release of glutamate, extravasation of blood components through a disrupted blood–brain barrier into the brain parenchyma and release of damage-associated molecular patterns (DAMPs, such as adenosine triphosphate [ATP], high mobility group box-1) from damaged cells ⁶ . Glutamate release induces further depolarization of still-viable neurons, Ca⁺⁺ influx, mitochondrial damage, and associated energy deficits would increase generation of ROS and RNS. DAMPs released from dead and damaged cells also initiate cytokine signaling cascades that orchestrate the inflammatory responses.

ROS and RNS, as well as multiple mediators and products of inflammation have been reported elevated in the brains of preclinical animal models and in humans following acute TBI. ROS, RNS, and inflammatory mediators can alter the activities of redox regulated enzymes and pathways and directly induce cytotoxic oxidative stress, initiating or exacerbating inflammation to alter and impair the function of neurons, glia and other brain cells, resulting in vicious cycles of inflammation, oxidative stress and cellular apoptosis and autophagy that drive secondary injury after TBI^5,7 Targeting on pathways leading to secondary brain injury would offer opportunities to develop novel treatments for acute TBI. These secondary processes provide opportunities for potential drug intervention^8,9.

TBI-induced neuronal damages in cortex and hippocampus are associated with deterioration of motor^5,10 and cognitive tasks^4,11. The vulnerability of the striatum to mild to moderate TBI is detailed in the literature^5,12–16. A number of preclinical papers indicate striatal tissue damage and abnormal dopaminergic transmission^13–17. In addition, several single photon emission computed tomography (SPECT) or positron emission tomography (PET) imaging studies in TBI patients also showed nigrostriatal dysfunction that may be associated with disruption of neuronal networks, despite relative structural preservation of the striatum^18–20.

The corticostriatal and nigrostriatal projections, particularly dopaminergic pathways, are critical for motor programming and coordination ²¹ . TBI results in altered dopamine (DA) neurotransmission^19,20, axonal degeneration ²² , oxidative stress, neuroinflammation ⁵ , and neuronal apoptosis ¹⁴ in the striatum. DA and dopaminergic signaling pathways are implicated in behavioral deficits following TBI^15,16,19,20. In addition, high levels of DA can be oxidized to form reactive DA quinones ²³ . Autoxidation of DA results in the formation of reactive oxygen species (ROS) ²⁴ . ROS include hydrogen peroxide (H₂O₂), superoxide (O₂•⁻), hydroxyl radical (•OH), and others containing unpaired electrons that may induce peroxidation of cellular structures, such as membrane lipids and proteins, and impairment of the mitochondrial electron transport chain ²⁵ . The brain is highly sensitive to ROS attack due to its high contents of polyunsaturated fatty acids resulting in lipid peroxidation, ²⁶ which often exacerbates neuronal damages, to result in the secondary injuries of TBI^25,27–30.

Brain levels of nitric oxide (NO) increase immediately after TBI and then again several hours–days later ³¹ . NO is the most common molecule of reactive nitrogen species (RNS) and can be formed intracellularly. Excessive production of highly reactive NO may also be responsible for the neurotoxicity, particularly in the presence of ROS. NO can react with superoxide anion (O2–•) to form peroxynitrite (ONOO–), the highly reactive oxidizing agent. Peroxynitrite-mediated oxidative damage has been previously demonstrated in a mouse model of TBI ³² . The lipid peroxidation resulted from ROS attack produce 4-hydroxynonenal (4-HNE), which can bind to cell membrane to impair their structural and functional integrities ³³ . The nitration of tyrosine residues at the 3-position in proteins to form 3-nitrotyrosine (3-NT) is mediated by RNS, such as peroxynitrite anion and nitrogen dioxide ³⁴ . Accordingly, the detection of 4-HNE and 3-NT accumulation thus provides us as an index of oxidative and nitrosative damages^33,34. Neurons are the most vulnerable cells for oxidative injury, while glial cells (mainly astrocytes and microglia) are much more resistant ³⁵ . The reasons for such differential susceptibility appear to be due to lower levels of antioxidant enzymes, including superoxide dismutase (SOD) and catalase ³⁶ , as well as glutathione ³⁷ , in neurons than in glial cells. The differences in the metabolic handling of iron and higher energy demands also contribute to higher vulnerability in neurons to oxidative injury than in glial cells.

Neuroinflammation is well accepted as a critical pathway that can beneficially recruit cells to initiate a reparative response following an injury but, on the other hand, when excessive or unregulated can detrimentally promote secondary damage in the pathogenesis of TBI. Tumor necrosis factor-α (TNF-α) is considered a master regulator of multifunctional proinflammatory cascades associated with central nervous system (CNS) injury and plays a vital role in initiation and propagation of inflammatory responses ³⁸ . Furthermore, there is considerable cross-talk between TNF-α, ROS, RNS, and neuroinflammation forming a vicious cycle. Therefore, oxidative stress also plays a pivotal part in neuroinflammation-involved secondary brain injury^25,30,39. Pomalidomide (Pom), a third-generation immunomodulatory drug and a TNF-α synthesis inhibitor, is a Food and Drug Administration (FDA)-approved thalidomide analogue with an additional amino group in the fourth carbon of the phthaloyl ring ⁴⁰ . Moreover, Pom, as an immune modulator clinically used in treating multiple myeloma ⁴¹ , has less adverse effects with neurotoxic, teratogenic, and anti-angiogenic properties than other thalidomide derivatives^42,43. We previously demonstrated that Pom reduces neuroinflammation, neuronal death induced by TBI in rat cerebral cortex^3,44. It has been shown that Pom significantly attenuated neuronal apoptosis in cortex by regulating Bax, Cytochrome c, and Poly (adenosine diphosphate [ADP]-ribose) polymerase ⁴⁵ . In addition to the high production of ROS resulting from DA auto-oxidation, the dopaminergic system within striatal and prefrontal cortex tissue is also exquisitely vulnerable to TBI and elevated TNF-α levels^46,47. Loss of tyrosine hydroxylase activity and DA release in the striatum of rats has been described following TBI ⁴⁸ and TBI-engendered striatal mitochondrial disruption leads to behavioral deficits ¹⁶ . Since oxidative injury is a major pathway driving neurodegeneration and neurons are more vulnerable than glial cells to oxidative injury, we further compared the effects of Pom on cortical and striatal tissue focusing on neurodegeneration, oxidative and nitrosative damages, and glia-mediated inflammation following TBI in this study.

Material and Methods

Pom Synthesis and Characterization

Pom synthesis and characterization were described in a previous study ³ . Briefly, Pom (4-amino-2-(2,6-dioxopiperidin-3-yl) isoindole-1,3-dione) was produced by a two-step synthetic procedure. Originally, 3-aminopiperidine-2,6-dione (CAS No. 24666-56-6) was condensed with 3-nitrophthalic anhydride (CAS No. 641-70-3) in refluxing acetic acid. Subsequent precipitation over ice water (0°C) provided nitro-thalidomide. Later hydrogenation by use of a palladium catalyst generated Pom as a bright yellow powder, whose appropriate structure was substantiated by chemical characterization and was verified with a purity of > 99.5%.

Animal Model of TBI by Controlled Cortical Impact

All animal experiments were performed in compliance with International Guidelines for Animal Research, and Institutional Animal Care and Use Committee (IACUC) of the affiliated university. Animals were housed in groups under 12 h light/dark cycle with the onset of light at 07:00 and offset of light at 19:00 and ad libitum access to pellet chow and water in a temperature-controlled (21–25°C) and humidity-controlled (45%–50%) room. Experiments were carried out during the daytime.

Male Sprague–Dawley rats (body weight: 250–300 g) were randomly divided into Sham and TBI groups with or without Pom treatment. We used a total of 24 rats with six animals per group for four groups based on a power analysis to determine “n size.” ⁴⁹ Sham group received a craniotomy without TBI under anesthesia. Previous studies indicated that there were no significant differences in behavioral and histological data between sham animals treated with either vehicle (Veh, 10% dimethyl sulfoxide [DMSO] in normal saline, i.v., n = 6) or Pom (0.5 mg/kg, i.v., n = 6) at 5 h post-injury^11,44. Therefore, data in Sham animals treated with Veh or Pom were combined and were designated as the Sham group (n = 12).

TBI group was subjected to TBI under anesthesia in a stereotaxic frame, using a controlled cortical impact (CCI) instrument with a rounded metal tip (5 mm in diameter) with velocity of 4 m/s and a deformation depth of 2 mm below as described previously^{3–5,10,11,50,51}. Such parameters of CCI resulted in moderate TBI with damages in cerebral cortex and stratum in rats^50,51. TBI animals further received either the Veh or Pom at 5 h post-injury and were designated the TBI + Veh or the TBI + Pom group, respectively. Throughout the surgery, body temperature of each animal was maintained at 37°C using a heating pad, and monitored with a rectal probe. No temperature changes were observed after administering Pom or vehicle in TBI or sham animals. Subsequent to the sham or TBI procedures, the rats were randomly allocated to Veh or Pom groups. The number of rats selected per group is detailed within each figure legend and is based on a power analysis ⁴⁹ . Behavioral testing and histological analyses were conducted by an experimenter blinded to the treatment conditions. All rats were sacrificed at 24 h after the surgery and behavior testing. Brain tissue was fixed in formalin and sectioned (10 µm) for further staining and morphological analysis.

Behavioral Testing of Neurological Outcomes

Behavioral evaluation was measured at 24 h post-TBI in the groups of sham, TBI + Veh, and TBI + Pom groups. The evaluations included a beam walk test for an evaluation of motor coordination and balance, an elevated body swing test (EBST) for motor asymmetry measurement, and a modified neurological severity score (mNSS) assessment that includes a composite of motor (muscle status and abnormal movement), sensory (visual, tactile, and proprioceptive), reflex, and balance tests. All procedures were completed as previously described, by blinded observers with some modification^3–5,52.

Contusion Volume Measured by Cresyl Violet Staining

Cresyl violet is a Nissl stain for evaluating neuronal cell body numbers and features and has become a standard histological stain for neurons. To measure the contusion volume resulted from TBI in the ipsilateral brain 24 h after TBI, cresyl violet-stained sections were digitalized and analyzed using a x1 objective and evaluated using ImageJ (National Institutes of Health, Bethesda, MD), as previously described^3,10,44,53. The volume was quantified by adding the injury regions and multiplying by the inter-slice distance (500 µm). Hemispheric tissue loss was expressed as a percentage that was calculated by the following formula ([contralateral hemispheric volume] × 100%) ⁵⁴ .

Fluoro-Jade C Staining of Degenerating Neurons

Fluoro-Jade C (FJC), as well as its predecessors, such as Fluoro-Jade and Fluoro-Jade B, is commonly used to stain degenerating neurons despite of types of the insult or mechanism of cell death. Among these three fluorochromes, FJC exhibited the greatest signal to background ratio, as well as the highest resolution ⁵⁵ .

Animals were sacrificed after behavior testing and brain tissues were then fixed in 10% neutral-buffered formalin, embedded in paraffin blocks, and sectioned (10 µm). For FJC staining, we used an FJC ready-to-dilute staining kit (TR-100-FJ, Biosensis) and processed in accordance with the manufacturer’s protocol with some modifications, as we previously described^3,4,10,44,53. Briefly, brain sections of each animal group were deparaffinized, rehydrated, and incubated in distilled water for 2 min. The sections were then incubated in a solution of potassium permanganate (1:15) for 10 min, rinsed in distilled water for 2 min, and then in the FJC solution (1:25) for 30 min. The brain sections were then washed and mounted on coverslips with VECTASHIELD mounting medium (Vector Laboratories). We utilized a fluorescent inverted microscope (IX70, Olympus, Japan) to observe and photograph the sections and SPOT image analysis software (Diagnostic Instruments) to count the number of FJC-positive cells in three randomly selected fields per slide. The numbers of FJC-positive cells on slides from individual animal in each group were counted to generate a mean number per group.

Double Immunofluorescence Staining and Immunohistochemistry Staining

The products of lipid peroxidation (4-HNE) and protein oxidation (3-NT) are regarded as indexes of oxidative and nitrosative damages. Neurons are the most vulnerable to oxidative and nitrosative damages in the early evolution of injury pathology after TBI. Double immunofluorescence (IF) staining of 4-HNE or 3-NT with a neuronal marker, neuronal nuclear protein (NeuN), was used to compare the neuroxidative stress among three groups.

Coronal sections (10 µm), obtained from ipsilateral peri-lesion cerebral cortex and striatal regions, were dried and then rehydrated in phosphate-buffered saline (PBS), and rinsed in PBS. The sections, blocked in 5% bovine serum albumin (BSA) for 60 min (BSA: PBS containing 5% BSA and 0.2% Triton X-100; Sigma, St Louis, MO, USA), incubated with the appropriate primary antibodies, including rabbit anti-4-HNE (1:250, Abcam, Cat# ab46545, RRID: AB_722490) and mouse anti-NeuN (1:500, GeneTex, Cat# GTX30773, RRID: 1949456), or mouse anti-3-NT (1:250, GeneTex, GTX41979, RRID: AB_10727902) and rabbit anti- NeuN (1:500, GeneTex, Cat# GTX16208) at 4°C overnight and with a secondary antibodies Alexa Fluo® 488 goat anti-rabbit immunoglobulin G (IgG) (Thermo Fisher Scintific, Cat# A11034, RRID: AB_2576217) and Alexa Fluo® 594 goat anti-mouse IgG (Jackson ImmunoResearch, Cat# 115-585-062, RRID: AB_2338876) or Alexa Fluo® 488 goat anti-mouse IgG (Jackson ImmunoResearch, Cat# 115-545-003, RRID: AB_2338840) and Alexa Fluo® 594 goat anti-rabbit IgG (Jackson ImmunoResearch, Cat# 115-585-003, RRID: AB_2338059) 1:250 at room temperature for 1 h. The sections were mounted with mounting medium (H-1000; Vector Laboratories, Burlingame, CA, USA). To perform the morphological identification of the glia cells, immunohistochemistry (IHC) staining were conducted, using primary antibodies rabbit anti-glial fibrillary acidic protein (GFAP) (1:1,000, GeneTex, Cat# GTX108711) to stain astrocytes or rabbit anti-Iba-1 (1:500, GeneTex, Cat# GTX100042, RRID: AB_1240434) to stain microglia. Before the incubation with the primary antibody, brain sections were blocked in 5% BSA for 60 min and were then quenched by incubation with a 3% H₂O₂ in PBS for 10 min. The sections were rinsed again and were then incubated with the primary antibody at 4°C overnight. After incubation, sections were rinsed three times for 5 min in PBS and incubate visualized using the avidin–biotin peroxidase complex method (ABC Elite kit; Vector Laboratories, Burlingame, CA). Negative controls of IF and IHC consisted of omission of the primary antibodies. The images of IF and IHC were viewed and saved by an Olympus IX 70 Fluorescence Microscope (Japan) equipped with a cooled CCD camera and SPOT advanced software (Diagnostic Instruments, Sterling Heights, MI, USA). The results were saved in a personal computer as digital micrographs and measured.

Measurement of Cytokines in Cortex and Striatum by Enzyme-Linked Immunosorbent Assay

Tissue contents of cytokines (TNF-α, interleukin [IL]-1β, and IL-6) in cortex and striatum were measured using enzyme-linked immunosorbent assay (ELISA) kits (BioSource International, Camarillo, CA, USA) according to the manufacturer’s protocol with some modification ³ .

Statistical Analysis

Statistical significance was assessed by one-way analysis of variance (ANOVA) test, pairwise statistical significance was assessed by Bonferroni post hoc test using SPSS software (Version 18, SPSS Inc.). Graphs are presented with mean ± SEM values and statistical significance was set at P < 0.05 or less, and is noted in all figures. The statistical data relating to each Figure are presented as a table and included as Supplemental Material.

Results

Post-Injury Administration of Pom (0.5 mg/kg, i.v.) Improves Motor Function Deficits at 24 h After TBI as Revealed by Behavioral Evaluations

Beam walking test was used to assess balance and motor coordination at 24 h after TBI. Vehicle-treated TBI animals took longer time (longer latency) to traverse the beam compared with the Sham group, indicating impaired motor coordination (Fig. 1A, 56.16 ± 4.14 vs 10.01 ± 2.21 s). TBI + Pom animals showed notably better test performance than TBI + Veh animals after injury (37.05 ± 6.93 s vs 56.16 ± 4.14 s, P < 0.01); however, the latency taken by Pom-treated animals was still longer than the Sham group (37.05 ± 6.93 vs 10.01 ± 2.21 s). EBST, which measures motor function asymmetry showed an increase in contralateral swing ratios in TBI + Veh animals, compared with the Sham group, indicating asymmetry motor function (Fig. 1B, 88.43 ± 3.59 s vs 51.15 ± 2.78, P < 0.001). Pom treatment after injury significantly alleviated functional asymmetry, compared with the TBI + Veh group (Fig. 1B, 69.1 ± 4.85 s vs 88.43 ± 3.59 s, P < 0.05). The mNSS assessment was used to evaluate overall sensory and motor functions. Compared with the Sham group, the TBI + Veh group had a significantly increased mNSS (Fig. 1C, P < 0.001), indicating greater neurological deficit severities. Pom improved neurological deficits, significantly lowering mNSS (Fig. 1C, P < 0.001).

Figure 1.

TBI caused motor function deficits and administration of Pom (0.5 mg/kg, i.v.) at 5 h post-TBI mitigated motor deficits, as revealed by various behavioral tests conducted at 24 h in the groups of Sham, TBI + vehicle (TBI + Veh), and TBI + Pom. (A) Balance and motor coordination measured by the beam walk test. (B) Motor asymmetry evaluated by the elevated body swing test (EBST). (C) Neurological function was evaluated by modified neurological severity score (mNSS) test. Data represent the mean ± SEM (n = 12 for Sham group, n = 6 for TBI + Veh and TBI + Pom, respectively). *P < 0.05 and ***P < 0.001 vs Sham group; ⁺P < 0.05, ⁺⁺P < 0.01, and ⁺⁺⁺P < 0.001 vs TBI + Veh group.

Post-Injury Pom Treatment Decreases Contusion Volume Caused by TBI

The contusion volume induced by TBI in the ipsilateral hemisphere was computed as percentage of the contralateral hemisphere at 24 h after injury with vehicle or Pom administered at 5 h. As illustrated in Fig. 2A, loss of volume in the ipsilateral hemisphere in the TBI + Veh animal was greater than the Sham animal. The contusion volume observed in the TBI + Veh group was 15.33% ± 1.39% of the contralateral hemisphere volume (Fig. 2B). Post-injury administration of Pom treatment notably reduced this volume to 7.74% ± 1.07% (Fig. 2B). This represents a 49.5% reduction in the contusion volume in comparison with the TBI + Veh group (P < 0.001).

Figure 2.

Post-injury Pom administration resulted in a decrease of contusion volume at 24 h. (A) Representative cresyl violet-stained coronal brain sections of Sham, TBI + Veh, and TBI + Pom rats at 24 h after TBI. (B) Pom administration significantly reduced the contusion volume at 24 h after TBI. Data are expressed as mean ± SEM (n = 12 for Sham group, n = 6 for TBI + Veh and TBI + Pom, respectively). ***P < 0.001 compared with the Sham group; ⁺⁺⁺P < 0.001 in comparison with the TBI + Veh group.

Post-Injury Administration of Pom Reduces TBI-Induced Neurodegeneration in Cerebral Cortex (54.58% Reduction) and Striatum (37.97% Reduction)

We performed FJC staining to visualize degenerating neurons. A low magnification hematoxylin and eosin (H&E)-stained coronal section from the sham animal showed the anatomical region in the cerebral cortex and dorsal striatum (Fig. 3A) where higher magnification images were taken for counting the FJC-positive cells (degenerating neurons) and statistical analysis (Fig. 3A). Many FJC-positive cells were observed in the striatal and cortical tissues at 24 h post-injury (Fig. 3B). The numbers of FJC-positive cells in cortex and striatum in Sham, TBI + Veh, and TBI + Pom groups were quantitatively compared (Fig. 3C). The number of degenerating cortical neurons was notably elevated in the TBI + Veh group (513.54 ± 48.05/mm², P < 0.001) in comparison with that in the Sham group (8.53 ± 0.91/mm²). In contrast, the TBI + Pom group exhibited significantly lower number of degenerating neurons (233.69 ± 15.46/mm², P < 0.001) compared with the TBI + Veh group; thereby providing a 54.58% reduction in degenerating neurons in cerebral cortex. A significantly elevated number of degenerating striatal neurons was also seen in the Veh-treated TBI group (158.54 ± 12.36/mm², P < 0.001) in comparison with that in the Sham group (10.83 ± 12.36/mm², P < 0.001). In contrast, the TBI + Pom group exhibited significantly lower number of degenerating striatal neurons (98.33 ± 31.07/mm², P < 0.001) in comparison with the TBI + Veh group; thereby providing a 37.97% reduction in degenerating neurons in striatum.

Figure 3.

Post-TBI administration of Pom reduced the number of FJC-positive cells in ipsilateral cortex and striatum at 24 h. (A) A lower magnification H&E-stained coronal brain section from the Sham animal showing the cerebral cortex and dorsal striatum. The black squares indicate the anatomical location where the higher magnification images were taken. Scale bar = 50 µm. (B) Representative photomicrographs revealing the presence of FJC-staining positive cells at 24 h in the Sham, TBI + Veh, and TBI + Pom groups. (C) Quantitative comparison of FJC-positive cells in different groups. The total number of FJC-positive cells is expressed as the mean number/mm². Data are expressed as mean ± SEM (n = 4 per group). ***P < 0.001 vs Sham group; ⁺⁺P < 0.05 vs TBI + Veh group.

Pom Treatment Reduces Oxidative and Nitrosative Damages in Cortical and Striatal Tissues of TBI Rats and is More Efficacious in Striatum Than in Cerebral Cortex

The products of lipid peroxidation (4-HNE) and protein nitroxidation (3-NT) have been regarded as markers for oxidative and nitrosative damages, respectively. Double IF staining of 4-HNE and NeuN indicates neurons with oxidative damage as a result of being attacked by ROS. Double IF staining of 3-NT and NeuN indicates neurons with nitrosative damage as a result of being attacked by peroxynitrite. Fig. 4 shows that 4-HNE-positive neurons in both cortical and striatal tissues in the Sham group were negligible (Fig. 4Aa–c in the cortex and 4Aj–l in the striatum). The number of 4-HNE-positive neurons in TBI + Veh group significantly increased in the cortex and striatum (Fig. 4Ad–f in the cortex and Fig. 4Am–o in the striatum), and these phenomena were alleviated by Pom treatment (Fig. 4Ag–I in the cortex and Fig. 4Ap–r in the striatum). Specifically, quantitative analysis revealed that the cortical and striatal 4-HNE-positive neuron numbers increased significantly in the TBI + Veh group at 24 h after TBI (Fig. 4B, 23.33 ± 2.6 cells/mm², P < 0.001 and 37.06 ± 2.59 cells/mm², P < 0.001, respectively) compared with that in the Sham group (Fig. 4B, 1.7 ± 0.61 cells/mm² and 1.83 ± 1.83 cells/mm², respectively). Pom administration reduced the number of cortical and striatal 4-HNE-positive neurons (13.49 ± 2.53 cells/mm², P < 0.01 and 2.7 ± 1.61 cells/mm², P < 0.001, respectively) compared with that in the TBI vehicle-treated group (Fig. 4B). Interestingly, the number of 4-HNE-positive neurons increased in striatal tissue (from 1.83 to 37.07 cells/mm², about 20 folds) more than that in the cortical tissue (from 1.70 to 23.32 cells/mm², about 14 folds) and the antioxidant effect of Pom was also more prominent in striatum (from 37.07 to 2.70 cells/mm², about 93% reduction) than in cerebral cortex (from 23.32 to 13.49 cells/mm², about 42% reduction).

Figure 4.

Post-TBI Pom treatment decreased 4-HNE-positive neuron numbers in cerebral cortex and striatum at 24 h after TBI. (A) Double IF staining of NeuN (red) in cortical (A: a, d, g) striatal brain tissues (A: j, m, p) and 4-HNE (green) in cortical (A: b, e, h) and striatal brain tissues (A: k, n, q). Yellow labeling (Merge) indicates NeuN and4-HNE double positive cell colocalization (A: c, f. i in cortex and l, o, r in striatum, respectively). Scale bar = 50 µm. (B) Quantitative comparison of NeuN-4-HNE-double positive cells in cerebral cortex and striatum in Sham, TBI + Veh, and TBI + Pom groups. Data are presented as mean ± SEM (n = 3 per group). ***P < 0.001, **P < 0.01; vs Sham group; ⁺⁺P < 0.01 vs TBI + Veh group.

Compared with the Sham group (Fig. 5Aa–c and 5Aj–l), numerous 3-NT-positive neurons were detected in the cortical and striatal tissue at 24 h after injury (Fig. 5Ad–f and 5Am–o, respectively). However, in comparison with the TBI + Veh group, the expression of 3-NT-positive neurons significantly decreased in cortex and striatum in the TBI + Pom group (Fig. 5Ag–i and 5Ap–r, respectively). A quantitative summary of 3-NT-positive neurons noticed in cortex and striatum in the Sham, TBI + Veh, and TBI + Pom groups is shown in Fig. 5B. Numbers of cortical and striatal 3-NT-positive neurons were notably increased in the TBI vehicle-treated group (22.59 ± 2.37 cells/mm², P < 0.001 and 33.86 ± 3.62 cells/mm², P < 0.001, respectively) compared with the Sham group (0.62 ± 0.25 cells/mm² and 5.06 ± 2.49 cells/mm², respectively). In contrast, the TBI + Pom group was observed a relatively lower number of cortical and striatal 3-NT-positive neurons (10.22 ± 1.55 cells/mm², P < 0.001 and 5.43 ± 0.43 cells/mm², P < 0.001, respectively) compared with the TBI + Veh group. Similar to the results with 4-HNE staining, the degree of increase in the number of 3-NT-positive neurons was greater in striatal tissue (from 5.07 to 33.87 cells/mm², about 6.68 folds) than that in the cerebral cortical tissue (from 0.62 to 22.60 cells/mm², about 36 folds) and the antioxidant effect of Pom was also more prominent in striatum (from 33.87 to 5.43 cells/mm², about 84% reduction) than in cerebral cortex (from 22.60 to 10.22 cells/mm², about 55% reduction).

Figure 5.

Pom treatment post-injury decreased 3-NT-positive neuron numbers in cerebral cortex and striatum at 24 h after TBI. (A) Double IF staining of NeuN (red) in cortical (A: a, d,g) striatal brain tissues (A: j, m, p) and 3-NT (green) in cortical (A: b, e, h) and striatal brain tissues (A: k, n, q). Yellow labeling (Merge) indicates NeuN and 3-NT double positive cell colocalization (A: c, f, i in cortex and l, o, r in striatum). Scale bar = 50 µm. (B) Quantitative comparison of NeuN-3-NT-double positive cells (3-NT-positive neurons) in cerebral cortex and striatum in Sham, TBI + Veh, and TBI + Pom groups. Data are presented as mean ± SEM (n = 3 per group). ***P < 0.001, **P <0.01; vs Sham group; ⁺⁺P < 0.01 vs TBI + Veh group.

Post-Injury Pom Treatment Reduces Microgliosis and Astrogliosis in Cortical and Striatal Tissues of TBI Rats

It is well known that TBI induces activation of microglial cells and astrocytes. Activated microglial cells can be recognized by their changes in their morphology from ramified (resting state morphology with long branching processes) into amoeboid (activated or reactive state morphology with shorter thicker processes) as well as increased numbers (microgliosis). We performed IHC staining, using primary antibody against Iba-1 to stain microglia. As illustrated in Fig. 6Aa and 6Ad, Iba-1-positive cells with a ramified morphology were observed in cortex and striatum in the Sham group. In comparison with the Sham group, the TBI + Veh group displayed Iba-1-positive cells with both ramified and amoeboid (a characteristic of reactive or activated microglia) morphology in cortex and striatum (Fig. 6Ab and 6Ae, respectively). However, compared with the TBI + Veh group, the TBI + Pom group displayed fewer amoeboid microglia and increased ramified microglia in cortex and striatum (Fig. 6Ac and 6Af, respectively). A quantitative analysis of Iba-1-positive cells in the Sham, TBI + Veh, and TBI + Pom groups is shown in Fig. 6B. Significantly increased numbers of cortical and striatal Iba-1-positive cells were noted in the TBI + Veh group (221.3 ± 2.71 cells/mm², P < 0.001 and 192.6 ± 20.6 cells/mm², P < 0.01, respectively) in comparison with the Sham group (88.1 ± 6.29 cells/mm² for cortical Iba-1-positive cells, and 81.3 ± 6.3 cells/mm² for striatal Iba-1-positive cells). In contrast, post-injury administration of Pom notably decreased the number of cortical and striatal Iba-1-positive (157.7 ± 18.01 cells/mm², P < 0.01, and 116.4 ± 6.14 cells/mm², P < 0.01, respectively) in comparison with the TBI + Veh group (Fig. 6B).

Figure 6.

Post-injury administration of Pom at 5 h after TBI attenuated microgliosis in cortex and striatum in TBI rats. (A) Representative photomicrographs showing results of IHC staining with antibody Iba-1 in cortical and striatal Iba-1 immunohistochemical staining, with microglial cell morphology magnified in lower right boxes. Scale bar = 30 µm. (B) Quantitative comparison of Iba-1-positive cells in different groups. Data are expressed as means (± SEM; n = 3 in each group). **P < 0.01 vs Sham group; ⁺⁺P < 0.01 vs TBI + Veh group.

Morphological change of activated or reactive astrocytes is characterized by hypertrophy of the main cellular processes, as well as increased numbers (astrogliosis). We performed IHC staining, using primary antibody against GFAP to stain astrocytes. As shown in Fig. 7, GFAP immunoreactivity exhibited a filament-like profile in the cytoplasm in cortex and striatum in the Sham group (Fig. 7Aa and 7Ad, respectively), whereas stellate soma and hypertrophic processes were observed in the vehicle-treated TBI group (Fig. 7Ab and 7Ae). Compared with the vehicle-treated TBI group, the TBI + Pom group exhibited fewer stellate soma and hypertrophic processes but more cells with filament-like immunoreactivity (Fig. 7Ac and 7Af). A quantitative analysis of GFAP-positive cells in the Sham, TBI + Veh, and TBI + Pom groups is shown in Fig. 7B. Numbers of cortical and striatal GFAP-positive cells significantly increased in the TBI vehicle-treated group (277 ± 8.04 cells/mm², P < 0.001 and 332.8 ± 11.63 cells/mm², P < 0.01, respectively) compared with the Sham group (141 ± 0.85 cells/mm² for cortical GFAP-positive cells, and 230 ± 7.48 cells/mm² for striatal GFAP-positive cells). In contrast, the TBI + Pom group had a notably lower number of cortical and striatal GFAP-positive cells (221 ± 16.29 cells/mm², P < 0.01 and 232 ± 23.21 cells/mm², P < 0.01, respectively) in comparison with the vehicle-treated TBI group.

Figure 7.

Pom treatment alleviated astrogliosis in cortex and striatum in TBI rats. (A) Representative photomicrographs showing cortical and striatal GFAP immunohistochemical staining across animal groups at 24 h post-TBI or sham, with astrocyte cell morphology magnified in lower right boxes. Scale bar = 30 µm. (B) Quantitative comparison of GFAP-positive cells in different groups. Data are expressed as means (±SEM; n = 3 in each group). **P < 0.01 vs Sham group; ⁺⁺P < 0.01 vs TBI + Veh group.

Post-TBI administration of Pom reduces neuroinflammation reflected by less elevation of tissue levels of TNF-α, IL-1β, and IL-6 in cerebral cortex and in striatum. Activation of microglial cells and astrocytes is well known to initiate a neuroinflammatory response through releasing large amounts of proinflammatory cytokines following TBI. We examined the effects of Pom on the concentration of proinflammatory cytokines (TNF-α, IL-1β, and IL-6) in cortical and striatal tissues at 24 h after TBI using ELISA. As illustrated in Fig. 8, the levels of TNF-α, IL-1β, and IL-6 were significantly elevated in the cortical tissue in TBI vehicle-treated group (4.05 ± 0.48 pg/mg protein, P < 0.001; 2.14 ± 0.7 pg/mg protein, P < 0.01; and 81.04 ± 6.48 pg/mg protein, P < 0.001, respectively) in comparison with the Sham group (2.4 ± 0.9, 1.31 ± 0.18, and 25.8 ± 2.03 pg/mg protein, respectively); these increments were significantly attenuated in the Pom-treated TBI group (2.51 ± 0.25 pg/mg protein, P < 0.01; 1.54 ± 0.05 pg/mg protein, P < 0.05, and 54.19 ± 6.87 pg/mg protein, P < 0.01, respectively) when Pom was administered at 5 h post-TBI. In line with this, levels of TNF-a, IL-1b, and IL-6 were significantly elevated in the striatum in the TBI vehicle-treated group (13.00 ± 0.28 pg/mg protein, P < 0.05; 30.48 ± 1.27 pg/mg protein, P < 0.001 and 36.56 ± 1.56 pg/mg protein, P < 0.05, respectively), whereas levels of IL-1β, and IL-6 (23.49 ± 1.48 pg/mg protein, P < 0.01 and 29.40 ± 1.28 pg/mg protein, P < 0.05, respectively) but not TNF-α (8.95 ± 0.85 pg/mg protein) were significantly attenuated by Pom treatment.

Figure 8.

Post-injury administration of Pom at 5 h after TBI reduced the levels of the pro-inflammatory cytokines in the cortical and striatal tissue at 24 h after TBI. Tissue levels of TNF-α (A), IL-1β(B), and IL-6 (C) in the cortex and striatum of Sham, TBI + Veh, and TBI + Pom groups were measured using ELISA kits. Data are expressed as means (±SEM; n = 5 in each group). *P < 0.05, **P < 0.01, and ***P < 0.001 vs Sham group; ⁺P < 0.05 and ⁺⁺P < 0.01 vs TBI + Veh group.

Discussion

The present study demonstrates that post-injury administration of Pom (0.5 mg/kg, i.v.), which is known as a TNF-α synthesis inhibitor and immunomodulatory drug^40–42, is able to mitigate motor behavioral deficits in experimental TBI in an animal model. It is noteworthy that a single i.v. injection of Pom ameliorated injury-induced neurological impairment and reduced lesion volume; i.v. and intraperitoneal (i.p.) injections are two most widely used routes for systemic administration of drugs in rodents. It is known that i.v. administration also provides less restricted systemic bio-distribution of drugs^56,57. The i.v. dose (0.5 mg/kg, i.v.) of Pom used in this study is smaller than the i.p. dose (20 mg/kg) used in rats for neuroprotection against brain ischemia/reperfusion ⁵⁸ . We previously demonstrated that the window of therapeutic intervention for Pom is up to 5 h after TBI ³ . This post-injury therapeutic time window is also longer than that (30 min after ischemia-reperfusion) provided by i.p. administration of Pom (20 mg/kg). Since Pom is an FDA-approved immunomodulatory drug clinically used in treating multiple myeloma, our results suggest that Pom is a good therapeutic candidate for clinical trials for a reposition of Pom in treating TBI.

Here, we showed that Pom reduces contusion volume in cerebral cortex and the number of degenerating neurons in striatal as well as cortical tissues. Unlike previous studies focusing on peri-lesion areas in cerebral cortex, we further compared the effects of Pom on TBI-induced cortical and striatal damage in the present study. TBI causes cell death mostly in the cerebral cortex but there is also additional secondary cell death in other brain regions, such as striatum and hippocampus^5,59,60. FJC, as well as its predecessors, such as Fluoro-Jade and Fluoro-Jade B, is commonly used to stain degenerating neurons despite of types of the insult or mechanism of cell death^3,55. This study demonstrated that Pom treatment reduces degenerating (FJC-positive) neurons in the cortical and striatal tissues at 24 h after TBI. Neuronal damage due to oxidative stress is generally accepted to be an important component of TBI^28–30. The brain is highly sensitive to oxidative damage due to its high content of polyunsaturated fatty acids and high levels of iron^26,32,61,62. Iron reacts with H₂O₂ causing hydrogen radical (•OH) formation that then acts as a potent initiator of peroxidative damage to polyunsaturated fatty acids^63,64. An aldehyde product of lipid peroxidation, 4-HNE, binds to cell membranes and intracellular proteins to modify their structures and functions, resulting in oxidative damage^65,66. The end product of protein nitration, 3-NT, as well as 4-HNE has been clearly demonstrated to be neurotoxic ³² . Moreover, it has been reported that peroxynitrite, formed by the reaction of superoxide, is involved in producing tissue damage in TBI^32,67,68. In addition, Pom reduces the number of 4-HNE-positive and/or 3-NT-positive neurons. It is interesting that Pom reduces oxidative and nitrosative damages in cortical and striatal tissues of TBI rats and is more efficacious in stratum than in cerebral cortex. Pom also attenuates microgliosis and astrogliosis as well as the levels of pro-inflammatory cytokines in cortex and striatum after TBI.

Oxidative and nitrosative stress are common features across neurodegenerative disorders, including Parkinson’s disease (PD), stroke, and TBI. Oxidative stress involves an imbalance between ROS generation and endogenous antioxidant defense mechanisms. Mitigation of oxidative and nitrosative stress can hence derive from lowering ROS/RNS levels or by augmenting antioxidant homeostatic compensatory mechanisms. It is likely that Pom, via multiple mechanisms, simultaneously impacts both of these mechanisms. For example, nitrosative stress, characterized by excess RNS, involves peroxynitrite formation by reaction of NO with superoxide anions (O₂•^–). Whereas NO has dominant physiological functions (mediating neurotransmission and vasodilation), excessive NO production directly attacks various antioxidant enzymes, inclusive of catalase ⁶⁹ , whereby post-translational modifications reduce its activity ⁷⁰ . Prior studies demonstrated that Pom can significantly lower NO generation and release from microglial cells (as evaluated by nitrite levels) ⁴⁴ . In addition, in relation to endogenous antioxidant proteins, recent studies by Tsai et al. demonstrated that a clinically translatable dose of Pom elevated the expression of nuclear factor erythroid-derived 2 (Nrf2) and superoxide dismutase 2 (SOD2)/catalase signaling pathways^45,58; these are important antioxidant mechanisms.

The striatum is an important component of the basal ganglia. Striatal neurons receive glutamate inputs from the cortex and DA innervation from the midbrain. The behavioral evaluations in this study involved motor learning and programming, critical striatal functions. Even postural asymmetry involves “axial set” in rats, an important striatal function. The role of striatal DA transmission in such functions is further documented by the behavioral deficits seen in preclinical rodent models of PD where the nigrostriatal DA system is specifically targeted. Striatal dysfunction is also related to deficiencies in learning, memory, and executive function after TBI^21,71. Moreover, DA release in the striatum after fluid percussion injury in rats with moderate or mild TBI, examined by in vitro fast-scan cyclic voltammetry, demonstrated that both tonic and bursting DA release were significantly inhibited in moderate TBI ⁷² . Quantitative IF detection of the oxidative and nitrosative damages products (4-HNE and 3-NT) was used to investigate the effect of Pom on TBI-induced oxidative and nitrosative damages in the cortical and striatal tissues. The number of cortical and striatal 4-HNE-positive and 3-NT-positive neurons, significantly elevated in the TBI + Veh group, as compared with the Sham group, was mitigated in the TBI + Pom group. This evidence demonstrated that Pom has an antioxidant effect. Interestingly, the degree of increase in the numbers of 4-HNE or 3-NT positive neurons was greater in striatum than that in the cortex and the antioxidant effect of Pom was also more prominent in striatum than in cortex. Accordingly, these results underlined the important effects of Pom for attenuation of TBI-induced oxidative and nitrosative damages and prominent antioxidant effects in striatum compared with the cerebral cortex.

Neuroinflammation can exhibit dual effects to either promote cellular protection or induce tissue injury through the beneficial but sometimes damaging potential of inflammatory mediators, depending on their time-dependent concentration. Activated glia, chiefly astrocytes and microglia, produce large amounts of pro-inflammatory and anti-inflammatory cytokines following TBI^8,73. TNF-α and IL cytokines (IL-1β, IL-6) display pleiotropic effects in the brain. At low physiological levels, TNF-α is intimately involved in regulating synaptic scaling, neurogenesis, sleep as well as immune surveillance; however, following a bacterial challenge or alternative insult, its synthesis and release can be massively scaled up, particularly in CNS injury, to thereby initiate and propagate inflammatory responses ³⁸ . This dual role of TNF-α as an inflammation initiator and physiological regulator to optimize brain function requires a complex interaction of feedback mechanisms and neuron-glia crosstalk, as well as integration through redox sensors present in enzymes, receptors, and transcription factors. If inappropriately balanced, the equilibrium between maintaining cellular homeostasis and cell survival fails and a vicious cycle of unregulated neuroinflammation, cellular dysfunction and cell death can ensue ⁷⁴ . Several studies^75–78 have suggested that levels of TNF-α and other cytokines elevate within 1 h, and reach the peak at 4 h post-TBI; afterwards, levels returned to baseline with a secondary smaller elevation at 72 h post-TBI ⁷⁹ . Pom and similar drugs are reported to inhibit TNF-α generation by targeting the 3ʹ-UTR of TNF-α mRNA, ultimately reducing its transcription and translation ⁴⁰ , and thereby lowering cytokine/chemokine levels. Our data (Fig. 8) showed tissue concentrations of TNF-α in cerebral cortex and striatum increased at 24 h after TBI. In addition, tissue concentrations of IL-1 and IL-6 also increased in cerebral cortex and striatum within 24 h. These results are in line with our previous studies using Pom to treat TBI^3,44,53.

In the present study, we clearly demonstrate that post-injury Pom treatment reduces oxidative and nitrosative damages in cortical and striatal tissues of TBI rats and is more efficacious in stratum than in cerebral cortex (Fig. 9). Focusing here on the striatum and cortex, in the light of increasing evidence showing that TBI alters dopaminergic neurotransmission^19,20,60 and induces striatal axonal degeneration ²² , oxidative stress, and neuroinflammation ⁵ that can result in neurological deficits^16,20,21,80, we observed that Pom reduced the number of cortical and striatal degenerating neurons after TBI with associated mitigation of oxidative and nitrosative damages and neuroinflammation.

Figure 9.

Schematic summary of the results in this study. (A) TBI caused motor function deficits as revealed by various behavioral tests, including beam walk test, elevated body swing test, and mNSS. The post-injury administration of Pom alleviated these motor behavioral deficits. (B) TBI resulted in oxidative and nitrosative damages in cerebral cortex and striatum, which might contribute to the disruption of neuronal projections and neurotransmission and subsequent motor behavioral deficits. Pom reduced the number of cortical and striatal degenerating neurons after TBI. Pom also inhibited the generation of oxidative and nitrosative damages products (4-HNE and 3-NT) in neurons. In addition, Pom attenuated microgliosis, astrogliosis, and elevations of proinflammatory cytokines in cortical and striatal tissue after TBI.

Pom is an FDA-approved drug that is widely used in the treatment of multiple myeloma. The routine clinical dose for multiple myeloma in patients is Pom 4 mg post-operatively per day on Days 1–21 of repeated 28-day cycles^81,82. The conventional dosage of Pom 5 mg daily is used for Karposi’s sarcoma (Micromedex Drug Reference 2.0). Translating the dose of Pom (0.5 mg/kg) used in the present rat study to an equivalent dose in humans in accord with FDA guidelines (Conversion of Animal Doses to Human Equivalent Doses Based on Body Surface Area), would result in a human dose of approximately 5 mg in a 60 kg human (https://www.semanticscholar.org/paper/Guidance-for-Industry-Estimating-the-Maximum-Safe-Alert/). Although Pom was administered intravenously in our study, as opposed to the conventional oral route, use of the i.v. route has been described in human studies and disparities between the different administration methods are potentially minimal, given the drug’s high oral bioavailability (an absorption of > 70% is reported in one study ⁸³ . In addition, we administered a single bolus dose at an acute TBI stage in our study, in contrast to the multiple daily dosing regimen of Pom for the treatment of multiple myeloma or Karposi’s sarcoma. Nonetheless, we acknowledge that our study in rats employed a slightly higher (5 vs 4 mg daily) drug dose compared with current regimens. Further investigations to establish the optimal Pom dosing regimen for TBI patients are therefore warranted.

Conclusion

Taken together, the main findings of the present study provide evidence that Pom possesses promising striatal and cortical neuroprotective effects against TBI, which may result from attenuation of oxidative and nitrosative damages, and neuroinflammation. These findings support the promise of Pom as a new strategy for the therapy of TBI. However, further evaluation and optimization of Pom should be conducted to support the safe translation from the preclinical to clinical arena in the repurposing of this FDA-approved drug.