J. Neurochem. (2010) 115, 921–929.
Antagonism of tumor necrosis factor-alpha with etanercept has proved to be effective in the treatment of spinal cord injury and centrally endotoxin-induced brain injury. However, etanercept may offer promise as therapy for traumatic brain injury (TBI). In this study, anesthetized rats, immediately after the onset of TBI, were divided into two major groups and given the vehicle solution (1 mL/kg of body weight) or etanercept (5 mg/kg of body weight) intraperitoneally once per 12 h for consecutive 3 days. Etanercept caused attenuation of TBI-induced cerebral ischemia (e.g., increased cellular levels of glutamate and lactate-to-pyruvate ratio), damage (e.g., increased cellular levels of glycerol) and contusion and motor and cognitive function deficits. TBI-induced neuronal apoptosis (e.g., increased numbers of terminal deoxynucleotidyl transferase αUTP nick-end labeling and neuronal-specific nuclear protein double-positive cells), glial apoptosis (e.g., increased numbers of terminal deoxynucleotidyl transferase αUTP nick-end labeling and glial fibrillary acidic protein double-positive cells), astrocytic (e.g., increased numbers of glial fibrillary acidic protein positive cells) and microglial (e.g., increased numbers of ionized calcium-binding adapter molecule 1-positive cells) activation and activated inflammation (e.g., increased levels of tumor necrosis factor-alpha, interleukin-1β and interleukin-6) were all significantly reduced by etanercept treatment. These findings suggest that etanercept may improve outcomes of TBI by penetrating into the cerebrospinal fluid in rats.
fluid percussion injury
glial fibrillary acidic protein
ionized calcium-binding adapter molecule 1
neuronal-specific nuclear protein
phosphate buffer saline
traumatic brain injury
tumor necrosis factor-alpha
terminal deoxynucleotidyl transferase dUTP nick-end labeling
Traumatic brain injury (TBI), the leading cause of morbidity and mortality in young adults and children, is a major public health problem globally. TBI is associated with microglial and astrocytic activation as well as the release of proinflammatory cytokines (Lucas et al. 2006). The most studied proinflammatory cytokines related to development of cerebral edema, breakdown of the blood-brain barrier, and secondary neuronal injury are interleukin (IL)-1, tumor necrosis factor (TNF)-α and IL-6 (Hailer et al. 2005). Using anti-inflammatory agents for therapy of TBI to reduce blood-brain barrier dysfunction, intracranial neutrophil infiltration and neuronal cell death and to improve neurological outcome is a promising treatment (Stahel et al. 2000).
Etanercept is a recombinant dimer of human TNF-α receptor proteins fused and bound to human IgG 1. It is a TNF antagonist with anti-inflammatory effects. Etanercept, when administered systemically at the dosage approved for its licensed indications (∼ 50 mg/week in humans), would not be expected to achieve therapeutic levels in the CSF because of its high molecular weight (Francis et al. 2004). More recent studies showed that etanercept was able to intervene in the brain inflammatory response and to protect brain and spinal cord from secondary damage caused by infiltrating leukocytes (Campbell et al. 2007; Genovese et al. 2006). This raised the possibility that experimental TBI could be affected by etanercept therapy.
To deal with the question, current experiments were conducted to assess the motor and cognitive function, cerebral ischemia and damage, neuronal and glial apoptosis, gliosis, and expression of TNF-α, IL-1β and IL-6 in TBI-affected rats treated with or without etanercept.
Male Sprague–Dawley rats (weight, 249 ± 11 g) were obtained from the Animal Resource Center of the National Science Council of Republic of China (Taipei, Taiwan). The animals were housed four in a group at an ambient temperature of 22 ± 1°C with a 12-h light-dark cycle. Pelleted rat chow and tap water were available ad libitum. All protocols were approved by the Animal Ethics Committee of the Chi Mei Medical Center (Tainan, Taiwan) to minimize discomfort in the animals during surgery and in the recovery period.
Animals were anesthetized with sodium pentobarbital (25 mg/kg, intraperitoneally; Sigma Chemical, St. Louis, MO, USA) and a mixture containing ketamine (4.4 mg/kg, intramuscularly; Nankuang Pharmaceutical, Tainan, Taiwan), atropine (0.02633 mg/kg, intramuscularly; Sintong Chemical, Ind., Taoyuan, Taiwan) and xylazine (6.77 mg/kg, intramuscularly; Bayer, Leverkusen, Germany). Both the femoral artery and vein on the right side were cannulated with PE50 polyethylene tubing for monitoring blood pressure and analyzing blood gas. After cannulation, the wound was sutured and animals were turned prone. The animals were placed in a stereotaxic frame, and the scalp was incised sagittally. The animals were subjected to a lateral fluid percussion injury (FPI) (McIntosh et al. 1987). After the scalp was incised, a 4.8-mm circular craniotomy was performed midway between lambda and bregma 3.0 mm to the right of the central suture. A modified Luer-lock connector (trauma cannula), 2.6 mm inner diameter, was secured into the craniotomy with cyanoacrylate adhesive and dental acrylic. A moderate FPI (2.2 atm) was produced by rapidly injecting a small volume of saline into the closed cranial cavity with a fluid percussion device (VCU Biochemical Engineering, Richmond, VA, USA). The animal was removed from the device, the acrylic removed, and the incision sutured. Each injured and sham-injured animal for the FPI model was closely evaluated immediately after FPI for behavioral recovery. Mean arterial pressure, heart rate and core temperature were continuously monitored during 60 min after FPI.
Rats were randomly allocated into the following groups: (i) TBI + saline group – rats were subjected to TBI plus an intraperitoneal (i.p.) dose of normal saline (mL/kg body weight) once per 12 h for consecutive 3 days (n = 8); (ii) TBI + etanercept group – this group was the same as the TBI + saline group, with the exception that etanercept at the dose of 5 mg/kg was administered intraperitoneally once per 12 h for consecutive 3 days (n = 8); (iii) Sham + saline group – rats were subjected to the surgical procedures as the above groups, with the exception that the FPI was not applied (n = 8); and (iv) Sham + etanercept group – this group was the same as group 3, with exception that etanercept at the dose of 5 mg/kg was administered intraperitoneally once per 12 h for consecutive 3 days (n = 8).
In Experiment 1, an i.p. dose of etanercept (5 mg/kg) or saline (1 mL/kg) was randomly administered immediately after TBI, and their effects on core temperature, mean arterial pressure, heart rate, and the hippocampus level of glutamate, glycerol, and the lactate-to-pyruvate ratio were assessed after TBI for 120 min. ENBRELTM (etanercept) (Wyeth Pharmaceuticals, New Lane, Havant, Hampshire, UK) was reconstituted with normal saline according to the manufacturer’s instructions.
In Experiment 2, an i.p. dose of etanercept (5 mg/kg) or saline was randomly administered immediately after TBI once per 12 h for consecutive 3 days, and their effect on the maximal angle animals could cling to an inclined plane as well as passive avoidance performance was assessed 4 days after TBI.
In Experiment 3, an i.p. dose of etanercept (5 mg/kg) or saline was randomly administered immediately after TBI once per 12 h for consecutive 3 days, and their effect on cerebral contusion zone was assessed 4 days after TBI. An i.p. dose of 5 mg/kg of etanercept was found to be effective for treating spinal cord injury in rats (Genovese et al. 2006). Therefore, the dosage of etanercept was adopted in the present experiments.
In Experiment 4, an i.p. dose of etanercept (5 mg/kg) or saline was randomly administered immediately after TBI once per 12 h for consecutive 3 days, and their effect on the extent of neuronal and glial apoptosis as well as gliosis was assessed 4 days after TBI.
In Experiment 5, an i.p. dose of etanercept (5 mg/kg) or saline was randomly administered immediately after TBI once per 12 h for consecutive 3 days, and their effect on the relative levels of selected rat cytokines in serum was assessed 4 days after TBI.
Monitoring cellular ischemia and damage markers in hippocampus
The animal’s head was mounted in a stereotoxic apparatus (David Kopf Instruments, Tujunga, CA, USA) with the nose bar positioned 3.3 mm below the horizontal line. After a midline incision in the skull, a dialysis probe (4 mm in length CMA/2; Carnegie Medicine, Stockholm, Sweden) was inserted. The microdialysis probe was stereotaxically and obliquely (anterior 4.3 mm) implanted into the right hippocampus (or dentate gyrus) according to the atlas and coordinates of Paxinos and Watson (1982): P, 8 mm; R, 3 mm; H, 5 mm. According to the methods described previously (Chio et al. 2007; Kuo et al. 2007), the micro-dialysis probe was perfused at 2.0 μL/min and the dialysates were sampled in microvials. The dialysates were collected every 20 min in a CMA/140 fraction collector (Carnegie Medicine). Aliquots of dialysates (5 μL) were injected onto a CMA 600 microdialysis analyzer (Carnegie Medicine) for measurement of lactate, glycerol, pyruvate and glutamate.
The inclined plane was used to measure limb strength. The animals were placed, facing right and then left, perpendicular to the slope of a 20 × 20-cm ruffer ribbed surface of an inclined plane starting at an angle of 55o (Chang et al. 2008). The angle was increased or decreased in 5°C increments to determine the maximal angle an animal could hold to the plane. The data for each day were the mean of left and right side maximal angles. All behavioral tests were examined and independently scored by two observers who were unaware of prior treatment. These scores were averaged to arrive at one score for each animal for the behavioral session.
Passive avoidance performance
This passive avoidance task is a one trial fear-motivated avoidance task in which the rat learns to refrain from stepping through a door to an apparently safer but previously punished dark compartment. The latency to refrain from crossing into the punished compartment serves as an index of the ability to avoid, and allows memory to be assessed. The apparatus (Shuttlebox-Passive Avoidance, Accuscan Instrument, Inc., Columbus, Ohio, USA) consists of a rectangular chamber divided into two compartments. One compartment is lighted by an overhead stimulus light and the other is black. The two compartments are separated by an automatic guillotine door and each has a grid floor placed through which a foot shock can be delivered. On test day (24 h after training), the rat is returned to the lighted compartment, facing away from the dark compartment. After 5 s, the guillotine door is lifted. When the rat enters the dark compartment with four paws, the guillotine is closed, and the latency to enter the dark compartment is recorded (from the time the door is lifted). The rat is removed and returned to the homecage.
Cerebral contusion assay
The triphenyltetrazolium chloride (TTC) staining procedures followed those described elsewhere (Wang et al. 1997). All the animals were killed 4 days after TBI. Under deep anesthesia (sodium pentobarbital, 100 mg/kg, intraperitoneally), the animals were perfused intracardially with saline. The brain tissue was then removed, immersed in cold saline for 5 min, and sliced into 2.0-mm sections with a tissue slice. The brain slices were incubated in 2% TTC dissolved in phosphate buffer saline (PBS) for 30 min at 37°C and then transferred to 5% formaldehyde solution for fixation. The volume of infarction, as revealed by negative TTC stains indicating dehydrogenase-deficient tissue, was measured in each slice and summed using computerized planimetry (Image-Pro Plus 5.0 software, Media Cybernetics, Inc., Bethesda, MD, USA). The volume of infarction was calculated as 2 mm (thickness of the slice) × [sum of the infarction area in all brain slices (mm2)] (Wang et al. 1997).
Terminal deoxynucleotidyl-transferase-mediated and dUTP-biotin nick end-labeling assay
The terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay was performed using the same brain tissues used in histologic verification. The color was developed using 3,3-diamino-benzidine tetra chloride. The sections were xylene-and ethanol-treated for paraffin removal and for dehydration. They were then washed with PBS and incubated in 3% H2O2 solution for 20 min. The sections were treated with 5 μg/mL proteinase k for 2 min at 26°C and rewashed with PBS (0.1 M, pH 7.4). The sections were then treated with a TUNEL reaction mixture (terminal deoxynucleotidyl transferase nucleotide; Roche, Mannheim, Germany) at 37°C for 1 h, and then the sections were washed with distilled water. They were then re-incubated in an anti-fluorescein antibody-cojugated with horseradish peroxidase at 26°C for 30 min, rewashed, and then visualized using the avidin-biotin-peroxidase complex technique and 0.05% 3, 3-diamino-benzidine tetrachloride (Sigma Chemical) as a chromogen. The numbers of TUNEL-positive cells were pathologist-counted in 30 fields/sections (×200 magnification). The blinding was performed for the pathology grading of the results.
The concentrations of TNF-α, IL-1β, IL-6 and IL-10 in the serum were determined using double-antibody sandwich enzyme-linked immunosorbent assay (R&D systems, Minneapolis, MN, USA) according to the manufacturer’s instructions.
Adjacent 50-μm sections, corresponding to coronal coordinates 0.20–0.70 mm anterior to the bregma, were incubated in 2 mol/L HCl for 30 min, rinsed in 0.1 mol/L boric acid (pH 8.5) for 3 min at 26°C, and then incubated with primary antibodies in PBS containing 0.5% normal bovine serum at 4°C overnight. The following primary antibodies were used in the present study: mouse anti-glial fibrillary acidic protein (GFAP for astrocytes), mouse anti-ionized calcium-binding adapter molecule 1 (Iba1; for microglia) and mouse anti-neuronal-specific nuclear protein (NeuN for neurones). The sections were then detected with Alexa-Flour 568 goat anti-mouse (IgG) antibody. The number of labeled cells was calculated in a coronal sections from each rat and expressed as the mean number of cells per section. For negative control sections, all procedures were performed in the same manner without the primary antibodies.
The data are presented as mean ± SD. The repeated measures analysis of variance was conducted to test the treatment-by-time interactions and the effect of treatment over time on each score. The Duncan’s multiple range test was used for post hoc multiple comparisons among means. p < 0.05 was considered evidence of statistical significance.
Etanercept did not affect physiological parameters changes during TBI
The physiological responses (including arterial blood pressure, heart rate and core temperature) observed for (TBI plus saline)-treated group were insignificantly different from (TBI plus etanercept)-treated group. (The data are not shown here).
Etanercept attenuated increased hippocampal levels of cellular hypoxia and damage markers during TBI
The values for hippocampal levels of glutamate, lactate-to-pyruvate ratio and glycerol in (TBI + saline)-treated rats were all significantly higher at 20–120 min after the start of TBI than they were for the sham-operated controls (Fig. 1). Resuscitation with etanercept immediately after TBI significantly attenuated the TBI-induced alteration in microdialysis concentrations of cellular ischemia (e.g., glutamate and lactate-to-pyruvate ratio) and damage (e.g., glycerol) markers in hippocampus (Chio et al. 2007; Kuo et al. 2007; Chen et al. 2009).
Etanercept improved motor and cognitive function during TBI
At 4 days after TBI, behavioral tests revealed that the (TBI plus saline)-treated rats had significantly lower performance in both motor and cognitive function tests than they were for sham-operated controls (Fig. 2). The TBI-induced motor and cognitive dysfunction could be significantly reduced by etanercept treatment.
Etanercept decreased contusion during TBI
The TTC-stained sections at 4 days after TBI showed a significant decrease in the contusion area of the etanercept-treated TBI group (Fig. 3). Both the cortical and hippocampal contusion areas were distinctly smaller in the etanercept-treated rats than in saline-treated TBI ones.
Etanercept attenuated systemic inflammation during TBI
The serum levels of TNF-α, IL-1β and IL-6 decreased and IL-10 increased, respectively, in the etanercept-treated TBI group (p < 0.05; Fig. 4) compared with the (TBI + saline)-treated group at 4 days after TBI.
Etanercept decreased neuronal and glial apoptosis during TBI
Both the TUNEL-stained and NeuN-stained sections at 4 days after TBI revealed a significant decrease of the numbers of TUNEL-NeuN double positive cells in the ischemic brain of the Etanercept-treated TBI group (Fig. 5). In addition, both the TUNEL-stained and GFAP-stained sections at 7 days after TBI revealed a significant decrease in the numbers of TUNEL-GFAP double positive cells in the ischemic regions of the etanercept-treated TBI group (Fig. 6).
Etanercept decreased TBI-induced gliosis in core region during TBI
The GFAP-stained sections at 4 days after TBI revealed a significant decrease of the numbers of GFAP-positive cells in the contusion region of the etanercept-treated group compared with the (TBI + saline)-treated group (Fig. 7). In particular, the GFAP-positive cells in the contusion area of the vehicle-treated group became swollen, which could be reduced by etanercept therapy. In addition, Iba1-stained sections at 4 days after TBI showed a significant decrease of the numbers of Iba1-positive cells of the etanercept-treated TBI group compared with the (TBI + saline)-treated group (Fig. 8).
Antagonism of TNF-α with etanercept has proved to be effective in the treatment of acute spinal cord injury (Genovese et al. 2006) and centrally recombinant interleukin-1 or endotoxin-induced brain injury (Campbell et al. 2007). The present results further showed that systemic delivery of etanercept significantly improved outcomes of TBI in rats. It should be stressed that the etanercept doses used in the current experimental set-up, namely 5 mg/kg i.p. per 12 h for three consecutive days (total dose of 30 mg/kg) are far higher than the normal subcutaneous dose used for rheumatoid arthritis (about 0.4 mg/kg per 3 days) (Elliott et al. 1994). The experimental dose is therefore more than 70 times higher than the usual human therapeutic dose, which one might expect to result in a higher dose of etanercept delivered into the CSF, as there is experimental data that suggests that ∼ 0.1–0.5% of the dose of a systemically administered large molecule will reach the CSF (Banks 2004). Therefore, administration of a dose of etanercept 70 times larger than the therapeutic dose in humans might result in significant penetration of etanercept into the CSF; particularly in an experimental setting such as that involving TBI, in which the blood-cerebrospinal fluid barrier might be damaged. Nevertheless, the present study is valuable and provides scientific data that supports a potential therapeutic role of etanercept in TBI. However, methods of administration that may enhance delivery of etanercept into the CSF (such as perispinal administration) (Tobinick 2010) may be necessary to provide adequate and selective delivery of etanercept to the CSF for therapeutic purposes. Our data are supported by several previous reports. For example, substantial basic science and clinical evidence suggests that excess TNF-α is centrally involved in the pathogenesis of Alzheimer’s disease (Tobinick and Gross 2008). Etanercept, a biologic antagonist of TNF-α, delivered by perispinal administration over a period of 6 months, has been reported to induce rapid cognitive improvement in a patient with late-onset Alzheimer’s disease. A well characterized mouse model where a systemic injection of interleukin-1β during the first five postnatal days (inflammatory insult) is combined with an intracerebral injection of the glutamatergic analogue ibotenate (excitotoxic insult) at postnatal day 5 was used (Aden et al. 2010). It was found that TNF-α blockade by etanercept given after the combined inflammatory and excitotoxic insult reduced brain damage by 50%. Using a rat model, it was found that systemic administration of etanercept prior to IL-1β-microinjection into the anterior hypothalamus inhibited certain IL-1β-mediated sickness behavior, such as the depression of open-field activity and reduced glucose consumption (Jiang et al. 2008). Intracerebroventricular administration of a selective TNF-α antagonist-soluble TNF-α receptor fusion protein at 15 min before and 1 h after TBI, improved performance in a series of standard motor tasks after injury (Knoblach et al. 1999).
In the current results, the TBI-induced overproduction of IL-1β, TNF-α and IL-6 in serum was significantly reduced by etanercept. In contrast, etanercept therapy significantly increased the serum levels of IL-10 during TBI in rats. In fact, IL-10 had important anti-inflammatory properties through inhibiting the production of the pro-inflammatory cytokines TNF-α, IL-1β and IL-6, while upgrading the expression of IL-1 receptor antagonist (Cartmell et al. 2003). Elevated plasma IL-10 levels had been reported in animals with lipopolysaccharide injection (Wang et al. 2001) and in patients with sepsis (van der Poll et al. 1997). In addition, systemic delivery of recombinant IL-10 protected mice form lethal endotoxaemia via reducing TNF-α release (Marchant et al. 1994). Neutralization of endogenously produced IL-10 resulted in an increased production of several proinflammatory cytokines and enhanced mortality in endotoxaemic mice (Standiford et al. 1995). IL-10-knockout mice had also been shown to have an increased likelihood of inflammatory bowel disease (Rennick et al. 1997), a higher sepsis-induced mortality rate (Berg et al. 1995) and an exacerbated and prolonged endotoxin-induced fever (Leon et al. 1999). Collectively, these data implied that etanercept may have improved outcomes of TBI in rats by inhibiting overproduction of pro-inflammatory cytokines.
Our data are confirmed by many previous results. Immediate anti-TNF-α (etanercept) therapy enhanced the rate of axonal regeneration after nerve injury in rats (Kato et al. 2010). TNF-α inhibited neurite outgrowth of cultured dorsal root ganglia (Larsson et al. 2005) and hippocampal (Neumann et al. 2002) neurons. An estrogen sulfate ameliorated systemic inflammation during TBI by decreasing TNF-α but increasing IL-10 production in serum (Chen et al. 2009). In addition, indomethacin enhanced neurogenesis after experimental stroke (Hoehn et al. 2005) and reduced intracranial hypertension in patients of TBI (Rosmussen 2005). Endotoxin-induced inflammation impaired neurogenesis (Ekdahl et al. 2003), whereas blockade of inflammation restored neurogenesis in adult brain following ischemic insult (Hoehn et al. 2005). IL-10 had direct neuronal effects with important implications for neuroprotection (Zhou et al. 2009).
In addition to inflammation, glial scar formation was a significant blockade to actual functional neuronal regeneration. An excellent recent review (Silver and Miller 2004) had proposed that to overcome the inhibitory environment of the glial scar might allow long-distance functional regeneration after TBI. Indeed, as demonstrated in the current results, the TBI-induced gliosis of both astrocytes and microglia were significantly reduced by etanercept therapy. Apparently, etanercept might enhance actual functional neuronal regeneration via inhibiting glial scar formation during TBI.
It has been well documented that the lactate/pyruvate ratio and glutamate are well-known markers of cellular ischemia, whereas glycerol is a marker of how severely cells are affected by the ongoing pathology (Chou et al. 2003). Excessive concentrations of glutamate have been noted in ischemic brain tissue (Bullock et al. 1995; Nilsson et al. 1996). Our previous study (Chio et al. 2007) showed that the animals with TBI had higher values of extracellular levels of glutamate, lactate-to-pyruvate ratio, and glycerol in ischemic cortex and intracranial pressure, but lower values of cerebral perfusion pressure. In the present study, we further demonstrated that etanercept therapy significantly attenuated the TBI-induced increased cerebral ischemia and injury markers in the ischemic cortex as early as 20 min post-injection. Etanercept might cause attenuation of the TBI-induced cortical ischemia and damage via reducing intracranial hypertension and cerebral hypoperfusion that occurred during the early phase of TBI. It should be stated that our experimental results, using doses 70 times larger than the usual human therapeutic dose, cannot properly be extrapolated to infer that etanercept administered systemically at its usual therapeutic dose can be used to treat TBI in humans. In addition, levels of etanercept in the CSF following systemic administration of etanercept should be measured in future studies.
In summary, the following behavioral, biochemical and histopathologic characteristics in the contusion brain were noted after TBI: (i) motor and cognitive dysfunction; (ii) cerebral ischemia and damage (evidenced by increased levels of cellular glutamate, lactate-to-pyruvate ratio and glycerol) (Chio et al. 2007; Kuo et al. 2007; Chen et al. 2009); (iii) cerebral contusion; (i.v.) systemic inflammation (evidence by increased levels of serum IL-1β, TNF-α and IL-6); (v) neuronal and glial apoptosis; and (vi) gliosis of both astrocytes and microglia. All these destructive characteristics were favorably influenced by etanercept therapy during TBI. In particular, etanercept may improve the outcomes of TBI by penetrating into the cerebrospinal fluid.
This study was funded in part by the National Science Council of the Republic of China (Grant Nos. NSC 98-2314-B-218 -001 -MY2 and NSC 99-2314-B-384 -004 -MY3, and DOH99-TD-B-111-003, Center of Excellence for Clinical Trial and Research in Neuroscience.