TT-301 Inhibits Microglial Activation and Improves Outcome after Central Nervous System Injury in Adult Mice

• Microglial inhibition may reduce secondary tissue injury and improve functional outcome following acute brain injury

  However, specific pharmacological strategies have not yet been developed

• A putative microglial inhibitor, TT-30 improved histological and functional parameters in murine models of traumatic brain injury and intracerebral hemorrhage

THE acute central nervous system (CNS) response to injury, such as traumatic brain injury (TBI) and intracerebral hemorrhage (ICH), is characterized by activation of microglia and astrocytes, which leads to acute up-regulation of proinflammatory cytokines, production of free radicals, and neuronal excitotoxicity.1,2These neuroinflammatory responses contribute to breakdown of the blood brain barrier (BBB), development of cerebral edema, and secondary neuronal injury, resulting in progressive intracranial hypertension, cerebral hypoperfusion, uncal herniation, and death, if left unchecked. In contrast, during the subacute or chronic period after injury, it has been postulated that glial activation may play an adaptive role by releasing neurotrophic mediators that promote adaptive synaptogenesis and repair.3,4Thus, modulation of acute neuroinflammatory pathways may represent a target for therapeutic development. However, nonspecific approaches designed to blunt the inflammatory response, such as the administration of glucocorticoids, have shown no benefit in the setting of acute brain injury.5Further, no pharmacological strategies have been shown to improve outcomes after acute brain injury in humans, and, presently, there remains a compelling unmet clinical need.

In the current study, we test the efficacy of TT-301, (6-phenyl-4-pyridin-4-yl-3-[4-pyrimidin-2-yl-piperazin-1-yl]-pyridazine hydrochloride) in clinically relevant murine models of TBI and ICH. TT-301 is a brain-penetrable, small molecule that was built on an aminopyridazine scaffold and rationally designed from a related analog, minozac. Minozac attenuates cytokine induction from activated microglia in culture and protects against neuronal synaptic dysfunction and cerebral edema in rodent models of chronic neuroinflammation.6,7Thus, although the exact mechanism is currently unknown, it is plausible that TT-301 may down-regulate the acute inflammatory response to CNS injury by inhibiting recruitment of activated microglia and suppressing their production of proinflammatory cytokines.

We tested the hypothesis that TT-301 would suppress microglial activity in the acute phase of neurologic injury, be associated with reduction in secondary neuronal injury, and improve neurologic function. As a step toward addressing the need for novel therapeutic approaches in this area, we explored the potential utility of TT-301 administered postinjury in our murine models of closed-skull impact TBI4and collagenase-induced ICH.3 

All procedures were performed under protocols approved by the Duke University Institutional Animal Care and Use Committee, Durham, North Carolina.

Physiologic measurements for arterial blood gas and glucose at 15 min before and 1 h after TT-301 (1 mg/kg) delivery via  intraperitoneal injection with concomitant arterial blood pressure and rectal temperature recorded continuously 15 min before, during, and after treatment for 1 h (n = 6).

Quantitative stereology for F4/80 (microglia), glial fibrillary acidic protein (GFAP; astrocytes), neuronal nuclei (NeuN; neurons), and Flouro Jade B (FJB; degenerating neurons) positive cells at baseline (preinjury), and at 1, 10, and 28 days after TBI in mice treated with TT-301 (1 mg/kg) or vehicle administered via  intraperitoneal injection at 30 min and 6 h after injury and continued twice daily for 5 days (n = 6).

Functional performance of mice treated with TT-301 (1 mg/kg) or vehicle administered via  intraperitoneal injection at 30 min and 6 h after TBI and continued twice daily for 5 days; testing included Rotorod over first 7 days and Morris water maze (MWM) at 4 weeks after injury (n = 12).

Functional performance of mice after prolonged treatment with TT-301 (1 mg/kg) or vehicle administered via  intraperitoneal injection at 30 min and 6 h after TBI and continued twice daily for 5 or 28 days after injury; testing included Rotorod over first 7 days, MWM at 4 weeks after injury, and F4/80 staining at day 28 (n = 12).

Determination of therapeutic window through functional performance of mice treated with TT-301 (1 mg/kg) or vehicle administered via  intravenous injection at 30 min, 3 h, or 6 h after TBI with a second intraperitoneal dose at 6 h after the first dose and then continued twice daily via  intraperitoneal injection for 5 days; testing included Rotorod over first 7 days and MWM at 4 weeks after injury (n = 12).

Extrapolation to other acute CNS injury mechanisms through functional performance of mice treated with TT-301 (1 mg/kg) or vehicle administered via  intraperitoneal injection at 30 min and 6 h after ICH and continued twice daily for 5 days after injury; testing included Rotorod over first 7 days (n = 12) and volumetric assessment of hematoma lesion and brain water content at 24 h (n = 6) after injury.

Determination of differential gene expression in mice treated with TT-301 (1 mg/kg) or vehicle administered via  intraperitoneal injection at 30 min and 6 h after TBI and continued twice daily for 5 days. Four treatment groups (n = 3) sacrificed 10 days post-TBI consisted of Control Vehicle (no injury, vehicle-treated), Control TT-301 (no injury, TT-301-treated), TBI 10 days TT-301 (TBI injury, TT-301-treated), and TBI 10 days Vehicle (TBI injury, vehicle-treated).

Our murine TBI model8was adapted from a previously described model of closed cranial trauma for the rat.4C57BL/6J mice (Jackson Laboratory, Bar Harbor, ME) were randomized to receive drug or vehicle immediately before injury. After tracheal intubation under 4.6% isoflurane, the lungs were mechanically ventilated with 1.6% isoflurane in 30% O²and 70% N². Rectal temperature was maintained at 37 ± 0.2°C. After securing the animal's head, a stereotactic device, a concave 3-mm metallic disc, was adhered to the skull immediately caudal to bregma. A 2.0-mm diameter pneumatic impactor (Air-Power, Inc., High Point, NC) was used to deliver a single midline impact to the disc surface. Analgesic regimens consisted of intraoperative administration of subcutaneous bupivacaine (0.125%, 0.2 ml) at the site of incision. Animals were allowed to recover spontaneous ventilation and then extubated with free access to food and water.

Our murine ICH model3,9was adapted from a previously described model of ICH in rats.10Briefly, C57BL/6J mice (Jackson) mice were randomized before treatment or vehicle groups. After tracheal intubation under 4.6% isoflurane, the lungs were mechanically ventilated with 1.6% isoflurane in 30% O²and 70% N². Rectal temperature was maintained at 37 ± 0.2°C by underbody circulating warm waterbed. After securing the animal's head in a stereotactic frame and a burr hole placement was created 2 mm left lateral to bregma, a 0.5 μl syringe needle (Hamilton, Reno, NV) was advanced to a depth of 3 mm from cortex. Type IV-S Clostridial collagenase (Sigma–Aldrich, St. Louis, MO) was injected over 5 min (0.1 U in 0.4 μl normal saline). The incision was then closed, and animals were extubated with free access to food and water. Analgesic regimens consisted of intraoperative administration of subcutaneous bupivacaine (0.125%, 0.2 ml) at the site of incision.

Mice were anesthetized in a chamber with 5% isoflurane in 30% O²and 70%N². The trachea was intubated with a 20-gauge Insyte-W intravenous catheter (Becton-Dickinson, Sandy, UT). The inspired isoflurane concentration was reduced to 1.6%, and the lungs were mechanically ventilated at a rate of 110 breaths/min with a delivered tidal volume of 0.70 ml. Rectal temperature was monitored and servoregulated with a surface heating/cooling system to a target of 37.0 ± 0.2°C. The right femoral artery was cannulated (PE10 catheter, Becton-Dickinson, Sparkes, MD) to monitor arterial blood pressure and to collect arterial blood samples. Blood gas and glucose were measured at 15 min before and at 1 h after drug delivery by ABL800 FLEX (Radiometer, Copenhagen, Denmark). Arterial blood pressure and rectal temperature were recorded continuously before, during, and after treatment for 1 h by Chart V3.3.8 program (MacLab, ADInstruments Pty Ltd., Bella Vista, Australia).

TT-301 (6-phenyl-4-pyridin-4-yl-3-[4-pyrimidin-2-yl-piperazin-1-yl]-pyridazine hydrochloride) has an empirical formula of C²³H²¹N⁷and a molecular weight of 395.47 (free-base). The product is available as a sterile concentrated solution in 0.9% NaCl at 2.5 mg base/ml. Before injury, mice were randomized to receive TT-301 (1 mg/kg) or vehicle (sterile saline) via  intraperitoneal injection at 30 min and 6 h after injury and then twice daily for 5 or 28 days, except in the experiment to determine therapeutic window, for which mice received the first dose via  IV injection at predefined intervals after injury (See Experimental Animal Groups). Drug or vehicle was administered in a volume of 100 μl sterile normal saline.

As previously described,11brains were dissected in the midsagittal plane after intracardial perfusion under anesthesia. One hemisphere was flash-frozen in liquid nitrogen and stored at −80°C; the other hemisphere was immersion fixed in 10% formaldehyde for 24 h, transferred into 1× phosphate-buffered saline, and stored at 4°C. For immunostaining, tissue was microwaved in saline sodium citrate and antigen retrieval buffer (Vector Labs, Burlingame, CA) according to the manufacturer's instructions, then incubated in 1% H²O², permeablized by 0.1% Saponin, and blocked with 10% goat serum. Brain sections were mounted using DPX mounting media (Fluka, Milwaukee, WI). FJB staining was performed using standard protocols.12Antimouse F4/80 antibody (rat monoclonal, 1:10,000; Serotec, Raleigh, NC), NeuN antibody (mouse monoclonal, 1:30,000; Chemicon, Temecula, CA), and GFAP (polyclonal, 1:3,000, Dako, Glostrup, Denmark) were used for immunohistochemical staining.13 

Cell counting was conducted using a Nikon 218912 light microscope interfaced with the StereoInvestigator software package (MicroBrightField, Williston, VT). The number of stained cells per volume of hippocampus was estimated by using the optical fractionator method.14The optical fractionator is an unbiased counting method, which is independent of the size, shape, and orientation of the cells to be counted. The parameters of the fractionator-sampling scheme were established in a pilot experiment and were uniformly applied to all animals. Before counting, all slides were coded to avoid experimenter bias. As determined by StereoInvestigator, we chose six sagittal sections (40 μm) spaced eight sections apart along the dorsal hippocampal formation by systematic random sampling. This number of sections proved sufficient to provide a coefficient of error between 0.09–0.11. On each section, the whole hippocampal area was delineated. For microglial quantification, the sampling grid was 399.027 (X) × 367.92 (Y) μm, and cells were counted within a probe volume defined by the counting frame (80 × 80 μm) and the dissector height (11 μm). For neuron quantification the sampling grid was 168.36 (X) × 111.84 (Y) μm, and cells were counted within a probe volume defined by the counting frame (50 × 50 μm) and the dissector height (11 μm). Only cells within the counting frame or overlapping the right or superior border of the counting frame, and for which nuclei came into focus while focusing down through the dissector height, were counted. The total number of F4/80, NeuN, FJB, or GFAP immunopositive cells was calculated per hippocampal volume of 1,920 μm thickness.

An automated Rotorod (Ugo Basile, Comerio, Italy) was used to assess the effects of therapeutic intervention on vestibulomotor function.15On the day before injury, mice underwent two consecutive conditioning trials at a set rotational speed (16 revolutions/min) for 60 s, followed by three additional trials with an accelerating rotational speed. The average time to fall from the rotating cylinder in the latter three trials was recorded as baseline latency. After injury, mice underwent consecutive daily testing for TBI or every-other-day testing for ICH with three trials of accelerating rotational speed (intertrial interval of 15 min). Average latency to fall from the rod was recorded. Mice unable to grasp the rotating rod were given a latency of 0 s.

As in our prior publication,9the MWM16was used to assess the effects of therapeutic intervention on spatial learning and memory at 4 weeks after injury by an examiner blinded to treatment assignment. Before injury, mice were trained on the visible platform (1 day; platform flagged, located in a different quadrant for each trial to minimize quadrant habituation, and no extra-maze visual cues) and hidden platform (4 days; platform submerged in western quadrant for all trials with several extra-maze visual cues) versions of the MWM task to habituate the mice to handling and to swimming, as well as to teach them the goal of the task, which was to escape from the water by climbing onto a platform. After injury, performance was evaluated in a black aluminum pool (105 cm in diameter, 60 cm in depth) filled with water opacified with powdered milk containing a platform (7.5 cm in diameter) submerged 1 cm below the water surface (25–27°C). The maze was kept in a room dedicated to behavioral testing to decrease stress.17A testing day consisted of one of four different quadrants for each trial randomly ordered each day and an intertrial interval of 20 min. Mice were allowed to search for the platform for 90 s. If unable to locate the platform, they were guided to it. Computerized video tracking system (KeilSoft LLC, Chapel Hill, NC) recorded latency to find the platform and swimming speed.

On the final day, a probe trial was conducted, which consisted of removal of the escape platform and release of the mouse at a point diagonally opposite from the previous location of the platform. The time spent searching all four quadrants and the number of crossings into the western quadrant were recorded.

Mice were anesthetized and euthanized at 24 h after injury with ICH. Brains were dissected and sectioned midsagittally with removal of cerebellum and brainstem, and each hemisphere weighed immediately (“wet” weight). Hemispheres were allowed to dehydrate over 24 h at 100°C and then reweighed (“dry” weight). Cerebral edema was expressed as water content calculated as a percentage of wet weight (wet weight − dry weight)/(wet weight) × 100.18 

At 24 h after injury with ICH, mice were euthanized, and brains were removed and frozen at −20°C. Coronal sections of 20 μm thickness were taken at 320 μm intervals over the rostral-caudal extent of the lesion. Sections were stained with hematoxylin and eosin, and lesion volume measured by digitally sampling stained sections with an image analyzer (M2 Turnkey System, Imaging Research, Inc., St. Catharines, Ontario, Canada). Lesion volumes (mm³) were computed as running sums of lesion area multiplied by the known interval (i.e. , 320 μm) between sections over the extent of the lesion expressed as an orthogonal projection.

The mouse BV2 microglial cell line was provided by Joanne McLaurin (University of Toronto, Ontario, Canada) and was maintained in Dulbecco's Modified Eagle Medium (Invitrogen, Carlsbad, CA) containing 5% fetal bovine serum, 100 U/ml penicillin, and 100 mg/ml streptomycin. The mouse epithelial ovarian cancer (EOC) 20 microglial cell line (American Type Culture Collection, Manassas, VA, cat. no. CRL-2469) was maintained in Dulbecco's Modified Eagle Medium containing 10% fetal bovine serum and 20% conditioned medium derived from the bone marrow cell line LADMAC (American Type Culture Collection, cat. no. CRL-2420). EOC20 cells are dependent on growth factor colony stimulating factor-1, which is secreted by the LADMAC cells. LADMAC cells were maintained in Modified Eagle Medium containing 10% fetal bovine serum, 100 U/ml penicillin, and 100 mg/ml streptomycin. Conditioned medium containing colony stimulating factor-1 was collected from confluent LADMAC cells after 5–7 days.

BV2 cells were plated at a density of 2 ×10⁵cells per well into 12-well plates and allowed to grow for 1 day. EOC20 cells were plated at a density of 4 ×10⁵cells per well into 12-well plates and allowed to grow for 2 to 3 days. For both microglial lines, growth medium was aspirated and cells were washed with serum-free medium before the addition of serum-free medium containing TT-301 at various concentrations. Control wells contained the same final concentration of dimethyl sulfoxide (0.5%) as compound containing wells. After 90 min of compound incubation, vehicle phosphate-buffered saline or lipopolysaccharide (10 mg/ml Escherichia coli  0111:B4; Sigma-Aldrich) was added to appropriate wells and incubated for a further 18 h. At the end of the 18-h incubation, media was removed from the wells, centrifuged (1,000 x g) for 10 min, and stored at −20°C until analysis.

TT-301 was dissolved in dimethyl sulfoxide to prepare a 100 mM stock. The 100 mM stock was serially diluted in dimethyl sulfoxide to prepare working solutions at 200× the concentration to be tested (1, 10, 25, and 60 μM). Compound stocks were freshly diluted into serum-free medium just before use in cell culture.

The concentrations of tumor necrosis factor-α, interleukin-6, and monocyte chemotactic protein-1 in conditioned media were measured using Quantikine ELISA kits (R&D Systems; Minneapolis, MN) according to the manufacturer's instructions. The concentration of interleukin-1β in cell lystates was determined using DuoSet ELISA kits (R&D Systems). Briefly, BV2 and EOC20 cells were washed once with cold phosphate-buffered saline before lysis on ice in 20 mM Tris pH 7.4, 150 mM NaCl, 0.5% NP-40 (Igepal CA-630, Sigma–Aldrich), and protease inhibitor cocktail III (Calbiochem, San Diego, CA). Cells were scraped, collected, and centrifuged (14,000 x g) for 10 min. Supernatants were quantified for protein concentrations using bicinchoninic assay (Pierce ThermoFisher Scientific, Rockford, IL) and stored at −20°C until analysis.

Microglial cell lines (BV2 and EOC20) were plated at a density of 5 ×10⁴cells per well into 96-well plates and allowed to grow for 2 days. Growth medium was aspirated and replaced with serum-free medium containing compounds at the test concentrations described above. After 18 h of incubation, cell viability was assessed using Cell Titer 96 AQ^ueousOne Solution Cell Proliferation Assay (Promega, Madison, WI), according to the manufacturer's instructions. Absorbance (490 nm) was measured using a SPECTRAmax Microplate Reader (Molecular Devices, Sunnyvale, CA).

Frozen, pulverized whole brain tissue was processed for RNA extraction (12 samples total, n = 3) according to manufacturer instructions (RNeasy Lipid Tissue Mini Kit, Qiagen, Valencia, CA). Four treatment groups, sacrificed 10 days after TBI, were assessed and compared: Control Vehicle (no injury, vehicle-treated), Control TT-301 (no injury, TT-301-treated), TBI 10 days TT-301 (TBI injury, TT-301-treated), and TBI 10 days Vehicle (TBI injury, vehicle-treated). RNA quantity and quality was assessed with the NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Inc., Wilmington, DE) and by agarose gel electrophoresis. Only samples with a 260/280 ratio between 1.9–2.1, and a 260/230 ratio greater than 2.0, were further processed. First strand complementary DNA was generated from 2 μg total RNA using the RT²First strand kit (SABiosciences, Frederick, MD), according to the manufacturer's instructions. Gene expression was measured using the Mouse Neurotrophin and Receptors polymerase chain reaction (PCR) Array (SABiosciences), which profiles the expression of 84 genes related to neuronal processes. RT-PCR was performed according to manufacturer's instructions using the 384-well plate format (4 samples, 96 wells per sample). One sample from each experimental group was run per plate to minimize potential batch effect between RT-PCR runs. Quality of the complementary DNA and PCR efficiency was verified by housekeeping genes and RT-PCR controls included in the PCR Array.

Raw RT-PCR data were analyzed using the Web-Based PCR Array Data Analysis software (SABiosciences). ΔC^tvalues and ΔΔC^t-based fold-change were calculated from raw threshold cycle data, using β-actin and glyceraldehyde 3-phosphate dehydrogenase as internal standards for normalization. Correlation, uncentered average linkage hierarchical cluster analysis was performed on gene expression values and visualized using Cluster/Treeview software programs (Eisen Software, Berkeley, CA). Gene pathway analysis was conducted using Gene Network Central Pro (SABiosciences).

Rotorod performance and MWM latencies were compared with two-way repeated-measures ANOVA with day as the repeated variable. The F  values were calculated, and if the probability distribution of F  with the appropriate degrees of freedom suggested a significant group effect. In this case, the two independent treatment factors (treatment effect and time) each had two or more levels. The interpretation of significance implies that the F-value demonstrated significance for treatment effect as well as time effect. Pairwise testing was performed between groups using Scheffé F test post hoc  method to correct for multiple comparisons. The numbers of FJB, F4/80, GFAP, and NeuN cell counts were compared between groups by the Mann–Whitney U test. An independent, two-tailed Student t  test was used to compare mortality, hemorrhage volume, and brain water content after testing for normality of the data. Statistical significance was assumed with P < 0.05. All values were expressed as mean ± SEM (SEM) and were performed on JMP (v7.0.1, SAS, Cary, NC). There were no data excluded, lost to observation, or missing from analyses.

To assess the physiologic impact of TT-301 treatment in these murine models, we performed hemodynamic monitoring with blood sampling before and after administration of TT-301 or vehicle. No difference was seen in pH, PaCO², P^aO², glucose, mean arterial pressure, or temperature before or after treatment at 15 min and 1 h. P^aO²was controlled at 163 ± 13.89 mmHg, PaCO²at 38.5 ± 10.55 mmHg, pH at 7.36 ± 0.02, glucose 156.75 ± 46.39 gm/dl, and mean arterial pressure at 80.43 ± 4.97 mmHg. There were no changes in any parameters over the 1 h of monitoring.

To investigate whether administration of TT-301 twice daily for 5 days after TBI was associated with a reduction in glial response to injury, we performed immunohistochemistry with F4/80, a marker of activated microglia and cells of monocyte lineage,19followed by quantitative stereology at baseline (preinjury), and at 1, 10, and 28 day postinjury (fig. 1). No F4/80 staining was present before injury. At 24 h after TBI, F4/80 staining was present, but TT-301-treated animals had a statistically significant reduction in F4/80-positive cells. (figs. 1A–D; P = 0.0481) This effect was more pronounced at 10 days following injury (5 days following the final treatment), when we noted a reduction in F4/80 immunoreactivity in animals previously treated with TT-301 as compared with vehicle (fig. 1G; P < 0.0001). Interestingly, these effects were not present at 28 days, nor was there an effect of TT-301 on astrocytosis as measured by GFAP-positive cells at any time point after TBI (24 h after injury: TT-301 [1 mg/kg]vs.  vehicle: 2,668 ± 172 astrocytes/mm³vs.  2,994 ± 223 astrocytes/mm³, P = 0.6038; figs. 1E and F).

To evaluate whether this reduction in microgliosis resulted in neuroprotection, FJB staining was performed to identify injured neurons in areas of the brain that are preferentially susceptible to closed head injury, such as the dentate gyrus and CA3 region of hippocampus. At 24 h after injury, treatment with TT-301 was associated with a reduction of injured neurons in hippocampus (fig. 1H; P = 0.0301), suggesting that TT-301 conferred short-term neuroprotection. To assess for the possibility of delayed neuronal injury beyond the acute phase of injury, we immunostained brains at 28 days after TBI with NeuN, a neuron-specific marker, and quantified neuronal density in the polymorphic (i.e. , superficial layer of the dentate gyrus) and CA3 regions of the hippocampus. We found that administration of TT-301 for 5 days after TBI resulted in increased surviving neurons at 28 days after injury when compared with vehicle-treated animals as assessed by unbiased stereology (TT-301 [1 mg/kg]vs.  vehicle: 63,025 ± 2190 neurons/mm³vs.  57,585 ± 1012 neurons/mm³, P = 0.0462).

We next evaluated whether the histologic evidence of reduced microgliosis and neuronal injury witnessed after administration of TT-301 was associated with improved functional outcomes. First, vestibulomotor deficit was assessed by measuring Rotorod latency on days 1–7 following closed head injury. Treatment with TT-301 was associated with improved Rotorod performance compared with vehicle treatment; this effect was durable over the 7-day testing period (fig. 2A; P = 0.0225). As neurocognitive deficits are also common sequelae of TBI, we performed MWM testing 4 weeks after TBI, and found treatment with TT-301 to be associated with improved neurocognitive performance on water maze latencies (fig. 2B; P = 0.0110). There was no difference in swimming speeds between groups (TT-301 [1 mg/kg]vs.  vehicle: 29.6 + 4.3 vs.  28.2 + 5.1 cm/s). Further, probe testing confirmed these data, demonstrating that mice receiving TT-301 treatment spent a similar amount of time in the appropriate (west) quadrant of the maze compared with sham animals and statistically significantly more time compared with vehicle-treated animals (TT-301 [1 mg/kg]vs.  sham vs.  vehicle: 62.3 ± 3.4 vs.  64.8 ± 3.0 vs.  34.8 ± 3.1 s; P = 0.0003).

Since the 5-day, twice daily treatment regimen was associated with a reduction of microgliosis at 10 days, but had no effect on microglial numbers at 28 days postinjury, we next sought to establish whether a prolonged treatment regimen would result in durability of microglial suppression with concomitant improved neurologic function. Thus, animals were treated with TT-301 or vehicle for 28 days after injury with TBI, using the same twice daily dosing regimen as the 5-day treatment. Interestingly, there remained no reduction in hippocampal microgliosis at 28 days postinjury, despite being treated up to the day of sacrifice (7,221 ± 550 microglia/mm³in animals treated with TT-301, vs.  6,518 ± 533 microglia/mm³in vehicle-treated animals.) This was consistent with functional testing, in which animals treated for 28 days did not demonstrate a greater treatment effect compared with 5-day treated animals on MWM (P = 0.3184), although both demonstrated greater treatment effects compared with vehicle-treated animals (fig. 3; P < 0.0187). This was again confirmed by probe testing where TT-301-treated animals, regardless of dosing regimen, spent statistically significantly greater time in the appropriate (west) quadrant of the maze when compared with vehicle-treated (TT-301 [1 mg/kg] 5-day vs.  28-day v. vehicle: 54.1 ± 3.2 vs.  57.7 ± 3.0 vs.  45.0 ± 4.0 s; P = 0.0491), despite similar swimming speeds.

To address the time window for which treatment might be effective postinjury, we repeated the above 5-day, twice daily treatment regimen of TT-301 after TBI but administered the first dose up to 6 h after injury, as opposed to 30 min in prior experiments. In this paradigm of delayed initiation of dosing, administration of TT-301 retained statistically significant functional benefit when compared with vehicle-treated animals, as demonstrated by improved performance on the Rotorod and MWM testing (fig. 4; P < 0.0001). (All TT-301-treated groups [initiation of dosing at 30 min, 3 h, and 6 h] retained similar latencies and demonstrated statistically significant improvement compared with vehicle-treated animals [data not shown; P = 0.0030].) MWM testing was confirmed by probe testing demonstrating that TT-301 treatment initiated at 6 h after injury resulted in mice spending statistically significantly greater time in the appropriate (west) quadrant of the maze when compared with vehicle-treated animals (TT-301 [1 mg/kg]vs.  vehicle: 56.8 ± 2.3 vs.  44.8 ± 3.0 s; P = 0.0032), despite similar swimming speeds.

Since the repertoire of CNS responses to injury is limited, we next addressed the question of whether the benefit from TT-301 administration may be extrapolated from a model of TBI to a model of ICH, which is also associated with microglial activation, BBB breakdown, cerebral edema, and secondary neuronal injury.20Using the same postinjury twice daily, 5-day treatment regimen as in the TBI experiments, administration of TT-301 had no effect on the development or volume of the intracerebral hematoma at 24 h following the injection of collagenase (TT-301 [1 mg/kg]vs.  vehicle: 0.689 ± 0.08 mm³vs.  0.655 ± 0.1 mm³; P = 0.8255). However, treatment with TT-301 was associated with a statistically significant decrease in cerebral edema, as quantified by brain water content at 24 h postinjury (TT-301 [1 mg/kg]vs.  vehicle: 61.52 ± 0.03% brain water vs . 63.64 ± 0.02% brain water; P = 0.0064). Moreover, treatment of TT-301 was associated with a reduction in vestibulomotor deficit, as assessed by Rotorod testing. This effect was sustained over the 7-day testing period (fig. 5; P = 0.0288) and was consistent with the effects of TT-301 in reducing neuroinflammatory responses and improving neurologic function found in our model of TBI.

In order to corroborate our in vivo  findings demonstrating the antiinflammatory potential of TT-301 on microglial activation, the pharmacological effects of TT-301 were investigated in vitro  in two mouse microglial cell lines, BV2 and EOC20. In these studies, concentrations of 1, 10, 25, and 60 μM of TT-301 were investigated. Cytokine ELISA analysis of BV2 microglial cells showed that, in comparison with vehicle-treated cells, increasing concentrations of TT-301 inhibited the lipopolysaccharide-induced increase of the early cytokine marker tumor necrosis factor-α in a dose-dependent fashion (fig. 6A). Treatment of activated BV2 microglia with TT-301 also resulted in concentration-dependent suppression of additional lipopolysaccharide-induced proinflammatory cytokines and chemokines, including interleukin-1β, monocyte chemotactic protein-1, and interleukin-6. The same concentrations of TT-301 showed no suppression on the induction of antiinflammatory cytokines (e.g. , interleukins and so forth) in lipopolysaccharide-stimulated BV2 cells or in Aβ-stimulated primary microglia, suggesting that TT-301 suppressive action is not a generalized effect (data not shown). Cell toxicity assays, performed in parallel, indicated that TT-301 inhibited lipopolysaccharide-induced proinflammatory cytokine levels at concentrations that were without effect on the metabolic integrity of BV2 microglia (data not shown). The effect of TT-301 on proinflammatory cytokine production in a second microglial line, EOC20, was investigated. Compared with vehicle treatment, treatment of lipopolysaccharide-activated EOC20 microglia with TT-301 resulted in concentration-dependent suppression of proinflammatory cytokine release, similar to the results observed in lipopolysaccharide-activated BV-2 cells (fig. 6B). Routine toxicity assays run in parallel indicated that exposure of EOC20 microglia to TT301 did not compromise cell viability (data not shown), suggesting that TT301 specifically inhibited lipopolysaccharide-stimulated cytokine production and/or secretion in this line.

Although the exact mechanism by which TT-301 suppresses proinflammatory cytokines and improves neuronal survival is not known, evidence from TT-301-related analogues6and our data suggest it modulates microglial signaling pathways associated with inflammatory phenotypes. Therefore, gene expression was measured using the Mouse Neurotrophin and Receptors PCR Array (SABiosciences), which profiles the expression of 84 genes involved with inflammation and neurotrophic activity (see Supplemental Digital Content 1, https://links.lww.com/ALN/A842, which is a table listing all genes used in this study). Using the same postinjury, twice daily 5-day treatment regimen as in the prior experiments, four treatment groups, sacrificed 10 days after TBI, were assessed and compared: Control Vehicle (no injury, vehicle-treated), Control TT-301 (no injury, TT-301-treated), TBI 10 days TT-301 (injury, TT-301-treated), and TBI 10 days Vehicle (injury, vehicle-treated). Eighteen genes were differentially expressed by a fold change of 2.0 or greater between the TBI 10 days TT-301 and TBI 10 days Vehicle groups (table 1). Pathway analysis revealed 12 of the differentially expressed genes were directly involved in the Janus kinase–Signal Transducer and Activator of Transcription (JAK-STAT) pathway, and four additional genes have been proposed to interact with or modify components of the JAK-STAT cascade (artemin, neuregulin 1, neurotrophic tyrosine kinase receptor (type 1), neurofibromin 1). Six of the genes interact directly with STAT3, which showed the largest down-regulation in the TBI 10 days TT-301-treated group (fig. 7A and B; values normalized to the Control Vehicle group). Hierarchical cluster analysis of differentially expressed genes demonstrated the highest degree of similarity between Control Vehicle and the TBI 10 days TT-301 groups (fig. 7C). The results of the pathway analysis suggest down-regulation of STAT3 and changes in the JAK-STAT cascade as a potential mechanism of action for TT-301 in our model of TBI.

In the current study, we demonstrate that administration of TT-301 inhibits microglial activation and improves both histologic and functional outcomes in clinically relevant murine models of TBI and ICH. Our data suggest that 5 days of treatment with TT-301 after TBI results in a reduction in microgliosis at 1 and 10 days, but not 28 days following injury. It should be noted that increased numbers of F4/80-positive cells are a consequence of CNS inflammation and may represent both proliferation of resident microglia and hematogenous recruitment of monocytes. In our model, this reduction in F4/80-positive cells was associated with a reduction of short-term hippocampal injury and an increase in long-term neuronal density in the CA3 region of the hippocampus. Consistent with the reduction in hippocampal pathology, treatment with TT-301 was associated with improved long-term neurocognitive performance on MWM testing. Further, this benefit from TT-301 administration was demonstrated in a second form of CNS injury using our model of ICH. In this model, ICH-injured mice demonstrated improved neurologic function by Rotorod latency after treatment with TT-301. Importantly, although there was a primary effect of both time and treatment with TT-301 on improved performance on Rotorod and MWM in both models, there was no interaction between time and treatment effects; thus, although TT-301 treatment improved outcomes throughout the testing period, it did not significantly alter the trajectory of recovery. Further, these in vivo  models and cell culture studies suggest these functional affects are because of microglial inhibition. Finally, the results of the pathway analysis showing down-regulation of STAT3 and changes in the JAK-STAT cascade suggest a potential mechanism of action for the reduction in microglial activation seen at 10 days following administration of TT-301.

TBI is associated with a spectrum of neuroinflammatory responses characterized by the release of reactive oxygen species, inflammatory mediators, apoptosis, and ischemic cascades. In particular, inflammatory cytokines, modulated by microglial and astrocytic activation, have been implicated in early events that mediate BBB breakdown and subsequent development of cerebral edema.21Despite the development of pharmacological strategies in other areas of acute CNS injury (e.g. , nimodipine for subarachnoid hemorrhage and thrombolytics for ischemic stroke), treatment after TBI remains largely supportive and is directed toward supportive management of hemodynamics and cerebral physiology.22Thus, microglial inhibition represents an innovative approach to this unmet clinical need.

Our current findings demonstrate that administration of the novel, CNS-penetrable small molecule TT-301 resulted in a reduction in microglial activation with associated reduction in injury to selectively vulnerable hippocampal neurons at 24-h postinjury and increased hippocampal neuronal density at 28 days postinjury. Furthermore, the reduction in microgliosis and neuronal injury paralleled the functional improvements in vestibulomotor skills and spatial memory and learning associated with treatment. Rotorod latency testing following CNS injury is considered to be a sensitive and reliable test of vestibulomotor function that models everyday skills in the clinical setting.23In addition to motor deficits, TBI-induced cognitive impairment is often associated with a substantial negative impact on quality of life, and may be mediated by the preferential susceptibility of hippocampal neurons to diffuse brain injury.24,25Recent reports of spatial navigation deficits of TBI survivors in a virtual water maze support the validity of MWM as a TBI outcome measure for learning and memory.26TT-301-mediated improvements in these clinically meaningful measures of functional outcome make this strategy attractive for clinical translation, especially when coupled with the sustained benefit seen with TT-301 administration up to 6 h following trauma.

The neuroprotective effects of TT-301 administration seen in TBI appear to be relevant to other forms of CNS injury associated with acute neuroinflammation. ICH is a devastating and relatively common form of cerebrovascular disease, which remains associated with poor outcome.27As with TBI, ICH is associated with glial activation and release of inflammatory mediators, which contribute to BBB breakdown, enhancement of secondary neuronal injury, and development of cerebral edema.28Thus, a therapeutic strategy that blunts microglial activation would also be expected to improve outcomes in this disease. In fact, although administration of TT-301 was not associated with any reduction in hematoma volume, it was associated with a reduction in cerebral edema and improved motor outcomes after injury with ICH.

Though the exact mechanism by which TT-301 inhibits microglial activition is not currently known, TT-301 treatment of stimulated microglial lines was effective at suppressing the upregulation of several proinflammatory cytokines relevant to CNS injury. By targeting microglia, the main cellular source of proinflammatory cytokines in the CNS, TT-301 may alter disease progression through attenuation of subsequent neuronal synaptic dysfunction, the cellular basis of clinical symptoms and behavioral alterations. In addition, our differential gene expression studies may provide insight into the mechanism(s) by which TT-301 modulates microglial function. Administration of TT-301 following injury was associated with down-regulation of STAT3 and changes in the JAK-STAT cascade. STAT signaling is associated with inflammatory pathways in the brain, and regulation of JAK-STAT expression has been shown to protect against inflammation-induced brain injury.29,–,31This is consistent with observations that STAT3 activation was associated with glial activation and secondary neuronal injury in a preclinical model of cerebral ischemia, and that injection of small interfering RNA specific for STAT3 was neuroprotective in murine models.32Moreover, recent observations indicate that thrombin-induced activation of JAK-2 and phosphorylation of STAT-3 lead to activation of cultured microglia, and subsequent degeneration of dopaminergic neurons.33This latter observation is especially relevant in ICH, as thrombin is implicated in initiating the neuroinflammatory responses in this setting. Given the known dissociation between gene expression and protein transcription, the findings from our study emphasize the need to further investigate the role of the inflammatory pathways through focused timed experiments evaluating both proteins involved in this pathway, as well as confirmatory Northern analysis to confirm candidate genes that were differentially expressed. However, it is promising that modulation of pathways, such as the JAK-STAT3 cascade, may serves as therapeutic targets in microglial-dependent neuroinflammatory responses.

There are several unexpected findings from this study. First, although 5 days of treatment with TT-301 after TBI was associated with a durable reduction in microgliosis for the first 10 days, microglial counts returned to normal by 28 days after injury. There also was no reduction in microgliosis at 28 days, despite extending treatment with TT-301 until the time of sacrifice. Though speculative, this finding may underscore emerging evidence, which suggest microglia respond acutely to CNS injury by up-regulating proinflammatory pathways that exacerbate primary CNS damage, but eventually play a beneficial adaptive role in the subacute to chronic setting by promoting neuronal survival and synapse formation.34,35Of course it is not possible to rule out the influence of systemic macrophage involvement after BBB breakdown in these models. However, the observed correlation between TT-301-mediated histologic and functional improvement (both in vestibulomotor and cognitive deficits) could be explained by TT-301 preferentially targeting the cells and pathways that drive the acute phase of CNS injury. Also, the dissociation between the effects of TT-301 on microglial versus  astrocytic function was unanticipated. Although microglia are considered the primary resident immunoeffector cells in the CNS, astrocytes also contribute to neuroinflammatory responses via  secretion of inflammatory mediators, and both astrocytic and microglial activation may be found concurrently.36,–,39In the current study, however, we found that, although treatment with TT-301 reduced microglial activation for the first 10 days postinjury, there was no appreciable effect on astrocytosis, as visualized by GFAP staining.

In summary, in the current study, we demonstrate that short-term administration of TT-301 postinjury improved histologic and functional outcomes in murine models of TBI and ICH. Moreover, administration of TT-301 was associated with functional improvement even when the initial dose was delayed until 6 h after injury. These results would suggest that a therapeutic strategy targeting pathways involved in acute microglial activation may have potential to be translated into improved clinical outcomes in human trials.