Robust neuroprotective effects of 2-((2-oxopropanoyl)oxy)-4-(trifluoromethyl)benzoic acid (OPTBA), a HTB/pyruvate ester, in the postischemic rat brain

Postischemic brain damage in stroke is proceded with complicated pathological events, and so multimodal drug treatments may offer better therapeutic means for improving clinical outcomes. Here, we report robust neuroprotective effects of a novel compound, 2-((2-oxopropanoyl)oxy)-4-(trifluoromethyl)benzoic acid (OPTBA), a 2-hydroxy-4-trifluoromethyl benzoic acid (HTB, a metabolite of triflusal)-pyruvate ester. Intravenous administration of OPTBA (5 mg/kg) 3 or 6 h after middle cerebral artery occlusion (MCAO) in Sprague-Dawley rats reduced infarct volumes to 38.5 ± 11.4% and 46.5 ± 15.3%, respectively, of that of MCAO controls, and ameliorated motor impairment and neurological deficits. Importantly, neuroprotective effects of OPTBA were far greater than those afforded by combined treatment of HTB and pyruvate. Furthermore, OPTBA suppressed microglial activation and proinflammatory cytokine inductions more effectively than HTB/pyruvate co-treatment in the postischemic brain and LPS-treated cortical slice cultures and also attenuated NMDA-induced neuronal death in hippocampal slice cultures. LC-MS analysis demonstrated that OPTBA was hydrolyzed to HTB and pyruvate with a t1/2 of 38.6 min in blood and 7.2 and 2.4 h in cortex and striatum, respectively, and HTB was maintained for more than 24 h both in blood and brain tissue. Together these results indicate OPTBA acts directly and via its hydrolysis products, thus acting as a multimodal neuroprotectant in the postischemic brain.

In the postischemic brain, neuronal cell damage and subsequent neurological dysfunction are caused by complicated pathological events occurring in a spatiotemporally-regulated manner. Excitotoxicity and Zn 2+ toxicity cause massive neuronal cell damages in the ischemic core during the acute phase 1 and this is followed by inflammation and apoptosis within a few hours to days that exacerbate brain injury 2 . It is for this reason that combinatorial or multimodal drug treatments are believed to be most effective for stroke treatment. In this respect, it has been reported combination treatment with edaravone and borneol confers synergistic neuroprotective effects in the postischemic brain via anti-oxidative and anti-inflammatory mechanisms, respectively 3 . In addition, co-treatment with recombinant tissue plasminogen activator (rtPA) and minocycline (a PARP-1 inhibitor) was found to enhance protective effects by suppressing inflammation, infarction formation, brain swelling, and hemorrhage in focal embolic stroke 4 . In a previous study, we also reported combination treatment with ethyl pyruvate and aspirin acted synergistically to afford neuroprotection in the postischemic brain via the differential modulation of NF-κ B signaling 5 . Subsequently, we introduced a multimodal neuroprotectant, OBA-09, a salicylic acid/pyruvate ester, which conferred robust neuroprotective effects in the postischemic brain by reducing ROS generation and suppressing excitotoxicity and Zn 2+ toxicity 6 .
OPTBA suppressed neuronal cell death in NMDA-treated hippocampal slice cultures. Next, we investigated whether OPTBA also confers neuroprotective effect against excitotoxicity. In hippocampal slice cultures treated with NMDA (10 μ M, 24 h), propidium iodide (PI) staining revealed significant increase of neuronal cell death, however, it was suppressed by OPTBA (250 μ M) (Fig. S3) and the efficacy was higher than those of HTB/PY (250 μ M each) or TF/PY co-treatment (250 μ M each) (Fig. 7A,B). NMDA receptors regulate neuronal PARP 1 expression and activity, which causes cell death via depletions of NAD and ATP 19 . NAD depletion in NMDA-treated hippocampal slice cultures was suppressed by co-treating OPTBA (250 μ M) and replenished NAD level in OPTBA (250 μ M)-treated culture was higher than those in HTB/PY (250 μ M each)-or TF/PY (250 μ M each)-co-treated cultures (Fig. 7C). Furthermore, OPTBA suppressed PARP-1 protein induction far more effectively than TF/PY-or HTB/PY-co-treatment and PAR formation (Fig. S4) was also significantly suppressed by OPTBA (Fig. 7D,E). These results indicated that OPTBA exerted a marked neuroprotective effect against excitotoxic stress and it might also contribute to a robust neuroprotection observed in the postischemic brain.

Discussion
In the present study, we showed OPTBA, a HTB-pyruvate ester, exerts a robust neuroprotective effect in the postischemic brain and that this was achieved by its anti-inflammatory and anti-excitotoxic effects probably accomplished directly by OPTBA and by its hydrolysis products, HTB and pyruvate. Among various cells involved in induction and aggravation of inflammation in the postischemic brain 20 , activated microglia probably play a key role by producing various neurotoxins, such as, nitric oxide, reactive oxygen species, and cytokines [21][22][23][24] . In the present study, OPTBA was found to suppress microglial activation more efficiently than the combined treatment of HTB and pyruvate in rat MCAO model (Fig. 5D,H) and in LPS-treated cortical slice cultures (Fig. 6A). HTB is a long-lasting active metabolite of triflusal and has been shown to inhibit COX-2 activity and the translocation of NF-κ B, and thus, inhibit the de novo expressions of genes targeted by NF-κ B 8,11 . Regarding the protective potency, it has been reported that overall neuroprotective effect of HTB in rat MCAO model is superior to triflusal and salicylic acid and that in particular, anti-inflammatory effects are also greater than them (Kim et al., Submitted). Since pyruvate is also known to block the infiltration of immune cells into the postischemic brain and inhibit LPS-induced microglial activation 16 , the anti-inflammatory potency of OPTBA could be contributed by the dual and complementary anti-inflammatory effects of HTB and pyruvate released from OPTBA after its hydrolysis. In this regard, it is worthy of noting that HTB and pyruvate were released from OPTBA hydrolysis with a prolonged time window; the t 1/2 was 38.6 min in blood and 7.2 and 2.4 h in brain parenchyma (Fig. 1). More importantly, plasma HTB level surged to the level higher than that of OPTBA 15 min after OPTBA injection, and the enhanced HTB level lasted for longer than 24 h both in blood and brain tissue (Fig. 1). These observations agree well with a previous report showing that HTB is stably detected in rat plasma (t 1/2 of 21.5 h) 25 . We found that when OPTBA is administered intravenously as a single bolus, it can be hydrolyzed into HTB and pyruvate in blood and then enter the brain, and alternatively, it is possible for OPTBA to enter the brain parenchyma and then it is hydrolyzed. Considering that the high HTB level was rapidly achieved after OPTBA injection in blood (Fig. 1B) and that OPTBA accumulated almost immediately but its hydrolysis was slower in brain parenchyma (Fig. 1C), the neuroprotective effect of OPTBA is likely to be achieved through the rapid and sustained provision of HTB and pyruvate from the OPTBA hydrolysis and also by the prolonged action of HTB thanks to its remarkable stability (Fig. 1).
It has been previously reported pyruvate markedly reduced infarct formation in the postischemic brain 15,26 and that it effectively replenished NAD levels in Zn 2+ -treated cortical neurons and scavenged hydrogen peroxide 13,27 . However, in previous studies, high doses of pyruvate (62.5-1000 mg/kg, i.p.) were required to suppress brain damage in the postischemic rat brain 15,16,28 and millimolar concentrations of pyruvate were required to obtain neuroprotective effects in NMDA-or Zn 2+ -treated neuronal cells 13,29,30 . However, in the present study, OPTBA at 5 mg/kg reduced infarct volumes to 35.5 ± 12.3% versus MCAO controls when administered 6 h after MCAO (Figs 2 and 3), and micromolar concentration of OPTBA conferred neuroprotective effects in NMDA or LPS-treated slice cultures (Figs 6 and 7). In view of the fact that pyruvate is spontaneously converted to parapyruvate, an inhibitor of a key step in the tricarboxylic acid cycle 31,32 , the efficacy of OPTBA at lower dosages appears to add a distinct advantage. In addition to its anti-inflammatory effect, OPTBA was found to exhibit superior anti-excitotoxic effects compared to that of combined treatment of HTB and pyruvate in NMDA-treated hippocampal slice cultures. It has been previously reported that exogenous pyruvate prevented neuronal degeneration by replenishing NAD in Zn 2+ -treated cortical neurons 13 and ATP levels in NMDA-treated slice cultures 30 . In a separate study, we also found that both triflusal and HTB inhibited neuronal cell death in NMDA-treated cortical neurons and inhibition by HTB was greater than that by TF (Kim et al., Submitted). Hence, the higher efficacy of OPTBA in suppressing neuronal cell death in NMDA-treated hippocampal slice cultures was contributed by both pyruvate and HTB via suppressing PARP-1 expression and PAR formation (Fig. 7D,E). Thereby, we speculate that in NMDA-treated slice cultures, OPTBA supplies energy metabolites (ATP and NAD) in a sustained manner by stably producing pyruvate and HTB.
Since the brain damage caused by ischemic stroke is due to diverse pathophysiological events 33 , therapeutic strategies of choice are multimodal or combinatorial drug treatment. In the present study, we reported that OPTBA confers a robust neuroprotective effect in the postischemic brain, which was afforded by anti-inflammatory and anti-excitotoxic effects, and that the prolonged provisions of pyruvate and HTB by OPTBA hydrolysis enhance these effects. Pyruvoyl chloride (3.12 g, 29.3 mmol) was added to a solution of 2-hydroxy-4-(trifluoromethyl)benzoic acid (2.01 g, 9.75 mmol) and K 2 CO 3 (4.04 g, 29.3 mmol) in acetone (150 mL) at 0 °C. The reaction mixture was stirred at room temperature for 4 h and quenched with 1 N HCl solution. The solution was extracted with ethyl acetate and the organic layer was washed with water and brine, dried over Na 2 SO 4 , and evaporated in vacuo. The crude residue was purified by column chromatography to give the title compound as a white solid (1.85 g, 68.8% yield). MP 275 °C; 1  OPTBA, triflusal, HTB, or pyruvate injection to MCAO-operated rats. Sodium pyruvate, HTB, or triflusal (2.5 mg/kg each) was dissolved in 70% DMSO (50 μ l) and administered intravenously at 6 h after MCAO. OPTBA (1, 2.5, 5, or 10 mg/kg) was administered intravenously in 50 μ l of 70% DMSO at indicated time points. Animals were randomly divided into 10 groups, as follows: a Normal group (n = 10), treatment-naïve controls; DMSO (50 μ l of 70% DMSO) + OPTBA group (n = 46), OPTBA-administered rats; a Sham group (n = 14), animals underwent surgery but were not subjected to MCAO; a MCAO group, treatment-naive MCAO controls (n = 41, 50 μ l of DMSO (70%)-treated); a MCAO + PY group, pyruvate-administered MCAO rats (n = 6); a MCAO + TF group (n = 9), triflusal-administered MCAO rats; a MCAO + HTB group (n = 10), HTB-administered MCAO rats; a MCAO + TF/PY group (n = 25), triflusal/pyruvate-co-administered MCAO rats; a MCAO + HTB/PY group (n = 25), HTB/pyruvate-co-administered MCAO rats; and a MCAO + OPTBA group (n = 67), OPTBA-administered MCAO rats. No animal died during surgery, but overall mortality after surgery was 4.5% (12/ week. MCAO was carried out as previously described 6 . In brief, male Sprague-Dawley rats (250-300 g) were anesthetized with 5% isoflurane in a 30% oxygen/70% nitrous oxide mixture, anesthesia was maintained during procedures using 0.5% isoflurane in the same gas mixture. Animals were randomly allocated to the 10 treatment groups described in the previous section. MCA occlusion was performed for 60 min using a nylon suture (4-0; AILEE, Busan, South Korea) and was followed by reperfusion. During the procedure, the left femoral artery was cannulated to obtain a blood sample, which was analyzed for pH, PaO 2 , PaCO 2 , and blood glucose concentration (I-STAT; Sensor Devices, Waukesha, WI). Laser Doppler flowmetry (Periflux System 5000; Perimed, Jarfalla, Sweden) was used to monitor regional cerebral blood flow (CBF) and relative CBF during the experiment. Operated rats which did not show > 70% reduction in CBF during MCAO were excluded from the experimental groups. A thermoregulated heating pad and a heating lamp were used to maintain a rectal temperature of 37.0 ± 0.5 °C during procedures. Investigators blinded to the experimental groups performed the behavioral analysis and the data analysis. Animals in the sham group were operated in an identical manner but the MCA was not occluded.

Infarct volume assessment.
Rats were decapitated at 2 days post-operation and whole brains were dissected coronally into 2-mm brain slices using a metallic brain matrix (RBM-40000, ASI, Springville, UT). Slices were immediately incubated in saline containing 2, 3, 5-triphenyl tetrazolium chloride (TTC, 2%) at 37 °C for 15 min and then in 4% paraformaldehyde. Areas of infarcted tissue were measured using the Scion Image program (Scion Image program, Frederick, MD). To adjust for edema and shrinkage, areas of ischemic lesions were calculated using (contralateral hemisphere volume x measured injury volume/ipsilateral hemisphere volume) and infarct volumes were quantified (in mm 3 ) by multiplying summed infarct areas of sections by section thickness.
Evaluation of modified neurological severity scores. Neurological deficits were evaluated using modified Neurological Severity Scores (mNSS) at 2 days after MCAO. The mNSS system consists of motor, sensory, balance, and reflex tests, all of which are graded using a scale of 0 to 18 (normal: 0, maximal deficit: 18) 34 .
Wire hanging test. The wire hanging test procedure has been previously described 35 , and was used to measure forelimb strength and grasping ability at 2 days after MCAO. Briefly, a rat was suspended by its forelimbs on a horizontal steel wire (50 cm long, 2 mm diameter And after grasping the time to falling off was measured using a stopwatch up to a cutoff time of 60 s. Rota-rod test. One day before surgery, rats were trained on a rota-rod unit (Daejon Instruments, Seoul, Korea) at a constant 3 rpm until they were capable of remaining on the rotating spindle for 180 s. At 2 days after surgery, residence times on the spindle were recorded at spindle speeds of 5 and 10 rpm with a 1 h rest period after testing at 5 rpm.
Immunohistochemistry. The animals were sacrificed at 2 days (n = 3 per group) after MCAO and brains were fixed using 4% paraformaldehyde (PFA) by transcardiac perfusion and post-fixed in the same solution overnight at 4 °C. Brain sections (40 μ m) were prepared using a vibratome, and then immunologically stained using a previously described floating method 6 . Primary antibodies were diluted as follows; 1:300 for anti-ionized calcium binding adaptor molecule-1 (Iba-1) (Wako Pure Chemicals, Osaka, Japan) and 1:250 for anti-Mac2 (Abcam, Cambridge, UK). The images shown are representative of the results obtained from three animals for each group.
RNA preparation and RT-PCR. Total RNA was prepared using TRIzol reagent (Gibco BRL, Gaithersburg, MD), and 1000 ng aliquots of RNA samples were used for cDNA synthesis, which was conducted using a RT-PCR kit (Roche, Mannheim, Germany). The sequences of the rat interleukin-1β (IL-1β ), TNF-α , IL-6, and GAPDH primers used were described previously 36 .
Organotypic hippocampal and cortical slice cultures. Rats were sacrificed at postnatal days 3 (cortex) or 7 (hippocampus). Brains were aseptically removed and cortices or hippocampi were dissected from hemispheres and cut into 350 μ m slices using a McIlwain Tissue Chopper (The Mickle Laboratory Engineering Co., Surrey, UK) and subsequent procedures for organotypic Slice cultures were conducted as described previously 37  Propidium iodide (PI) staining. Neuronal cell death in OHSCs was determined by propidium iodide (PI) staining. Briefly, OHSCs were treated with NMDA (10 μ M) for 24 h, PI (1 μ g/mL) was then added and incubation continued for 30 min. OHSCs were then fixed in 4% paraformaldehyde (PFA) for 15 min and fluorescence was visualized under a Zeiss fluorescence microscopy (Axio Observer, Oberkochen, Germany). PI fluorescence intensities were measured in the hippocampal CA1 and CA3 regions using Image J software (National Institutes of Health, Bethesda, ML) and presented as fold increases versus NMDA non-treated control.