Abstract
The activation of poly(ADP-ribose) polymerase-1 (PARP-1) after exposure to nitric oxide or oxygen-free radicals can lead to cell injury via severe, irreversible depletion of NAD. Genetic deletion or pharmacological inhibition of PARP-1 attenuates brain injury after focal ischemia and neurotoxicity in several neurodegenerative models in animals. FR247304 (5-chloro-2-[3-(4-phenyl-3,6-dihydro-1(2H)-pyridinyl)propyl]-4(3H)-quinazolinone) is a novel PARP-1 inhibitor that has recently been identified through structure-based drug design. In an enzyme kinetic analysis, FR247304 exhibits potent and competitive inhibition of PARP-1 activity, with a Ki value of 35 nM. Here, we show that prevention of PARP activation by FR247304 treatment protects against both reactive oxygen species-induced PC12 cell injury in vitro and ischemic brain injury in vivo. In cell death model, treatment with FR247304 (10–8-10–5 M) significantly reduced NAD depletion by PARP-1 inhibition and attenuated cell death after hydrogen peroxide (100 μM) exposure. After 90 min of middle cerebral artery occlusion in rats, poly(ADP-ribosy)lation and NAD depletion were markedly increased in the cortex and striatum from 1 h after reperfusion. The increased poly(ADP-ribose) immunoreactivity and NAD depletion were attenuated by FR247304 (32 mg/kg i.p.) treatment, and FR247304 significantly decreased ischemic brain damage measured at 24 h after reperfusion. Whereas other PARP inhibitors such as 3-aminobenzamide and PJ34 [N-(6-oxo-5,6-dihydro-phenanthridin-2-yl)-N,N-dimethylactamide] showed similar neuroprotective actions, they were less potent in in vitro assays and less efficacious in an in vivo model compared with FR247304. These results indicate that the novel PARP-1 inhibitor FR247304 exerts its neuroprotective efficacy in in vitro and in vivo experimental models of cerebral ischemia via potent PARP-1 inhibition and also suggest that FR247304 or its derivatives could be attractive therapeutic candidates for stroke and neurodegenerative disease.
Activation of nuclear enzyme poly(ADP-ribose) polymerase (PARP) promotes cell death through processes involving energy depletion. Reactive oxygen species (ROS)-mediated damage of DNA can activate PARP (Szabo et al., 1996; Eliasson et al., 1997) and consumes NAD and consequently ATP, culminating in cell dysfunction or necrosis (Ha and Snyder, 1999). In addition, PARP plays a central role in the caspase-independent apoptosis pathway mediated by apoptosis-inducing factor. Translocation of apoptosis-inducing factor from the mitochondria to the nucleus is dependent on PARP activation in neurons treated with various DNA-damaging stimuli, including hydrogen peroxide (Yu et al., 2002). This cellular suicide mechanism of both necrosis and apoptosis by PARP activation has been implicated in the pathogenesis of ischemic brain injury and neurodegenerative disorders, and PARP inhibitors have been shown to be effective in animal models of stroke, traumatic brain injury and Parkinson's disease (Cosi et al., 1996; Endres et al., 1997; Abdelkarim et al., 2001; LaPlaca et al., 2001; Iwashita et al., 2004).
The brain is particularly susceptible to radical-mediated neuronal damage because of high levels of oxygen consumption, unsaturated fatty acids, and iron stores, combined with low antioxidant resources. Oxidative stress is a critical step in neuronal degeneration, including cerebrovascular injuries such as stroke and in neuropathology such as Alzheimer's disease, Parkinson's disease, and amyotrophic lateral sclerosis, as well as in normal aging (Coyle and Puttfarchem, 1993; Facchinetti et al., 1998). For example, ROS formation is increased after both permanent and reversible middle cerebral artery occlusion in rats (Nelson et al., 1992; Chan, 1996; Peters et al., 1998). Infarct volume after focal cerebral ischemia is reported to be reduced by overexpressing extracellular superoxide dismutase in mice (Sheng et al., 1998) and by the administration of free radical scavengers (Yamamoto et al., 1983; Yang et al., 2000; Iwashita et al., 2003). Thus, free radicals highly contribute to brain damage, especially after focal cerebral ischemia.
There is a significant increase of extracellular glutamate after stroke that causes neuronal damage through an excitotoxicity process. Increased glutamate in cerebral ischemia activates N-methyl-d-aspartate receptors, resulting in increased intracellular calcium concentration and neuronal nitric-oxide (NO) synthase (nNOS) activation. NO produced by nNOS reacts with superoxide to form the highly toxic peroxynitrite, and both NO and peroxynitrite can cause DNA damage, followed by PARP activation (Zhang et al., 1994). This mechanism of toxicity is supported by the report that PARP activation is markedly diminished in nNOS-deficient mice subjected to middle cerebral artery occlusion-reperfusion (Endres et al., 1998). There are several other sources of ROS such as metabolism of arachidonic acid via the cyclooxygenase and lipoxygenase pathways, oxidation of hypoxanthine and xanthine by xanthine oxidase, and mitochondrial electron transport, which can contribute to the DNA damage and PARP activation (Siesjo et al., 1989).
Direct evidence for the involvement of PARP in the pathogenesis of neuronal damage in stroke models has been reported; ischemic injuries were markedly decreased in PARP (–/–) mice (Eliasson et al., 1997; Endres et al., 1997) and was also attenuated by the treatment of widely used PARP inhibitor 3-aminobenzamide (3-AB) in wild-type mice (Endres et al., 1997). Furthermore, poly(ADP-ribose) formation was detected in both permanent and transient focal ischemia models, and 3-AB, nicotinamide, and a new series of PARP inhibitors such as 3,4-dihydro-5-[4-1(1-piperidynil)buthoxy]-1(2H)-isoquinolinone and PJ34 ameliorated ischemic brain damage as a result of PARP inhibition in the brain (Takahashi et al., 1997; Tokime et al., 1998; Abdelkarim et al., 2001).
We have recently identified FR247304 as a novel potent PARP-1 inhibitor. The purpose of the present study was, first, to evaluate PARP-1 inhibitory activity of FR247304 by using enzyme kinetics analysis. The second purpose of the present study was to determine the PARP-1 inhibitory properties and the neuroprotective properties of FR247304 in in vitro experimental neuronal cell death models, in which PARP is markedly activated by H2O2 exposure. Finally, the neuroprotective properties of FR247304 were investigated using the transient focal ischemia model in rats. In this model, PARP activation was also assessed by both poly(ADP-ribose) immunohistochemistry and NAD level to clarify the mechanism of the neuroprotective action of this compound. To compare the neuroprotective activity of FR247304 with other known PARP inhibitors, we also evaluated 3-AB and PJ34 in some models as references.
Materials and Methods
Materials
Rat pheochromocytoma PC12 cells were purchased from American Type Culture Collection (Manassas, VA). FR247304 [5-chloro-2-[3-(4-phenyl-3, 6-dihydro-1(2H)-pyridinyl)propyl]-4(3H)-quinazolinone; chemical structure shown in Fig. 1] and PJ34 [N-(6-oxo-5,6-dihydrophenanthridin-2-yl)-N,N-dimethylactamide] were synthesized at Fujisawa Pharmaceutical Co., Ltd., (Osaka, Japan). 3-AB was purchased from Sigma-Aldrich (St. Louis, MO). Vitamin E was purchased from Nakarai Tesque (Kyoto, Japan). Hydrogen peroxide (30%), lactate dehydrogenase (LDH) assay kit was from Wako Pure Chemicals (Tokyo, Japan). Tissue culture medium and fetal bovine serum were purchased from Sigma-Aldrich (St. Louis, MO), and tissue culture dishes were from Sumitomo (Osaka, Japan). Recombinant human PARP enzyme was purchased from Trevigen (Gaithersburg, MD), and recombinant mouse PARP-2 enzyme was purchased from Alexis Biochemicals (San Diego, CA). Unless otherwise stated, all other materials were purchased from Sigma-Aldrich.
Measurement of PARP Inhibitory Activity and Specificity of FR247304
PARP Inhibitory Activity in Vitro. To assess the PARP-1 or PARP-2 inhibitory activity of FR247304, 3-AB, and PJ34, PARP activity was evaluated as described previously (Banasik et al., 1992) with minor modifications. PARP enzyme assay was carried out in a final volume of 100 μl consisting of 50 mM Tris-HCl (pH 8.0), 25 mM MgCl2, 1 mM dithiothreitol, 10 μg activated salmon sperm DNA, 0.1 μCi of [adenylate-32P]NAD, 0.2 units of recombinant human PARP for PARP-1 assay or 0.1 units of recombinant mouse PARP-2 for PARP-2 assay, and various concentrations of FR261529 or 3-AB. The reaction mixture was incubated at room temperature (23°C) for 15 min, and the reaction was terminated by adding 200 μl of ice-cold 20% trichloroacetic acid (TCA) and incubated at 4°C for 10 min. The precipitate was transferred onto GF/B filter (Unifilter-GF/B; PerkinElmer Life and Analytical Sciences, Boston, MA) and washed three times with 10% TCA solution and 70% ethanol. After the filter was dried, the radioactivity was determined by liquid scintillation counting.
Determination of Radical Scavenging Activity. For measurement of lipid peroxidation, thiobarbituric acid reactive substances (TBARS) were estimated using the modified method of Buege and Aust (1978) and Callaway et al. (1998). Briefly, brain synaptosomes were prepared from Wistar rats (Charles River, Hino, Japan). To evaluate the inhibitory activity of FR247304, different concentrations of the compounds were dissolved in 50% dimethyl sulfoxide (DMSO), and then 5 μl of diluted solution (the final is 1% DMSO) was added to each rat brain synaptosome and incubated with ammonium ferric sulfate (100 μM) at 37°C for 30 min. The reaction was stopped with addition of 20% TCA, and the precipitated proteins were removed by centrifugation at 10,000g for 15 min. The aliquots of supernatant were then added to an equal volume of thiobarbituric acid. The samples were heated at 95°C for 30 min and then cooled on ice before reading absorbance at 532 nm. Concentrations of TBARS were calculated using standard curve obtained with malondialdehyde. Percentage of inhibition of TBARS production was calculated as follows: % inhibition = [(Max–Drug)/(Max–Base)] × 100, where Max represents values in the presence of ammonium ferric sulfate, Base represents the values in the absence of ammonium ferric sulfate, and Drug represents the values of test compounds.
Determination of NOS Inhibitory Activity. NOS catalytic activity was assayed by measuring the Ca2+-dependent conversion of [3H]arginine to [3H]citrulline as described by Huang et al. (1993). For this assay, dissected rats brain was homogenized in 20 volumes (w/v) of 25 mM Tris buffer (pH 7.4) containing 1 mM EDTA and 1 mM EGTA. After centrifugation (20,000g for 15 min at 4°C), 25 μl of supernatant was added to 75 μl of 50 mM Tris buffer (pH 7.4) containing 1 mM NADPH, 1 mM EDTA, 3 mM CaCl2, and 0.1 μCi of [3H]arginine (specific activity 64 Ci/mmol; PerkinElmer Life and Analytical Sciences) in the absence or presence of FR247304 solution and incubated for 15 min at 37°C. The reaction was terminated by the addition of 250 μl of Dowex AG50WX-8 (Pharmacia, Peapack, NJ) and cooled on ice. After centrifugation, [3H]citrulline was quantified by liquid scintillation counting of 100-μl supernatant. No significant [3H]citrulline production occurred in the absence of calcium.
Preparation of Nuclear Extracts from PC12 Cells and the Rat/Mouse Brain
For preparation of nuclear extracts, the published methods were used for preparation of nuclear extracts, with minor modifications (Lahiri and Ge, 2000). To prepare the nuclear extracts from PC12 cells, 2 × 106 cells cultured in F25 flask were washed with 10 ml of phosphate-buffered saline, and cells were resuspended in 500 μl of cold buffer A (10 mM HEPES, pH 7.6, 15 mM KCl, 2 mM MgCl2, 0.1 mM EDTA, and 0.1% Nonidet P-40) and homogenized gently. The homogenate was centrifuged at 5000g for 30 s, and the supernatant containing cytoplasm and RNA was removed. The nuclear pellet was resuspended in 50 μl of ice-cold buffer B (50 mM HEPES pH 7.9, 400 mM KCl, 0.1 mM EDTA, and 10% glycerol). The tube was mixed thoroughly and placed on a micro tube mixer for 15 min at 4°C. The nuclear extract was centrifuged at 11,000g for 10 min. The supernatant containing the proteins from the nuclear extract was removed carefully to a fresh tube. The protein was measured in the nuclear extract and then used for the PARP-1 assay immediately.
For preparation of nuclear extracts from rat and mouse brain, normal whole brains were dissected and transferred to Teflon homogenizer. The buffer A was added at 300 mg of brain tissue per 1 ml and 10 strokes of homogenization were performed. The whole suspension was transferred equally to the Eppendorf tubes followed by centrifugation in a microcentrifuge (1600g) in a microcentrifuge at 4°C for 1 min. The supernatant contains mostly cytoplasmic constituents were removed, and 300 μl of buffer B was added to the nuclear pellet in each of the Eppendorf tubes. The tubes were mixed thoroughly and placed on a micro tube mixer for 15 min. The supernatant containing the proteins from the nuclear extract was removed carefully to a fresh tube. The protein was measured in the nuclear extract and then the crude-solution containing PARP enzyme and DNA was used for the PARP-1 assay immediately.
Neuroprotective Efficacy in PC12 Cells
Cells and Hydrogen Peroxide Treatment. PC12 cell cultured were grown in Dulbecco's modified Eagle's medium supplemented with 5% (v/v) fetal calf serum, 5% (v/v) horse serum, and a 1% (v/v) penicillin-streptomycin antibiotics mixture. Cells were grown in an atmosphere of 95% air and 5% CO2 at 37°C. For all experiment, cells were seeded at a density of 4 × 104cells/well in 96-well culture plates and allowed to attach overnight.
Drug Treatment. FR247304 or PJ34 was dissolved in 100% DMSO at 10–2 M and then diluted in DMEM without serum. 3-AB was dissolved in 10% DMSO at 1 M and then diluted in DMEM without serum. Each solution was added to culture plate 0.5h before H2O2 exposure.
Determination of Cellular NAD Level. To determine NAD level in cultured cells, PC12 cells were seeded at 2 × 105 cell/well in 24-well plates and cultured for 24 h. FR247304 was added to cell media at several concentrations. Thirty minutes later, cells were exposed 100 μM H2O2 for 30 min, and cells were detached using by cell scraper and then collected in microcentrifuge tube by centrifugation for 5 min × 100g at 4°C. To quantify the NAD level in rat brain, brain homogenate (10 mg tissue/150 μl in phosphate-buffered saline) dissected from the cerebral cortex and the striatum were prepared, respectively. Cells or brain homogenate was extracted with 200 μl of 0.5 M HClO4 for 15 min and then 60 μl of 2 M KOH/0.2 M K2HPO4-KH2PO4, pH 7.5 was added to the acidic supernatant obtained by centrifugation. NAD level in the supernatant was measured using enzymatic conversion to NADH by alcohol dehydrogenase.
Determination of Cell Viability. For assessment of cell viability, hydrogen peroxide-induced cytotoxicity was quantified by a standard measurement of LDH release with the use of the LDH assay kit (Wako Pure Chemicals). Briefly, 6 h after hydrogen peroxide exposure, 20 μl of medium of each well was collected, and the solution prepared from LDH assay kit was added. After incubation at room temperature for 30 min, the reaction was stopped by addition of 1 N HCl, and absorbance was measured at 450 nm using a microplate reader (Molecular Devices, Sunnyvale, CA).
Pharmacokinetic Study in Rats
Measurement of the concentration of FR247304 in plasma and brain were performed in rats following intraperitoneal administration at 32 mg/kg. FR247304 was suspended in 0.5% methylcellulose because it was not soluble in saline, and administered i.p. in a volume of 2 ml/kg. The plasma and brain was collected at 0.5, 1, 2, 4, 6 h after dosing, and the plasma and brain level of FR247304 were measured using high-performance liquid chromatography.
Focal Cerebral Ischemia Models in Rats
Animals. For transient focal ischemia, 9- to 10-week-old male Wistar rats (weighing 274–380 g) from Charles River were used. All animals were housed in a room maintained at 23 ± 2°C with 55 ± 5% humidity, and with a 12-h light/dark cycle (light on at 7:00 AM). The minimum quarantine period was at least 1 week before the experiment. Animals were housed five per cage and allowed free access to food and water. All experiments in the present study were performed under the guidelines of the Experimental Laboratory Animal Committee of Fujisawa Pharmaceutical. and were in strict accordance with the principles and guidelines of the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All efforts were made to minimize both the number of animals used and stress to the animals during experimental procedures.
Laser-Doppler Flowmetry. Changes of regional cerebral blood flow (rCBF) were recorded at the surface of the right parietal cortex using a laser-Doppler flowmetry (Omegaflow, FLO-N1; Neuroscience, Osaka, Japan) before, during, and after MCA occlusion. After rats were placed in a stereotaxic frame, craniectomy (2 mm in diameter, 4–6 mm lateral and 1–2 mm caudal to the bregma) was performed with extreme care over the MCA territory. The dura was left intact. The probe of the laser-Doppler flowmeter was lowered to the bottom of the cranial burr hole using a micromanipulator. The probe was held stationary by a probe holder secured to the skull with dental cement. Changes in rCBF were expressed as a percentage of the baseline value.
Transient Focal Ischemia in Rats. Transient occlusion of the right MCA was induced by insertion of a nylon suture through the right internal carotid artery to the origin of the MCA as described previously (Furuichi et al., 2003). Rats were anesthetized with a mixture of 4.0% halothane and oxygen-nitrogen (30% oxygen and 70% nitrogen) and maintained with a mixture of 1.5% halothane and oxygen-nitrogen during the procedure. Each animal was placed in the supine position, and a midline incision was made to the skin of the neck. The right common carotid artery (CCA) was exposed with careful protection of the vagus nerve. The external carotid artery, internal carotid artery (ICA), and CCA were carefully isolated and maintained in a Y-shape by use of silk suture. After ligation of the external carotid artery and CCA, an incision to insert a monofilament was made at the bifurcation. The monofilament was a 19-mm-long, 4-0 nylon surgical suture (Nicchou, Tokyo, Japan) coated with silicone (Xantopren L; Heraeus Kulzer, Dormagen, Germany) to thicken the distal 5 mm to about 0.4 mm in diameter. The proximal tip of the monofilament was heated to create a globular stopper for easy removal. The monofilament was introduced into the lumen of the ICA. In this way, the monofilament passed through the origin of the MCA and thereby occluded it. A silk suture was toed around the ICA to immobilize the monofilament. The neck wound was closed and each animal was allowed to recover from anesthesia. Ninety minutes after MCA occlusion, each animal was reanesthetized, and the neck wound was reopened to remove the monofilament and to resume blood flow to the MCA. The neck wound was closed again.
Drug Administration. FR247304, PJ34, or 3-AB, which was suspended with 0.5% methylcellulose, was administered at doses of 10 and 32 mg/kg for FR247304, 3.2 and 10 mg/kg for PJ34, or 32 and 100 mg/kg for 3-AB intraperitonially twice at 10 min before MCA occlusion and 10 min before recirculation. The administration volume was adjusted to 2 ml/kg.
Measurement of Brain Damage. Twenty four hours after the reperfusion, each rat was anesthetized with intraperitoneal administration of sodium pentobarbital (50 mg/kg), and the brain was fixed by transcardial perfusion with heparinized saline and then with 10% neutral formalin solution. The brain was cut into 2-mm coronal slices using a brain microslicer at +4, +2, 0, –2, –4, and –6 mm from the bregma. Each coronal slice was embedded in paraffin, and 3- to 4-μm sections were prepared from each slice with a microtome. These sections were stained with hematoxylin and eosin. To correct swelling due to brain edema or atrophy caused by tissue damage, the damaged area was compensated by the ratio of the whole area of the left cerebral hemisphere to that of the right cerebral hemisphere. The area with ischemia brain damage was delineated under light microscopy and quantified using an image analyzer system. The total damaged area was determined by summing up the damaged area from the six sections and presented as percentage of the damaged area compared with the whole coronal area.
Determination of NAD Content in Rat Brain. The NAD level in the cortex and striatum after ischemia-reperfusion in rats were quantified as described above. For time-course study, the brain was removed at 0, 1, 3, and 24 h after reperfusion and cut into 2-mm coronal slices using a brain microslicer at 0 mm from the bregma. The cerebral cortex and striatum of damaged hemisphere were carefully dissected, and the NAD level was quantified as described above. To evaluate the effect of FR247304, the cerebral cortex and striatum of damaged hemisphere were dissected at 3 h after reperfusion.
Poly(ADP-Ribose) Immunohistochemistry. To detect the distribution of poly(ADP-ribosyl)ation after focal brain ischemia, rats were anesthetized with sodium pentobarbital (50 mg/kg) and transcardially perfused with 4% paraformaldehyde at 1 and 6 h after induction of ischemia. After immediate removal, brains were stored in the buffer containing 20% sucrose overnight at 4°C. Frozen coronal sections (40 μm) were prepared on a freezing microtome and then endogenous peroxidase was quenched in 0.3% H2O2 in methanol. After blocking with 10% donkey serum, sections were incubated with rabbit anti-poly(ADP-ribose) polyclonal antibody (BIOMOL Research Laboratories, Plymouth Meeting, PA) at 4°C overnight. This antibody was used at a concentration of 1:100. The sections were washed with 0.05% Triton X in phosphate-buffered saline and then incubated with biotinylated goat anti-rabbit IgG antibody at 1:150 dilutions for 30 min and with a peroxidase-streptavidin for 30 min. Peroxidase was developed with aminoethyl carbazole (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) for 30 min and colored red. Finally, to enhance nuclear staining, the sections were counterstained with hematoxylin (blue color).
Statistical Analysis
The IC50 values obtained from studies in vitro were calculated using GraphPad Prism 3.3 software (GraphPad Software Inc., San Diego, CA). All values are expressed as mean ± S.E.M. Statistical analysis was carried out using Student's t test comparing the drug-treated group with control group and by using one-way analysis of variance (ANOVA) followed by Dunnett's multiple comparison test for the drug-treated groups versus the control group. P values less than 0.05 are considered to be significant.
Results
Enzyme Kinetics Study and PARP-1/PARP-2 Inhibitory Activity of FR247304
FR247304 exhibited typical competitive inhibition of enzyme kinetics (Fig. 2A) with a Ki value of 35 nM (Fig. 2B). To compare the species differences of PARP inhibitory activity of FR247304, human recombinant PARP, and nuclear extract from rat and mouse brain were used as rat PARP and mouse PARP-1. FR247304 potently inhibited the enzyme activity with an IC50 of 65 ± 0.9, 68 ± 2.1, and 63 ± 1.8 nM in human, rat, and mouse PARP, respectively (Fig. 3A). To compare PARP inhibitory activity with other well known PARP inhibitors, 3-AB and PJ34 were also evaluated. They inhibited the PARP enzyme activity with an IC50 of 11200 ± 810 and 110 ± 1.9 nM, respectively (Table 1). Furthermore, to confirm the selectivity of FR247304, PARP-2 enzyme assay was also conducted. In our assay system, FR247304 was shown to be about 10-fold selective for PARP-1 (IC50 value of 68 nM) over PARP-2 (IC50 value of 620 nM).
Specificity of FR247304
To determine whether FR247304 has properties to reduce ROS-induced cytotoxicity directly, radical scavenging activity and NOS inhibitory activity were evaluated using TBARS assay and NOS catalytic activity assay. FR247304 did not inhibit TBARS production, even at a concentration of 10–5 M, although vitamin E showed radical scavenging activity from 10–6 M (Fig. 3B). In the NOS assay, 7-nitroindazole, a selective nNOS inhibitor, prevented NOS catalytic activity assessed by [3H]citrulline production at concentrations ranging from 10–7 M to 10–5 M, although FR247304 had no inhibitory activity up to 10–5 M (Fig. 3C). Thus, these results suggest that FR247304 can potently and competitively inhibit PARP enzyme but does not scavenge free-radicals or inhibit NOS activity.
Neuroprotective Action in PC12 Cells
In this in vitro study, we confirmed whether H2O2 treatment induced PARP activation, as well as concomitant NAD depletion and cell death in rat pheochromocytoma PC12 cells. PARP activation by H2O2 exposure was evaluated by NAD assay. Excessive PARP activation by H2O2 (100 μM) exposure for 30 min resulted in massive NAD depletion (Fig. 4A), and this NAD depletion was concentration dependently inhibited by FR247304 treatment (10–8–10–5 M). Furthermore, exposure of H2O2 for 6 h induced severe cell damage, although FR247304 treatment at a concentration ranging from 10–8 to 10–5 M significantly and concentration dependently attenuated cell death, as assessed by LDH release assay (Fig. 4B). To compare the neuroprotective properties of other PARP inhibitors in PC12 cells, 3-AB and PJ34 were evaluated using by LDH assay. 3-AB and PJ34 treatment also significantly and concentration dependently attenuated cell death at a concentration ranging from 10–4 to 10–2 M and 10–7 to 10–5 M, respectively (Table 1).
Pharmacokinetic Study
The plasma and brain concentrations of FR247304 in rats were determined at 0.5, 1, 2, 4, and 6 h after intraperitoneal administration at 32 mg/kg. Mean plasma and brain concentrations of Cmax were 0.15 μg/ml and 0.74 μg/g tissue, respectively, and even at 6 h after dosing, the plasma and brain concentration were relatively highly sustained (0.10 μg/ml in the plasma and 0.36 μg/g tissue in the brain) as shown in Fig. 5. Thus, the concentration of FR247304 in the brain was significantly higher than that in the blood, and the brain/plasma concentration ratio was 4.35 at the Cmax time point (1 h after dosing).
Neuroprotective Action of FR247304 in Rat Focal Cerebral Ischemia
Effect of FR247304 on Ischemic Damage after Transient Focal Cerebral Ischemia. rCBF decreased to 21.3 ± 3.7% (n = 9) of the baseline level immediately after MCA occlusion and sustained during 90 min of ischemia in the control group. After reperfusion, rCBF increased to 95 to 100% of baseline within 5 min. FR247304 (10 and 32 mg/kg i.p.) did not alter rCBF; rCBF was 21.9 ± 4.6% (n = 9) of the baseline level immediately after the MCA occlusion and was 95 to 100% of baseline after reperfusion. In the control group, the dorsolatelal cortex and basal ganglia showed extensive damage that could be clearly differentiated from normally perfused area (representative pictures shown in Fig. 6A). The volume of ischemic brain infarction in the cerebral cortex and striatum in the control group was 126.85 ± 15.58 and 52.64 ± 1.98 mm2, respectively. FR247304 dose dependently reduced the size of infarcted cortical area when administered 10 min before MCA occlusion and 10 min before reperfusion (Fig. 6, B and C). Cortical damage was reduced by 27 and 47% at the doses of 10 and 32 mg/kg, respectively, and the protection at the dose of 32 mg/kg was statistically significant both in total cortical infarct size and in the area at serial coronal sections (2-mm sections 2–4) compared with control group (P < 0.05 by one-way ANOVA followed by Dunnett's multiple comparison test). In the striatum, FR247304 slightly but significantly reduced striatal infarct size by 14% compared with control group. As shown in Fig. 7, MCA occlusion resulted in an increase of body temperature. However, FR247304 treatment had no influence on body temperature both at 1.5 h after occlusion and at 24 h after reperfusion. To compare the potency and efficacy with other PARP inhibitors, 3-AB and PJ34 were evaluated at a dose of 100 mg/kg or at the doses of 3.2 and 10 mg/kg, respectively. Cortical damage was slightly reduced by 11% with 3-AB treatment. PJ34 at the dose of 3.2 mg/kg significantly reduced cortical damage by 33%; however, 10 mg/kg dosing showed reversed effect (17% reduction) (Table 1).
Poly(ADP-Ribose) Polymer Formation and Effect of FR247304. Increased immunoreactivity for poly(ADP-ribose) polymer was observed in ischemic cerebral cortex 1 h after reperfusion (Fig. 8A). This poly(ADP-ribose) polymer formation was still detected 6 h later (Fig. 8B); however, at 15 h after reperfusion, the formation was decreased to below detectable levels (data not shown). In the striatum, marked poly(ADP-ribose) polymer formation was also observed 1 h after reperfusion (Fig. 8C), and we confirmed that poly(ADP-ribose) polymer-positive cells were mostly neurons by morphological examination. To evaluate the PARP inhibitory effect of FR247304 in this model, sections of the cerebral cortex from 1 h after reperfusion were prepared. The formation of poly(ADP-ribose) polymer was strongly inhibited by administration of FR247304 at a dose of 32 mg/kg (Fig. 8D).
NAD Depletion and Effect of FR247304. NAD contents in the cerebral cortex and striatum were determined at 1, 3, and 24 h after ischemia-reperfusion. Marked NAD depletion was observed from 1 h after reperfusion in the cerebral cortex. In the striatum, significant decrease of NAD content was observed from 3 h after reperfusion (Fig. 9A). The NAD depletion after reperfusion was prevented by FR247304 treatment at a dose of 32 mg/kg both in the cerebral cortex and in the striatum (Fig. 9B). This result was consistent with that of PARP immunostaining, suggesting that FR247304 exerts neuroprotective effect on transient focal ischemia in rats via its potent PARP inhibition.
Discussion
The aim of the present study was to evaluate the neuroprotective potential of a novel PARP-1 inhibitor FR247304, which was recently designed using X-ray analysis and a structure-based drug design system, on ROS-mediated neuronal damage in vitro and on brain injury associated with transient focal ischemia-reperfusion in rats. FR247304 competes with NAD to inhibit at the catalytic site of PARP, consistent with the report that quinazolinone derivatives, including FR247304, tightly bind to the nicotinamide-ribose binding site (NI site) by hydrogen bonds (Ser904 and Gly863) and by a sandwich hydrophobic interaction (Tyr907 and Tyr869), and also that these compounds bind to the adenineribose binding site (AD site), in which known inhibitors do not bind (Kinoshita et al., 2004). Interestingly, FR247304 was more selective for PARP-1 than PARP-2 (10-fold higher selectivity) compared with nonselective general PARP-1 inhibitors such as 3-AB (IC50 for PARP-1 = 11.2 μM and for PARP-2 = 9.8 μM; unpublished data) and PJ34 (IC50 for PARP-1 = 110 nM and for PARP-2 = 86 nM). FR247304 also showed similar inhibitory activity both for human recombinant PARP and for rat brain PARP, suggesting that this compound can exert potent PARP-1 inhibitory activity in neuronal death models of rat origin tested in the present study, such as rat PC12 cells and a transient focal ischemia model in rats.
To determine PARP inhibitory properties and the neuroprotective properties of FR247304 in cultured cells, cell damage was induced by H2O2 exposure in PC12 cells, and PARP activation was measured by NAD depletion. In our cell assay conditions, H2O2 exposure generated marked NAD reduction followed by severe cell damage. As expected given the potent PARP-1 inhibitory activity of FR247304, this compound markedly attenuated both NAD depletion and cell damage, and the potency was consistent with its PARP-1 inhibitory activity in rat PARP enzyme assay. Other quinazolinone derivatives, which we have recently synthesized, showed an equivalent correlation between the potency in inhibiting PARP-1 activity and neuroprotective properties in cultured cells (unpublished data), suggesting that FR247304 possesses superior neuroprotective properties via its potent PARP-1 inhibitory activity combined with sufficient cell membrane permeability.
In addition to in vitro efficacy, FR247304 provides significant cerebroprotection after transient cerebral ischemia. Cerebral ischemia-reperfusion is accompanied by enhanced poly(ADP-ribosyl)ation (Endres et al., 1997; Tokime et al., 1998), and infarct volume is reduced by the administration of PARP inhibitors (Takahashi et al., 1997; Abdelkarim et al., 2001). The neuroprotective effects of PARP-1 inhibitors indicate that PARP play an important role in brain damage after transient ischemia. Consistent with the neuroprotective properties of other known PARP-1 inhibitors, FR247304 markedly reduced cortical infarct volume by 47% compared with vehicle-treated control. The neuroprotective effect of FR247304 was more prominent in the cortex than in the striatum, in accordance with previous reports concerning neuroprotectants in stroke models (Lo et al., 1998; Schmid-Elsaesser et al., 2000; Furuichi et al., 2003; Iwashita et al., 2003) and also with our observation that NAD depletion in focal cerebral ischemia-reperfusion occurs predominantly in the cortex immediately after reperfusion and that FR247304 treatment markedly attenuates NAD depletion. These results suggest that FR247304 exerted potent neuroprotective properties in focal ischemia model via its potent PARP-1 inhibitory activity, showing compatibility with results in the in vitro cell death model.
In focal ischemia models, nitric oxide and superoxide generation on reperfusion have been demonstrated (Nelson et al., 1992; Peters et al., 1998), and concomitant generation of these radicals can lead to formation of the strong oxidant peroxynitrite during reperfusion. Infarct volume after focal cerebral ischemia is reported to be reduced by overexpressing extracellular superoxide dismutase in mice (Sheng et al., 1998) and by the administration of free radical scavengers (Yamamoto et al., 1983; Yang et al., 2000; Iwashita et al., 2003). 7-Nitroindazole, a selective neuronal NOS inhibitor, has also been reported to ameliorate ischemic brain injury (Kamii et al., 1996; Gursoy-Ozdemir et al., 2000). These results raised the possibility that drugs that have a radical scavenging activity or neuronal NOS inhibitory activity could attenuate the infarct volume induced by reperfusion after focal cerebral ischemia. However, in our in vitro assay, FR247304 showed no antioxidant property and nNOS inhibitory activity even at a concentration of 10–5 M, indicating that FR247304 does not modulate the NO and/or ROS-mediated pathway, but instead that the neuroprotective properties is likely due to the result of its specific PARP-1 inhibitory activity.
It has been proposed that PARP activation may be a late common step responsible for oxidative cell injury (Ha and Snyder, 1999) and PARP-induced cell death correlates well with cellular NAD depletion (Ying et al., 2001; Yu et al., 2002). NADPH oxidase and nNOS contribute to increased oxidative stress with subsequent activation of PARP, and NADPH oxidase and nNOS inhibitors attenuate neuronal death without direct inhibition of PARP (Hwang et al., 2002). The DNA alkylating agent-induced NAD depletion is the cause of glycolytic failure and cell death after PARP activation, thus, the maintenance of NAD can mitigate cell death (Ying et al., 2003), suggesting that NAD depletion plays an essential role in the PARP-induced cell death pathway. Consistent with these reports, neuroprotection by FR247304 observed in our study was the consequence of the maintenance of NAD by PARP inhibition; thus, the activation of PARP could be causative factor and PARP inhibition is probably responsible for the neuroprotection in an oxidative neuronal death pathway.
As demonstrated by the pharamacokinetic study, the brain concentration of FR247304 at 32 mg/kg i.p. was found to be 0.75 μg/g at 1 h postdosing, the concentration that is calculated to be higher than 10–6 M. Furthermore, even at 6 h after drug administration, the brain concentration was sustained at the level of half maximum concentration (Cmax), which is estimated as about 10–6 M. This dosing regimen yielded high brain levels of FR247304 to sufficiently inhibit PARP-1 inhibitory activity in the brain. As expected from a good pharmacokinetic profile, treatment with FR247304 before the ischemia-reperfusion insult produced robust and significant neuroprotection, suggesting that ischemia-reperfusion induced PARP activation persists for some time and that the neuroprotective effects of FR247304 presumably are related to its PARP-1 inhibiting properties. The present study showed the temporal pattern of brain PARP activation after ischemia demonstrating that PARP activation evaluated by poly(ADP-ribose) polymer formations were sustained from immediately after reperfusion to 6 h later, and then the formation was decreased to below detectable levels at 15 h after reperfusion.
Although the putative mechanism of neuroprotection by FR247304 is PARP-1 inhibition, infarct volume can also be reduced by improving rCBF. However, there was no enhancement of rCBF to cortical area of the infarct in 10 and 32 mg/kg FR247304-treated rat as assessed by rCBF monitored during the MCA occlusion and monitoring period. Therefore, it is unlikely that any heomodynamic actions of FR247304 contributed to the neuroprotective action via improvement of rCBF. Furthermore, reductions of infarct volume and neurological deficits in PARP-deficient mice or in mice treated with the PARP-1 inhibitor 3-AB does not depend on changes in CBF, but rather are correlated with reduced PARP activation in ischemic brain tissue (Eliasson et al., 1997; Endres et al., 1997). Moreover, although hypothermia can also reduce infarct volume, body temperature does not change after ischemia-reperfusion in PARP-deficient mice (Onesti et al., 1991; Chen et al., 1992; Eliasson et al., 1997). Because FR247304 treatment did not influence body temperature in the present study, the beneficial effects of this compound are unlikely to result from the hypothermic activity.
Several classes of competitive PARP-1 inhibitors have been reported to date and treatment with PARP-1 inhibitors such as 3-AB or PJ34 reportedly decreased ischemic injury in a rat model of MCA occlusion (Endres et al., 1997; Abdelkarim et al., 2001). In the present study, we have compared the potency and neuroprotective efficacy of these reference PARP inhibitors with FR247304 in a PARP-1 enzyme assay, a cell death assay and a rat transient focal ischemia model. FR247304 showed the most potent PARP inhibitory activity compared with 3-AB and PJ34, and the rank order of their cytoprotective properties against H2O2-induced cell death paralleled their PARP inhibitory actions. In our in vivo experimental design, FR247304 treatment (32 mg/kg i.p.) exerted robust reduction in infarct volume by 47%, and was more efficacious than those of 3-AB and PJ34 treatment. There is a report demonstrating that an intravenous administration of PJ34 markedly reduced infarct volume by over 70% (Abdelkarim et al., 2001), raising the possibility that intravenous administration might provide a high brain concentration, and thus, the extent of neuroprotection may depend on the paradigm differences (e.g., administration route and time point). Furthermore, although we have observed a dose-dependent neuroprotection with FR247304 treatment, PJ34 showed a bell-shaped dose-response relationship for its protective actions and it lost the efficacy at a higher dose (10 mg/kg i.p., 17% reduction), being consistent with a case of another PARP inhibitor 3,4-dihydro-5-[4-1(1-piperidynil)buthoxy]-1(2H)-isoquinolinone (Takahashi et al., 1997). Although 3-AB reduced infarct volume from relatively low dose (32 mg/kg) in spite of its low potency in vivo, higher dosing (100 mg/kg) was less neuroprotective. As it is not clear why high doses of PARP-1 inhibitors are less neuroprotective, extensive investigations could be required if excessive PARP inhibition induces cell death and low selective compound possessing secondary effects can affect cell death/survival. Our comparative studies taken together indicate that FR247304 exerts significant and dose-dependent neuroprotection on brain injury associated with focal brain ischemia, consistent with other PARP-1 inhibitors in previous reports (Endres et al., 1997; Takahashi et al., 1997). However, considering with the reduction of brain infarction size following focal ischemia in PARP-1 deficient mice (Eliasson et al., 1997) and also that FR247304 possessed superior potential on PARP inhibitory activity in vitro, the efficacy and potency of FR247304 in focal ischemia in vivo were below expectations. Therefore, it would warrant further detailed studies to address efficacy with intravenous administration, specificity, cell permeability and pharmacokinetic property of each drug.
In conclusion, a newly synthesized PARP-1 inhibitor, FR247304, exhibited potent PARP-1 inhibition both in vitro and in vivo, with significant neuroprotective properties after ischemia-reperfusion in rats, suggesting that this compound and/or a water-soluble derivative of FR247304 could be not only an important tool for investigation of the physiological role of PARPs in neurodegenerative pathways but also an attractive therapeutic candidate for stroke and neurodegenerative disease.
Acknowledgments
We thank Dr. Raymond D. Price for helpful comments in preparing the manuscript.
Footnotes
-
DOI: 10.1124/jpet.104.066944.
-
ABBREVIATIONS. PARP-1, poly(ADP-ribose) polymerase-1; ROS, reactive oxygen species; NO, nitric oxide; nNOS, nitric-oxide synthase; 3-AB, 3-aminobenzaminde; FR247304, 5-chloro-2-[3-(4-phenyl-3, 6-dihydro-1(2H)-pyridinyl)propyl]-4(3H)-quinazolinone; PJ34, N-(6-oxo-5,6-dihydrophenanthridin-2-yl)-N,N-dimethylactamide; LDH, lactate dehydrogenase; TBARS, thiobarbituric acid reactive substances; DMSO, dimethyl sulfoxide; rCBF, regional cerebral blood flow; MCA, middle cerebral artery; CCA, common carotid artery; ICA, internal carotid artery; ANOVA, analysis of variance.
- Received February 11, 2004.
- Accepted April 6, 2004.
- The American Society for Pharmacology and Experimental Therapeutics