초록/요약

Abstract Part I. Stroke is the third leading cause of death around the world, primarily affecting individuals who are more than 65 years old. Despite the significant advances that have been made in understanding the pathophysiology of cerebral ischemia at the cellular molecular level, the thrombotic tissue plasminogen activator (t-PA) is approved by the FDA only for use in patients with acute ischemic stroke. Unfortunately, the utilization of t-PA is limited by its short time window of efficacy. Neuroprotective agents as well as antioxidant or/and anti-inflammatory drugs have generated as much interest due to their potentially long-term efficacy for the prevention of stroke. The only neuroprotectant drug for the treatment of stroke, is edaravone, which is a free radical scavenger approved by the Japanese Ministry of Health, Labor and Welfare in 2001 for use within 24 h after onset for the purpose of improving the neurological symptoms. In our efforts to find a novel neuroprotective agent, two analogues, oxopropanoyloxy benzoic acid 1 and 2-((2-oxopropanoyl)oxy)-4-(trifluoromethyl) benzoic acid 2, were synthesized by esterification of salicylic acid and 2-hydroxy-4-trifluoromethylbenzoic acid (HTB) with pyruvate, respectively, and their neuroprotective effects were evaluated by measuring the infarct volume in the postischemic brain in animal models. When 2 was administrated in the MCAO (middle cerebral arterial occlusion) model, the efficacy of infarct suppression by 2 was greater than treatment with 1, which indicates that HTB has better neuroprotection effect in the postischemic brain than salicylic acid. Based on the above results, we then synthesized the dimeric compounds, where the two molecules of HTB were conjugated via succinic acid and adipic acid, for the purpose of improving the neuroprotective effect of HTB. A series of dimeric HTB compounds was synthesized using aliphatic linkers with different carbon numbers. Among the HTB conjugates with linkers, 8, 10, and 11 showed inhibitory activities (49.5 ± 7.0 %, 49.0 ± 12.3% and 49.7 ± 8.7%, respectively) comparable to triflusal (42.8 ± 12.3%). To find compounds that are more neuroprotective compared to 2, we synthesized the dimeric compounds, where the two molecules of HTB were directly conjugated to form anhydride 12. In addition, two analogues of the HTB anhydride, substituting anhydride (-O-) with either an amino group (-NH-) 17 or a carbon linker (-CH2-) 28, were synthesized based on bioisosterism. An analysis of these compounds showed a significant difference in the cortical infarct volume (81.5 ± 1.5%, 59.9 ± 9.1% and 55.9 ± 7.8% in 12, 17 and 28, respectively). Among the synthesized compounds, the efficacy of infarct volume suppression by 12 was much greater than it was by 2. The intrinsic activity of anhydride 12 was confirmed by investigating the stability of 12 in a pH 7.4 buffer and in biological media such as rat blood. Unlike the compounds with linkers, 12 was slowly hydrolyzed at pH 7.4 buffer (T1/2 83.3 h) compared to triflusal (T1/2 40.0 h) and 12 was fairly stable in rat plasma (T1/2 22.2 h) compared to triflusal (T1/2 21.9 min). Furthermore, 12 was stable against enzymatic hydrolysis in Sprague-Dawley rats due to long half-life (3.6 h). Therefore, we concluded that 12 has the intrinsic capability of infarct volume suppression regardless of the HTB formed by the hydrolysis of anhydride linkages. The radical scavenging activity levels of all synthesized compounds were investigated using a spectrometric method to prove whether the inhibitory effects of triflusal and HTB conjugates on the infarct volume of the MCAO model in rats were related to their antioxidant activity. Reactive oxygen species (ROS) with different scavenging activities, such as hydrogen peroxide, superoxide, and nitric oxide, were used for the test, and the compounds of 12 showed the highest scavenging activities against superoxide and nitric oxide radicals. The protective effect against excitotoxicity was investigated in primary cortical neurons treated with NMDA. 12 reduced neuronal cell death in dose-dependent manner. This result indicated that 12 has potent anti-excitotoxic effect. Together these results showed that 12 has potent therapeutic potential as a neuroprotectant in the postischemic brain and these effects were conferred by intrinsic activity. Part II. Among the numerous approaches to treat obesity, an important development has been the CB1 receptor antagonists. The first-in-class cannabinoid (CB) receptor antagonist rimonabant was withdrawn from the market due to side effects such as anxiety and suicide ideation, related to the central nervous system, as the CB1 receptor is located in the brain. Therefore, in order to reduce the side effects of a CB1 receptor antagonist related to the central nervous system, we designed an anti-obesity drug which acts on the peripheral nerve rather than on the central CB1 receptor. In the present study, we synthesized a novel series of oxoacetamide analogues which is structurally related to rimonabant. Among the important structures of rimonabant, the pyrazole ring, the 4-chorophenyl ring, the 2,4-dicyclophenyl ring were maintained. In addition, 32a, with a modification of the amine side chain and extension of the carbonyl group, was initially examined. Through replacement with various amines, we studied the SAR (structure-activity relationships) the thiomorpholino derivative 32j, where the oxygen atom in 32i was replaced with a sulfur atom, revealing moderate binding affinity with a 2.5-fold increase compared to the corresponding piperidinyl group in 32g. Substitution of the piperidinyl ring in 32a by cyclohexyl 32n clearly increased affinities to the CB1 receptor (IC50 21.8 nM). The extension of the carbon chain length from a two-carbon chain to an eight-carbon chain (32r-32aa) increases the binding affinity to the CB1 receptor. 32v showed CB1 antagonistic activity (IC50 52.8) comparable to that of 32j (IC50 65.3 nM). Five-week-old male C57BL/6 mice were fed a high-fat (HF) diet for 4 months to induce obesity. From day 1, the body weight decreased in the rimonabant, 32j, 32n and 32v treated groups with a concentration of 10mg/kg and 30mg/kg compared to the HF-DIO control group. The final body weights of the 32j, 32n and 32v treated groups with a concentration of 10mg/kg and 30mg/kg were reduced by 1.7, 2.1, 3.4, 4.8, 3.6 and 5.6%, respectively. Rimonabant reduced the body weight by 13.9% compared to the weight loss in the HF-DIO control group. These results indicated that while 32j, 32n and 32v acted on reducing weight and depressing appetite, it had less of a pharmaceutical effect than rimonabant.