3-(2,6-Dimethylphenyl)-2-Selenoxo-1,3-Thiazolidin-4-One Suppresses Hydrogen Peroxide–Induced Cytotoxicity on PC12 Cells Via Activation of MAPK

Abstract

We newly synthesized organic selenium compounds (5-membered ring compounds) including 2-selenoxo-1,3-thiazolidin-4-ones (compounds A) and 3-alkoxy-4,5-dihydro-5-selenoxo-1H-1,2,4-triazole-1-carboxylates (compounds B). To address whether these compounds show antioxidative effects, we also examined their superoxide radical (O₂ ⁻)-scavenging effects. Moreover, we examined the effects of compound Aa on the activation of mitogen-activated protein kinase/extracellular signal-regulated protein kinases (MAPK/ERK1/2) and suppression of hydrogen peroxide-induced cytotoxicity in rat pheochromocytoma cells (PC12 cells). We evaluated the O₂ ⁻-scavenging activities of the compounds by a chemiluminescence method, and activation of ERK1/2 in PC12 cells was evaluated by Western blot analysis. At 166 μmol/L, the O₂ ⁻-scavenging activities were markedly different among compounds A and B. 3-(2,6-Dimethylphenyl)-2-selenoxo-1,3-thiazolidin-4-one (compound Aa) exhibited the strongest superoxide anion-scavenging activity among compounds A and B. The concentration necessary for 50% inhibition of the activity (IC₅₀) of compound Aa was 25.9 μmol/L. Compound Aa activated ERK1/2 of the PC12 cell, as did ebselen, and suppressed hydrogen peroxide-induced cytotoxicity more potently than ebselen. In addition, the toxicity of compound Aa was less than that of ebselen. From these results, it is assumed that compound Aa is a candidate drug to prevent oxidative stress-induced cell death.

Reactive oxygen species ¹ including superoxide (O₂ ⁻), hydrogen peroxide (H₂O₂), and hydroxyl radicals (·OH) are generated in aerobic cells. ² Reactive oxygen species may react with various pivotal biomolecules such as DNA and proteins, and the excessive reactive oxygen species generation in the cells may lead to denaturation of the molecules. ³ In addition, endogenous reactive oxygen species generators such as granulocytes, monocytes, macrophages, eosinophils, and basophils produce reactive oxygen species and damage the tissue/organ due to abnormal inflammation. ^{4 –6} To avoid toxicity of reactive oxygen species under the normal physiological conditions, the reactive oxygen species production may be regulated. ⁷ As another effect of reactive oxygen species, recent reports suggest that the low concentration of reactive oxygen species in the cell may regulate signaling pathways. ⁸

Superoxide dismutases (SODs), catalase, glutathione peroxidases (GPXs), and some vitamins are representative antioxidants that protect against reactive oxygen species toxicity in cells. ^4,5 Glutathione peroxidases are believed to be an important antioxidant enzyme and effectively reduce the toxicity of H₂O₂ in vitro and in vivo. ^5,9 Furthermore, the active domain of GPX essentially contains selenium atoms. ⁹ Various studies have reported that selenoproteins, including GPX, reduce oxidative stress in cells. ^10,11 Thus, various organic selenium compounds may be candidates for reactive oxygen species scavengers.

Ebselen is a 5-membered ring selenium-containing heterocyclic compound showing GPX-like activity ¹² and is a synthetic antioxidant with reactive oxygen species scavenger effects. ¹³ The antioxidant effects of ebselen are due to its selective blockade of leukocyte infiltration and activation, leading to the elimination of H₂O₂. ¹³ This compound is a multifunctional antioxidant and a potential chemopreventive agent in inflammation-associated carcinogenesis. ¹⁴ As such, various types of organic selenium compounds may be applicable to the reduction of oxidative stress ^{15 –17} and may be used as antioxidative/anti-inflammatory substances for therapeutic drugs against inflammatory diseases (ie, atopic dermatitis) and stroke. ¹³ Morey et al demonstrated that some selenoproteins, such as selenoprotein P, regulate the redox potential in cells, resulting in the modulation of various phosphorylation pathways including Ras/mitogen-activated protein kinase (MAPK) signaling. ¹⁸ On the other hand, we demonstrated that ebselen also influenced the redox potential in rat pheochromocytoma cells (PC12 cells) and induced activation of MAPK/extracellular signal-regulated protein kinases ([ERK1/2], MAPK/ERK1/2) and neural differentiation via regulation of kinases or phosphatases involved in intracellular signaling. ¹⁹ Thus, such selenium compounds are thought to play roles not only as reactive oxygen species scavengers but also as modulators of intracellular signaling.

Against this research background, we sought to synthesize various organic selenium compounds, such as selenoamides, ¹⁷ thio- and selenoureas, ¹⁶ and selenocarbamates, ¹⁵ in order to evaluate their biological potential. It is already known that most of these compounds are also highly efficient in vitro O₂ ⁻ scavengers. ^15,16 Selenoureas in particular showed pronounced antiinflammatory effects with low toxicity in various human cell lines. ²⁰ In this study, we newly synthesized 5-membered ring compounds such as 2-selenoxo-1,3-thiazolidin-4-one (compound A and its analogues: compounds Aa-d) and 3-alkoxy-4,5-dihydro-5-selenoxo-1H-1,2,4-triazole-1-carboxylate (compound B and its analogues: compounds Be-j), and their superoxide radical (O₂ ⁻)-scavenging effects, as antioxidants were evaluated by a highly sensitive and quantitative real-time kinetic chemiluminescence method. Moreover, to evaluate the biological effects such as cytotoxicity, cell death inhibition, and cell death/cell proliferation-related signaling pathways of 3-(2,6-dimethylphenyl)-2-selenoxo-1,3-thiazolidin-4-one (compound Aa) and ebselen, we examined whether they could activate MAPK/ERK1/2 and suppress H₂O₂-induced cytotoxicity in PC12 cells.

Materials and Methods

Materials

Compounds Aa-d and Be-j were prepared according to previously reported procedures. ^21,22 A cypridina luciferin analog, 2-methyl-6-(4-methoxyphenyl)-3,7-dihydroimidazo-[1,2-] pyrazin-3-one hydrochloride (MCLA), obtained from Tokyo Kasei (Tokyo, Japan), was used as a chemiluminescent probe for superoxide radicals. 2-Methyl-6-(4-methoxyphenyl)-3,7-dihydroimidazo-[1,2-] pyrazin-3-one hydrochloride was dissolved in twice-distilled water and stored at −80°C prior to use. The concentration of 2-methyl-6-(4-methoxyphenyl)-3,7-dihydroimidazo-[1,2-] pyrazin-3-one hydrochloride solution was determined by absorbance at 430 nm using an absorbance coefficient value of ∊ = 9600 M⁻¹ cm⁻¹, as described. ²³ Superoxide dismutase (lyophilized powder, 3400 units/mg protein) and xanthine oxidase (grade III) were purchased from Sigma Chemical Co (St Louis, Missouri). Hypoxanthine was purchased from Wako Pure Chemical Industries Ltd (Osaka, Japan) and used without further purification. All other chemicals and solvents were of analytical grade and used without further purification.

Synthetic methods for the preparation of compounds Aa-d

A round-bottom flask equipped with a condenser and a magnetic stirrer was charged with an aryl isoselenocyanate (1.0 mmol) in 20 mL of a mixture of ethyl alcohol and water (2:1). An equimolar amount of the respective mercapto carboxylic acid was then added dropwise. The mixture was stirred at room temperature for several hours before being poured into 50 mL of cold water. After 2 hours of stirring, the precipitate was filtered and purified, if necessary, by recrystallization from ethyl alcohol.

3-(2,6-Dimethylphenyl)-2-selenoxo-1,3-thiazolidin-4-one (compound Aa)

Yield 56%. Orange powder. Mp: 123°C to 125°C. ¹ H-NMR (300 MHz, CDCl₃): δ 2.10 (s, 2 Me), 3.95 (s, CH₂), 7.18–7.33 (m, 3 arom. H). ¹³C-NMR (75 MHz, CDCl₃): δ 17.5, 37.3, 127.5, 127.9, 130.0, 136.1, 173.2, 200.2. CI-MS: 303 (60, [M+NH₄]⁺), 286 (100, [M+H]⁺). Theoretically calculated values were for C₁₁H₁₁NOSSe: C, 46.48; H, 3.90; N, 4.93; S, 11.28. Found: C, 46.33; H, 4.12; N, 5.00; S, 11.32.

3-(4-Methylphenyl)-2-selenoxo-1,3-thiazolidin-4-one (compound Ab)

Yield 91%. Mp: 201°C to 203°C (ethyl alcohol). ¹ H-NMR (300 MHz, CDCl₃): δ 2.42 (s, Me), 3.90 (s, CH₂), 7.10, 7.35 (AA′BB′, J = 8.2, 4 arom. H). ¹³C-NMR (75 MHz, CDCl₃): δ 21.3, 37.4, 128.0, 130.3, 133.2, 140.0, 173.2, 203.3. CI-MS: 289 (31, [M+NH₄]⁺), 272 (100, [M+H]⁺). Theoretically calculated values were for C₁₀H₉NOSSe: C, 44.45; H, 3.36; N, 5.18; S, 11.87. Found: C, 44.30; H, 3.21; N, 5.03; S, 11.78.

3-(4-Methoxyphenyl)-2-selenoxo-1,3-thiazolidin-4-one (compound Ac)

Yield 93%. Mp: 182°C to 184°C (ethyl alcohol). ¹ H-NMR (300 MHz, CDCl₃): δ 3.85 (s, MeO), 3.89 (s, CH₂), 7.03, 7.14 (AA′BB′, J = 8.2, 4 arom. H). ¹³C-NMR (75 MHz, CDCl₃): δ 37.3, 55.4, 114.9, 128.2 (1 arom. C), 129.4, 160.3, 173.3, 203.6. CI-MS: 305 (32, [M+NH₄]⁺), 288 (100, [M+H]⁺). Theoretically calculated values were for C₁₀H₉NO₂SSe: C, 41.96; H, 3.17; N, 4.89; S, 11.20. Found: C, 42.23; H, 3.33; N, 4.88; S, 11.12.

3-(4-Bromophenyl)-2-selenoxo-1,3-thiazolidin-4-one (compound Ad)

Yield 89%. Mp: 111°C to 113°C (ethyl alcohol). ¹ H-NMR (300 MHz, CDCl₃): δ 3.92 (s, CH₂), 7.11, 7.67 (AA′BB′, J = 8.1, 4 arom. H). ¹³C-NMR (75 MHz, CDCl₃): δ 37.4, 123.3, 130.0, 132.9, 138.5, 171.8, 203.8. CI-MS: 353 (35, [M+NH₄]⁺), 336 (100, [M+H]⁺). Theoretically calculated values were for C₉H₆NOSSeBr: C, 32.26; H, 1.80; N, 4.18; S,9.57. Found: C, 32.41; H, 2.02; N, 3.98; S, 10.00.

Synthetic methods for the preparation of compounds Be-j

A 25-mL round-bottom flask equipped with magnetic stirrer and condenser was charged with a mixture of diethyl or bis(tert-butyl) azodicarboxylate (0.92 mL, 2.0 mmol) and Ph₃P (524 mg, 2.0 mmol) in dichloromethane (20 mL). The mixture was stirred under N₂ atmosphere at 0°C (ice bath) for 30 minutes. The appropriate amount of isoselenocyanate (2.0 mmol) was added in 1 portion. The mixture was stirred for 15 hours at room temperature and then evaporated to dryness under reduced pressure. The crude product was purified by column chromatography on SiO₂ using hexane/ethyl acetate (100/0 to 50/50) as the eluent and recrystallized in ethyl acetate.

Ethyl 3-ethoxy-4,5-dihydro-4-phenyl-5-selenoxo-1H-1,2,4-triazole-1-carboxylate (compound Be)

Yield: 552 mg (81%). Yellowish crystals. Mp: 142°C to 144°C (ethyl acetate). IR (cm⁻¹): 3422w (br), 2981w, 2931w, 1773s, 1619s, 1595w, 1501w, 1451m, 1386m, 1368m, 1328s, 1308s, 1220s, 1175w, 1155w, 1109w, 1088w, 1066w, 1028m, 1005m, 979w, 902w, 861w, 847w, 776w, 710w, 689w. ¹ H-NMR (300 MHz, CDCl₃): δ 1.36 (t, J = 7.1, CH₃), 1.49 (t, J = 7.1, CH₃), 4.49 to 4.62 (m, 2 CH₂), 7.35 (d, J = 8.1, 2 arom. H), 7.52 to 7.59 (m, 3 arom. H). ¹³C-NMR (75 MHz, CDCl₃): δ 14.0, 14.1, 65.0, 68.5, 128.1, 129.4, 130.0, 132.6, 148.2, 156.3, 167.8. ESI-MS: 360 (14), 361 (13), 362 (52), 363 (3), 364 (100, [M+Na]⁺), 365 (11), 366 (15). Theoretically calculated values were for C₁₃H₁₅N₃O₃Se: C, 45.89; H, 4.44; N, 11.86. Found: C, 45.58; H, 4.58; N, 12.64.

Ethyl 3-ethoxy-4,5-dihydro-4-(4-methylphenyl)-5-selenoxo-1H-1,2,4-triazole-1-carboxylate (compound Bf)

Yield: 554 mg (78%). Yellowish crystals. Mp 120°C to 122°C (ethyl acetate). IR (cm⁻¹): 3442w (br), 2979w, 2956w, 1768s, 1619s, 1514m, 1477w, 1454m, 1389m, 1368m, 1327s, 1307s, 1294s, 1220s, 1173w, 1152w, 1109w, 1099w, 1064w, 1029m, 981w, 905w, 850w, 817w, 753w, 712w, 622w. ¹ H-NMR (300 MHz, CDCl₃): δ 1.37 (t, J = 7.1, CH₃), 1.49 (t, J = 7.1, CH₃), 2.43 (s, CH₃), 4.48 to 4.60 (m, 2 CH₂), 7.22, 7.34 (AA′BB′, J = 8, 4 arom. H). ¹³C-NMR (75 MHz, CDCl₃): δ 14.0, 14.3, 21.3, 65.0, 68.4, 127.7, 130.0, 127.3, 140.2, 149.0, 156.0, 167.1. ESI-MS: 374 (12), 375 (14), 376 (48), 377 (4), 378 (100, [M+Na]⁺), 379 (11), 380 (13), 733 (3). Theoretically calculated values were for C₁₄H₁₇N₃O₃Se: C, 47.46; H, 4.84; N, 11.86. Found: C, 47.25; H, 4.85; N, 11.85.

Ethyl 3-ethoxy-4-(4-fluorophenyl)-4,5-dihydro-5-selenoxo-1H-1,2,4-triazole-1-carboxylate (compound Bg)

Yield: 603 mg (84%). Yellowish crystals. Mp 151°C to 153°C (ethyl acetate). IR (cm⁻¹): 3442w (br), 3085w, 2982w, 1769s, 1621s, 1512s, 1475w, 1454m, 1386m, 1367m, 1311s, 1290s, 1216s, 1171w, 1150w, 1112w, 1088w, 1063m, 1024m, 1004w, 977m, 905w, 853m, 821w, 759w, 725w, 711w, 635w, 622m. ¹ H-NMR (300 MHz, CDCl₃): δ 1.35 (t, J = 7.1, CH₃), 1.49 (t, J = 7.1, CH₃), 4.41 to 4.51 (m, 2 CH₂), 6.95, 7.19 (AA′BB′, J = 8, 4 arom. H). ¹³C-NMR (75 MHz, CDCl₃): δ 14.0, 14.3, 65.1, 68.6, 116.4, 130.1, 128.4, 148.9, 155.8, 162.3, 167.1. ESI-MS: 378 (15), 379 (12), 380 (55), 381 (4), 382 (100, [M+Na]⁺), 383 (15), 384 (17). Theoretically calculated values were for C₁₃H₁₄N₃O₃SeF: C, 43.59; H, 3.94; N, 11.73. Found: C, 43.37; H, 4.00; N, 11.88.

Ethyl 4-(4-chlorophenyl)-3-ethoxy-4,5-dihydro-5-selenoxo-1H-1,2,4-triazole-1-carboxylate (compound Bh)

Yield: 726 mg (97%). Yellowish crystals. Mp 142°C to 144°C (ethyl acetate). IR (cm⁻¹): 3442w (br), 2982w, 1765s, 1625s, 1495m, 1469w, 1442w, 1373m, 1310s, 1289s, 1218s, 1173w, 1148w, 1089m, 1025m, 977m, 907w, 836w, 753w, 703w. ¹ H-NMR (300 MHz, CDCl₃): δ 1.31 (t, J = 7.1 Hz, CH₃), 1.41 (t, J = 7.1, CH₃), 4.39 to 4.51 (m, 2 CH₂), 7.24, 7.43 (AA′BB′, J = 8, 2H, 4 arom. H). ¹³C-NMR (75 MHz, CDCl₃): δ 14.1, 14.3, 65.1, 68.7, 124.6, 129.5, 129.7, 131.0, 148.9, 155.6, 166.9. ESI-MS: 394 (13), 395 (14), 396 (51), 397 (4), 398 (100, [M+Na]⁺), 399 (18), 400 (49). Theoretically calculated values were for C₁₃H₁₄N₃O₃SeCl: C, 41.67; H, 3.77; N, 11.21. Found: C, 41.63; H, 3.84; N, 11.34.

Ethyl 4-(4-bromophenyl)-3-ethoxy-4,5-dihydro-5-selenoxo-1H-1,2,4-triazole-1-carboxylate (compound Bi)

Yield: 686 mg (82%). Yellowish crystals. Mp 147°C to 149°C (ethyl acetate). IR (cm⁻¹): 3444w (br), 2982w, 2935w, 1755s, 1718m, 1638s, 1591w, 1493m, 1451m, 1392m, 1366m, 1327s, 1303s, 1286s, 1215w, 1170w, 1154w, 1104w, 1068w, 1030m, 1011m, 988w, 909w, 850w, 754w, 711w. ¹ H-NMR (300 MHz, CDCl₃): δ 1.31 (t, J = 7.1, CH₃), 1.41 (t, J = 7.1, CH₃), 4.41 to 4.51 (m, 2 CH₂), 7.20, 7.58 (AA′BB′, J = 8, 4 arom. H). ¹³C-NMR (75 MHz, CDCl₃): δ 14.1, 14.3, 65.1, 68.7, 124.2, 129.8, 132.7, 131.5, 148.9, 155.6, 166.8. ESI-MS: 438 (15), 439 (12), 440 (55), 441 (4), 442 (100, [M+Na]⁺), 443 (14), 444 (12). Theoretically calculated values were for C₁₃H₁₄N₃O₃SeBr: C, 37.25; H, 3.37; N, 10.03. Found: C, 37.12; H, 3.23; N, 9.99.

Tert-Butyl-3-(tert-butoxy)-4-(4-bromophenyl)-4,5-dihydro-5-selenoxo-1H-1,2,4-triazole-1-carboxylate (compound Bj)

Yield: 855 mg (90%). Yellowish crystals. Mp 118°C to 120°C (ethyl acetate). IR (cm⁻¹): 3443w (br), 2985w, 2931w, 1764s, 1701w, 1612s, 1488m, 1453w, 1408w, 1372m, 1337m, 1320s, 1297m, 1278s, 1223m, 1143s, 1065w, 1000m, 895w, 848m, 836m, 804w, 761w, 655w. ¹ H-NMR (300 MHz, CDCl₃): δ 1.53 (s, 3 CH₃), 1.67 (s, 3 CH₃), 7.21, 7.62 (AA′BB′, J = 8, 4 arom. H). ¹³C-NMR: δ 27.6, 27.8, 86.6, 88.1, 123.8, 129.8, 132.5, 132.1, 147.1, 153.5, 165.1. ESI-MS: 494 (17), 495 (15), 496 (55), 497 (25), 498 (100, [M+Na]⁺), 499 (18), 500 (75), 501 (12), 502 (10). Theoretically calculated values were for C₁₇H₂₂N₃O₃SeBr: C, 42.96; H, 4.67; N, 8.83. Found: C, 43.22; H, 4.95; N, 9.15.

Assay of superoxide anion-scavenging activities

The superoxide anion-scavenging activities of compounds Aa-d and Be-j were measured by a previously reported method. ²³ In brief, hypoxanthine (5 × 10^{− 5} mol/L) and EDTA (0.1 mmol/L) were prepared in 3 mL of 50 mmol/L tris(hydroxymethyl)aminomethane-HCl (Tris-HCl) buffer, pH 7.8 (the standard solution), with or without various concentrations of compounds Aa-d and Be-j. To evaluate certain superoxide anion-scavenging activities, we prepared a calibration curve using standard solutions of SOD (0.6-30 ng/mL). Compounds Aa-d and Be-j (25 mmol/L) were dissolved in dimethyl sulfoxide and stored at −80°C prior to use. Chemiluminescence was measured with a luminometer (Aloka, BLR201) at 25°C. The reaction was initiated by the addition of 2-methyl-6-(4-methoxyphenyl)-3,7-dihydroimidazo-[1,2-] pyrazin-3-one hydrochloride (5.8 × 10^{− 7} mol/L) to the standard in solution excluding xanthine oxidase, continued for 2.5 minutes and then for a further 2 minutes after the addition of xanthine oxidase (6.5 U). A typical result for compound Aa is shown in Figure 1 . Superoxide anion-scavenging activity was found to be too strong when the compounds were tested primarily at 166 μmol/L; thus, the activity was measured at 8.3, 16.6, 33.3, or 41.0 μmol/L. The percentage inhibition of 2-methyl-6-(4-methoxyphenyl)-3,7-dihydroimidazo-[1,2-] pyrazin-3-one hydrochloride–dependent chemiluminescence was calculated as previously described. ²³ The concentration for 50% inhibition (IC₅₀) was calculated from the values obtained at 4 different concentrations of compound Aa (8.3, 16.6, 33.3, and 41.0 μmol/L).

Figure 1.

Effect of compound Aa on 2-methyl-6-(4-methoxyphenyl)-3,7-dihydroimidazo-[1,2-] pyrazin-3-one hydrochloride (MCLA)–dependent luminescence. A, Chemiluminescence inhibition curves by compound Aa. Arrows indicate the addition of MCLA or xanthine oxidase (XOD). B, Concentration-dependent change in superoxide anion-scavenging activity to determine the IC₅₀ value of compound Aa.

Measurement of Uric Acid Generation in Hypoxanthine–Xanthine Oxidase System

The amounts of uric acid were measured using a uric acid measurement kit (L type WAKO UA M; Wako Pure Chemical Industries, Ltd).

Cell Culture

In the present study, we used rat PC12 cells because previous reports suggested that this cell line is a useful tool as various models of neurological dysfunctions. ²⁴ Pheochromocytoma cells were cultured as described previously. ²⁵ In brief, the cells were maintained in Dulbecco modified Eagle medium (DMEM; Sigma) supplemented with 10% heat-inactivated horse serum ([HS] Gibco BRL, Grand Island, New York) and 5% heat-inactivated fetal bovine serum ([FBS] Sanko Junyaku, Co, Ltd, Tokyo, Japan; serum-containing medium) or in DMEM supplemented with 1% bovine serum albumin ([BSA] serum-free medium).

Detection of Phosphorylated Proteins

Each compound was suspended in the serum-free medium and sonicated until fully emulsified. Pheochromocytoma cells were seeded at 2 × 10⁶ cells per well onto collagen-coated 6-well plates (Corning Incorporated Life Sciences, Lowell, Massachusetts) in the serum-containing medium and precultured for 2 days at 37°C in an atmosphere of 95% air/5% CO₂. The cells were then washed with phosphate-buffered saline (PBS) and incubated with the above-mentioned culture medium containing the various agents shown in Figure 2 for 10 minutes at 37°C. The culture plates were then placed on ice and each well was washed with 3 mL of 2 mmol/L Tris-HCl buffer (pH 8.0) containing 0.33 mol/L NaF and 6.25 mol/L Na₃VO₄ and subsequently lysed with 150 μL of 20 mmol/L Tris-HCl buffer (pH 8.0) containing 150 mmol/L NaCl, 2 mmol/L EDTA, 1% Nonidet P-40 (w/v), 1% sodium deoxycholate (w/v), 0.1% sodium dodecyl sulfate (SDS; w/v), 50 mmol/L NaF, 0.1% aprotinin (w/v), 0.1% leupeptin (w/v), 1 mmol/L Na₃VO₄, and 1 mmol/L phenylmethylsulfonylfluoride (PMSF). Cell lysates were collected using a cell scraper and centrifuged at 15 000g for 30 minutes at 4°C. The supernatant was collected, and the overall protein concentration was determined by a BCA Protein Assay Reagent Kit (Pierce, Rockford, Illinois) with BSA as the standard.

Figure 2.

Effects of compound Aa, ebselen, or NGF on the activation of ERK1/2. The final concentration of each reagent used for the culture is shown. The phosphorylated ERK1/2 and ERK1/2 proteins were visualized by Western blot, and the intensity of each band was densitometrically calculated. Ratios of the relative intensity of the phosphorylated and nonphosphorylated ERK1 or 2 bands are expressed as the means ± SD of 3 independent determinations. The control group is indicated as O. Levels of significance, *P < 0.05; **P < 0.01; ***P < 0.005. NGF indicates nerve growth factor; ERK1/2, extracellular signal-regulated protein kinase; SD, standard deviation.

Supernatant fluids containing proteins (20 μg) were mixed with lithium dodecyl sulfate (LDS) sample buffer (Invitrogen Corp, Carlsbad, California) and incubated for 5 minutes at 80°C. Proteins in samples were separated on SDS-polyacrylamide gel electrophoresis, and the proteins in gels were electroblotted onto polyvinylidene fluoride (PVDF) filters (Fluorotrans membrane W, 0.2 μm; Nihon Genetics, Tokyo, Japan). Immunoblotting analysis was performed using monoclonal antibodies against p44/42 ERK or phospho p44/42 ERK (Cell Signaling Technology, Lake Placid, New York) as primary antibodies, followed by reaction with horseradish peroxidase-conjugated anti-rabbit or anti-mouse immunoglobulin G (IgG) antibodies from Promega Co (Fitchburg, Wisconsin) as the secondary antibody. The blots were developed by the enhanced chemiluminescence method (Hyperfilm-ECL plus; Amersham Biosciences Corp, Piscataway, New Jersey).

The Measurement of Cytotoxicity by 5-Dimethyl-2-Thiazolyl)-2,5-Diphenyl-2-Htetrazolium Bromide Assay

Pheochromocytoma cells were cultured onto collagen-coated 96-well plates (2 × 10⁶ cells per well) in serum-containing medium for 2 days at 37°C in an atmosphere of 95% air/5% CO₂. Culture medium was replaced with 50 μL of the serum-free medium containing each test agent after washing with PBS, and the cells were cultured with each test agent for a range of days. The cytotoxicity was determined by the 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) reduction assay. ²⁶ The cells were incubated with 0.25 ng of MTT/mL (final concentration) for 2 hours, and the reaction was stopped by adding 50 μL of 50% (volume/volume [v/v]) dimethylformamide (DMFA) containing 20% (w/v) SDS. The amount of MTT formazan product was determined photometrically using a microplate reader (Model Ultrospec Visible Plate Reader II of Amersham Biosciences, Tokyo, Japan) by measuring the absorbance at 562 nm with a reference wavelength of 630 nm. The relationships between absorbance and cell numbers were clarified, and the number of cells in each well was calculated.

Measurement of H₂O₂ Concentrations

The concentration of H₂O₂ was determined by the horseradish peroxidase-catalyzed oxidation of fluorescent scopoletin method with minor modifications. ²⁷ Briefly, compound Aa or ebselen (100 μmol/L) and H₂O₂ (400 μmol/L) were added to 100 μL of distilled water or serum-free medium in a 96-well plate followed by 2-hour incubation at 37°C. Scopoletin (10 μmol/L) and horseradish peroxidase (50 μg) were added, and the decrease in scopoletin fluorescence was measured using a microplate fluorometer (1420 Multilabel counter, ARVO; excitation and emission at 335 and 460 nm, respectively; Wallac, Turuk, Finland).

Statistical Analysis

The results were expressed as means ± standard deviation (SD). The significant differences between the groups compared were determined using analysis of variance (ANOVA) followed by Dunnett test.

Results

Superoxide Anion-Scavenging Activities of Compounds A and B and Ebselen

The chemical structures, superoxide anion-scavenging activities, and IC₅₀ values of compounds Aa-d and Be-j and ebselen are shown in Table 1 . The superoxide anion-scavenging activities were nearly dose-dependent for all compounds; and Aa, Ac, and Ad showed the highest level of superoxide anion-scavenging activity at 166 μmol/L. As compound Aa had the lowest IC₅₀ value at 25.9 μmol/L (Table 1), it was selected and used for further experiments. The superoxide anion-scavenging activities of these compounds did not change by the addition of 50 μmol/L of U0126, a selective mitogen-activated protein kinase kinase (MEK) inhibitor (data not shown).

Table 1.

Chemical Structures, Superoxide Anion-Scavenging Activities, and IC₅₀ Values of 2-Selenoxo-1,3-Thiazolidin-4-Ones 1, 3-Alkoxy-4,5-Dihydro-5-Selenoxo-1H-1,2,4-Triazole-1-Carboxylates and Ebselen ^a

^a IC₅₀ values were determined using linear regression analysis of the dose–response curves of the inhibition of chemiluminescence by the compounds. The superoxide anion-scavenging activities were calculated by a rate of decrease in chemiluminescence value after addition of xanthine oxidase. The superoxide anion-scavenging activities of each compound were expressed as a percentage of that of compound Aa.

We investigated whether compounds Aa-d and Be-j could eliminate superoxide anions generated by xanthine oxidase. For this purpose, the amount of uric acid, which is the by-product of superoxide anions in the mixture for superoxide anion-scavenging activity assay, was measured with or without compound Aa (data not shown). The generation of uric acid was not affected by the addition of a relatively high concentration of compound Aa (41.0 μmol/L), suggesting that this compound did not inhibit xanthine oxidase. In addition, the effects of compound Aa on cultured PC12 cells were investigated.

Effects of Compound Aa, Ebselen, or Nerve Growth Factor on the Activation of ERK1/2

The activation of ERK1/2 is one of the checkpoints to assess the activation of the classical Ras/MAPK cascade, ²⁸ which is triggered by an engaged tyrosine kinase receptor or G-protein-coupled receptor and results in proliferation and/or differentiation. We previously confirmed that ebselen is an activator of the Ras/MAPK cascade and an initiator of neuronal differentiation. ²⁵ Therefore, we examined whether compounds A and B could activate ERK1/2 and found that ERK1/2 was activated only by compound Aa (data not shown). Thereafter, we focused on the biological activity of compound Aa.

The effect of ebselen, compound Aa, or nerve growth factor (NGF) on the phosphorylation of ERK1/2 was examined (Figure 2). Nerve growth factor markedly induced phosphorylation of ERK1/2. Compound Aa and ebselen similarly activated ERK1/2; however, they had a much weaker effect than NGF. On the other hand, phosphorylation of ERK1/2 was not induced in the presence of 400 μmol/L of H₂O₂ (data not shown).

Effects of Compound Aa or Ebselen on Survival of PC12 Cells Exposed to H₂O₂

The oxidative stress-induced cytotoxicity was evaluated by determining the viability of PC12 cells exposed to H₂O₂ (Figure 3 ). First, we studied the relationship between cell death and H₂O₂ concentration. Two hours after the addition of 200, 400, 1000, or 2000 μmol/L H₂O₂, the cell viability was 100%, 36%, 8%, or 0%, respectively. At 400 μmol/L of H₂O₂, a significant decrease in cell viability was observed, suggesting the potent oxidative stress-induced cytotoxicity of H₂O₂.

Figure 3.

Effects of compound Aa or ebselen on survival of PC12 cells exposed to H₂O₂. Pheochromocytoma cells were treated with compound Aa, ebselen, or NGF for 10 minutes, and then the cells were washed with PBS. The cells were then exposed to 400 μmol/L of H₂O₂ for 2 hours. In some experiments, the cells were pretreated for 4 hours with U0126 (50 μmol/L) before treatment with the reagents. The amount of surviving cells was measured by MTT assay as described in the text. Results shown are means ± SD (n = 3). The significance of differences between the groups compared was determined using the analysis of variance (ANOVA) followed by Dunnett test. The control group is indicated as O. Levels of significance, ***P < 0.005; -, no significant difference. NGF indicates nerve growth factor; PBS, phosphate-buffered saline; MTT, 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide; SD, standard deviation.

Pretreatment of PC12 cells with ebselen or compound Aa attenuated this oxidative stress-induced cytotoxicity. However, NGF did not show this effect. Compound Aa was more potent than ebselen, and a concentration of even 50 μmol/L was enough to suppress H₂O₂-induced cell death completely. However, such protective effects of both compound Aa and ebselen disappeared on treatment with U0126.

Effects of Cell Viability of Compound Aa, Ebselen, or NGF on PC12 Cells

Ebselen prevents ischemia- or lead-induced cell death, ^1,2 so we compared the effects of compound Aa, ebselen, and NGF on serum deprivation-induced cytotoxicity (Figure 4 ). Cytotoxicity occurred on addition of more than 100 μmol/L of ebselen, when tested over 6 days after exposure. Compound Aa was nontoxic over 6 days at a concentration of 100 μmol/L but toxic at 1 mmol/L in PC12 cells. These results demonstrated that the toxicity of compound Aa was weaker than that of ebselen. Contrary to these results, NGF increased the cell viability after serum deprivation from culture medium, as described previously (Figure 4C). ²⁹

Figure 4.

Effects of cell viability of compound Aa, ebselen, or NGF on PC12 cells. The PC12 cells were cultured for several days in DMEM containing 1% BSA and various concentrations of compound Aa (A), ebselen (B), or NGF (C). The amount of cells in the well was measured by MTT assay as described in the text. Results shown are means ± SD (n = 3). NGF indicates nerve growth factor; DMEM, Dulbecco modified Eagle medium; BSA, bovine serum albumin; MTT, 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide; SD, standard deviation.

Discussion

In the present study, we have systematically synthesized various organic selenium compounds including selenocarbamates, selenoureas, thioureas, tertiary selenoamides, and 2-amino-1, 3-selenazoles and bis-(2-amino-5-selenazoyl) ketones and found that these compounds have superoxide anion-scavenging activities, and their IC₅₀ values ^{15 –17,30,31} range approximately from 0.1 to 100 μmol/L. From the results shown in Table 1, it was confirmed that the superoxide anion-scavenging activities of these newly synthesized compounds, 2-selenoxo-1,3-thiazolidin-4-ones compounds Aa-d and some of the 3-alkoxy-4,5-dihydro-5-selenoxo-1H-1,2,4-triazole-1-carboxylates compounds Be-j were similar to those of other previously synthesized organic selenium compounds such as selenocarbamates and selenoureas. The results suggest that compound Aa and some of the B compounds are organic selenium compounds and useful reactive oxygen species scavengers. The generation of uric acid was not significantly affected by the addition of a relatively high concentration of compound Aa (41.0 μmol/L), suggesting that our compounds act as scavengers of superoxide anions rather than inhibitors of xanthine oxidase.

A low concentration of O₂ ⁻ in the human body generally plays a beneficial role in biological defenses and intercellular signal transduction. ⁵ On the other hand, excessive O₂ ⁻ production is involved in the pathogenesis of a number of disorders, including inflammation, rheumatoid arthritis, and asthma. ^32,33 Oxidative stress may be defined as an imbalance between the cellular production of reactive oxygen species, a key component of inflammation and inflammatory disorders, and antioxidant defense mechanisms. ⁵ The processes associated with inflammatory responses are complex and often involve the production of reactive oxygen species including O₂ ⁻. In this study, our newly synthesized compounds acted as effective O₂ ⁻ scavengers in vitro. Thus, these compounds may eliminate excessive O₂ ⁻ and lead to suppression of reactive oxygen species overproduction.

Compound Aa and ebselen activated ERK1/2, although their effect was much weaker than that of NGF (Figure 2). Both compound Aa and ebselen suppressed H₂O₂-induced cytotoxicity (Figure 3), but this effect was inhibited by pretreatment with U0126. It was confirmed that activation of the MAPK cascade is required for suppression of cytotoxicity by compound Aa and ebselen. However, H₂O₂-induced cytotoxicity was not inhibited by NGF. As NGF is known to activate the MAPK cascade strongly, ^34,35 it was suggested that activation of only the MAPK cascade was insufficient for suppression of H₂O₂-induced rapid cytotoxicity and another unknown factor is required.

It is known that serum deprivation induces apoptosis in PC12 cells. ²⁵ Although NGF increased cell viability under serum-free cultivation, compound Aa and ebselen did not suppress serum deprivation-induced cell death. Recent studies suggested that upregulation of nuclear factor κB ³⁶ or inhibition of p38 MAPK and Jun-amino-terminal kinase (JNK) activation ³⁷ suppressed apoptosis induced by H₂O₂ in PC12 cells. Therefore, it might be that both compound Aa and ebselen can inhibit rapid apoptosis, such as that induced by H₂O₂. On the other hand, although NGF induced phosphorylation of Akt, one of a serine/threonine protein kinase that plays a key role in apoptosis inhibition, and JNK 2/3, the addition of compound Aa was irrelevant to the phosphorylation of Akt or JNK 2/3 (data not shown). From this, it was supposed that Aa suppressed cytotoxicity by a mechanism different to that of NGF.

It was suggested that H₂O₂ can induce cell death by apoptosis in PC12 cells. ³⁸ In the present study, vesicles were formed and nucleus fragmentation was observed by the addition of Hoechst 33342 solution in the dead cells; therefore, H₂O₂ (400 μmol/L)-induced cell death in the cells may be apoptosis. On the other hand, serum deprivation induced apoptosis in PC12 cells. ²⁵ Because H₂O₂-induced cell death was suppressed by Aa but serum deprivation-induced cell death was not, of the several types of apoptosis, Aa may suppress H₂O₂-induced apoptosis in PC12cells.

We examined whether Aa or ebselen (100 μmol/L) could react with H₂O₂ (400 μmol/L). After a 2-hour reaction, the concentration of H₂O₂ did not change, demonstrating Aa or ebselen did not directly react with H₂O₂. Next, we examined whether Aa or ebselen could react with H₂O₂ (400 μmol/L) in culture medium with PC12 cells. Pheochromocytoma cells were treated for 10 minutes with 100 μmol/L of Aa following which the cells were removed and washed with PBS. The cells were then exposed to 400 μmol/L of H₂O₂ for 2 hours. The H₂O₂ concentration of the supernatant was similar with or without Aa. To exclude the effects of serum in these experiments, a control group utilizing 1% BSA instead of serum was added to the DMEM. H₂O₂ was expected to decrease if Aa or ebselen showed GPX-like activity or if GPX was induced by the addition of Aa or ebselen. Since the presence or absence of Aa or ebselen did not affect the concentration of H₂O₂, Aa or ebselen was believed to be independent of GPX concentration or GPX-like activity in this study.

Oxidative stress is one of the common causal factors of neuronal cell death. ³⁹ H₂O₂-induced cytotoxicity was clearly suppressed by compound Aa and the effect was more potent than that of ebselen. From the results of cytotoxicity measurements (Figure 4), it was confirmed that the toxicity of compound Aa was lower than that of ebselen. It appears that the toxicity of compound Aa will not hinder its usage as a drug, and its clinical application against oxidative stress is hopeful.

Antioxidants such as ebselen regulate intercellular signaling via regulation of kinase and phosphatase. ¹⁹ In this study, we found that compound Aa showed superoxide anion-scavenging activity, induction of ERK1/2 phosphorylation, and suppression of H₂O₂-induced cell death. It has been suggested that H₂O₂-induced cell death is partially associated with ERK phosphorylation. ^{40 –42} However, this mechanism may not be exactly known at present. To address these relationships, further studies may be needed.

On the other hand, NGF did not suppress H₂O₂-induced cell death, although ERK1/2 proteins in the PC12 cells were phosphorylated, and serum deprivation-induced apoptosis was suppressed. Thus, it was assumed that compound Aa activated another pathway in addition to the MAPK cascade. Both compound Aa and NGF inhibited cell death. However, since the mechanisms were different, complementary activity could be expected using both together. Moreover, Aa might be able to cross the blood–brain barrier because of its low molecular weight compared with NGF. However, the potential of compound Aa to prevent and treat some neurological disorders seems to be high enough to warrant future evaluation because the promotion of morphological differentiation was not observed in PC12 cells with the addition of Aa.

In this study, we found that 3-(2,6-dimethylphenyl)-2-selenoxo-1,3-thiazolidin-4-one, compound Aa, had the highest superoxide anion-scavenging activity among the related compounds tested. Furthermore, it was clarified that compound Aa activated ERK1/2 of cultured PC12 cells and suppressed H₂O₂-induced cytotoxicity. However, the cytotoxicity of compound Aa was lower than that of ebselen, an analogous active selenium-containing compound. Compound Aa has interesting medical potential to protect against and treat some neurological disorders and may even be effective in treating depression induced by stress.