Journal of Inorganic Biochemistry 103 (2009) 432–440
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry journal homepage: www.elsevier.com/locate/jinorgbio
Dinuclear copper(II) complexes containing 6-(benzylamino)purines as bridging ligands: Synthesis, characterization, and in vitro and in vivo antioxidant activities Pavel Štarha a, Zdeneˇk Trávnícˇek a,*, Radovan Herchel a, Igor Popa a, Pavel Suchy´ b, Ján Vancˇo a,c a
´ University, Krˇízˇkovského 10, CZ-771 47 Olomouc, Czech Republic Department of Inorganic Chemistry, Faculty of Science, Palacky Department of Human Pharmacology and Toxicology, Faculty of Pharmacy, University of Veterinary and Pharmaceutical Sciences, Palackého 1–3, CZ-612 42 Brno, Czech Republic c Department of Chemical Drugs, Faculty of Pharmacy, University of Veterinary and Pharmaceutical Sciences, Palackého 1–3, CZ-612 42 Brno, Czech Republic b
a r t i c l e
i n f o
Article history: Received 13 October 2008 Received in revised form 4 December 2008 Accepted 17 December 2008 Available online 27 December 2008 Keywords: Copper(II) 6-(Benzylamino)purine SOD-mimic activity Antidiabetic activity Dinuclear complexes
a b s t r a c t A series of dinuclear copper(II) complexes involving 6-(benzylamino)purine derivatives, (HLn), as bridging ligands were synthesized, characterized and tested for both their in vitro and in vivo antioxidant activities. Based on results of elemental analyses, temperature dependence of magnetic susceptibility measurements, UV–vis, FTIR, EPR, NMR and MALDI-TOF mass spectroscopy, conductivity measurements and thermal analyses, the complexes with general compositions of [Cu2(l-HLn)4Cl2]Cl2 2H2O (1–4) and [Cu2(l-HLn)2(l-Cl)2Cl2] (5–7) were prepared {where n = 1–4; HL1 = 6-[(2-methoxybenzyl)amino]purine, HL2 = 6-[(4-methoxybenzyl)amino]purine, HL3 = 6-[(2,3-dimethoxybenzyl)amino]purine and HL4 = 6[(3,4-dimethoxybenzyl)amino]purine}. In the case of complexes 2, 3, 5 and 7, the antioxidant activities were studied by both in vitro {superoxide dismutase-mimic (SOD-mimic) activity} and in vivo {cytoprotective effect against the alloxan-induced diabetes (antidiabetic activity)} methods. The obtained IC50 value of the SOD-mimic activity for the complex 5 (IC50 = 0.253 lM) was shown to be even better than that of the native bovine Cu,Zn-SOD enzyme (IC50 = 0.480 lM), used as a standard. As for the antidiabetic activity, the pretreatment of mice with complexes 3 and 7 led to the complete elimination of cytotoxic attack of alloxan and its free radical metabolites, used as a diabetogenic agent. The cytoprotective effect of these compounds was proved by the preservation of the initial blood glucose levels of the pretreated animals, as against the untreated control group. Ó 2008 Elsevier Inc. All rights reserved.
1. Introduction 6-(Benzylamino)purine {N6-(benzyl)adenine; Bap} belongs to a family of plant hormones called cytokinins. Some of the compounds involving the Bap skeleton, as well as its suitable derivatives, influence both stimulatory and inhibitory processes in the plant tissues [1]. Another important feature of the Bap derivatives, especially those with suitable substituent at the C2 and N9 positions of the purine moiety, is the fact that some of them may behave as the inhibitors of cyclin dependent kinases (CDK) in various phases of the human cell cycle [2]. One of the best-known CDK inhibitors derived from the 6-(benzylamino)purine moiety is (R)Roscovitine, i.e. 2-[(R)-(1-ethyl-2-hydroxyethylamino)]-6-(benzylamino)-9-isopropylpurine (named also Seliciclib or CYC202), that is currently in the IIb phase of clinical trials in patients with nonsmall cell lung cancer (NSCLC) [3,4]. Recently, we reported the preparation and characterization of several copper(II) complexes bearing Bap derivatives, i.e. [Cu2(lBapH)2(l-Cl)2(BapH)2Cl2] 2H2O [5,6], [Cu2(l-BapH)2(l-Cl)2Cl2] * Corresponding author. Tel.: +420 585 634 944; fax: +420 585 634 954. E-mail address:
[email protected] (Z. Trávnícˇek). 0162-0134/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.jinorgbio.2008.12.009
[5,7], [Cu2(l-Cl)2(BapH)2Cl2] [6], [Cu2(l-Bap)2(l-Cl)2(H2O)2] [6] and [Cu(BapH+)2Cl3]Cl nH2O [6,7]. Moreover, the in vitro cytotoxicity of these complexes against selected human cancer cell lines {osteogenic sarcoma (HOS), breast adenocarcinoma (MCF7), chronic myelogenous erythroleukemia (K562), malignant melanoma (G361), mouse melanoma (B16-F0)} has been also evaluated, and their structure–activity relationships have been reviewed by E.R.T. Tiekink in the literature [8]. The most of the above-mentioned complexes showed moderate cytotoxicity, with IC50 values ca 30–60 lM, or were found to be non-cytotoxic. However, the notable cytotoxicity was achieved for the complex [Cu2(l-2-ClBapH)2(l-Cl)2(2-ClBapH)2Cl2] 2H2O, where 2-ClBapH = 6-[(2-chlorobenzyl)amino]purine, with IC50 = 8.2 lM on B16-F0 cell line and IC50 = 17 lM on G361 cell line. For that reason, the last-mentioned type of dinuclear complexes became the object of this study. Next motivation to realize the presented work was to extend the spectrum of biological tests relating to these types of copper(II) complexes, and as a consequence of this, to make attempts to find any biological property of such compounds which could be applicable in practice, e.g. in medicine. Thus, although some of the structural types of the above-mentioned copper(II) complexes have been already prepared, we decided to prepare some of them
P. Štarha et al. / Journal of Inorganic Biochemistry 103 (2009) 432–440
again, however to change formerly used Bap derivatives for the ligands HL1–HL4, and moreover, to test them for their both in vitro (SOD-mimic) and in vivo (antidiabetic) antioxidant activities. The reason to choose such type of the biological tests is fact that the Cu,Zn-SOD enzyme (SOD = superoxide dismutase) may be considered as one of the most important naturally occurring copper(II) proteins of aerobic organisms, protecting cells against the toxic effects of oxygen free radicals [9]. It has the ability to decrease toxicity of the superoxide anion radical ( O 2 ) in the animal and human organisms. Cu,Zn-SOD converts O 2 to the less reactive agents (such as H2O2, further eliminated by a catalase), thus helping to protect our bodies against the one of the most important representatives of free radicals, playing the crucial role in many cells in connection with the production of more reactive oxygen or nitrogen reactive species (RONS) with high deleterious consequences to the tissues. The Cu(II) ions present in the active site of this enzyme were found to be responsible for the redox-based dismutation process. Such information led to syntheses and antioxidant activity testing of many copper(II) complexes [10–13], among them, e.g. [Cu2(tppen)(H2O)2](ClO4)4 and [Cu2(tppen)Cl4] {tppen = N,N,N’,N’-tetrakis(6-pivalamido-2-pyridylmethyl)ethy- lenediamine} were found to be also significantly active in case of the SOD-mimic activity testing, with IC50 values of 0.76 and 0.54 lM, respectively, [10]. The next type of antioxidant activity testing that we used in the present study is based on the properties of alloxan (2,4,5,6 [1H,3H]-pyrimidinetetrone; A) [13–15]. This compound is toxic to the b-cells of Langerhans islets of pancreas due to several free radicals {e.g. O 2 , hydroxyl radicals ( OH) or C-based alloxan radi cals ( AH)} produced in the b-cells while alloxan is metabolically reduced to dialuric acid (AH2), which could be re-oxidized back to alloxan. Thus, the free radical scavengers with ability to transform these toxic agents, and thus to protect b-cells, show the antidiabetic activity in this animal model. For instance, some of copper(II) Schiff base complexes were prepared and in vivo tested on alloxan-induced diabetic mice. The antidiabetic activity of some of them, namely [Cu2(L)2(H2O)]H2O and [Cu(L)(H2O)]n {L = N-salicylidene-b-alanine(2-)}, was found to be significant since the glucose levels of mice pretreated with these complexes were lower than those determined for TROLOXÒ (a soluble analogue of the vitamin E) [13,14]. This work brings several new findings on the field of the syntheses, physical properties and biological activities of dinuclear copper(II) complexes involving benzyl-substituted derivatives of 6(benzylamino)purine. We prepared and characterized copper(II) complexes of the general formula [Cu2(l-HLn)4Cl2]Cl2 2H2O (1–
433
4) and [Cu2(l-HLn)2(l-Cl)2Cl2] (5–7), where HL1 = 6-[(2-methoxybenzyl)amino]purine, HL2 = 6-[(4-methoxybenzyl)amino]purine, HL3 = 6-[(2,3-dimethoxybenzyl)amino]purine, HL4 = 6-[(3,4-dimethoxybenzyl)amino]purine. The syntheses of the complexes 1–4 were performed in 0.1 M HCl. This concentration of HCl was used for the first time for the preparation of copper(II) complexes with the Bap derivatives. Surprisingly, no protonation of the Bap ligands occurred during the syntheses of the complexes in contrast to zinc(II) complexes involving 6-(benzylamino)purine-based CDK inhibitors as ligands, where the ligands were single protonated [16]. The copper(II) complexes 5–7 were synthesized in ethanol. As a continuation of the biological activity screening of the copper(II) complexes with Bap derivatives, the compounds 2, 3, 5 and 7 were tested for both in vitro (SOD-mimic) and in vivo (cytoprotective effect against alloxan-induced diabetes) antioxidant activities. 2. Experimental 2.1. Materials Chemicals and solvents were obtained from the commercial sources (Sigma–Aldrich Co., Acros Organics Co., Fluka Co., Lachema Co.) and used without further purification. The organic compounds 6-[(2-methoxybenzyl)amino]purine (HL1), 6-[(4-methoxybenzyl)amino]-purine (HL2), 6-[(2,3-dimethoxybenzyl)amino]purine (HL3) and 6-[(3,4-dimethoxybenzyl)amino]purine (HL4) (see Fig. 1) were synthesized according previously published procedure [17]. These organic compounds were characterized by elemental analysis (C, H, N), melting point measurements, thin layer chromatography (TLC) and FTIR and NMR spectroscopies. 2.2. Methods Elemental analyses (C, H, N) were performed on a Flash EA1112 Elemental Analyser (Thermo Finnigan). The content of copper was determined using the chelatometric titration with murexide as an indicator. Determinations of the melting points were performed using a Melting Point B-540 apparatus (Büchi) with the gradient of 5 °C min1 and were uncorrected. It is necessary to emphasize at this point, that the determined melting points are connected with degradation of the complexes in all cases. Conductivity measurements of 103 M DMF solutions of prepared complexes were made using a Cond 340i/SET (WTW) conductometer at 25 °C. Electronic absorption spectra of 103 M DMF solutions and diffuse-reflectance spectra were recorded with
Fig. 1. 6-(Benzylamino)purine derivatives HL1–HL4 used for the syntheses of the copper(II) complexes 1–7; given with the atoms numbering.
434
P. Štarha et al. / Journal of Inorganic Biochemistry 103 (2009) 432–440
a Lambda 40 spectrometer (Perkin Elmer Instruments) in the range of 200–1000 nm. Nexus 670 FT–IR spectrometer (ThermoNicolet) was used to record infrared spectra of all compounds using the KBr pellets (400–4000 cm1) and Nujol technique (150– 600 cm1). EPR spectra of polycrystalline samples were recorded on a MiniScope MS200 spectrometer (Magnettech) at both laboratory and liquid nitrogen temperatures (300 and 77 K); standard TEMPONE (2,2,6,6-tetramethylpiperidone-N-oxyl) [18]. The magnetic susceptibility measurements were performed using a SQUID magnetometer (Quantum Design) in the temperature range of 2–300 K with an applied field of 1 T. The diamagnetic corrections were calculated using Pascal constants [19] and the correction for the temperature-independent paramagnetism vTIP = +0.75 m3 mol1 per Cu(II) ion was applied. Thermogravimetric (TG) and differential thermal analyses (DTA) were performed using a thermal analyzer Exstar TG/DTA 6200 (Seiko Instruments Inc.) in dynamic air conditions (150 mL min1) between the room temperature (20 °C) and 1100 °C (gradient 5 °C min1). Positive ion mass spectra were measured in the reflectron mode using Microflex MALDI-TOF LRF20 mass spectrometer (Bruker Daltonics) equipped with a nitrogen laser operating at 337 nm (laser repetition rate of 10 Hz, number of laser shots of 100–200); the peptide mixture was used as an external standard. The samples for MALDI-TOF-MS were prepared as follows. A saturated solution of a-cyano-4-hydroxycinnamic acid in 50% acetonitrile/0.2% trifluroacetic acid (TFA) was used as a MALDI matrix. 0.5 lL of this mixture was pipetted on the target plate and dried at laboratory temperature. 0.5 lL of 1 mM acetone solutions of the complexes 4 and 5 was added to the place of dried-up matrix and was allowed to dry at laboratory temperature. Measurements of 1H and 13C NMR spectra of DMSO-d6 solutions of complexes 4 and 5 were performed at 27 °C on Bruker Avance300 spectrometer at 300.13 MHz and 75.47 MHz for 1H and 13C, respectively, and with tetramethylsilane (TMS) as an internal standard. 2.3. Synthesis of Cu(II) complexes 2.3.1. Synthesis of [Cu2(l-HL1)4Cl2]Cl2 2H2O (1), [Cu2(l-HL1)4 Cl2]Cl2 2H2O (2), [Cu2(l-HL2)4Cl2]Cl2 2H2O (3), [Cu2(l-HL4)4 Cl2]Cl2 2H2O (4) 1 mmol (for 1) or 2 mmol (for 2, 3 and 4) of the corresponding 6-(benzylamino)purine derivative HL1 (1, 2), HL2 (3) or HL4 (4) was suspended in 20 mL of 0.1 M HCl. The suspension was added to a solution of CuCl2 2H2O (1 mmol) in 20 mL of 0.1 M HCl. The reaction mixture was stirred for 6 h at 60 °C. Light blue powdered product formed which was removed by filtration, washed by a small amount of 0.1 M HCl, distilled water, ethanol and diethyl ether and dried at 40 °C under infrared lamp. 1: Yield: 50%, m.p.: 224–225 °C (the temperature of melting point nearly equals to the temperature of the decomposition in all next cases). Anal. Calc. for Cu2C52H52N20O4Cl4 2H2O: C, 47.1; H, 4.3; N, 21.1; Cu, 9.6. Found: C, 47.2; H, 4.4; N, 21.1; Cu, 9.2%. KM (DMF solution, S cm2 mol1): 21.5. FTIR (Nujol, cm1): 282s, 527vs. FTIR (KBr, cm1): 527w, 1028w, 1117m, 1246m, 1314m, 1349m, 1464m, 1494m, 1539m, 1637vs, 2834m, 3124m, 3273m, 3396m. UV–vis (solid state, nm): 261, 303, 380sh, 644. UV–vis (DMF solution, nm): 641. emax (M1 cm1): 282. 2: Yield: 50%, m.p.: 225–226 °C. Anal. Calc. for Cu2C52H52N20O4Cl4 2H2O: C, 47.1; H, 4.3; N, 21.1; Cu, 9.6. Found: C, 47.1; H, 4.3; N, 21.4; Cu, 9.5%. KM (DMF solution, S cm2 mol1): 25.0. FTIR (Nujol, cm1): 281s, 526vs. FTIR (KBr, cm1): 526w, 1028w, 1117m, 1246m, 1314m, 1349m, 1464m, 1494m, 1539m, 1637vs, 2834m, 3118m, 3272m, 3385m. UV–vis (solid state, nm): 257, 305, 379sh, 646. UV–vis (DMF solution, nm): 642. emax (M1 cm1): 231.
3: Yield: 60%, m.p.: 227–228 °C. Anal. Calc. for Cu2C52H52N20O4Cl4 2H2O: C, 47.1; H, 4.3; N, 21.1; Cu, 9.6. Found: C, 46.8; H, 4.3; N, 20.8; Cu, 9.1%. KM (DMF solution, S cm2 mol1): 24.8. FTIR (Nujol, cm1): 282s, 531vs, 569s. FTIR (KBr, cm1): 531w, 571w, 1032m, 1120m, 1177m, 1247s, 1315s, 1349m, 1462s, 1513s, 1539m, 1640vs, 2835m, 3127m, 3252m. UV–vis (solid state, nm): 258, 305, 384sh, 644. UV–vis (DMF solution, nm): 646. emax (M1 cm1): 256. 4: Yield: 60%, m.p.: 240–241 °C. Anal. Calc. for Cu2C56H60N20O8Cl4 2H2O: C, 46.5; H, 4.5; N, 19.4; Cu, 8.8. Found: C, 46.0; H, 4.2; N, 19.6; Cu, 8.6%. KM (DMF solution, S cm2 mol1): 23.2. FTIR (Nujol, cm1): 208m, 284m, 532s, 565s. FTIR (KBr, cm1): 542w, 567w, 1025m, 1161m, 1245s, 1266s, 1317s, 1354m, 1455s, 1468s, 1519s, 1642vs, 2835m, 3122m, 3223m, 3301m. UV–vis (solid state, nm): 268, 301, 381, 635. UV–vis (DMF solution, nm): 635. emax (M1 cm1): 371. 1H NMR (DMSO-d6, ppm): d/Dd (Dd = dcomplex dligand) 9.49/1.37 (bs, 1H, C2H), 9.22/1.10 (bs, 1H, C8H), 6.87/0.15 (s, 1H, C14H), 6.66/0.20 (bs, 2H, C11H, C15H), 4.72/0.10 (bs, 2H, C9H), 3.58/0.12 (s, 3H, C17H), 3.53/0.15 (s, 3H, C16H). 13C NMR (DMSO-d6, ppm): d/Dd 163.24/10.91 (C2), 152.82/1.30 (C6), 148.13/0.41 (C13), 147.91/0.25 (C12), 145.51/2.87 (C4), 142.56/3.66 (C8), 130.94/ 1.61 (C10), 120.07/0.65 (C15), 119.00/0.41 (C5), 112.01/0.38 (C14), 111.13/0.42 (C11), 55.53/0.02 (C17), 55.27/0.10 (C16), 43.35/0.53 (C9). 2.3.2. Synthesis of [Cu2(l-HL1)2(l-Cl)2Cl2] (5), [Cu2(l-HL2)2(l-Cl)2Cl2] (6), [Cu2(l-HL3)2(l-Cl)2Cl2] (7) The suspension of the corresponding compound HL1–HL3 (1 mmol) in ethanol (10 mL) was added to a solution of 1 mmol of CuCl2 2H2O in 10 mL of ethanol during stirring. Yellow–green powders formed during 6 h of stirring and heating (50 °C). The products were filtered off, washed with a small amount of ethanol and diethyl ether, and dried at 40 °C under infrared lamp. 5: Yield: 60%, m.p.: 262–264 °C. Anal. Calc. for Cu2C26H26N10O2Cl4: C, 40.1; H, 3.4; N, 18.0; Cu, 16.3. Found: C, 39.7; H, 3.4; N, 17.7; Cu, 15.8%. KM (DMF solution, S cm2 mol1): 29.9. FTIR (Nujol, cm1): 318s, 534m, 577m, 600vs. FTIR (KBr, cm1): 533w, 577w, 600m, 1029m, 1116m, 1249s, 1310s, 1349m, 1465s, 1494m, 1539m, 1632vs, 2831w, 3149m, 3253m, 3322m. UV–vis (solid state, nm): 272, 302, 330, 415sh, 846. UV–vis (DMF solution, nm): 834. emax (M1 cm1): 353. 1 H NMR (DMSO-d6, ppm): d/Dd 9.09/0.92 (bs, 1H, C2H), 8.94/0.82 (bs, 1H, C8H), 7.02/0.18 (bs, 1H, C13H), 7.02/0.15 (bs, 1H, C15H), 6.76/0.21 (bs, 1H, C12H), 6.60/0.24 (bs, 1H, C14H), 4.60/0.10 (bs, 2H, C9H), 3.57/0.26 (s, 3H, C16H). 13C NMR (DMSO-d6, ppm): d/Dd 161.26/8.43 (C2), 155.48/1.65 (C11), 152.67/1.75 (C6), 147.12/2.58 (C4), 145.26/6.30 (C8), 127.91/0.26 (C13), 127.370.42 (C15), 126.67/ 0.65 (C10), 119.83/1.36 (C5), 118.28/2.23 (C14), 109.00/1.81 (C12), 54.09/1.67 (C16). 6: Yield: 65%, m.p.: 237–238 °C. Anal. Calc. for Cu2C26H26N10O2Cl4: C, 40.1; H, 3.4; N, 18.0; Cu, 16.3. Found: C, 40.2; H, 3.3; N, 17.7; Cu, 15.9%. KM (DMF solution, S cm2 mol1): 19.2. FTIR (Nujol, cm1): 305m, 516m, 534m, 598vs. FTIR (KBr, cm1): 517w, 533m, 598s, 1032m, 1113m, 1179m, 1245s, 1309s, 1344m, 1407m, 1430m, 1513s, 1545m, 1640vs, 2828w, 3146m, 3273m, 3326s. UV–vis (solid state, nm): 278, 302, 328, 416sh, 792. UV– vis (DMF solution, nm): 830. emax (M1 cm1): 94. 7: Yield: 75%, m.p.: 260–261 °C. Anal. Calc. for Cu2C28H30N10O4Cl4: C, 40.1; H, 3.6; N, 16.7; Cu, 15.1. Found: C, 39.9; H, 3.5; N, 16.2; Cu, 14.7%. KM (DMF solution, S cm2 mol1): 30.0. FTIR (Nujol, cm1): 315m, 532m, 594vs. FTIR (KBr, cm1): 531m, 594s, 1073m, 1107m, 1141w, 1244m, 1310s, 1430m, 1474s, 1585m, 1625vs, 2837w, 3153m, 3249s, 3349m. UV–vis (solid state, nm): 277, 302, 332, 428sh, 822. UV–vis (DMF solution, nm): 831. emax (M1 cm1): 351.
435
P. Štarha et al. / Journal of Inorganic Biochemistry 103 (2009) 432–440
2.4. Biological activity testing
Table 1 Selected FTIR spectral data (cm1) for the complexes 1–7.
The selected complexes 2, 3, 5 and 7 were studied for their antioxidant activities in both in vitro (SOD-mimic activity) and in vivo (cytoprotective effect against alloxan-induced diabetes) tests by the modifications of the formerly published procedures [13,14].
Compound
2.4.1. SOD-mimic activity testing The SOD-mimic activities were determined by the indirect method based on the competitive reaction of the tested compounds (2, 3, 5 and 7) and XTT dye [2,3-bis(2-methoxy-4-nitro5-sulfophenyl)-2h-tetrazolium-5-carboxanilide natriumsalt] with a saturated DMSO solution of potassium superoxide (KO2). The interaction of XTT dye with superoxide anion radical led to the formation of the orange XTT-formazane, which concentration was determined by the spectroscopic measurements at 480 nm. The tested complexes, which acted as the scavengers of superoxide anion radicals, decreased the absorbance at 480 nm. The SODmimic activity was expressed as the IC50 value, calculated from the concentration-dependent curve of the inhibition of the absorbance at 480 nm at four concentration levels. The required amounts of DMSO solutions of 2, 3, 5 and 7 were added to 1.25 mL of 10 mM potassium phosphate buffer (pH 7.4) to provide 100, 50, 10 and 5 lM solutions. Subsequently, 50 lL of XTT solution in DMSO were added. The resulting solution was mixed thoroughly. The reaction was started by the addition of 100 lL of a saturated KO2 solution in DMSO. The absorbance at 480 nm was measured against a blank sample prepared without the XTT dye (value A). The same procedure was used to prepare the control sample without the tested copper(II) complexes and the absorbance at 480 nm was measured against the solution of XTT dye before the KO2 addition (value B). All the samples were incubated for 30 min at laboratory temperature. Three samples of each concentration level of all the copper(II) complexes were tested (n = 3).
200–600 cm1
400–4000 cm1
m(CuAN)
m(CarAO)
m(C@N)
m(CmetAH)
[Cu2(l-HLn)4Cl2]Cl2 2H2O 1 282s 2 281s 3 282s 4 284m
527vs 526vs 531vs 532s
1246m 1246m 1247s 1245s
1637vs 1637vs 1640vs 1642vs
2834m 2834m 2835m 2835m
[Cu2(l-HLn)2(l-Cl)2Cl2] 5 318s 6 305m 7 315m
534m 534m 532m
1249s 1245s 1244m
1632vs 1640vs 1625vs
2831w 2828w 2837w
m(CuACl)
The percentages of inhibition (% INH) values were determined according the formula (1):
A % INH ¼ 100 1 B
ð1Þ
The values of IC50 were calculated from the linearized logarithmic curve of the concentration dependence of the percentage inhibition (% INH) and represent the concentrations of the tested copper(II) complexes reducing the formazane formation to 50%. 2.4.2. Cytoprotective effect against alloxan-induced diabetes In vivo tests of antioxidant activity studied the protective effect of the complexes 2, 3, 5 and 7 in mouse model of alloxan-induced diabetes mellitus. Alloxan (A) transported into b-cells of Langerhans islets of mouse pancreas is reduced (by e.g. ascorbate, glutathione) to dialuric acid (AH2), whose re-oxidation leads back to alloxan [13–15]. Several free radicals, such as ( O 2 ), hydroxyl radicals (OH) or alloxan radicals (AH), are formed during this intracellular redox process. These free radicals selectively destroy the bcells, which results in lower production of insulin by the treated mice. The female ICR (imprinting control region) albino mice
Table 2 Selected characteristics of the studied complexes 1–7 and Cu(II) complexes with structurally relative organic ligands. Compounda
Chromophore
Geometry (s)b
Binding site
J (cm1)
giso
{Cu(l-L)4Cu} moiety [Cu2(l-AzabH)4Cl2]Cl2 3CH3OH [Cu2(l-AdeH)4Cl2]Cl2 6H2O [Cu2(l-AdeH)4(H2O)2](ClO4)4 2H2O {[Cu2(l-Ade)4(H2O)2][Cu(oda)(H2O)]4} 6H2O [Cu2(l-Ade)4(H2O)2] 5H2O [Cu2(l-HL1)4Cl2]Cl2 2H2O (1) [Cu2(l-HL1)4Cl2]Cl2 2H2O (2) [Cu2(l-HL2)4Cl2]Cl2 2H2O (3) [Cu2(l-HL4)4Cl2]Cl2 2H2O (4)
CuN4Cl CuN4Cl CuN4O CuN4O CuN4O CuN4Cl CuN4Cl CuN4Cl CuN4Cl
SQ SQ SQ SQ SQ SQ SQ SQ SQ
l-N1,N7g l-N3,N9 l-N3,N9 l-N3,N9 l-N3,N9 l-N3,N9 l-N3,N9 l-N3,N9 l-N3,N9
329 285 312 274
2.12 2.08 2.17 2.10
h
{Cu(l-L)2(l-Cl)2Cu} moiety [Cu2(l-Nphtd)2(l-Cl)2Cl2] [Cu2(l-HL1)2(l-Cl)2Cl2] (5) [Cu2(l-HL2)2(l-Cl)2Cl2] (6) [Cu2(l-HL3)2(l-Cl)2Cl2] (7) [Cu2(l-3-ClBapH)2(l-Cl)2Cl2] [Cu2(l-4-ClBapH)2(l-Cl)2Cl2] [Cu2(l-2-ClBapH)2(l-Cl)2(2-ClBapH)2Cl2]2H2O [Cu2(l-3-ClBapH)2(l-Cl)2(3-ClBapH)2Cl2]2H2O [Cu2(l-2-ClBap)2(l-Cl)2(H2O)2] [Cu2(l-3-ClBap)2(l-Cl)2(H2O)2]
CuN2Cl3 CuN2Cl3 CuN2Cl3 CuN2Cl3 CuN2Cl3 CuN2Cl3 CuN3Cl3 CuN3Cl3 CuN2Cl2O CuN2Cl2O
SQ (0.44) TB TB TB TB TB Oh Oh TB TB
l-N1,N8g l-N3,N9 l-N3,N9 l-N3,N9 l-N3,N9 l-N3,N9 l-N3,N9; N9 l-N3,N9; N9 l-N3, N9 l-N3, N9
(0.00) (0.00) (0.01) (0.01) (0.00)
xPI
gxy
gz
|DAB|/hc
Reference [31]c,d,e [22]c,[29]d,e [29]d,e,[30]c [36]c,d [36]c This work This work This work This work
0.003
h
h
0.248f
h
h
h
h
h
h
h
h
h
h
h
h
h
h
h
h
h
266 256 261 269
2.01 2.05 2.26 2.18
0.028 0.035 0.045 0.0045
2.053 2.054 2.050 2.052
2.297 2.333 2.255 2.234
0.249 0.246 0.243 0.237
139 78.9 114 119 82 114 111 123 77 100
2.11 2.15 2.15 2.13 1.98 2.00 2.23 2.00 2.00 2.00
h
h
h
h
0.0064 0.0093 0.0093 0.0003 0.0030
2.076 2.086 2.073
2.335 2.205 2.251
0 0 0
h
h
h
h
h
h
h
h
h
h
0.017
h
h
h
h
h
h
h
h
h
h
h
[33]c,[34]d This work This work This work [5]d [7]d [6]d [5]d [6]d [6]d
a Adenine (AdeH); 4-azabenzimidazole (AzabH); 6-[(2-chlorobenzyl)amino]purine (2-ClBapH); 6-[(3-chlorobenzyl)amino]purine (3-ClBapH); 6-[(4-chlorobenzyl)amino]purine (4-ClBapH); 6-[(2-methoxybenzyl)amino]purine (HL1); 6-[(4-methoxybenzyl)amino]purine (HL2); 6-[(2,3-dimethoxybenzyl)amino]purine (HL3); 6-[(3,4dimethoxybenzyl)amino]purine (HL4); 1,8-naphthyridine (Nphtd); diglycolic acid (H2oda). b SQ = square–pyramidal, TB = trigonal–bipyramidal, Oh = octahedral; s = (b a)/60, b and a are the largest angles around the central atom. c Structure determined by single crystal X-ray analysis. d Magnetic studies. e EPR studies. f DAB was recalculated from originally reported D(S = 1) as DAB = 2 D(S = 1). g Different numbering compared to purine derivatives. h Not reported to date.
436
P. Štarha et al. / Journal of Inorganic Biochemistry 103 (2009) 432–440
(30–40 g; Velak, Czech Republic) were used in the experiment. Mice were divided into eight groups: six pretreated with the tested compounds, one positive control (only alloxan solution administered) and one negative control (isotonic saline solution administered). Every group consisted of eight mice (n = 8). Each of the copper(II) complexes and alloxan monohydrate were dissolved in 10% (v/v) DMSO diluted by an aqua pro injectione water. The solutions of the tested compounds were administered intraperitoneally (at doses of 40 lmol Cu/kg). The alloxan solution was injected into the tail vein (0.1 mL/10 g body weight) 30 min after the application of the tested copper(II) compound solution. The initial glucose levels were measured in the intact mice (Table 2, Day 1). During the next 4 days of the experiment the glucose levels were measured in the morning after at least 3 h of fasting (Day 2–5). One-drop glucose oxidase test and blood glucose reflective photometer GlucotrendÒ 2 with GlucotrendÒ Glucose (max. concentration 33.3 mM; Roche, Germany) test strips were used to determine glucose concentrations (mM) in the venous blood. The observed changes in glucose levels were evaluated with the ANOVA method [20].
and DMF solutions (103 M). Three (for 1–4) and four (for 5–7) bands observed in the solid state spectra between 257 and 428 nm can be assigned to the intraligand charge-transfer (CT) and ligand-to-copper(II) CT transitions [23,24]. The maxima in the region of 635–846 nm may be assigned to the typical d–d transitions. It should be also emphasized that the positions of these bands are different in case of complexes 1–4 (635–646 nm) compared to compounds 5–7 (792–846 nm), which indirectly pointed to the various stereochemistry and structures of both series of the complexes, i.e. [Cu2(l-HLn)4Cl2]Cl2 2H2O (1–4) and [Cu2(lHLn)2(l-Cl)2Cl2] (5–7). The d–d transitions were also detected in the electronic spectra measured in 103 M DMF solutions of 1–7 with molar absorption coefficients, emax, ranges from 94 to 371 M1 cm1. Regarding both the solid state and solutions spectra, it has to be concluded that no significant changes in maxima positions were observed for the d–d transitions in the framework of each of two groups of complexes.
3. Results and discussion 3.1. General properties Modifications of formerly described syntheses of copper(II) complexes involving 6-(benzylamino)purine derivatives [5–7] were used to prepare the complexes [Cu2(l-HLn)4Cl2]Cl2 2H2O (1–4) and [Cu2(l-HLn)2(l-Cl)2Cl2] (5–7). The reactions were performed in ethanol or hydrochloric acid. The pH value of the reaction medium was found to be the crucial factor for the composition of the prepared complexes. Their proposed structures are depicted in Fig. 2. The reaction of CuCl2 2H2O with HL1 in 0.1 M HCl in the 1:1 molar ratio led to the light blue powdered complex [Cu2(lHL1)4Cl2]Cl2 2H2O (1). If the same copper(II) salt and the organic compounds HL1, HL2, or HL4 were left to react in the 1:2 molar ratio in the same reaction medium, the light blue complexes [Cu2(lHL1)4Cl2]Cl2 2H2O (2), [Cu2(l-HL2)4Cl2]Cl2 2H2O (3) and [Cu2(lHL4)4Cl2]Cl2 2H2O (4) with identical composition were obtained. The yellow–green unsolvated complexes [Cu2(l-HL1)2(l-Cl)2Cl2] (5), [Cu2(l-HL2)2(l-Cl)2Cl2] (6) and [Cu2(l-HL3)2(l-Cl)2Cl2] (7) were synthesized in ethanol in the 1:1 molar ratio of the reactants. The complexes 1–7 seem to be well soluble in N,N’-dimethylformamide (DMF) and dimethyl sulfoxide (DMSO) at room temperature. Limited solubility was observed in acetone, methanol, ethanol, chloroform and distilled water. The values of molar conductivity in DMF solutions are ranging from 19.2 to 30.0 S cm2 mol1 (Section 2.3). Surprisingly, the values regarding complexes 1–4 are significantly lower than those anticipated for a 1:2 electrolyte type (130–170 S cm2 mol1) and even for that of a 1:1 type (65–90 S cm2 mol1) [21]. Following these results, it could be deduced that the composition of the complexes 1–4 should be interpreted as [Cu2(l-HLn)4Cl4] 2H2O. However, we have reason to believe that the composition of complexes 1–4 was proposed correctly, mainly in connection with structural similarity of the complexes with that of [Cu2(l-AdeH)4Cl2]Cl2 6H2O (AdeH = adenine) which structure was determined by a single crystal X-ray analysis [22]. Generally, it may be concluded that significantly reduced conductivity values might be also ascribed to an imperfect dissociation of the discussed complexes in the solvent used. 3.2. UV–vis and FTIR spectroscopies The electronic spectra of the prepared copper(II) complexes 1–7 were measured in the 200–1000 nm region both in the solid state
Fig. 2. Proposed composition and structures of the prepared complexes 1–7.
P. Štarha et al. / Journal of Inorganic Biochemistry 103 (2009) 432–440
The FTIR spectra measured in the 400–4000 cm1 region confirmed the presence of the organic molecules HL1–HL4 in the corresponding copper(II) complexes since the vibrations characteristic for these organic compounds also appeared in the spectra of 1–7 (Table 1) [25]. The intensive bands observed at 1625–1642 cm1 can be attributed to the m(C@N) vibrations. Moreover, they are shifted about ca 20 cm1 as compared to free ligands, generally showing on the coordination of the HLn ligands to Cu(II) ions. The maxima observed at 1455–1545 cm1 can be associated with m(C@C) vibrations of the purine moiety, while the methoxy-group vibrations were observed at 1025–1049 cm1 for m(CmetAO), 1245–1249 cm1 for m(CarAO), and 2828–2837 cm1 for m(CmetAH). The bands at 3118–3153 and 3249–3273 cm1 belong to the m(CarAH), and m(NAH) vibrations, respectively, of the organic ligands in complexes 1–7. Far FTIR spectra (200–600 cm1) of 1–7 revealed several new peaks as compared to the spectra of the corresponding free organic molecules HL1–HL4. These maxima fall within 281–318 and 526– 534 cm1 intervals and can be assigned to the m(CuACl), and m(CuAN) vibrations, respectively (Table 1) [26]. The existence of the m(CuAN) and m(CuACl) vibrations supported our assumption of the coordination of the organic molecules HL1–HL4 as well as a part of chloride ions to Cu(II) centers in the complexes 1–7. Results following from UV–vis and FTIR spectroscopies indicated the existence of two different types of complexes, i.e. [Cu2(l-HLn)4Cl2]Cl2 2H2O (1–4) and [Cu2(l-HLn)2(l-Cl)2Cl2] (5– 7). Moreover, it has been found that changes in molar ratios of reactants (1:1 or 1:2) had no impact on the product composition (viz complexes 1 as compared to 2–4). 3.3. 1H and
13
C NMR spectroscopies
437
Fig. 3. Temperature dependence of the magnetization (h) and the effective magnetic moment (s) for 1 (experimental data); full lines – calculated data.
^ ¼ Jð~ H SA ~ SB Þ þ lB Bg iso ð^SAz þ ^SBz Þ
ð2Þ
where the J parameter characterizes the energy gap between the singlet (S = 0) and triplet state (S = 1), resulting from the coupling of two local spins SA = SB = 1/2. The experimental data were fitted using the sum of the molar magnetization for a dimer and a monomer, Mmol = (1 xPI) Mdimer + 2xPIMPI. The magnetization for the dimer, i.e. for [Cu2(l-HLn)4Cl2]Cl2 2H2O, was calculated using the formula,
Mmol ¼ lB g iso NA ½expððJ þ xÞ=kTÞ expððJ xÞ=kTÞ=½1 þ expððJ þ xÞ=kTÞ þ expðJ=kTÞ þ expððJ xÞ=kTÞ
ð3Þ
Complexes 4 and 5 (DMSO-d6 solutions), as representatives of [Cu2(l-HLn)4Cl2]Cl2 2H2O (1–4) and [Cu2(l-HLn)2(l-Cl)2Cl2] (5– 7), were studied by 1H and 13C NMR spectroscopies. Observed chemical (d) and calculated coordination shifts (Dd = dcomplex dligand) are given in Section 2.3. The changes in the C2, C4 and C8 chemical shifts were the most apparent in the 13C NMR spectrum of 4 as compared to the free HL4 ligand. Their coordination shifts (Dd) equaled to 10.91 ppm (for C2), 2.87 ppm (for C4) and 3.66 ppm (for C8). It may be caused by a bidentate coordination of the HL4 ligand into two Cu(II) centers in a dinuclear complex 4, most likely through the N3 and N9 atoms of the purine skeleton (see Fig. 2). The results following from 1H NMR spectra of complex 4 and the free ligand, with Dd for C2H = 1.30 ppm and Dd for C8H = 1.10 ppm, also supported the above-mentioned conclusions. As for complex 5, the coordination shifts of C2 (Dd = 8.43 ppm), C4 (Dd = 2.58 ppm) and C8 (Dd = 6.30 ppm) were also observed to be higher than those belonging to the remaining carbon atoms within the complex 5 and free HL1 ligand. Moreover, the coordination shifts of C2H (Dd = 0.25 ppm) and C8H (Dd = 0.30 ppm) were significant as well, which indirectly proved the bidenatate bridging coordination of HL1 to two Cu(II) centers through the N3 and N9 atoms.
where x = lBgisoB [28]. The magnetization of the monomeric paramagnetic impurity with SPI = 1/2 was evaluated as MPI = NAlBgPISPI BðgÞ, where BðgÞ is Brillouin function and g = lBgPIB /kT [28]. Three parameters were left to vary during the fitting procedure, J, giso and xPI, with fixed gPI = 2.2. The resulting J values are very close to 260 cm1 and are comparable with those reported for [Cu2(l-AdeH)4Cl2]Cl2 6H2O [22,29], [Cu2(l-AdeH)4(H2O)2](ClO4)4 2H2O [29,30] and [Cu(l-AzabH)4Cl2]Cl2 3CH3OH [31] complexes (AzabH = 4-azabenzimidazole), as summarized in Table 2. Owing to high similarity in magnetic properties of the latter complexes as compared to 1–4 ones, and mainly based on known molecular X-ray structures of the above-mentioned, we suppose that the same arrangement occurs for the complexes 1–4, i.e. both Cu(II) ions are penta-coordinated in a square–pyramidal geometry and bridged by four N3–C4–N9 bridges originating from the Bap derivatives (see Fig. 2). Four nitrogen atoms of bridging Bap derivatives form basal plane and one chloride ion is bonded in the apical position. The EPR powder spectra for 1–4 at the room temperature show typical pattern for the triplet state splitted by the asymmetric exchange for which the following spin Hamiltonian can be written as
3.4. Magnetic properties and EPR spectroscopy
^ ¼~ SB þ lB~ SA þ ~ SB Þ H SA DAB ~ B g ð~
The magnetic susceptibility (calculated from the temperature dependence of the magnetization at B = 1 T) for [Cu2(lHLn)4Cl2]Cl2 2H2O (1–4) is slightly increasing on cooling from the room temperature up to 220–240 K, then it starts to decrease up to approximately 50 K and again increases on lowering the temperature (Fig. 3). This magnetic behavior is typical for antiferromagnetic coupled dimers with a small amount of the monomeric paramagnetic impurity [27]. The magnetic data were successfully fitted with the following spin Hamiltonian,
In the case of the axial symmetry, the tensors DAB have one parameter DAB and the tensor g has two parameters gxy, gz. The spectra were simulated using EasySpin [32] package (Fig. 4, Table 2). The resulting parameters DAB for 1–4 are almost identical, and moreover, they are very similar to that found for a dinuclear copper(II) complex, [Cu(l-AzabH)4Cl2]Cl2 3CH3OH, bridged by four 4-azabenzimidazoles [31]. The EPR spectra at 77 K were dominated by the paramagnetic impurity, since at such low temperature the triplet state of the dimer is almost depopulated.
ð4Þ
438
P. Štarha et al. / Journal of Inorganic Biochemistry 103 (2009) 432–440
3.5. MALDI-TOF mass spectroscopy Matrix assisted laser desorption/ionization time-of-flight mass spectroscopy (MALDI-TOF-MS) with positive ionization mode was performed for [Cu2(l-HL4)4Cl2]Cl2 2H2O (4) and [Cu2(lHL1)2(l-Cl)2Cl2] (5) complexes, as representatives of complexes 1–4 and 5–7. The peak at 286 m/z may be attributed to the 6-[(3,4-dimethoxybenzyl)amino]purine (HL4). This molecule is further fragmented to [adenineH]+ (136 m/z) and [purineH]+ (121 m/z) which peaks were also found in a mass spectrum of 4. [Cu(HL4)2]+ and [Cu(HL4)]+ fragments were detected at 634, and 350 m/z, respectively. Moreover, the presence of [Cu(HL4)]+ fragment was confirmed by a peak at 537 m/z assignable to [Cu(HL4) + matrix]+. The results obtained for complex 5 were similar as in the case of the above discussed complex 4. The peaks assignable to [purineH]+, [adenineH]+, [HL1]+, [Cu(HL1)]+, [Cu(HL1) + matrix]+ and [Cu(HL1)2]+ fragments were found at 121, 136, 256, 320, 507, and 574 m/z, respectively. Fig. 4. The X-band powder EPR spectrum for 1 measured at room temperature. Top – experimental data, bottom – simulated data.
3.6. TG/DTA thermal studies
For the complexes 5–7, the maxima of the magnetic susceptibility were observed between 75 and 100 K and the experimental data were successfully fitted to the spin Hamiltonian as given in the Eq. (2). The antiferromagnetic exchange for 5–7 was found to be smaller as compared to 1–4, with J ranging from 78 to 120 cm1. Similar value of J (139 cm1) was found for complex (Nphtd = 1,8-naphthyridine) [33], [Cu2(l-Nphtd)2(l-Cl)2Cl2] where the Cu(II) ions are bridged by the two N–C–N bridges of the two organic molecules and two chlorides, as determined by a single crystal X-ray analysis [34]. We consider the similar structure for complexes 5–7. Finally, it should be mentioned that the value of s parameter, s = 0.44, as determined for Cu2(l-Nphtd)2(l-Cl)2Cl2], indicated a highly distorted square–pyramidal geometry for the mentioned compound [35]. Despite of that we suppose that the central Cu(II) ions adopt a distorted trigonal–bipyramidal geometry with higher probability in case of the complexes 5–7 (Table 2, Fig. 2). The room temperature EPR powder spectra for 5–7 show only an axial spectrum without any asymmetric exchange present (DAB = 0). The data were simulated using two different gxy, gz parameters (Table 2). The liquid nitrogen spectra were very similar and no new feature was observed.
Thermogravimetric (TG) and differential thermal (DTA) analyses were used to describe thermal behavior of the prepared complexes 1–7. As representatives, the complexes of [Cu2(lHL4)4Cl2]Cl2 2H2O (4) and [Cu2(l-HL1)4(l-Cl2)Cl2] (5) were chosen for the thermal decomposition study. The decay of the complex 4 (Fig. 5) started at 30 °C and is associated with the loss of the crystal water molecules (calc./found for 2H2O: 2.5/2.4%). The dehydration was finished at 114 °C. This process is accompanied by a very small endo-effect observed on the DTA curve with maximum at 71 °C. The anhydrous complex was thermally stable up to 189 °C, when it started to decay. Further, the complex decomposed without formation of thermally stable intermediates up to 612 °C. The degradation was accompanied by three exo-effects with maxima at 270, 447 and 528 °C, which can be associated with the oxidation of organic parts of 4. No weight changes were found above the temperature of 612 °C when a thermally stable final product was formed. The final product was determined to be CuO using powder diffraction technique (Inorganic Crystal Structural Database (ICSD), ver. 1.4.2, Karlsruhe, Germany; ICSD No. 87122). The total weight loss, as determined from the TG curve for 4, was found to be 89.9% (calculated 89.0%). The complex 5 was thermally stable up to 170 °C, which indicated that this complex was not solvated. Then, it decomposed without formation of thermally stable intermediates up to 433 °C
Fig. 5. TG and DTA curves of the complexes 4 (left) and 5 (right).
439
P. Štarha et al. / Journal of Inorganic Biochemistry 103 (2009) 432–440 Table 3 Results of the in vitro (SOD-mimic) and in vivo (antidiabetic) antioxidant activity tests. Compound
In vitro SOD-mimic activity testing
In vivo antidiabetic activity testing
a
% INH
[Cu2(l-HL1)4Cl2]Cl2 2H2O (2) [Cu2(l-HL2)4Cl2]Cl2 2H2O (3) [Cu2(l-HL1)2(l-Cl)2Cl2] (5) [Cu2(l-HL3)2(l-Cl)2Cl2] (7) SOD Controlc Alloxand
100 lM
50 lM
10 lM
5 lM
92.2 ± 0.3 92.6 ± 0.1 94.5 ± 0.2 95.1 ± 0.2 – – –
91.7 ± 0.1 92.1 ± 0.3 93.4 ± 0.2 94.6 ± 0.2 – – –
71.6 ± 0.3 72.3 ± 0.1 78.5 ± 0.3 nm – – –
64.6 ± 0.2 68.5 ± 0.0 72.5 ± 0.3 64.3 ± 0.4 – – –
IC50
Glucose level ± SEM (mM)b Day 1
Day 2
Day 3
Day 4
Day 5
1.090 0.687 0.253 1.250 0.480 – –
4.7 ± 0.1 5.1 ± 0.3 5.2 ± 0.2 4.6 ± 0.2 – 5.1 ± 0.2 4.6 ± 0.2
7.3 ± 0.5 7.2 ± 0.4 6.3 ± 0.8 7.4 ± 0.3 – 8.0 ± 0.2 8.0 ± 0.8
16.3 ± 3.6 7.4 ± 0.2* 7.8 ± 0.1* 6.8 ± 0.3* – 7.7 ± 0.2 16.3 ± 2.7
17.3 ± 3.3 7.8 ± 0.3* 8.9 ± 0.7* 7.2 ± 0.2** – 8.6 ± 0.2 18.6 ± 3.1
17.4 ± 3.4 8.1 ± 0.4* 9.4 ± 0.4 7.7 ± 0.4* – 9.2 ± 0.3 15.2 ± 2.5
nm, not measured. qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Pn a 1 2 % IHN, i.e. percentage of inhibition of XTT formazane production given in mean values (± standard deviation; SD ¼ n1 i¼1 ðXi XÞ c; n ¼ 3Þ observed in in vitro SODmimic activity testing and calculated IC50 values together with the Cu,Zn-superoxide dismutase (SOD) used as a standard. b Mean values of blood glucose concentrations of mice tested in vivo for cytoprotective effect against alloxan-induced diabetes; SEM = standard error of the mean SD ffiffiffi; SD is standard deviation; n = 8). (SEM ¼ p n c Negative control (non-treated mice). d Positive control (mice treated with alloxan). * Significance level (p < 0.05). ** Significance level (p < 0.01).
which was accompanied by intensive exo-effects with maxima at 262, 309 and 344 °C. No weight changes were observed above the temperature of 433 °C. CuO was found to be as a final product of the degradation (a weight loss found/calc.: 79.8/81.0%). 3.7. Biological activity testing The antioxidant activities of 2, 3, 5 and 7 were tested by both in vitro (SOD-mimic) and in vivo (cytoprotective effect against alloxan-induced diabetes) methods. The results are summarized in Table 3 and graphically displayed in Fig. 6 and 7. 3.7.1. In vitro SOD-mimic activity The determinations of 2, 3, 5 and 7 SOD-mimic activities showed promising antioxidant properties of the tested copper(II) complexes. The method allowed us to use the native bovine Cu,Zn-SOD as a standard. The SOD-mimic activity of complex 5, expressed as IC50 value, was found to be 0.253 lM (Table 3, Fig. 6). This value is significantly lower than that determined for the reference sample, i.e. the native bovine Cu,Zn-SOD enzyme (IC50 = 0.480 lM, as determined by the XTT method [13]). It means that complex 5 had higher SOD-mimic activity in comparison with the used standard. Similar results were reported in case of mononuclear and dinuclear copper(II) complexes with tripodal polypyridyl-amine ligands [10].
Fig. 6. The results of in vitro SOD-mimic activity testing of the complexes 2, 3, 5 and 7 together with the native bovine Cu,Zn-superoxide dismutase (SOD) used as a standard, expressed as IC50 values (lM).
It was proved that dimeric complexes (IC50 = 0.54–0.76 lM) with this type of ligands showed higher antioxidant activities in comparison with the native Cu,Zn-SOD enzyme (IC50 = 2.81 lM, as determined by the NBT method [10]), and in contrast to monomeric ones (IC50 = 5.02–140.0 lM). Higher SOD-mimic activity of dimeric copper(II) complexes could be explained by the possible cooperation of both Cu(II) centers in free radical binding and electron transfer. The SOD-mimic activity of the complex 3 (IC50 = 0.687 lM) was found to be slightly lower as compared to Cu,Zn-SOD. However, IC50 values of 2 (1.090 lM) and 7 (1.250 lM) exceeded that one of the native enzyme more than two-times. Due to the significant SOD-mimic activity of the tested complexes, we expect the redox mechanism of superoxide dismutation, similar to the Cu,Zn-SOD. 3.7.2. In vivo cytoprotective effect against alloxan-induced diabetes The venous blood glucose concentration was measured in the morning after 3 h of fasting. The glucose levels in the alloxangroup rose significantly over those measured for the control group (respresenting the normal values) already in the second day of the experiment. The glycaemic control in the mice pretreated with the complexes 3, 5 and 7 was completely preserved. It can be demonstrated by constant level of the initial values of the blood glucose concentration during the next 4 days of the experiment. This indi-
Fig. 7. The mean blood glucose levels [mM], together with standard deviations, of treated mice (n = 8) during five testing days; control (non-treated mice), alloxan (alloxan-treated mice), 2, 3, 5 and 7 (mice treated by the corresponding copper(II) complex).
440
P. Štarha et al. / Journal of Inorganic Biochemistry 103 (2009) 432–440
cates high in vivo antidiabetic activity of the complexes 3 and 7 against toxic free radicals produced by alloxan metabolism in this type of animal testing. On the other hand, mice pretreated with complex 2 showed only insignificant changes in blood glucose levels compared with the positive control. This fact, in connection with the high SOD-mimic activity of these complexes, agrees also with our hypothesis, that the superoxide elimination itself could be beneficial in preventing some of the late-stage diabetic complications, but it does not give the ability to the complexes to prevent the cytotoxic attack of alloxan and its metabolic intermediates, predominantly the hydroxyl radical. 4. Conclusions We prepared and characterized a series of dinuclear copper(II) complexes (1–7) with derivatives of 6-(benzylamino)purine (HL1–HL4) as ligands bridging two strongly antiferromagnetically coupled Cu(II) ions. The pH value of the reaction medium (either 0.1 M HCl or ethanol) was found to play crucial role regarding the composition of the prepared copper(II) complexes. Thus, the complexes of two types having the general formula [Cu2(lHLn)4Cl2]Cl2 2H2O (1–4) and [Cu2(l-HLn)2(l-Cl)2Cl2] (5–7) were synthesized. Further, the results of in vitro (SOD-mimic) and in vivo (cytoprotective effect against alloxan-induced diabetes) antioxidant activity tests of the selected copper(II) complexes 2, 3, 5 and 7 are presented. Both types of testing have been performed for the first time for copper(II) complexes involving 6-(benzylamino)purine derivatives, thereby our knowledge of biological properties of these complexes has been intensively extended. The results of the in vitro SOD-mimic activity tests proved complex 5 to be significantly more active (IC50 = 0.253 lM) than a standard (i.e. native bovine Cu,Zn-SOD enzyme, IC50 = 0.480 lM). The complexes 3 and 7 were found to be effective free radical scavengers also in in vivo testing of cytoprotective effect against alloxan-induced diabetes. They were able to maintain completely the glycaemic control in the pretreated animals and thus eliminate the cytotoxic effects of alloxan and its metabolic intermediates, especially the most reactive one, i.e. the hydroxyl radical. Acknowledgements This work was supported by the Ministry of Education, Youth and Sports of the Czech Republic (a Grant No. MSM6198959218). The authors would like to thank to Prof. Marek Šebela for MALDI-TOF-mass spectra and Dr. Miroslava Matíková-Malˇarová for FTIR spectra measurements, and Mr. Lukáš Dvorˇák for performing of CHN elemental analyses.
References [1] P.J. Davies, Plant Hormones, Springer, Dordrecht, 1997. [2] L. Meijer, A. Borgne, O. Mulner, J.P.J. Chong, J.J. Blow, N. Inagaki, M. Inagaki, J.G. Delcros, J.P. Moulinoux, Eur. J. Biochem. 243 (1997) 527–536. [3] C. Benson, S. Kaye, P. Workman, M. Garret, M. Walton, J. de Bono, Br. J. Cancer 92 (2005) 7–12. [4] . ˇ , J. Marek, K. Dolezˇal, J. [5] M. Malonˇ, Z. Trávnícˇek, M. Maryško, R. Zborˇil, M. Mašlán Rolcˇík, V. Kryštof, M. Strnad, Inorg. Chim. Acta 323 (2001) 119–129. [6] Z. Trávnícˇek, M. Malonˇ, Z. Šindelárˇ, K. Dolezˇal, J. Rolcˇík, V. Kryštof, M. Strnad, J. Marek, J. Inorg. Biochem. 84 (2001) 23–32. ˇ , Z. Trávnícˇek, M. Maryško, J. Marek, K. Dolezˇal, J. Rolcˇík, M. Strnad, [7] M. Malon Trans. Met. Chem. 27 (2002) 580–586. [8] M. Gielen, E.R.T. Tiekink, Metallotherapeutic Drugs and Metal-based Diagnostic Agents, Wiley, London, 2005, pp. 220–223. [9] J.M. McCord, I. Fridovich, J. Biol. Chem. 244 (1969) 6049–6055. [10] K. Jitsukawa, M. Harata, H. Arii, H. Sakurai, H. Masuda, Inorg. Chim. Acta 324 (2001) 108–116. [11] N. Ishimoto, T. Nemoto, K. Nagayoshi, F. Yamashita, M. Hashida, J. Contr. Rel. 111 (2006) 204–211. [12] H. Ohtsu, Y. Shimazaki, A. Odani, O. Yamauchi, W. Mori, S. Itoh, S. Fukuzumi, J. Am. Chem. Soc. 122 (2000) 5733–5741. [13] J. Vancˇo, O. Švajlenová, E. Racˇanská, J. Muselík, J. Valentová, J. Trace, Elem. Med. Biol. 18 (2004) 155–161. [14] J. Vancˇo, J. Marek, Z. Trávnícˇek, E. Racˇanská, J. Muselík, O. Švajlenová, J. Inorg. Biochem. 102 (2008) 595–605. [15] M. Elsner, E. Gurgul-Convey, S. Lenzen, Free Radic. Biol. Chem. 41 (2006) 825– 834. [16] Z. Trávnícˇek, V. Kryštof, M. Šipl, J. Inorg. Biochem. 100 (2006) 214–225. [17] J.A. Kuhnle, G. Fuller, J. Corse, B.E. Mackey, Physiol. Plantarum 41 (1977) 14– 21. [18] W. Snipes, J. Cupp, G. Cohn, A. Keith, Biophys. J. 14 (1974) 20–32. [19] E. König, Landolt-Börstein, Springer, Berlin, 1966. [20] Origin, Version 8, OriginLab Corporation, Northampton, MA, 2007. [21] W.J. Geary, Coord. Chem. Rev. 7 (1971) 81–122. [22] P. de Meester, A.C. Skalpski, Inorg. Phys. Theor. (1971) 2167–2169. [23] E.I. Solomon, A.B.P. Lever, Inorganic Electronic Structure and Spectroscopy, Applications and Case Studies, vol. 2, Wiley, New York, 1999. [24] P. Amudha, M. Kandaswamy, L. Govindasamy, D. Velmurugan, Inorg. Chem. 37 (1998) 4486–4492. [25] C.J. Pouchert, The Aldrich Library of Infrared Spectra, Aldrich Chemical Company Press, Milwaukee, 1981. [26] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination Compounds, Part B: Applications in Coordination, Organometallic and Bioinorganic Chemistry, Wiley, New York, 1997. [27] O. Kahn, Molecular Magnetism, Wiley, New York, 1993. [28] R. Bocˇa, Theoretical Foundation of Molecular Magnetism, Elsevier, Amsterdam, 1999. [29] D. Sonnenfroh, R.W. Kreilick, Inorg. Chem. 19 (1980) 1259–1262. [30] A. Terzis, A.L. Beauchamp, R. Rivest, Inorg. Chem. 12 (1973) 1166–1170. [31] G.A. van Albada, I. Mutikainen, U. Turpeinen, J. Reedijk, Polyhedron 25 (2006) 3278–3284. [32] S. Stoll, A. Schweiger, J. Magn. Reson. 178 (2006) 42–55. [33] C. Mealli, F. Zanobini, J. Chem. Soc. Chem. Commun. 2 (1982) 97–98. [34] K. Emerson, A. Emad, R.W. Brookes, R.L. Martin, Inorg. Chem. 12 (1973) 978– 981. [35] A.W. Addison, T.N. Rao, J. Reedijk, J. van Rijn, G.C. Verschoor, J. Chem. Soc. Dalton Trans. (1984) 1349–1356. [36] J.M. González-Pérez, C. Alarcórn-Payer, A. Castiñeiras, T. Pivetta, L. Lezama, D. Choquesillo-Lazarte, G. Crisponi, J. Niclós-Gutiérrez, Inorg. Chem. 45 (2006) 877–882.
