Protective effects of 2,3-diaryl-substituted indole-based cyclooxygenase-2 inhibitors on oxidative modification of human low density lipoproteins in vitro

Abstract

It has been suggested that 2,3-diaryl-substituted indole-based cyclooxygenase-2 (COX-2) inhibitors (2,3-diaryl-indole coxibs) do not only appear as potent anti-inflammatory agents but also show the ability to scavenge reactive oxygen species (ROS). This led to the hypothesis that 2,3-diaryl-indole coxibs also may act as potent inhibitors of oxidative modification of low-density lipoprotein (LDL), which is considered a key factor in atherogenesis. The aim of this study was to explore i) the reactivity of a series of new synthesized 2,3-diaryl-indoles with several well characterized LDL oxidation systems and ii) subsequent effects on an inflammatory/atherogenic microenvironment. The results demonstrate that under the present experimental conditions2,3-diaryl-indoles showed potent ROS scavenging activity and were able to markedly inhibit LDL oxidation. Subsequently, this led to a substantial decrease of modified LDL uptake by scavenger receptors in THP-1 macrophages in vitro and in rats in vivo. Moreover, modified LDL-mediated monocyte/neutrophil adhesion to endothelial cells, macrophage NFκB activation, as well as macrophage and endothelial cell cytokine release was diminished in vitro. The reduction of modified LDL-induced atherogenic effects by antioxidant 2,3-diaryl-indole coxibs may widen the therapeutic window of COX-2 targeted treatment.

Keywords

Antioxidants atherogenesis selective cyclooxygenase-2 (COX-2) inhibitors (coxibs)inflammation lipid peroxidation protein oxidation radical scavenger reactive oxygen species (ROS)

1 Introduction

Cyclooxygenase-2 (COX-2) is an inducible isoenzyme, whose overexpression is implicated in a number of inflammatory or inflammation-associated disease processes. Equal to the constitutively expressed COX-1 isoenzyme but more important in pathophysiological situations COX-2 catalyzes the conversion of arachidonic acid into prostaglandin H₂. Starting from this precursor molecule subsequent enzymatic and/or non-enzymatic reactions lead to the formation of dozens of eicosanoids like prostaglandins, prostacyclin, thromboxanes, and isoprostanes as potent regulators of inflammatory response. This includes recruitment of cells, activation of enzymes, generation of reactive oxygen species (ROS), vasodilation, endothelial fenestration, platelet aggregation, pain, and fever. Furthermore, an elevated COX-2 level is a prominent finding in neurodegenerative, cardiovascular, and neoplastic disorders and is correlated with certain factors influencing disease progression and response to therapy [50]. Consequently, selective blocking of the COX-2 isoenzyme provides a number of therapeutic strategies [35]. This also is generally accepted for atherosclerosis. All the more, as one key atherogenic process, postsecretory modification in the structure of low density lipoproteins (LDL) by ROS is strongly wedded to inflammatory processes. Besides other enzymes, which in an inflammatory microenvironment produce ROS, e.g., NADPH oxidase and myeloperoxidase, COX-2 itself has been suggested to be an important source of ROS [21]. There is experimental evidence that the anti-inflammatory effects of selective COX-2 blockade in part is caused by scavenging ROS. Most selective COX-2 inhibitors (coxibs) are characterized by a central monocyclic five- or six-membered, or bicyclic heterocyclic core structure with two adjacent aromatic rings bearing a methylsulfonyl or aminosulfonyl group. Among them the indole motif is a classical pharmacophore present in the non-selective COX inhibitor indomethacin. More recently, N-substituted indole carboxyclic acid esters as well as various 2,3-, 3,6-, and 2,6-disubstituted indole derivatives were described [5 , 32]. In line with this, a series of thirteen novel potential coxibs based on a 2,3-diaryl substituted indole chemical lead with high affinity and selectivity for COX-2 has been developed by us as appropriate templates for radiotracer design. For our efforts the high potential of both varying the C-2/C-3 substitution pattern in the indole scaffold, e.g., by introducing fluorine or methoxy groups, and using a highly effective synthesis route via McMurry cyclization was of particular interest [25, 33]. The indole motif, on the other hand, also is a prominent antioxidant pharmacophore [30]. Thus, not unexpectedly, these 2,3-diaryl-substituted indoles exhibited physicochemical properties like high lipophilicity and distinct redox activity suggesting them also to be promising compounds positively affecting susceptibility of LDL to oxidative modification [25, 33]. This prompted us to undertake a detailed in vitro study to explore i) the reactivity of a series of new synthesized 2,3-diaryl-indoles with several well characterized LDL oxidation systems and ii) the subsequent effects on an inflammatory/atherogenic microenvironment.

2 Materials and methods

2.1 Solvents and reagents

Melatonin, all other reagents and solvents were of the highest purity available from Sigma-Aldrich-Group (Taufkirchen, Germany), if not indicated otherwise.

2.2 Chemical synthesis of coxibs

For this investigation we selected five compounds (C1-C5) from a series of thirteen (physico)chemically and biochemically well characterized 2,3-diaryl-indole coxibs considering promising data on their a) inhibitory activity against COX-1/COX-2, b) selectivity forCOX-2 and c) potential redox activity as published elsewhere [33]. In this regard, synthesis of 2-[4-(aminosulfonyl)phenyl]-3-(4-methoxyphenyl)-1H-indole (C1), 2-[4-(aminosulfonyl)phenyl]-3-(4-fluorophenyl)-1H-indole (C2), 3-(4-fluorophenyl)-5-methoxy-2-[4-(methylsulfonyl)phenyl]-1H-indole(C3), 2-[4-(aminosulfonyl)phenyl]-3-(4-fluorophenyl)-5-methyl-1H-indole (C4), and 2-(4-methoxy-phenyl)-3-[4-(methylsulfonyl)phenyl]-1H-indole (C5) essentially followed the procedures published by us [25, 33]. Selective COX-2 inhibitor 4-[5-(4-methylphenyl)-3-(trifluoromethyl)pyrazol-1-yl]benzenesulfonamide (celecoxib) was synthesized according to the method published by Penning et al. [39]. The chemical structures of all compounds investigated in this work are given in Scheme 1.

2.3 Isolation of native LDL

Native albumin-free LDL (nLDL; density 1.006–1.063 g/mL) were isolated from the blood plasma of healthy, normolipidemic, normoglycemic male volunteers by sequential very fast ultracentrifugation (VFU) and prepared for oxidation experiments as previously described by us in detail [41]. Blood plasma and LDL were all processed in subdued light to prevent the photooxidation of LDL. All buffers and solutions were made oxygen-free by degassing and purging with argon. LDL apoB-100 was measured by immunoelectrophoresis using ‘ready-to-use’ agarose gels (Sebia, Issy-les-Moulineaux, France). Immediately before oxidation of LDL, EDTA, and salt from the density gradient were removed using a size exclusion column (Econo-Pac 10DG, Bio-Rad Laboratories, Munich, Germany) and phosphate-buffered saline (PBS, 10 mM sodium phosphate, 150 mM sodium chloride, pH 7.4) as the eluent [41].

2.4 LDL lipid and protein oxidation

In order to determine redox activity-structure relationships of 2,3-diaryl-substituted indole-based cyclooxygenase-2 (COX-2) inhibitors (2,3-diaryl-indole coxibs) and, as controls, the non-indole-derivative celecoxib and the indole-derivative melatonin, three well-characterized LDL oxidation models generating OH^•/O₂ ^• ⁻ or hypochlorite (OCl⁻) were used. All compounds were tested at a final concentration of 1μM. For lipid peroxidation, to 200μL-aliquots of nLDL (125μg apoB-100/mL, equal to 0.25μM LDL) in 96 well plates (UV-Star plates, UV transparent to 200 nm, Greiner Bio One, Frickenhausen, Germany) were added 25μL of an aqueous solution of CuSO₄ (16μM) and 25μL of stock solution of the compound (10μM) to be tested for redox-activity. The same preparation without CuSO₄ was used as control. The oxidative process was monitored using a Synergy 4 thermostatic micro plate reader (BioTek Instruments, Bad Friedrichshall, Germany) by following the formation of conjugated dienes at 234 nm every 5 min for 4 hours at 30°C. This approach results in a curve exhibiting a lag phase, during which the absorbance does not increase significantly, a propagation phase, during which the absorbance increases rapidly, and a degradation phase, characterized by a slow fall in the absorbance [14, 33]. There is a positive (negative) correlation between lag phase duration and the concentration of antioxidants (prooxidants) contained in the LDL sample [14]. For protein oxidation, 2 mL-aliquots of nLDL (125μg apoB-100/mL, equal to 0.25μM LDL) with or without 1μM of the compound tested were subjected to two other well characterized oxidation systems: A) hemin/H₂O₂ (10μM/100μM) and B) HOCl (100μM) at 37°C for 40 hours in the dark [26, 40]. Then, on completion of the oxidation process, LDL were delipidated. After isolation, reduction, and enzymatic hydrolysis of apoB-100 the oxidative process was monitored in A) by mass spectrometric determination of 5-hydroxy-2-aminovaleric acid (HAVA) that is formed by reduction of γ-glutamyl semialdehyde, which is a highly specific product of iron-(OH^•/O₂ ^• ⁻)-mediated protein oxidation and B) by mass spectrometric determination of formation of 3-chlorotyrosine (3Cl-TYR), a highly specific marker of OCl⁻-mediated protein oxidation, as described elsewhere in detail [33 , 43]. The results obtained by these LDL oxidation models are expressed as ratios between the duration of lag phase in the presence and in the absence of the compound (R_DIENE), between the apoB-100 HAVA content in the absence and in the presence of the compound (R_HAVA), and between the apoB-100 3-chlorotyrosine content in the absence and in the presence of the compound (R_3Cl - TYR). Thus, in these approaches R values (R_DIENE, R_HAVA or R_3Cl - TYR) higher than 1 indicate antioxidant activity, R values of about 1 mean that the compound has no effect and R values lower than 1 suggests prooxidant activity. For reason of comparability R_DIENE and R_HAVA values were adopted from [33]. R_3Cl - TYR values measured in this study correspond to four experiments, performed in triplicate.

2.5 OH^• scavenging assay

In order to further characterize the redox activity of 2,3-diaryl-indole coxibs, in particular, to discriminate possible OH^• scavenging properties, a simplified deoxyribose degradation assay was used according to the procedure described by Lapenna et al. [31]. Results were expressed as percentage of OH^• scavenging activity % SA_OH • of test sample compared to control without any antioxidant (100% deoxyribose oxidation). In this study the use of EDTA in the reaction system is essential because inhibition of iron-dependent deoxyribose degradation in the absence of EDTA does not only depend on OH^• scavenging, but also on its ability to form complexes with iron ions [17]. Each study corresponds to four experiments, performed in triplicate.

2.6 O₂^•⁻ scavenging assay

In order to further characterize the redox activity of 2,3-diaryl-indole coxibs, in particular, to discriminate possible O₂ ^• ⁻ scavenging properties, a spectrophotometric assay monitoring O₂ ^• ⁻-induced reduction of nitroblue tetrazolium chloride (NBT) was used. In this assay the generation of O₂ ^• ⁻ was performed using a nicotinamide adenine dinucleotide (NADH)/phenazine methosulfate (PMS) system according to the procedure described by Costa et al. [9]. Results were expressed as percentage of O₂ ^• ⁻ scavenging activity (% SA_O2 ^• _-) of test sample compared to control without any antioxidant (100% diformazan formation). The selective effect of superoxide on the reduction of NBT to diformazan was confirmed using superoxide dismutase. Each study corresponds to four experiments, performed in triplicate.

2.7 Radiolabeling of native and oxidized LDL

Native LDL and oxidatively modified LDL, which were obtained from both the hemin/H₂O₂ (oxLDL) and HOCl (OCl-LDL) LDL oxidation experiments, were radiolabeled with no-carrier added N-succinimidyl-4-[¹⁸F]fluorobenzoate ([¹⁸F]SFB) as published in detail elsewhere [41]. In a typical LDL radiolabeling experiment, approximately 200 MBq of purified and sterile [^18F]fluorobenzoylated LDL were obtained, which had an effective specific activity in the range of 400–500 GBq/μmol for both [¹⁸F]FB-nLDL and [¹⁸F]FB-oxLDL/[¹⁸F]FB-OCl-LDL (each related to apoB-100; Mr 516.000, without carbohydrate content) at the time of cell uptake studies. Furthermore, both [¹⁸F]FB-nLDL and [¹⁸F]FB-oxLDL/[¹⁸F]FB-OCl-LDL showed high in vitro stability at several time points (30 min, 2 h, and 4 h after radiolabeling) using PBS (pH 6.5 to 7.5) as the solvent at 37°C. At each time point after labeling more than 96% of total activity (decay-corrected) could be recovered in the intact apoB-100 molecule. Of note, radiolabeling of LDL with no-carrier added [¹⁸F]SFB did not lead either to adverse oxidation of native LDL particles or to additional adverse modification of oxidatively modified LDL particles [42].

2.8 Lipoprotein uptake experiments in vitro

The influence of 2,3-diaryl-indole coxibs C1, C3 and C4 on the biological activity of LDL particles in terms of specific cellular binding and uptake of both radiolabeled [¹⁸F]FB-nLDL and [¹⁸F]FB-oxLDL/[¹⁸F]FB-OCl-LDL obtained from hemin/H₂O₂ and HOCl LDL oxidation experiments was assessed in human monocyte cell line THP-1 in vitro [41, 42]. For lipoprotein uptake experiments, cells were seeded in 24-well plates at a density of 1×10⁵ cells/mL. THP-1 cells were propagated in RPMI 1640, 10% FCS, containing penicillin (100 U/mL), streptomycin (100μg/mL) at 37°C in 5% CO₂ [41]. For the uptake studies, cells were centrifuged and resuspended in fresh RPMI medium containing 10% FCS in 24-well plates at a density of 1×10⁴ cells/mL. Transformation of THP-1 monocytes to macrophages (THP-1Mφ) was performed by adding 64 nM phorbol myristate acetate for 72 h as described [41]. The cells were then washed extensively with serum-free RPMI medium and diluted to the appropriate density. In brief, for binding assays cells were incubated for 2 h at 4°C with either [¹⁸F]FB-nLDL or [¹⁸F]FB-oxLDL/[¹⁸F]FB-OCl-LDL species (2.5–50μg of protein/mL in 125μL of media with 50 mM Hepes, pH 7.4) in a total volume of 250μL. At 4°C, LDL bind to LDL receptors or scavenger receptors, but the lipoprotein-receptor complexes are not internalized. Non-specific binding was determined by the addition of an excess (500μg/mL) of unlabeled nLDL and oxLDL, respectively, and represented less than 15% of total lipoprotein binding in this cell type. The extent of specific binding was calculated bysubtracting the non-specific binding of [¹⁸F]FB-nLDL and [¹⁸F]FB-oxLDL/[¹⁸F]FB-OCl-LDL, respectively, from total binding. Assays for cell association of both [¹⁸F]FB-nLDL and [¹⁸F]FB-oxLDL/[¹⁸F]FB-OCl-LDL lasted for 2 h at 37°C, as for the binding studies but without Hepes. Cells incubated with LDL at 37°C contain both membrane-bound and internalized lipoproteins. Lipoprotein uptake for 2 hours was calculated as the difference between total cell-associated lipoprotein (37°C) and membrane-bound lipoprotein (4°C). At the end of all incubations the cells were washed twice with 1 mL of PBS containing 0.1% (w/v) bovine serum albumin, then twice with 1 mL of PBS. The cells were solubilized in 1.5 mL of 1 M NaOH, assayed for protein content and counted for [¹⁸F]-activity in a Cobra II gamma counter (Canberra-Packard, Meriden, CT, USA). Cell viability in this experimental setting was assessed using the trypan blue dye exclusion test. Cellular protein was determined using the bicinchoninic acid protein assay (Pierce, Rockford, IL, USA) using bovine serum albumin as protein standard. Lipoprotein uptake is expressed as percent of injected dose per μg protein (% ID/μg protein) after 2 hours.

2.9 Lipoprotein uptake experiments in vivo

The influence of 2,3-diaryl-indole coxibs C1, C3 and C4 on the biological activity of LDL in terms of blood clearance and specific organ uptake of radiolabeled [¹⁸F]FB-nLDL as well as [¹⁸F]FB-oxLDL and [¹⁸F]FB-OCl-LDL obtained from hemin/H₂O₂ and HOCl LDL oxidation experiments, respectively, was assessed in rats in vivo [41, 42]. For investigation of in vivo catabolism of [¹⁸F]FB-nLDL and [¹⁸F]FB-oxLDL/[¹⁸F]FB-OCl-LDL, animal experiments were carried out with male Wistar rats (Kyoto-Wistar strain; aged 6 weeks; 160–170 g; Harlan Laboratories, Venray, The Netherlands) according to the guidelines of the German Regulations for Animal Welfare. The protocol was approved by the local Ethical Committee for Animal Experiments and has been published elsewhere [42]. In brief, for biodistribution studies, the animals were injected [¹⁸F]FB-nLDL, [¹⁸F]FB-oxLDL or [¹⁸F]FB-OCl-LDL (0.5 mL; 0.8–1.2 MBq; radiochemical purity 96% ; PBS, pH 7.2) into the tail vein under desfluran anaesthesia. Biodistribution was determined in groups of eight rats sacrificed 5 and 60 min post injection, respectively, by heart puncture under desfluran anaesthesia. Organs and tissues of interest were rapidly excised, weighed, and the [¹⁸F]-activity was determined (Cobra II gamma counter, Canberra-Packard, Meriden, CT, USA). The accumulated activity in organs and tissues was calculated as the percentage of the injected dose per gram tissue (% ID/g tissue; corrected for decay).

2.10 Cell adhesion experiments in vitro

The potential of 2,3-diaryl-indole coxibs C1, C3 and C4 to influence the biological activity of oxLDL-promoted proinflammatory/proatherogenic processes in terms of adhesion of leukocytes to endothelial cells was assessed in human umbilical vein endothelial cells (HUVEC) in vitro. Adhesion experiments were predominantly performed as co-incubation assays with HUVEC, THP-1 monocytes, and lipoproteins following a protocol published elsewhere with some modifications [28, 57]. In brief, HUVEC were passaged onto 24-well plates, brought to confluence and treated simultaneously with native and oxidized LDL species obtained from hemin/H₂O₂ and HOCl LDL oxidation experiments (50μg/mL each final concentration) for 1 hour at 37°C. Then HUVEC were washed once with 25 mM-Hepes-buffered M199 (M199H) and, thereafter, incubated for further 60 min with THP-1 monocytes (10⁶ cells per well) at 37°C. Prior to adhesion THP-1 monocytes were labeled with the fluorescence dye BCECF/AM (Merck Group, Darmstadt, Germany). Therefore, 5×10⁶ cells/mL were incubated with 10μg/mL BCECF/AM at 37°C for 30 min. Then, cells were centrifuged and resuspended in M199H medium at 2×10⁶ cells/mL. For each labeled cell preparation a calibration curve was prepared. The calibration curve was linear in the range from 12.500 to 200.000 cells per vial. At the end of the incubation period the supernatants were carefully removed from the wells. HUVEC were washed once with M199H and the inverted test plate was centrifuged (50× g, 5 min). To lyse the cells 1 mL of 0.1 M Tris buffer plus 0.1% Triton X-100 was added to each well for 30 min. Fluorescence intensity (excitation 503 nm, emission 527 nm) of adhered cells was measured in triplicate, and the mean value was used for statistical analysis. Additionally, for the most effective compound (C1) a subsequent adhesion experiment using human polymorphonuclear leukocytes (PMN) was performed. This approach allows for distinction of effects related to the use of a human primary inflammatory cell from those observed in an immortalized (leukemia-derived) cell line. Therefore, human PMN were isolated from heparinized blood of healthy volunteers by density gradient centrifugation as described previously in detail [27, 29]. PMN then underwent the same procedure as reported above for THP-1 cells.

2.11 Cell inflammatory response in vitro

The potential of 2,3-diaryl-indole coxibs C1, C3 and C4 to influence the biological activity of oxLDL-mediated inflammatory/atherogenic processes in terms of specific cytokine release was assessed in THP-1Mφ (TNFα) and HUVEC (IL-8) [3]. Once differentiated, THP-1Mφ were incubated for 16 h (HUVEC for 4 h) with oxLDL obtained from hemin/H₂O₂ and HOCL LDL oxidation experiments in the absence or presence of 1μM of the compounds to be tested. At the end of each incubation, conditioned media were harvested and centrifuged, and TNFα and IL-8 levels were quantified on the supernatant fraction using a commercial enzyme-linked immunosorbent assays (PromoKine, PromoCell GmbH, Heidelberg, Germany) according to the manufacturer’s instructions. The results (concentrations amounted to ng/mL or pg/mL) were expressed as fold over control and all were corrected for cell viability. Cell viability in this experimental setting was determined by dimethylthiazoldiphenyltetrazolium bromide (MTT) assay [51]. Cell viability was calculated for cells upon every single/combined treatment (50μg/mL native or oxidized LDL with/without 1μM 2,3-diaryl-indole coxibs) and expressed as a percentage over the control conditions. Under these conditions and within the respective time windows for each individual experiment 2,3-diaryl-indole coxibs showed no sign of toxicity and cell viability of both THP-1 and HUVEC did not decrease below 90% (Data not shown in detail). Furthermore, nuclear factor κB (NF-κB) p56 subunit activity was measured in nuclear extracts from THP-1Mφ using an ELISA kit according to the manufacturer’s protocol (Pierce Biotechnology, Rockford, IL, USA). Chemoluminescence was measured on Synergy 4 thermostatic micro plate reader (BioTek Instruments, Bad Friedrichshall, Germany) [16].

2.12 Statistical analysis

Results are presented as means±SD from at least 3 independent experiments. Nonparametric statistical analyses were calculated by using the SPSS 20 software package (SPSS Inc., Chicago, IL, USA). Mann-Whitney test was used to compare 2 independent groups, Kruskal-Wallis test followed by Bonferroni post-hoc analysis was used to account for multiple testing. For all analyses a value of P < .05 was considered as statistically significant.

3 Results and discussion

The aim of this work was to explore i) the reactivity of 2,3-diaryl-indole coxibs with several well characterized LDL oxidation systems and ii) subsequent effects on a inflammatory/atherogenic microenvironment. The five compounds investigated were selected from a series of novel 2,3-diaryl-indole coxibs preferably bearing a fluorine or methoxy substituent, in order to gain a pool of versatile scaffold compounds promising for ¹⁸F or ¹¹C radiotracer development for functional characterization of COX-2 activity/expression in vivo using positron emission tomography [33]. The indole heterocycle is a well-characterized antioxidant pharmacophore [30]. Thus, in our previous study on 2,3-diaryl indole coxibs it consequently had to be considered that antioxidative properties might interfere with assays for determination of the COX inhibitory activity. Indeed, we observed disturbances of these assays, which clearly could be attributed to redox-active properties of the novel compounds [33]. This hypothesis was supported by another finding that one of the compounds (C1) was able to inhibit not only the formation of prostaglandin E₂ but also the formation of F_2α-isoprostanes after the exposure of endothelial cells in both monolayer and organo-type aortic ring models to ionizing radiation [44, 56]. F_2α-isoprostanes are prostaglandin-like products of nonenzymatic, free radical-catalyzed peroxidation of arachidonic acid, which are considered as potential biomarkers and have been correlated with conditions of oxidative stress, e.g., in inflammatory and atherogenic processes [38, 46]. Noteworthy, the combination of both COX-2 inhibitory and antioxidative potential in one compound is hypothesized to be very attractive. Compounds combining these properties potentially might overcome some adverse effects observed in long-term medication with coxibs, e.g., atherogenic properties in part associated with a prooxidative action [37, 58]. On the other hand, such compounds might act as double-edged swords in many pathophysiological situations, e.g., by inhibiting radiation-induced COX-2 expression and formation of ROS in parallel thus protecting normal tissue from adverse effects of radiation therapy [44, 56]. Therefore, we decided to evaluate the antioxidant capacity of the 2,3-diaryl-indole coxibs in more detail by focusing on potential prevention of LDL oxidation. There is experimental and clinical evidence supporting oxidative modification of LDL as an early and critical event in atherogenesis. Oxidation of LDL particles to atherogenic ones, on the one hand, generates high-uptake particles for macrophage and granulocyte scavenger receptor pathways leading to inappropriate accumulation of intracellular lipid deposits, and, on the other hand, generates a large panoply of lipid mediators promoting endothelial dysfunction and an inflammatory endothelial/subendothelial microenvironment [6 , 52]. The finding from earlier investigations that 2,3-diaryl indoles already show substantial antioxidative effects at pharmacologically relevant low levels prompted us to use consequently 1μM of each test compound in all experimental settings [33].

3.1 Redox activity of 2,3-diaryl-indoles

Table 1 shows data obtained from lipid and protein oxidation assays as well as from radical scavenging assays summarized here as redox activity of 2,3-diaryl indoles C1 to C5 at a concentration of 1μM in vitro. The well characterized antioxidative indole-derivative melatonin [2] and the potentially redox-neutral 1,5-diaryl-pyrazole-derivative celecoxib [58] served as reference compounds.

Except C4, the 2,3-diaryl-indoles demonstrate potent antioxidative/ROS scavenging activity (C1 > C3 > C5 ∼ C2 > > C4) towards OH^•, O₂ ^• ⁻, and OCl⁻. The more detailed investigations hence confirmed the earlier finding that various 2,3-diaryl indoles exhibit a substantial antioxidative behavior (R_DIENE/R_HAVA>1) at pharmacologically relevant low concentrations [33]. From these initial experiments three compounds were selected for further experiments: the most effective non-fluorine compound C1, the also very effective but fluorine compound C3, and, as control, the redox-neutral fluorine compound C4. Of note, in our investigations melatonin exhibited already at a concentration of 1μM a high redox activity with respect to the formation of dienes and HAVA that was comparable to the 2,3-diaryl-indoles C3 and C5. This is in contrast to the rather weak antioxidative activity of melatonin reported by others [49, 59]. On the other hand, more consistently with the literature, in this work melatonin at 1μM showed only very low OCl⁻ and O₂ ^• ⁻ scavenging activities when compared to the most potent 2,3-diaryl-indoles. The lack of aryl substituents and, logically, the higher hydrophilicity of melatonin might be one explanation for this observation.

3.2 Influence of 2,3-diaryl-indoles on lipoprotein uptake in vitro

When differentiated to macrophages after stimulation with phorbol esters, monocytic THP-1 cells are considered scavenger receptor bearing human cells and are an accepted model for the examination of cellular interaction with oxidatively modified LDL. In agreement with former results, the present lipoprotein uptake experiments revealed a significantly higher specific binding and uptake in THP-1 macrophages of both radiolabeled [¹⁸F]FB-oxLDL and [¹⁸F]FB-OCl-LDL compared to [¹⁸F]FB-nLDL (Fig. 1) [19, 42]. LDL that were oxidized in the presence of 1μM 2,3-diaryl-indole coxibs still show scavenger receptor-mediated uptake in THP-1Mφ cells, which, however, was lower compared to untreated oxLDL (C1 > C3; P < .05) and OCl-LDL (C1 ∼ C3). The observed antioxidative effects appear to be more pronounced in oxLDL when compared to OCl-LDL, which is consistent with the parameters characterizing redox activity of 2,3-diaryl-indoles towards OH^•/O₂ ^• ⁻-mediated and OCl⁻-mediated oxidation as shown in Table 1. Of interest, compound C4 did not influence the scavenger receptor mediated uptake of oxLDL/OCl-LDL in THP-1Mφ cells. As an essential prerequisite for this investigation it has been demonstrated by us earlier that the [¹⁸F]fluorobenzoylation procedure used does not lead either to adverse oxidation of nLDL particles or to additional adverse oxidative modification of oxLDL/OCl-LDL particles [41 , 49]. Moreover, 2,3-diaryl-indoles did not significantly influence the cellular uptake of nLDL (Fig. 1).

3.3 Influence of 2,3-diaryl-indoles on lipoprotein uptake in vivo

Figure 2 summarizes the distribution of activity (decay-corrected) in male Wistar rats after a single intravenous injection of [¹⁸F]FB-nLDL, [¹⁸F]FB-oxLDL or [¹⁸F]FB-OCl-LDL.

Data were obtained at 5 and 60 min post injection. In agreement with former results, the present biodistribution experiments revealed both a significantly faster blood clearance and a significantly higher specific association in liver, spleen, kidneys, and adrenals of both [¹⁸F]FB-oxLDL and [¹⁸F]FB-OCl-LDL compared to [¹⁸F]FB-nLDL. Oxidatively modified LDL that were oxidized in the presence of 1μM 2,3-diaryl-indole coxibs show a partly normalized blood clearance (P < .05) which is accompanied by significantly lowered uptake in liver and spleen (P < .05), and, in part, a normalized uptake in kidneys and adrenals when compared to untreated oxLDL (C1 > C3, P < .05) and OCl-LDL (C1 ∼ C3), respectively. In accordance with the in vitro uptake data this in vivo investigation indicates a substantial rerouting of treated oxidatively modified LDL particles from the scavenger receptor pathways back to the LDL receptor pathway. In contrast, compound C4 did not significantly influence the metabolic behavior of oxLDL/OCl-LDL in the rat model. As an essential prerequisite for this investigation it has been demonstrated by us that both [¹⁸F]FB-nLDL and [¹⁸F]FB-oxLDL/¹⁸F]FB-OCl-LDL exhibit high radiotracer stability in vivo.

3.4 Influence of 2,3-diaryl-indoles on oxidized lipoprotein-mediated cell adhesion in vitro

Incubation of HUVEC with 1×10⁶ THP-1 cells and PMN, respectively, and phosphate buffered saline showed a basal adhesion of 65,000±9,400 (THP-1) and 23,000±5,100 (PMN) cells/well. This basal adhesion amounted to 6.5% /2.3% of added THP-1/PMN cells and corresponds well to data reported by others and us [13, 28]. After co-incubation of HUVEC, THP-1 cells and LDL adhesion increased significantly (P < 0.05) (Fig. 3). Similar results were obtained using PMN instead of THP-1 monocytes. Addition of nLDL in the presence of 1μM 2,3-diaryl-indole coxibs did not show further effects. On the other hand, addition of both oxLDL and OCl-LDL instead of nLDL to the experimental system resulted in a significantly higher adhesion. This increment could be markedly decreased by 2,3-diaryl-indole coxibs (C1 ∼ C3) for both cell types. In contrast, the use of compound C4 revealed no effect on cell adhesion.

3.5 Influence of 2,3-diaryl-indoles on oxidized lipoprotein-mediated inflammatory response in vitro

Macrophages are the major source of proinflammatory cytokines like TNFα within the atherosclerotic plaque [11]. These cytokines, in turn, amplify the inflammatory response in the cellular microenvironment, e.g., by increasing macrophage adhesion to endothelial cells [61]. This process essentially contributes also to oxidized lipoprotein-mediated cell adhesion [62]. In consequence, exposure of THP-1Mφ to oxidatively modified LDL resulted in an increase of TNFα secretion compared to nLDL. This increment could be markedly decreased by 2,3-diaryl-indole coxibs (C1 ∼ C3) (Fig. 4). Vascular endothelial cells, on the other hand, usually are resistant to cholesterol accumulation by lipoprotein uptake. However, these cells also have shown to express receptors that recognize oxidatively modified LDL and to have the biochemical pathways for sterol synthesis and receptor-mediated endocytosis of lipoproteins [8, 10]. Furthermore, oxidatively modified LDL serve as important proinflammatory activators in HUVEC, e.g., by inducing and modulating expression/secretion of IL-8 [34, 54]. Consistently, exposure of HUVEC to oxidatively modified LDL resulted in a significant increase of IL-8 secretion compared to nLDL. This increment also could be markedly decreased by 2,3-diaryl-indole coxibs (C1 ∼ C3)(Fig. 4).

The expression of inflammatory cytokines TNFα and IL-8 is regulated transcriptionally by NF-κB, which among other factors is activated by oxidatively modified lipoproteins and reactive oxygen species [36, 48]. In line with this, exposure of THP-1Mφ to oxidatively modified LDL resulted in a substantially higher activity of NF-κB activity compared to THP-1Mφ incubated with nLDL. Oxidatively modified LDL that were oxidized in the presence of 1μM 2,3-diaryl-indole coxibs resulted in a significant decrement in NF-κB activation when compared to untreated oxLDL/OCl-LDL (C1 > C3; P < .05) (Fig. 5). These experiments also showed that compound C4 at a concentration of 1μM revealed no protective effects.

3.6 Redox activity of 2,3-diaryl-indoles: Structure-activity-relationships

Summarizing, the antioxidant potency of 2,3-diaryl-indole coxibs was investigated using accepted models simulating oxidative modification of LDL and, subsequently, experimental settings allowing for discrimination of selected atherogenic effects typically mediated by oxidatively modified LDL in an inflammatory environment. The data demonstrate that the examined 2,3-diaryl-indole coxibs exhibit scavenging activity for ROS with variable effectiveness, depending on the kind of ROS and the compounds’ chemical structure. In this work, compound C1 showed the overall most prominent antioxidative potential which was significantly higher than that of the known indole antioxidant melatonin used as positive control. The not further evaluated compounds C2 and C5, characterized by an unsubstituted 5-position in the indole heterocycle like C1, also showed an abundant antioxidative action that, however, was more comparable to that of melatonin. Regarding the individual redox activity of 2,3-diaryl indoles the impact of the substituent in 5-position of the indole system is noteworthy. An alkyl (methyl) substituent in this position seems to hinder efficient ROS scavenging of compound C4. Additionally, the methyl substituent in this position also diminishes the COX-2 inhibitory activity as previously published by us [33]. By contrast, a methoxy substituent in 5-position of the indole heterocycle seems to substantially support ROS scavenging as also demonstrated with compound C3 [15]. Moreover, a methoxy substituent also increases COX-2 inhibitory activity [33]. As expected, the 1,5-diaryl-pyrazole-derivative celecoxib did not exhibit any redox activity in the model systems used and was considered as negative control [33, 38]. A more detailed investigation to differentiate the ROS scavenging properties indicated i) a preferred inhibition of lipid oxidation compared to protein oxidation, ii) a preferred interaction with transition metal-catalyzed oxidation compared to HOCl-mediated oxidation, and iii) a preferred OH^• scavenging compared to O₂ ^• ⁻ scavenging. The overall antioxidative action of the compounds tested is in line with the hypothesized mechanism discussed in detail for other indole antioxidants like 2-phenylindoles [53]. This mechanism essentially would allow the presence of substituents like a methoxy group or a hydroxyl group at 5-position. Of interest, in most experiments performed here C1 showed a substantially higher antioxidative activity than did compound C3 that not alone can be explained by this mechanistic view [1, 23]. Of importance, this difference appeared to be more pronounced in the OH^•/O₂ ^• ⁻-mediated oxidation systems when compared to OCl⁻-mediated oxidation. In this regard, potential transition-metal binding properties of indole derivatives have to be taken into consideration [23, 24]. In compound C1 this possibly is supported by presence of both an aminosulfonyl and a methoxy group in para-position in the 3-phenyl and 2-phenyl substituents, respectively [4]. Regarding lipoprotein oxidation the antioxidative action of the compounds tested in part also seems to be influenced by their lipophilicity. The 2,3-diaryl-indoles, as other coxibs, show low aqueous solubility and high logPs, allocating them, beside other factors, to the low-solubility, high-permeability BCS Class II compounds and, on the other hand, strongly suggesting a natural predisposition for increased plasma lipoprotein binding [60]. This is consistent with the data reported here. In this regard, a detailed investigation on both potential transition-metal binding properties and the distribution of 2,3-diaryl-indole coxibs in plasma lipoproteins still should be performed. Considering the overall setting in this investigation most effects observed clearly can be attributed to the antioxidative action of 2,3-diaryl-indole coxibs on LDL oxidation. However, under the experimental conditions employed radical scavenging should not consume the total amount of compounds. Thus, lipoprotein-bound 2,3-diaryl-indole coxibs are suggested to exert additional, direct actions on the cells used in vitro or on tissues in the biodistribution experiment in vivo. This might include, logically, cyclooxygenase-2 inhibition but also direct antioxidative/regulative effects on intracellular pathways that, in turn, should contribute to the observed effects. This would further explain the considerable effects achieved at the concentration of 2,3-diaryl-indole coxibs used.

4 Conclusion

This work demonstrated ROS scavenging activity of novel 2,3-diaryl-indole coxibs with high potential to decrease susceptibility of LDL to oxidation in vitro. Subsequently, this led to a substantial decrease of LDL uptake by scavenger receptors in THP-1 macrophages in vitro and in a rodent model in vivo. Moreover, LDL-mediated macrophage adhesion to endothelial cells, macrophage NFκB activation, as well as macrophage and endothelial cell cytokine release in vitro was substantially diminished. Regarding the influence of most potent compounds C1 and C3 on these ‘downstream’ effects the tendency for each of them is the same. Individual variation in effect strength, however, is assumed to depend on the actual influence of the tested compounds on the overall pattern of oxidized lipids and apolipoproteins within each individual LDL particle. In line with the present data it can be hypothesized that lipophilicity and transition-metal binding may contribute to variability. The LDL-associated antioxidant properties of 2,3-diaryl-indole coxibs essentially should positively contribute to other known antiatherogenic effects of cyclooxygenase inhibition [7, 47]. This potentially may widen the therapeutic window of COX-2 targeted treatment. In this regard, antioxidant coxibs potentially provide an opportunity to overcome in part some of the known adverse cardiovascular effects of long-term cyclooxygenase inhibition [18]. On the other hand, antioxidant coxibs are promising adjuvant therapeutics for prevention of radiation-induced vascular dysfunction and atherogenesis [12 , 56].

Footnotes

Acknowledgments

The authors are grateful to Mareike Barth, Catharina Heinig, Uta Lenkeit, and Regina Herrlich for their expert technical assistance in human LDL preparation, radiolabeling, and characterization. We are also grateful to Sigrid Nitzsche from the former Lipoprotein Laboratory at the Department of Internal Medicine 3, Medical Faculty and University Hospital Carl Gustav Carus, Technische Universität Dresden, for her expert technical assistance in LDL characterization, her advice and many stimulating discussions. We also thank Steffi Kopprasch, Ph.D., from the Pathobiochemistry Unit at the Department of Internal Medicine 3, Medical Faculty and University Hospital Carl Gustav Carus, Technische Universität Dresden, for her expert advice regarding primary leukocyte isolation and handling as well as many fruitful discussions. This work was supported in part by the German Research Foundation (grant Pi 304/1-1). The authors also thank the Helmholtz Association for funding a part of this work through Helmholtz-Portfolio Topic “Technologie und Medizin –Multimodale Bildgebung zur Aufklärung des In-vivo-Verhaltens von polymeren Biomaterialien”. Franz-Jacob Pietzsch is graduate student member of the Integrated Research Training Group “Matrixengineering” (within the Transregional Collaborative Research Centre 67 “Functional biomaterials for controlling healing processes in bone and skin –from material science to clinical application” funded by German Research Foundation) at Medical Faculty and University Hospital Carl Gustav Carus, Technische Universität, Dresden.

References

Antosiewicz

, Damiani

, Jassem

, Wozniak

, Orena

and Greci

, Influence of structure on the antioxidant activity of indolinic nitroxide radicals, Free Radicals in Biology & Medicine 22 (1997), 249–255.

Bonnefont-Rousselot

, Chevé

, Gozzo

, Tailleux

, Guilloz

, Caisey

, Teissier

, Fruchart

J.C.

, Delattre

, Jore

, Lesieur

, Duriez

and Gardès-Albert

, Melatonin related compounds inhibit lipid peroxidation during copper or free radical-induced LDL oxidation, Journal of Pineal Research 33 (2002), 109–117.

Boudjeltia

K.Z.

, Legssyer

, Van Antwerpen

, Kisoka

R.L

., Babar

, Moguilevsky

, Delree

, Ducobu

, Remacle

, Vanhaeverbeek

and Brohee

, Triggering of inflammatory response by myeloperoxidase-oxidized LDL, Biochemistry and Cell Biology 84 (2006), 805–812.

Camí

G.E.

, Chacón Villalba

M.E.

, Colinas

, Echeverría

G.A.

, Estiu

and Soria

D.B.

, Crystal structure, spectroscopy and theoretical studies of p-cyanobenzenesulfonamide and a Cu(II) complex, Journal of Molecular Structure 1024 (2012), 110–116.

Campbell

J.A.

, Bordunov

, Broka

C.A.

, Browner

M.F.

, Kress

J.M.

, Mirzadegan

, Ramesha

, Sanpablo

B.F.

, Stabler

, Takahara

, Villasenor

, Walker

K.A.

, Wang

J.H.

, Welch

and Weller

, Rational design of 6-methylsulfonylindoles as selective cyclooxygenase-2 inhibitors, Bioorganic Medicinal Chemistry Letters 14 (2004), 4741–4745.

Capra

, Bäck

, Barbieri

S.S.

, Camera

, Tremoli

and Rovati

G.E.

, Eicosanoids and their drugs in cardiovascular diseases: Focus on atherosclerosis and stroke, Medicinal Research Reviews 33 (2013), 364–438.

Cipollone

, Cicolini

and Bucci

, Cyclooxygenase and prostaglandin synthases in atherosclerosis: Recent insights and future perspectives, Pharmacology & Therapeutics 118 (2008), 161–180.

Cominacini

, Garbin

, Pasini

A.F.

, Davoli

, Campagnola

, Pastorino

A.M.

, Gaviraghi

and Lo

, Cascio, Oxidized low-density lipoprotein increases the production of intracellular reactive oxygen species in endothelial cells: Inhibitory effect of lacidipine, Journal of Hypertension 16 (1998), 1913–1919.

Costa

, Moutinho

, Lima

J.L.F.C.

and Fernandes

, Antioxidant activity and inhibition of human neutrophil oxidative burst mediated by arylpropionic acid non-steroidal anti-inflammatory drugs, Biological and Pharmaceutical Bulletin 29 (2006), 1659–1670.

10.

Costa

, Zimetti

, Pedrelli

, Cremonesi

and Bernini

, Manidipine reduces pro-inflammatory cytokines secretion in human endothelial cells and macrophages, Pharmacological Research 62 (2010), 265–270.

11.

Crisafulli

, Galuppo

and Cuzzocrea

, Effects of genetic and pharmacological inhibition of TNF-alpha in the regulation of inflammation in macrophages, Pharmacological Research 60 (2009), 332–340.

12.

Eldor

, Fuks

, Matzner

, Witte

L.D.

and Vlodavsky

, Perturbation of endothelial functions by ionizing irradiation: Effects on prostaglandins, chemoattractants and mitogens, Seminars in Thrombosis and Hemostasis 15 (1989), 215–225.

13.

Erl

, Weber

P.C.

and Weber

, Monocytic cell adhesion to endothelial cells stimulated by oxidized low density lipoprotein is mediated by distinct endothelial ligands, Atherosclerosis 136 (1998), 297–303.

14.

Esterbauer

, Striegl

, Puhl

and Rotheneder

, Continuous monitoring of in vitro oxidation of human low density lipoprotein, Free Radical Research Communications 6 (1989), 67–75.

15.

Gozzo

, Lesieur

, Duriez

, Fruchart

J.C.

and Teissier

, Structure-activity relationships in a series of melatonin analogues with the low-density lipoprotein oxidation model, Free Radicals in Biology & Medicine 26 (1999), 1538–1543.

16.

Haase-Kohn

, Wolf

, Lenk

and Pietzsch

, Copper-mediated cross-linking of S100A4, but not of S100A2, results in proinflammatory effects in melanoma cells, Biochemical and Biophysical Research Communications 413 (2011), 494–498.

17.

Halliwell

, Gutteridge

J.M.

and Aruoma

O.I.

, The deoxyribose method: A simple test-tube assay for determination of rate constants for reactions of hydroxyl radicals, Analytical Biochemistry 165 (1987), 215–219.

18.

Harirforoosh

, Asghar

and Jamali

, Adverse effects of nonsteroidal antiinflammatory drugs: An update of gastrointestinal, cardiovascular and renal complications, Journal of Pharmacy & Pharmaceutical Sciences 16 (2013), 821–847.

19.

Hoppmann

, Steiniger

, Strobel

, Haase

and Pietzsch

In vivo catabolism of hypochlorite-modified low density lipoproteins (LDL): Insights from small animal positron emission tomography studiesstudies, In: Pietzsch

, (ed.), Research Signpost, Trivandrum, India, 2006, pp 197–219

20.

, Guo

, Chu

, Bai

, Yi

, Cheng

and Li

, Synthesis and biological evaluation of substituted 2-sulfonyl-phenyl-3-phenyl-indoles: A new series of selective COX-2 inhibitors, Bioorganic Medicinal Chemistry 11 (2003), 1153–1160.

21.

J.Y.

and Kim

., Paik

S. G.

, and Han

P. L.

, Cyclooxygenase-2-dependent neuronal death proceeds via superoxide anion generation, Free Radical Biology and Medicine 41 (2006), 960–972.

22.

Jawabrah Al-Hourani

, Sharma

S.K.

, Suresh

, and Wuest

, Cyclooxygenase-2 inhibitors: A literature and patent review -, Expert Opinion on Therapeutic Patents 21(9) (2011), 1339–1432.

23.

Kaminskaia

N.V.

, Ullmann

G.M.

, Fulton

D.B.

and Kostic

N.M.

, Spectroscopic, kinetic, and mechanistic study of a new mode of coordination of indole derivatives to platinum(II) and palladium(II) ions in complexes, Inorganic Chemistry 39 (2000), 5004–5013.

24.

Karekal

M.R.

, Biradar

, and Bennikallu Hire Mathada

Synthesis, characterization, antimicrobial, DNA cleavage, and antioxidant studies of some metal complexes derived from schiff base containing indole and quinoline moieties, Bioinorganic Chemistry and Applications 2013 (2013), 315972.

25.

Kniess

, Laube

, Bergmann

, Sehn

, Graf

, Steinbach

, Wuest

and Pietzsch

, Radiosynthesis of a 18F-labeled 2,3-diarylsubstituted indole via McMurry coupling for functional characterization of cyclooxygenase-2 (COX-2) in vitro and in vivo, Bioorganic Medicinal Chemistry 20 (2012), 3410–3421.

26.

Kopprasch

, Leonhardt

, Pietzsch

and Kühne

, Hypochlorite-modified low-density lipoprotein stimulates human polymorphonuclear leukocytes for enhanced production of reactive oxygen metabolites, enzyme secretion, and adhesion to endothelial cells, Atherosclerosis 136 (1998), 315–324.

27.

Kopprasch

, Pietzsch

and Grässler

, The protective effects of HDL and its constituents against neutrophil respiratory burst activation by hypochlorite-oxidized LDL, Molecular and Cellular Biochemistry 258 (2004), 121–127.

28.

Kopprasch

, Pietzsch

and Westendorf

, H-J Kruse and J. Grässler, The pivotal role of scavenger receptor CD36 and phagocyte-derived oxidants in oxidized low density lipoprotein-induced adhesion to endothelial cells, The International Journal of Biochemistry & Cell Biology 36 (2004), 460–471.

29.

Kopprasch

, Richter

, Leonhardt

, Pietzsch

and Grässler

, Urate attenuates oxidation of native low-density lipoprotein by hypochlorite and the subsequent lipoprotein-induced respiratory burst activities of polymorphonuclear leukocytes, Molecular and Cellular Biochemistry 206 (2000), 51–56.

30.

Kruk

, Aboul-Enein

H.Y.

, Michalska

, Lichszteld

, Kubasik-Kladna

and Olgen

, scavenging activity for reactive oxygen species by N-substituted indole-2-carboxylic acid esters, Luminescence 22 (2007), 379–386.

31.

Lapenna

, Ciofani

, Festi

, Neri

, Pierdomenico

S.D.

, Giamberardino

M.A.

and Cuccurullo

, Antioxidant properties of ursodeoxycholic acid, Biochemical Pharmacology 64(11) (2002), 1661–1667.

32.

Laube

, Kniess

and Pietzsch

, Radiolabeled COX-2 inhibitors for non-invasive visualization of COX-2 expression and activity–a critical update, Molecules 18 (2013), 6311–6355.

33.

Laube

, Tondera

, Sharma

S.K.

and Bechmann

, F-J Pietzsch, A. Pigorsch, M. Köckerling, F. Wuest, J. Pietzsch and T. Kniess, 2,3-Diaryl-substituted indole based COX-2 inhibitors as leads for imaging tracer development, RSC Advances 4 (2014), 38726–38742.

34.

Lussi

, Potapovich

, Riccardo

, Dellambra

, Pressi

, Kostyuk

, Dal Toso

, De Luca

, Pastore

, and Korkina

, Anti-inflammatory effects of concentrated ethanol extracts of Edelweiss (Leontopodium alpinum Cass.) callus cultures towards human keratinocytes and endothelial cells, Mediators of Inflammation 2012 (2012), 498373.

35.

Marnett

L.J.

, The COXIB experience: A look in the rearview mirror, Annual Review of Pharmacology and Toxicology 49 (2009), 265–290.

36.

Mazière

and Mazière

J.C.

, Activation of transcription factors and gene expression by oxidized low-density lipoprotein, Free Radicals in Biology & Medicine 46 (2009), 127–137.

37.

Mukherjee

, Nissen

S.E.

and Topol

E.J.

, Risk of cardiovascular events associated with selective COX-2 inhibitors, Journal oft the American Medical Association 286 (2001), 954–959.

38.

Oguogho

, Kritz

, Wagner

and Sinzinger

, 6-oxo-PGF1α and 8-epi-PGF2α in the arterial wall layers of various species: A comparison between intact and atherosclerotic areas, Prostaglandins, Leukotrienes & Essential Fatty Acids 64 (2001), 167–171.

39.

Penning

T.D.

, Talley

J.J.

, Bertenshaw

S.R.

, Carter

J.S.

, Collins

P.W.

, Docter

, Graneto

M.J.

, Lee

L.F.

, Malecha

J.W.

, Miyashiro

J.M.

, Rogers

R.S.

, Rogier

D.J.

, Yu

S.S.

, Anderson

G.D.

, Burton

E.G.

, Cogburn

J.N.

, Gregory

S.A.

, Koboldt

C.M.

, Perkins

W.E.

, Seibert

, Veenhuizen

A.W.

, Zhang

Y.Y.

and Isakson

P.C.

, Synthesis and biological evaluation of the 1,5-diarylpyrazole class of cyclooxygenase-2 inhibitors: Identification of 4-[5-(4-methylphenyl)-3-(trifluoromethyl)-1H-pyrazol-1-yl]benze nesulfonamide (SC-5, celecoxib), Journal of Medicinal Chemistry 40 (1997), 1347–1365.

40.

Pietzsch

, Measurement of 5-hydroxy-2-aminovaleric acid as a specific marker of iron-mediated oxidation of proline and arginine side-chain residues of low-density lipoprotein apolipoprotein B-100, Biochemical and Biophysical Research Communications 270 (2000), 852–857.

41.

Pietzsch

, Bergmann

, Rode

, Hultsch

, Pawelke

, Wuest

and van den Hoff

, Fluorine-18 radiolabeling of low density lipoproteins (LDL): A potential approach for characterization and differentiation of metabolism of native and oxidized LDL in vivo, Nuclear Medicine and Biology 31 (2004), 1043–1050.

42.

Pietzsch

, Bergmann

, Wuest

, Pawelke

, Hultsch

and van den Hoff

, Catabolism of native and oxidized low density lipoproteins: In vivo insights from small animal positron emission tomography studies, Amino Acids 29 (2005), 389–404.

43.

Pietzsch

, Kopprasch

J.S.

, and Bergmann

,Analysis of 3-chlorotyrosine as a specific marker of protein oxidation: The use of N(O,S)-ethoxycarbonyltrifluoroethyl ester derivatives and gas chromatography/mass spectrometry, Rapid Communications in Mass Spectrometry 17 (2003), 767–770.

44.

Pietzsch

, Laube

, Pietzsch

F.J.

, Bergmann

and Kniess

, Concomitant targeting of cyclooxygenase-2 and oxidant stress pathways for radioprotection of endothelial cells, In: Lewis

B.S.

, Flugelman

M.Y.

, and Halon

D.A.

, (eds.), Medimond S.r.l - Monduzzi Editore International, Bologna, Italy, 2011, pp. 107–110.

45.

Praticò

, Prostanoid and isoprostanoid pathways in atherogenesis, Atherosclerosis 201 (2008), 8–16.

46.

Praticò

, Lawson

J.A.

, Rokach

and FitzGerald

G.A.

, The isoprostanes in biology and medicine, Trends in Endocrinology & Metabolism 12 (2001), 243–247.

47.

Reis

E.D.

, Roque

, Dansky

, Fallon

J.T.

, Badimon

J.J.

, Cordon-Cardo

, Shiff

S.J.

and Fisher

E.A.

, Sulindac inhibits neointimal formation after arterial injury in wild-type and apolipoprotein E-deficient mice, Proceedings of the National Academy of Sciences USA 97 (2000), 12764–12769.

48.

Robbesyn

, Salvayre

and Negre-Salvayre

, Dual role of oxidized LDL on the NF-kappaB signaling pathway, Free Radical Research 38 (2004), 541–551.

49.

Seeger

, Mueck

A.O.

and Lippert

T.H.

, Effect of melatonin and metabolites on copper-mediated oxidation of low density lipoprotein, British Journal of Clinical Pharmacology 44 (1997), 283–284.

50.

Simmons

D.L.

, Botting

R.M.

and Hla

, Cyclooxygenase isozymes: The biology of prostaglandin synthesis and inhibition, Pharmacological Reviews 56 (2004), 387–437.

51.

Steffen

, Schewe

and Sies

, Epicatechin protects endothelial cells against oxidized LDL and maintains NO synthase, Biochemical and Biophysical Research Communications 331 (2005), 1277–1283.

52.

Steinberg

, Parthasarathy

, Carew

T.E.

, Khoo

J.C.

and Witztum

J.L.

, Beyond cholesterol. Modifications of low-density lipoprotein that increase its atherogenicity, The New England Journal of Medicine 320 (1989), 915–924.

53.

Suzen

, Bozkaya

, Coban

and Nebioğlu

, Investigation of the in vitro antioxidant behaviour of some 2-phenylindole derivatives: Discussion on possible antioxidant mechanisms and comparison with melatonin, Journal of Enzyme Inhibition and Medicinal Chemistry 21 (2006), 4405–4411.

54.

Terkeltaub

, Banka

C.L.

, Solan

, Santoro

, Brand

and Curtiss

L.K.

, Oxidized LDL induces monocytic cell expression of interleukin-8, a chemokine with T-lymphocyte chemotactic activity, Arteriosclerosis and Thrombosis 14 (1994), 47–53.

55.

Ullm

, Laube

, Bechmann

, Kniess

and Pietzsch

, Organotypical vascular model for characterization of radioprotective compounds: Studies on antioxidant 2,3-diaryl-substituted indole-based cyclooxygenase-2 inhibitors, Clinical Hemorheology and Microcirculation 58 (2014), 281–295.

56.

Ullm

, Sehn

, Laube

, Tondera

, Bechmann

, Mosch

, Kniess

and Pietzsch

, Targeting cyclooxygenase-2 and oxidant stress pathways for attenuation of radiation-induced vascular dysfunctionIn: Zarzuelo

, and Jimenez

, (eds.), Medimond S.r.l - Monduzzi Editore International, Bologna, Italy, 2013, pp. 8790.

57.

Vaporciyan

A.A.

, Jones

M.L.

and Ward

P.A.

, Rapid analysis of leukocyte-endothelial adhesion, Journal of Immunological Methods 159 (1993), 93–100.

58.

Walter

M.F.

, Jacob

R.F.

, Day

C.A.

, Dahlborg

, Weng

and Mason

R.P.

, Sulfone COX-2 inhibitors increase susceptibility of human LDL and plasma to oxidative modification: Comparison to sulfonamide COX-2 inhibitors and NSAIDs, Atherosclerosis 177 (2004), 235–243.

59.

Walters-Laporte

, Furman

, Fouquet

, Martin-Nizard

, Lestavel

, Gozzo

, Lesieur

, Fruchart

J.C.

, Duriez

and Teissier

, A high concentration of melatonin inhibits in vitro LDL peroxidation but not oxidized LDL toxicity toward cultured endothelial cells, Journal of Cardiovascular Pharmacology 32 (1998), 582–592.

60.

Wasan

K.M.

, Brocks

D.R.

, Lee

S.D.

, Sachs-Barrable

and Thornton

S.J.

, Impact of lipoproteins on the biological activity and disposition of hydrophobic drugs: Implications for drug discovery, Nature Reviews Drug Discovery 7 (2008), 84–99.

61.

You

, Wang

, Tong

, Hao

, Li

, Yang

, Huang

and Liao

, Anti-inflammatory effect of acetylharpagide demonstrated by its influence on leukocyte adhesion and transmigration in endothelial cells under controlled shear stress, Clinical Hemorheology and Microcirculation 56 (2014), 205–217.

62.

Zhang

, The role of inflammatory cytokines in endothelial dysfunction, Basic Research in Cardiology 103 (2008), 398–406.

Protective effects of 2,3-diaryl-substituted indole-based cyclooxygenase-2 inhibitors on oxidative modification of human low density lipoproteins in vitro

Abstract

Keywords

1 Introduction

2 Materials and methods

2.1 Solvents and reagents

2.2 Chemical synthesis of coxibs

2.3 Isolation of native LDL

2.4 LDL lipid and protein oxidation

2.5 OH • scavenging assay

2.6 O2 • − scavenging assay

2.7 Radiolabeling of native and oxidized LDL

2.8 Lipoprotein uptake experiments in vitro

2.9 Lipoprotein uptake experiments in vivo

2.10 Cell adhesion experiments in vitro

2.11 Cell inflammatory response in vitro

2.12 Statistical analysis

3 Results and discussion

3.1 Redox activity of 2,3-diaryl-indoles

3.2 Influence of 2,3-diaryl-indoles on lipoprotein uptake in vitro

3.3 Influence of 2,3-diaryl-indoles on lipoprotein uptake in vivo

3.4 Influence of 2,3-diaryl-indoles on oxidized lipoprotein-mediated cell adhesion in vitro

3.5 Influence of 2,3-diaryl-indoles on oxidized lipoprotein-mediated inflammatory response in vitro

3.6 Redox activity of 2,3-diaryl-indoles: Structure-activity-relationships

4 Conclusion

Footnotes

Acknowledgments

References

2.5 OH^• scavenging assay

2.6 O₂^•⁻ scavenging assay