Tryptamine-Gallic Acid Hybrid Prevents Non-steroidal Anti-inflammatory Drug-induced Gastropathy

Background: Non-steroidal anti-inflammatory drugs (NSAIDs) induce gastropathy by promoting mitochondrial pathology, oxidative stress, and apoptosis in gastric mucosal cells. Results: We have synthesized SEGA (3a), a tryptamine-gallic acid hybrid, which prevents NSAID-induced gastropathy by preventing mitochondrial oxidative stress, dysfunction, and apoptosis. Conclusion: SEGA (3a) bears an immense therapeutic potential against NSAID-induced gastropathy. Significance: This novel molecule is a significant addition in the discovery of gastroprotective drugs. We have investigated the gastroprotective effect of SEGA (3a), a newly synthesized tryptamine-gallic acid hybrid molecule against non-steroidal anti-inflammatory drug (NSAID)-induced gastropathy with mechanistic details. SEGA (3a) prevents indomethacin (NSAID)-induced mitochondrial oxidative stress (MOS) and dysfunctions in gastric mucosal cells, which play a pathogenic role in inducing gastropathy. SEGA (3a) offers this mitoprotective effect by scavenging of mitochondrial superoxide anion (O2̇̄) and intramitochondrial free iron released as a result of MOS. SEGA (3a) in vivo blocks indomethacin-mediated MOS, as is evident from the inhibition of indomethacin-induced mitochondrial protein carbonyl formation, lipid peroxidation, and thiol depletion. SEGA (3a) corrects indomethacin-mediated mitochondrial dysfunction in vivo by restoring defective electron transport chain function, collapse of transmembrane potential, and loss of dehydrogenase activity. SEGA (3a) not only corrects mitochondrial dysfunction but also inhibits the activation of the mitochondrial pathway of apoptosis by indomethacin. SEGA (3a) inhibits indomethacin-induced down-regulation of bcl-2 and up-regulation of bax genes in gastric mucosa. SEGA (3a) also inhibits indometacin-induced activation of caspase-9 and caspase-3 in gastric mucosa. Besides the gastroprotective effect against NSAID, SEGA (3a) also expedites the healing of already damaged gastric mucosa. Radiolabeled (99mTc-labeled SEGA (3a)) tracer studies confirm that SEGA (3a) enters into mitochondria of gastric mucosal cell in vivo, and it is quite stable in serum. Thus, SEGA (3a) bears an immense potential to be a novel gastroprotective agent against NSAID-induced gastropathy.

a result of MOS. SEGA (3a) in vivo blocks indomethacin-mediated MOS, as is evident from the inhibition of indomethacin-induced mitochondrial protein carbonyl formation, lipid peroxidation, and thiol depletion. SEGA (3a) corrects indomethacin-mediated mitochondrial dysfunction in vivo by restoring defective electron transport chain function, collapse of transmembrane potential, and loss of dehydrogenase activity. SEGA (3a) not only corrects mitochondrial dysfunction but also inhibits the activation of the mitochondrial pathway of apoptosis by indomethacin. SEGA (3a) inhibits indomethacin-induced down-regulation of bcl-2 and up-regulation of bax genes in gastric mucosa. SEGA (3a) also inhibits indometacin-induced activation of caspase-9 and caspase-3 in gastric mucosa. Besides the gastroprotective effect against NSAID, SEGA (3a) also expedites the healing of already damaged gastric mucosa. Radiolabeled ( 99m Tc-labeled SEGA (3a)) tracer studies confirm that SEGA (3a) enters into mitochondria of gastric mucosal cell in vivo, and it is quite stable in serum. Thus, SEGA (3a) bears an immense potential to be a novel gastroprotective agent against NSAIDinduced gastropathy.
NSAIDs 2 are the most popular drugs commonly used throughout the world for the treatment of pain, inflammation, rheumatic disorders, and osteoarthritis (1,2). Approximately 30 million patients consume NSAIDs on a daily basis (1). However, NSAIDs have limitations; they induces gastropathy, and ϳ107,000 patients are hospitalized every year due to NSAIDrelated gastrointestinal complications (3). Extensive studies have established that besides acid secretion, there are other important factors, such as gastric mucosal blood flow, mucusbicarbonate secretion, antioxidant level, reactive oxygen species (ROS), mitochondrial oxidative stress (MOS), apoptosis, and mucosal cell renewal, involved in the pathogenesis of gastroduodenal injury (4 -11). It is well established that the major cause of NSAID-induced gastropathy is the development of oxidative stress in gastric mucosa due to the excess generation of ROS. It has also been documented that ROS induces gastropathy through the induction of gastric mucosal cell apoptosis (4 -6, 12). NSAID acts as an inhibitory uncoupler in human mitochondria (13). Indomethacin (NSAID) interacts with the complex I of electron transport chain and results in the leakage of electrons, which in turn leads to the formation of superoxide anion radical (O 2 . ) (14). The dismutation of mitochondrial O 2 .
by superoxide dismutase leads to the formation of hydrogen peroxide (H 2 O 2 ) (15,16), and H 2 O 2 further reacting with O 2 .
generates highly reactive hydroxyl radical ( ⅐ OH) through the Haber-Weiss reaction (6). The excess O 2 . , if not dismutated, offers toxic insult by oxidatively damaging and inactivating mitochondrial aconitase, resulting in the release of iron from its Fe-S cluster (6,17). Again, the released iron in presence of H 2 O 2 generates ⅐ OH through the Fenton reaction (17). Heme oxygenase 1 may also generate free iron by catabolizing excess free heme. Heme oxygenase 1 translocates to mitochondria and decreases intramitochondrial free heme accumulated during gastric injury by NSAID. Excess free heme and overactivity of heme oxygenase 1 inside mitochondria may favor the accumulation of free iron in excess to ferritin sequestration (18). Free iron overload in cells has been shown to be associated with the development and progression of several pathological conditions (19 -21). Intramitochondrial free iron and ROS lead to MOS and consequent dysfunction (19 -25). MOS disrupts cellular integrity and promotes cell death (5,26,27), which ultimately leads to organ damage. The overproduction of ROS develops mitochondrial pathology (22,24,28,29), as indicated by the defect in electron transport chain and ATP synthesis, opening of mitochondrial permeability transition pore (MPTP), fall in transmembrane potential (⌬⌿ m ), oxidative damage of mitochondrial DNA, proteins, and phospholipids (30), and finally the activation of the mitochondrial pathway of apoptosis (6,31). Thus, mitochondrial dysfunction triggers the mitochondrial pathway of apoptosis (6,(32)(33)(34)(35)(36)(37)(38). Mitochondrial dysfunction and concurrent apoptosis play an important role in NSAID-induced gastropathy (4, 6 -7, 12). Hence, it is clear that the molecule that will prevent MOS and consequent mitochondrial dysfunction will be effective against NSAID-induced gastropathy.
The aim of the present study is to design a small molecule that will correct NSAID-induced mitochondrial pathology, apoptosis, and gastropathy. Here, we report the designing of a tryptamine-gallic acid hybrid molecule, SEGA (3a), which prevented NSAID-induced mitochondrial pathology, apoptosis, and gastropathy by blocking MOS, chelating intramitochondrial free iron, and correcting mitochondrial pathology entering into mitochondria.
General EDC Coupling Procedure for Formation of Esters or Amide-To a solution of 3,4,5-tris(benzyloxy)benzoic acid/4hydroxycinnamic acid/indole-3-acetic acid (1 eq, 12 mmol), amines (hydrochloride)/alcohol (1.2 eq, 14.4 mmol), 1-hydroxybenzotriazole (1 eq, 12 mmol, for amines), and Et 3 N (6 eq, 72 mmol, for amines)/DMAP (1 eq, 12 mmol, for alcohols) in N,N-dimethylform amide, EDC hydrochloride (1.2 eq, 14.4 mmol) was added at 0°C. Then the reaction mixture was stirred at room temperature overnight until completion of the reaction, monitored by thin layer chromatography (TLC). After that reaction, the mixture was quenched by the addition of ice-cold H 2 O and extracted with ethyl acetate. The combined organic phase was washed with brine and dried over Na 2 SO 4 . The organic phase was then reduced in vacuo; the concentrated ethyl acetate extracts were chromatographed over a silica gel column.
General Procedure of Debenzylation (Hydrogenolysis)-A mixture of compound and 10% palladium on carbon (catalytic) in a methanol/chloroform mixture (5:1) (10 ml) was hydrogenated at 40 p.s.i. for 2 h and was filtered through Celite, after removal of catalyst. The filtrate was evaporated in vacuo to dryness, to give the product. The details of the materials and methods for synthesis are described in the supplemental material.
Determination of in Vitro Antioxidant Property by Following Ferric Reducing Antioxidant Power (FRAP)-The assay was performed in a 96-well microplate as described earlier (4). FRAP reagent was prepared by mixing of 10 ml of acetate buffer (200 mM, pH 3.6), 1 ml of TPTZ solution (10 mM in 40 mM HCl), and 1 ml of ferric chloride solution (20 mM) in distilled water. The solution was kept for 1 h in a water bath at 37°C. In a 96-well microplate, 25 l of the compounds under investigation dissolved (in methanol or water) at different concentrations in the range 1-100 M were placed in triplicate, and freshly prepared FRAP solution (175 l) was added to this sample. Absorbance was monitored at 595 nm at different time intervals up to 150 min in a microplate reader. Absorbance of 175 l of FRAP solution and 25 l of methanol or water mixture was taken, which was subtracted from the absorbance of the samples at each time interval to calculate the absorbance change (⌬A). The FRAP value at time interval t (FRAP value t ) was calculated according to the formula, where ⌬a t T is the absorbance change after the time interval t (6 min) relative to the tested tryptamine derivatives at a concentration of 10 M, and ⌬a t Fe 2ϩ is the absorbance change at the same time interval relative to ferrous sulfate at the same concentration (4).
Free Radical-scavenging Activity by Following DPPH Radical Assay-DPPH is a stable free radical and shows absorbance at 517 nm. Antioxidant molecules scavenge the DPPH radical by donating hydrogen, as visualized by discoloration of the DPPH radical from purple to yellow (4, 39). The assay system contained 1 ml of compounds under investigation dissolved (in methanol or water) at different concentrations in the range 10 -100 M and 4 ml of DPPH (0.15 mM) in methanol (80% (v/v) in water) and mixed well. It was allowed to stand for 30 min at room temperature away from light. Ascorbic acid and gallic acid were used as positive control. The absorbance of the solution was measured at 517 nm.
Iron Chelating Activity in Vitro-The assay system has a total volume of 1 ml containing Fe(II) (10 M) in 20 mM phosphate buffer, pH 7.4. SEGA (3a) at different concentrations (500 nM to 100 M) and TPTZ (20 M solution) were added to the Fe(II) solutions in small volumes to the sample cuvette with the concomitant addition of the same volume of DMSO to the reference cuvette (SEGA (3a) was dissolved in DMSO). For the control group, the assay system is the same without SEGA (3a). Desferrioxamine, a well known iron chelator was used as a positive control. Iron chelating ability of SEGA (3a) at different concentrations was monitored by recording the absorbance of the Fe(II)-TPTZ complex immediately after each addition in the quartz cells of the 1-cm light path in a PerkinElmer Life Sciences Lamda 15 UV-visible spectrophotometer at 25 Ϯ 1°C. The contents were mixed well before the spectrum was recorded.
Animals and Indomethacin (NSAID)-induced Gastric Damage (Gastropathy)-All of the in vivo studies were done in accordance with the institutional animal ethical committee guidelines. Sprague-Dawley rats (180 -220 g) were used for this study. Each group (control or experimental) of animals was maintained at 24 Ϯ 2°C with a 12-h light and dark cycle. The animals were fasted for 24 h before the start of experiments to avoid food-induced increased acid secretion and its effect on gastric lesions. The rats were provided with water ad libitum. A gastric mucosal lesion was induced by indomethacin as described (5). Briefly, all of the animals were divided into control, indomethacin-treated, and drug-pretreated indomethacin-treated groups. Oral administration of indomethacin at a dose of 48 mg⅐kg Ϫ1 b.w. was given to the fasted animals to induce gastric injury. In the drug-pretreated groups, the animals were given SEGA (3a) (50, 20, 10, 5, 3, and 1 mg⅐kg Ϫ1 b.w.), intraperitoneally 30 min prior indomethacin treatment. The control group received vehicle only. After 4 h of indomethacin treatment, the animals were sacrificed under proper euthanasia, and stomachs were collected. The severity of mucosal lesions was scored as the injury index (40) according to the following scale: 0, no pathology; 1, a small injury (1-2 mm); 2, a medium injury (3-4 mm); 4, a large injury (5-6 mm); 8, a larger injury (Ͼ6 mm). The sum of the total scores in each group of rats divided by the number of the animals was expressed as the mean injury index (4 -6).
For the healing study, gastric mucosal injury was first induced with indomethacin treatment at a dose of 48 mg⅐kg Ϫ1 b.w. 4 h after the induction of mucosal injury, some of the animals were divided into two different groups (n ϭ 6) (i.e. autohealing and SEGA (3a)-induced healing). This time point was referred as 0 h of healing. At this point of time, in the SEGA (3a)-induced healing group, SEGA (3a) was administered (intraperitoneally) at a dose of 50 mg⅐kg Ϫ1 b.w. (this dose was selected from the dose-response curve). The animals, which were not treated with SEGA (3a) and received only indometh-acin, served as the autohealing group. Starting from 0 h, the stomach was dissected out from all groups at intervals of 2, 4, 8, and 24 h, respectively, for measuring injury index as described (6,12) and histological studies.
Histological Study-Stomach tissue from control and experimental groups was washed a number of times with phosphatebuffered saline (PBS, pH 7.4) and fixed in 10% buffered formalin for 12 h at 25°C. The fixed tissues were then dehydrated and embedded in paraffin for preparing semithin sections (4). A microtome was used to prepare the semithin sections, which were then taken over poly-L-lysine-coated glass slides for hematoxylin-eosin staining. The stained sections were observed under a microscope (Leica DM-2500, Leica Microsystems GmbH, Wetzlar, Germany) and were documented by a high resolution digital camera.
Soret Spectroscopy to Detect Hemoglobin Released in Stomach during Mucosal Injury-After opening the stomach, gastric mucosal tissues from control and experimental groups were washed with PBS (pH 7.4). This PBS solution was collected and clarified by centrifugation at 12,000 ϫ g for 20 min. The clarified PBS solution was monitored to detect hemoglobin by recording Soret absorbance immediately in quartz cells of 1-cm light path in a PerkinElmer Lamda 15 UV-visible spectrophotometer at 25 Ϯ 1°C (41,42).
Gastric Mucosal Cell Culture-Gastric mucosal cells were isolated and cultured as described earlier (4). Mucosa from the rat stomach was scraped in Hanks' balanced salt solution (HBSS) (pH 7.4) containing 100 units/ml penicillin, 100 units/ml streptomycin, and 10 g/ml gentamycin. The mucosa was then minced and suspended in HBSS (pH 7.4), containing 0.05% hyaluronidase and 0.1% collagenase type I. The suspension was incubated for 30 min at 37°C in a 5% CO 2 environment with shaking and then filtered through a sterile nylon mesh. The filtrate was centrifuged at 600 ϫ g for 5 min, and the cell pellet was washed with HBSS (pH 7.4) and further centrifuged. The pellet was incubated in 5 ml of Ham's F-12 medium in a T25 flask supplemented with 10% fetal bovine serum (FBS) and 100 units/ml penicillin, 100 units/ml streptomycin, and 10 g/ml gentamycin. Cells were cultured at 37°C with 5% CO 2 and grown to ϳ90% confluence before treatment. 90% of the cells obtained following this protocol possessed epithelial characteristics. For all in vitro experiments, cultured cells were first divided into three groups: control, indomethacin-treated, and SEGA (3a)-pretreated indomethacin-treated in 12-well plates with 10 6 cells/well. Each well of the "indomethacin group" was treated with indomethacin, each well of the "indomethacin plus SEGA (3a) group" was treated with SEGA (3a) 30 min prior to indomethacin treatment, and the control was treated with only vehicle.

Measurement of Intramitochondrial Superoxide Anion (O 2 . )
in Gastric Mucosal Cells-Mitochondrial O 2 . was detected by fluorescence microscopy using the specific dye MitoSOX. Equal numbers of gastric mucosal primary cultured cells from control and experimental groups were used for the detection of intramitochondrial O 2 . using MitoSOX, a superoxide-sensitive fluorescence probe, following the protocol as described in the product catalogue (6,43). Cells were stained with the fluorescent probe in HBSS (pH 7.4) and incubated for 15 min at 37°C in the dark. After the incubation, cells were washed with HBSS three times and used for fluorescence microscopy (Leica DM-2500). Staining of MitoSOX was visualized using a red filter.

Measurement of Intramitochondrial Free Iron in Gastric Mucosal
Cells-Equal numbers of gastric mucosal primary cultured cells from control and experimental groups were used for free iron localization using Phen Green SK, an iron-sensitive fluorescence probe, following the protocol as described in the product catalogue. Cells were first incubated with Phen Green SK (20 M) for 15 min at 37°C in the dark. After the incubation, cells were washed with HBSS and used for fluorescence microscopy (Leica DM-2500). The fluorescence of Phen Green SK was visualized using a green filter (6).
Isolation of Mitochondria-Mitochondria were isolated and purified by following the protocol reported earlier (18,44). In brief, the scraped gastric mucosa from the control, indomethacin-treated (48 mg⅐kg Ϫ1 b.w.), and SEGA (3a)-pretreated (50 mg⅐kg Ϫ1 b.w.) indomethacin-treated groups were minced and homogenized in isolation buffer, followed by centrifugation at 600 ϫ g for 10 min to remove the cell debris and nuclear pellet. This was further centrifuged at 12,000 ϫ g for 15 min to obtain the crude mitochondrial pellet. A 25-50% Percoll density gradient was prepared. Over the 25% Percoll, the crude mitochondrial pellet (resuspended in cold 15% Percoll solution) was layered. It was further centrifuged at 30,000 ϫ g at 4°C for 30 min to obtain pure mitochondria at the interface between the Percoll (25-50%) layers. The mitochondria were isolated from the interface. Further, they were washed with isolation buffer and centrifuged at 16,700 ϫ g at 4°C for 10 min. The supernatant was discarded, and 10 mg⅐ml Ϫ1 fatty acid-free BSA was added and mixed. Afterward, the mixture was centrifuged at 6,900 ϫ g at 4°C for 10 min. The purified mitochondria (pellet) were resuspended in storage buffer. Mitochondrial protein content was determined by using the method of Lowry (45).
Measurement of MOS-Isolated mitochondria from stomach tissue of control, indomethacin-treated (48 mg⅐kg Ϫ1 b.w.), and SEGA (3a)-pretreated (50 mg⅐kg Ϫ1 b.w.) and indomethacintreated (48 mg⅐kg Ϫ1 b.w.) rats were used to detect MOS. MOS was measured as described earlier through the quantification of total thiol, lipid peroxidation products, and protein carbonyl in mitochondria. In brief, thiol content was measured by its reaction with 5,5Ј-dithiobis(nitrobenzoic acid) to yield the yellow chromophore of thionitrobenzoic acid, which was measured at 412 nm. Mitochondrial lipid peroxidation was assayed by adding 1 ml of the mitochondrial fraction in 0.9% normal saline to 2 ml of thiobarbituric acid/TCA mixture (0.375% (w/v) and 15% (w/v), respectively) in 0.25 N HCl and was mixed and boiled for 15 min. The samples were then cooled, and after centrifugation, the absorbance of the supernatant was read at 535 nm. Tetraethoxypropane was used as a standard. Protein carbonyl was measured by following the standard colorimetric method that measures the binding of dinitrophenylhydrazine to the carbonyl group and was quantified by taking the absorbance at 362 nm (4,43).
Assessment of Mitochondrial Respiratory Function by Following Mitochondrial Oxygen Consumption-Mitochondrial oxygen consumption was measured using a Clark-type electrode in a liquid phase oxygen measurement system (Oxygraph, Hansatech, Norfolk, UK) (46). Complex I (state 3)-mediated oxygen consumption was initiated by the incorporation of glutamate and malate (5 mM each) to 1 ml of respiratory medium (250 mM sucrose, 5 mM KH 2 PO 4 , 5 mM MgCl 2 , 0.1 mM EDTA, and 0.1% BSA in 20 mM HEPES, pH 7.2). Basal respiration (state 2) was measured following the addition of mitochondrial suspension. The addition of ADP (1 mM) marks the initiation of state 3 respiration. State 4 respiration was recorded in the absence of ADP. The respiratory control ratio (RCR) was obtained from the ratio of state 3 respiration (nmol of O 2 consumed) and state 4 respiration (nmol of O 2 consumed) (18,47).
Measurement of ⌬⌿ m -Isolated mitochondria from stomach tissue of rats were used to detect ⌬⌿ m . Equal amounts of mitochondria (25 g) from control and experimental groups were taken in 100 l of JC-1 assay buffer (100 mM MOPS, pH 7.5, containing 550 mM KCl, 50 mM ATP, 50 mM MgCl 2 , 50 mM sodium succinate, 5 mM EGTA) and were incubated in the dark with JC-1 (300 nM) for 15 min at 25°C. The fluorescence of each sample was measured in a Hitachi F-7000 fluorescence spectrofluorimeter (excitation 490 nm, emission 530 nm for JC-1 monomer; emission 590 nm for JC-1 aggregates). ⌬⌿ m was expressed as fluorescence ratio of 590 nm/530 nm (4,43).
Measurement of Mitochondrial Dehydrogenase Activity-Mitochondrial metabolic function was studied by observing the ability of mitochondrial dehydrogenases to reduce 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) into formazan dye (4). Equal numbers of gastric mucosal primary cultured cells were taken (10 6 cells) in each well of a 12-well plate and were divided into control and experimental groups. After incubating for 16 h, cells were dissociated and centrifuged at 500 ϫ g for 5 min, and the supernatant was then discarded. Cell pellet was reconstituted in fresh cell culture medium. Equal numbers of cells (10 5 cells) in a final volume of 100 l of cell culture medium from each group were then taken in a 96-well plate in triplicates. MTT (0.1% final concentration) solution was added to each well of both control and experimental groups, mixed well, and incubated for 3 h at 37°C in a CO 2 incubator. After the incubation, 100 l of MTT solubilization solution containing 10% Triton X-100 plus 0.1 N HCl in anhydrous isopropyl alcohol was added to solubilize the insoluble formazan crystals at the bottom of the well. The MTT reduction (absorbance of formazan dye) was measured at 570 nm.
Assay of Caspase-9 and Caspase-3 Activities-Caspase-9 and caspase-3 activities were measured using a commercially available kit (Biovision and Sigma, respectively). In brief, stomach tissue from control, indomethacin-treated (48 mg⅐kg Ϫ1 b.w.), and SEGA (3a)-pretreated (5, 10, and 50 mg⅐kg Ϫ1 b.w.) indomethacin-treated rats were minced and homogenized in caspase lysis buffer (provided with the respective kits). The homogenate was centrifuged at 16,000 ϫ g for 15 min. For the caspase-9 assay, the supernatant was collected, containing an equal amount of protein for each sample, and mixed with 50 l of 2ϫ reaction buffer (provided with the kit). This was followed by the addition of the substrate, LEHD-p-nitroanilide (200 M final concentration) for caspase-9. In case of caspase-3, 5 l of the supernatant was taken together with 1ϫ reaction buffer and 10 l of substrate (provided with the kit), Ac-DEDV-p-nitroa-nilide (200 M final concentration). The mixture was incubated at 37°C for 1 h, and absorbance was taken at 405 nm (4,43).
RT-PCR for Proapoptotic and Antiapoptotic Genes-Equal amounts of stomach tissue (30 mg) from control, indomethacin-treated (48 mg⅐kg Ϫ1 b.w.), and SEGA (3a)-pretreated (50 mg⅐kg Ϫ1 b.w.) indomethacin-treated (48 mg⅐kg Ϫ1 b.w.) rats were used for total RNA isolation using a commercially available kit (RNeasy kit, Qiagen). RNA (2 g) was used to prepare cDNA using oligo(dT) 18 . Equal amounts of cDNA were used for PCR amplification using specific forward and reverse primers of bcl-2, bax, and actin. The PCR-amplified products were resolved in 2% agarose gel and documented in a gel documentation system (Alpha Infotech). The intensity of each band was quantified with densitometric software (Lab Image beta version, Kapelan GmbH, Germany). The intensity of each band was normalized with that of actin (6).
Terminal Deoxynucleotidyltransferase dUTP nick end labeling (TUNEL) Assay in Vitro in Cultured Gastric Mucosal Cells and in Vivo in Rat Gastric Mucosa-In vitro apoptosis was detected in the cultured gastric mucosal cells using a commercially available APO-BrdU TM TUNEL assay kit (Invitrogen). In brief, cultured cells were first divided into control, indomethacin-treated and SEGA (3a)-pretreated indomethacin-treated groups. After 16 h of incubation, the cells were washed with PBS twice and fixed with 1% paraformaldehyde in PBS (pH 7.4), followed by treatment with 70% ethanol in ice. The cells were then loaded with DNA labeling solution, containing terminal deoxynucleotidyltransferase. Cells were then stained with Alexa Fluor 488 dye-labeled anti-BrdU antibody. The cells were finally stained with propidium iodide (PI) solution containing RNase A. The cells were then visualized under a fluorescence microscope (Leica DM-2500) using appropriate filters for Alexa Fluor 488 and PI (4,6,43).
For the detection of apoptosis in vivo, the gastric mucosa from control, indomethacin-treated, and SEGA (3a)-pretreated indomethacin-treated rats were collected. Then these tissues were fixed in 10% buffered formalin for 12 h at 25°C and processed as described under "Immunohistological Studies." The semithin sections (5 m) were used for TUNEL staining using a commercially available kit (Promega).
Immunohistochemical Studies-The semithin sections (5 m) of mucosal tissues from control and experimental groups were deparaffinized in xylene and rehydrated in graded ethanol. After antigen retrieval, slides were rinsed in water and washed twice with Tris-buffered saline (TBS) (pH 7.4) plus Triton X-100 (0.025%) with gentle agitation. The sections were then blocked with 1% BSA in TBS for 2 h at 25°C. Primary anti-active caspase-3 was added at a dilution of 1:500 to the sections and kept at 4°C overnight. The next day, the sections were rinsed twice for 5 min in TBS plus Triton X-100 (0.025%) with gentle agitation. To block the endogenous peroxidases, the slides were incubated with 0.3% H 2 O 2 . Then the slides were incubated with HRP-labeled anti-rabbit IgG secondary anti-rat antibody at a dilution of 1:1000 in 1% BSA in TBS for 2 h. Finally, the slides were rinsed three times in TBS. Slides were stained with diaminobenzidine and counterstained with hematoxylin. The slides were viewed under the 10ϫ objective of a Leica DM-2500 microscope.
Radiolabeling of SEGA (3a)-SEGA (3a) was labeled with 99m Tc by a standard stannous reduction method (48) as per Reaction 1. Nitrogen-purged water was used for the preparation of aqueous 99m TcO 4 Ϫ solution and stannous chloride solution. Briefly, aqueous 99m TcO 4 Ϫ (2 mCi/ml) was mixed with 0.03 ml of freshly prepared stannous chloride solution (1 mg/ml) at pH 3.2 and further mixed with 1 ml of SEGA (3a) solution (3 mg/ml). The mixture was incubated separately for 20 min at room temperature (30°C).
The effect of stannous chloride on the labeling efficiency at different concentrations was studied to find the optimum concentration needed for maximum labeling. After adding SEGA (3a) to the mixture of 99m TcO 4 Ϫ and SnCl 2 adjusted at the optimum pH (pH 3.2), the solution was incubated for various time periods to see the effect of incubation time on the yield of labeling. The extent of labeling of SEGA (3a) was determined by ascending TLC using 2.5 ϫ 10-cm silica gel strips as the stationary phase and either acetone, methanol, or brine solution as the mobile phase. The test sample (incubation mixture) (2-3 l) was applied 1 cm from the base of the TLC plate and dried at room temperature. The plates were then developed in appropriate solvent systems. Acetone was used for the determination of free pertechnetate, whereas either methanol or brine solution was used for the determination of radiocolloid. After developing, the plates were dried, and the distribution of radioactivity was determined by cutting the portion of the strips and counting it in a ␥-scintillation counter (Electronic Corp. of India, model LV4755 (Hyderabad, India)) at 140 keV).
Transchelation with Diethylene Triamine Pentaacetic Acid (DTPA)-This study was performed to check the stability and strength of binding of 99m Tc with the SEGA (3a). Radiolabeled preparations of 0.5 ml were challenged against three different concentrations (10, 30, and 50 mM) of DTPA in 0.9% saline by incubating at 37°C for 2 h. The effect of DTPA on labeling was measured by TLC on a silica gel plate using normal saline and acetone as the mobile phase, which allowed the separation of free pertechnetate and DTPA chelate (R f ϭ 0.9) from that of the 99m Tc-labeled SEGA (3a), which remained at the point of application (R f ϭ 0).
Determination of Mitochondrial Uptake of SEGA (3a)-Mitochondrial uptake was carried out according to the following method. All animal experiments were carried out in compliance with the relevant national laws relating to the conduct of animal experimentation and with the approval of institutional animal ethics committees. Sprague-Dawley rats (180 -220 g) were used for this study. The animals were fasted for 24 h before the start of experiments as described above. After 24 h of fasting, all of the rats were well hydrated by intraperitoneal administration of saline (0.9%, 2 ml) for 1 h. After another 1 h, the 99m Tc-chelate SEGA (3a) in a total volume 0.03 ml (5)(6)(7)(8) was administered through an intraperitoneal route in each rat. After 30 min, indomethacin (48 mg⅐kg Ϫ1 b.w.) was adminis-trated orally in each rat. The rats were sacrificed at 4 h postinjection. Mitochondria of stomach mucosa were isolated as described above. Mitochondrial protein was estimated, and the radioactivity of 99m Tc-chelate SEGA (3a) was counted in a ␥-scintillation counter against suitably diluted aliquots of the injected solution as a standard. The data were expressed as percentage dose/mg of mitochondrial protein (Mean Ϯ S.E).
Stability Studies-The stability of 99m Tc-labeled SEGA (3a) was determined in vitro using 0.9% sodium chloride and serum (from rat) by ascending TLC. The labeled complex (0.5 ml) was mixed with 1.5 ml of normal saline or rat serum and incubated at 37°C. The samples were withdrawn at regular intervals up to 24 h, monitored by TLC, and analyzed in a ␥-counter.
Statistical Analysis-All data are presented as mean Ϯ S.E. The level of significance was determined by unpaired Student's t test with one-way analysis of variance as applied. A p value of Յ0.05 was considered as significant.

Synthesis of Tryptamine-Gallic Acid Hybrid Molecule-A
small molecule having the iron-chelating property and the capability of preventing MOS from entering into mitochondria is necessary to protect gastric mucosa against NSAID-induced gastric mucosal injury or gastropathy. To design such a small molecule, we began the synthesis using 5Ј-hydroxytryptamine (5HT), a hydrophobic amine that enters inside mitochondria (49). However, 5HT is toxic at high concentration (50). In contrast, free amine and hydroxyl groups of 5HT offer a scope to conjugate a powerful antioxidant bearing an iron-chelating property to make a non-toxic antioxidant hybrid molecule retaining mitochondrial penetration. Gallic acid (GA), a natural polyphenol and antioxidant (4), possesses the iron-chelating property (4,51), and that is why we selected it to make a hybrid molecule with 5HT. The strategy might give double benefits because the conjugation of GA with 5HT is expected to enhance the bioavailability of GA in body fluid (lack of bioavailability is a common problem of bioactive polyphenol), or 5HT may be detoxified by GA through toxic group protection. 5HT was conjugated with GA through an amide linkage to synthesize SEGA (3a) and through an ester linkage to synthesize GASE (4d) (Fig. 1) (supplemental Schemes S1 and S2). Both conjugates were tested first for their antioxidant property in vitro by following FRAP and DPPH free radical-scavenging activity. The FRAP assay is based on the measurement of the ability of a substance to reduce Fe(III) to Fe(II); the greater the reducing ability, the better the antioxidant property. Antioxidants reduce the colorless Fe(III)-TPTZ to a blue Fe(II)-TPTZ complex, which results in an increase in the absorbance at 595 nm, giving a FRAP value. A higher FRAP value indicates a greater reducing (i.e. antioxidant property) ability of the compound. FRAP values at 6 min were calculated from the equation described above. At 6 min, the absorbance change takes place abruptly due to reduction of Fe(III) into Fe(II). The results clearly indicate that SEGA (3a) shows a reducing ability (Fe(III) to Fe(II)) that is much better than GASE (4d) ( Table 1). In the DPPH assay, the decrease of absorbance is correlated with the antioxidant potency of the compounds. The greater the decrease in absorbance, the higher is the DPPH scavenging potency (i.e. the antioxidant potency of different synthesized compounds). The results indicate that SEGA (3a) also shows greater DPPH scavenging potency compared with GASE (4d) ( Table 1). These results indicate that when 5HT is conjugated with GA through amide linkage, it appears to be more effective than when conjugated by ester linkage.
Our next objective was to synthesize different types of tryptamine-antioxidant conjugates through amide linkage using other antioxidants replacing GA and to evaluate their activities for comparative efficacy. We replaced GA with 4-hydroxycinnamic acid and indole-3-acetic acid to synthesize other tryptamine-antioxidant conjugates, such as 2b and 2c, respectively (Fig. 1). These compounds were synthesized by successive condensation of 5HT with 4-hydroxycinnamic acid and indole-3-acetic acid, respectively through amide linkage

Tryptamine-GA Hybrid against NSAID-induced Gastropathy
(supplemental Scheme S1). Next, we searched to find out whether 5HT is the best possible tryptamine for our purpose. We replaced 5HT with other tryptamines to synthesize several other tryptamine-antioxidant conjugates, such as TRGA (3b), MEGA (3c), 2f, 2h, and 2i, by successive condensation of tryptamine, 5-methoxytryptamine with GA, 4-hydroxy cinnamic acid, and indole-3-acetic acid, respectively ( Fig. 1) (supplemental Scheme S3). 5HT itself showed little antioxidant activity in vitro (Table 1). We were interested in investigating whether in SEGA (3a), 5HT has any individual antioxidant property. To explore this, we synthesized dimer of 5HT (Fig. 1) (supplemental Scheme S4). Now antioxidant potencies were evaluated of all of the synthesized compounds by FRAP and DPPH free radical scavenging assays in vitro. For the preliminary screening of antioxidant activity, all of the synthesized compounds were taken at high concentration (100 M). From the above results, it is evident that SEGA (3a) shows antioxidant property in FRAP as well as DPPH assays (Table 1). We were interested in checking the antioxidant property of SEGA (3a) at different concentrations. Results indicate that SEGA (3a) shows excellent antioxidant property in the FRAP assay as well as in the DPPH-scavenging assay in a concentration-dependent fashion (data not shown). Now, we tested whether SEGA (3a) could chelate free iron in vitro. The iron-chelating property was assessed by a TPTZ assay. Fe(II) solution in the presence of TPTZ gave a broad peak at 595 nm (Fig. 2). This peak was obtained due to Fe(II)-TPTZ complex formation. When SEGA (3a) was added to the Fe(II) solution, no such broad peak was observed after the addition of TPTZ solution (Fig. 2). Thus, from this experiment, it is evident that SEGA (3a) chelates free iron in vitro. Now, SEGA (3a), because of its maximum antioxidant potential and iron-chelating property, was subjected to further detailed biological evaluation and mechanistic studies on NSAID-induced gastropathy.

SEGA (3a) Prevents Indomethacin (NSAID)-induced
Gastric Mucosal Damage-We tested whether SEGA (3a) could protect indomethacin (an NSAID)-induced MOS-mediated mitochondrial pathology and gastropathy in vivo. SEGA (3a) protected gastric mucosa from indomethacin-induced gastric injury in a dose-dependent manner (ED 50 ϭ 6.9 mg⅐kg Ϫ1 b.w.), as evident from the gastric injury index (Fig. 3A). For rapid visualization of the protective effect of SEGA (3a), we present the real morphological data obtained by opening the stomach interior. From the morphology, it is very clear that SEGA (3a) protected the injury, and the oozing out of blood (which appeared black due to oxidation of released hemoglobin under an acidic environment) in indomethacin exposed rat gastric mucosa (Fig. 3B). The gastroprotective effect of SEGA (3a) was also verified by following the changes in microscopic structure of the actual histology of the gastric mucosa. SEGA (3a) restored normal architecture of gastric mucosa from indomethacin-induced increased gastric mucosal cell death and cell shedding in the superficial mucosa (Fig. 3C). Excessive gastric mucosal injury by NSAID leads to the release of blood in the stomach. In an indomethacin-treated rat, a sharp Soret peak (417 nm) was observed, indicating the presence of hemoglobin in the stomach due to a mucosal injury. But in the case of SEGA (3a) pretreatment, we did not find any Soret peak. The data  . and free iron was induced in cultured gastric epithelial cells by indomethacin (Fig. 4, A and B 4A). Mitochondrial free iron was measured by Phen Green SK, a specific fluorescent probe used to assay chelatable iron (Fig. 4B). Mitochondria of gastric mucosal cells were tagged by Mitotracker Red, a specific fluorescent probe for mitochondria. From the experiment, it is evident that indomethacin treatment resulted in increased intramitochondrial free iron accumulation, but pretreatment with SEGA (3a) significantly inhibited indomethacin-mediated free iron accumulation as revealed by decreased fluorescence of Phen Green SK (Fig. 4B). Mitochondrial O 2 . and free iron are responsible for MOS (6). SEGA (3a), by scavenging O 2 . and free iron, protected mitochondria from MOS and restored the mitochondrial functions in gastric mucosal cells during indomethacin-induced gastropathy (Fig.  4C). SEGA (3a) significantly prevented indomethacin-induced mitochondrial lipid peroxidation, thiol depletion, and protein carbonyl formation (Fig. 4C), which are the markers for MOS. SEGA (3a) Corrects Indomethacin-induced Mitochondrial Dysfunction-Because SEGA (3a) scavenges intramitochondrial O 2 . , chelates intramitochondrial free iron, and prevents MOS, we were interested to find out whether SEGA (3a) could prevent mitochondrial pathology or dysfunction. The fall of the mitochondrial RCR, collapse of mitochondrial transmembrane potential (⌬⌿ m ), and loss of dehydrogenase activity are hallmarks or indicators for mitochondrial pathology or dysfunction. The functional integrity of mitochondria in the presence or absence of SEGA (3a) in gastric mucosal cells after indomethacin treatment was investigated. Mitochondria from indomethacin-treated gastric mucosa showed severe inhibition of complex I-mediated state 3 (in the presence of ADP) respiration and mild inhibition of state 4 (in the absence of ADP) respiration. As a consequence, the RCR (the ratio of state 3 and state 4 respiration) was significantly decreased, indicating impairment of mitochondrial respiration. Interestingly, administration of SEGA (3a) restored the altered RCR value (Table 2). SEGA (3a) prevented indomethacin-induced loss of ⌬⌿ m . Indomethacin showed an about 40% decrease of ⌬⌿ m , as measured by the ratio of fluorescence values measured at 590 nm (JC-1 aggregate) and 530 nm (JC-1 monomer). SEGA (3a) pretreatment restored the indomethacin-induced fall of ⌬⌿ m almost to the control level (Table 2). SEGA (3a) protected the mitochondrial dehydrogenase from indomethacin-induced inactivation. Indomethacin significantly inhibited mitochondrial dehydrogenases activity (as measured by MTT reduction) in gastric mucosal cells, indicating functional impairment of mitochondria, but pretreatment with SEGA (3a) under similar conditions restored the loss of mitochondrial dehydrogenase activity ( Table 2). SEGA (3a) Prevents Activation of Mitochondrial Pathway of Apoptosis-The activation of the mitochondrial pathway of apoptosis in gastric mucosal cells is a consequence of indomethacin-induced MOS. Because SEGA (3a) prevents MOS as well as mitochondrial dysfunction, SEGA (3a) should protect the mitochondrial pathway of apoptosis in gastric mucosal cells by indomethacin. The data indicate that pretreatment with SEGA (3a) significantly attenuated indomethacin-induced activation of casapse-9 (marker of mitochondrial pathway of apoptosis) (Fig. 5A) as well as caspase-3 (general marker for apoptosis) (Fig. 5B) in gastric mucosa in a dose-dependent manner. Indomethacin stimulated about 2-fold activation of caspase-9 in the gastric mucosal cells (Fig. 5A). Pretreatment with SEGA (3a) significantly attenuated the activation of casapse-9 and brought the activity close to the normal level (Fig.  5A). Indomethacin activated caspase-3 by more than 2-fold in the gastric mucosal cells (Fig. 5B). Pretreatment with SEGA (3a) significantly blocked the activation of caspase-3 (Fig. 5B). bcl-2 and bax play a critical role in the mitochondrial pathway of apoptosis. Indomethacin was found to down-regulate the expression of antiapoptotic bcl-2 and up-regulate the expression of bax compared with control ( Fig. 5, C-E). SEGA (3a) pretreatment was found to block indomethacin-induced upregulation of bcl-2 as well as down-regulation of bax (Fig. 5, C-E). The antiapoptotic effect of SEGA (3a) was further tested by following DNA fragmentation performing a TUNEL assay in gastric mucosal cells in vitro (Fig. 6A) and in vivo in the presence of indomethacin (Fig. 6B). In control cells, the absence of green signal (DNA fragmentation was indicated by green fluorescence of Alexa Fluor 488) indicated no apoptotic DNA fragmentation. However, in indomethacin-treated gastric mucosal cells, green fluorescence was prominent, which was co-localized with PI-stained nuclei, indicating apoptotic DNA fragmentation. In SEGA (3a)-pretreated cells, the intensity as well as the total number of cells showing green fluorescence were much less compared with indomethacin-treated cells (Fig. 6A). SEGA (3a) pretreatment also significantly prevented indomethacin-induced gastric mucosal apoptosis in vivo in mucosal tissue (Fig. 6B). The TUNEL-positive cells (dark brown staining, indicated by an arrow) were abundant in the gastric mucosal tissue in the presence of indomethacin, whereas the TUNEL-positive cells were decreased significantly in the SEGA (3a)-pretreated indomethacin-treated group (Fig. 6B). The antiapoptotic effect of SEGA (3a) was further confirmed by immunohistochemistry using anti-active caspase-3 antibody (Fig. 6C). Active caspase-3 immunolabeled mucosal cells (dark brown staining, indicated by an arrow) were found after indomethacin treatment. But SEGA (3a) pretreatment significantly decreased the active caspase-3-immunolabeled cells, indicating the antiapoptotic role of SEGA (3a) (Fig. 6C). Thus, the data indicate that SEGA (3a) prevents indomethacin-induced gastric mucosal cell apoptosis. SEGA (3a) Accelerates Healing of Indomethacin-induced Damage of Gastric Mucosa-Prevention of MOS and the mitochondrial pathway of apoptosis expedites the healing process (5). Because SEGA (3a) prevents both MOS and apoptosis, we were interested in discovering whether SEGA (3a) could accelerate the healing of indomethacin-induced already damaged gastric mucosa. Interestingly, in addition to the gastroprotective effect, SEGA (3a) also accelerated healing of already injured mucosa by indomethacin (Fig. 7). Although autohealing takes place in the case of damaged mucosa, SEGA (3a) treatment accelerates the healing process. SEGA (3a)-induced healing of gastric mucosal injury was checked by histological analysis (Fig.  7). At 4, 8, and 20 h, the mucosa shows gastric injury with an injury index of 52, 28, and 20, respectively, whereas after treatment with SEGA (3a), damage of gastric mucosa was gradually repaired, as evident from an injury index of 14, 8, and 0, respectively. At 20 h, SEGA (3a) completely restored normal architecture of gastric mucosa, whereas in the case of the indomethacin group, there was significant injury. The results indicate that mucosa shows a time-dependent autohealing of the indomethacin-induced gastric damage in the absence of SEGA (3a). However, SEGA (3a) treatment significantly expedites healing with the progress of time, as evident from the restoration of gastric mucosa (Fig. 7). In indomethacin-treated animals, the autohealing at 4 h was negligible, as evident from the distorted mucosal histology, but SEGA (3a) treatment restored healthy mucosal architecture at 4 h, with almost complete restoration at 20 h (Fig. 7).
Quantitation of SEGA (3a) Entering Mitochondria-Because SEGA (3a) scavenged intramitochondrial O 2 . , chelated intramitochondrial iron, and prevented MOS, we were interested in quantitating how much of the administered SEGA (3a) entered the mitochondria under in vivo conditions. For this purpose, SEGA (3a) was radiolabeled with 99m Tc isotope as reported (48) and administered to rats. The data indicated that 0.05% of the administered dose of SEGA (3a) entered per mg of mitochondria of gastric mucosal tissue (Fig. 8A). The stability or the structural integrity of SEGA (3a) in physiological saline as well as in serum was checked. SEGA (3a) was found to be very stable at 37°C (Fig. 8B).

DISCUSSION
The present study describes the designing and synthesis of a small molecule, tryptamine-gallic acid hybrid (SEGA (3a)), which prevents NSAID-induced mitochondrial pathology, apoptosis, and gastropathy by blocking MOS through scavenging of intramitochondrial O 2 . and free iron and correcting mitochondrial dysfunction. The mitochondria are a potential subcellular therapeutic drug target against NSAID-induced gastropathy because they produce O 2 . and free iron, which play an important role in triggering this pathological condition. A mitochondriatargeted molecule is required for this purpose. This molecule must be small and lipophilic and be an ROS scavenger in nature. Moreover, free iron is known to generate ROS through the Fenton reaction. Thus, the iron-chelating property would be an additional advantage in controlling oxidative stress. All of these criteria were considered while designing the molecule. Several antioxidants and iron chelators have been reported, but none of them can satisfy all of the above criteria. Thus, a new molecule is essential, which will satisfy all of these criteria in preventing NSAID-induced gastropathy. Keeping this in mind, we have synthesized a series of tryptamine-antioxidant hybrid molecules. GA, when conjugated with 5HT through amide linkage, shows greater activity both in vitro and in vivo. Thus, all other tryptamineantioxidant hybrid molecules were generated through the amide linkage. For the structure-activity relationship studies, we synthesized different tryptamine-antioxidant derivatives. Because SEGA (3a) appears to be the most active among all of the tryptamine-antioxidant conjugates, it is suggested that the presence of the 5-hydroxy group in the indole moiety of SEGA (3a) plays an important role for its gastroprotective activity. When the 5-hydroxy group in the indole moiety of SEGA (3a) was replaced by hydrogen and the methoxy group in the molecules TRGA (3b) and MEGA (3c), the activity was decreased. Indomethacin was selected as the representative NSAID over others because it is the most frequently used NSAID in gastrointestinal toxicity studies in experimental animals (13). The dose of indomethacin was selected as 5 mM and 48 mg⅐kg Ϫ1 b.w. for in vitro and in vivo studies, as reported earlier (4,6,7). The role of MOS and consequent apoptosis behind NSAID-induced gastric mucosal injury is already well established and is considered to be the major player in the acid-independent (5) and COX-independent pathway of NSAID-mediated gastric injury (53,54). Indomethacin with its acidic carboxyl group (pK a ϭ 4.5) and lipid solubility has been found to damage both rat and human mitochondria (13). Moreover, indomethacin enhances mitochondrial ROS, which disrupts mitochondrial function (6). , which leads to the generation of ROS. Increased ROS develops MOS by oxidizing protein and lipid, including cardiolipin and protein thiol. Iron (Fe 2ϩ ) is released from the Fe-S cluster of aconitase due to ROS-mediated damage and further aggravates oxidative stress by producing hydroxyl radical ( ⅐ OH). The MOS results in mitochondrial dysfunction or pathology and activation of the mitochondrial pathway of apoptosis, which plays a pathogenic role in gastropathy. The tryptamine-gallic acid hybrid molecule, SEGA (3a), enters the mitochondria and prevents NSAID-induced gastropathy.
We propose the whole gastroprotective mechanism of SEGA (3a) through a schematic representation (Fig. 9). Indomethacin interacts with complex I of the electron transport chain and results in the leakage of electrons, which in turn leads to the generation of ROS. SEGA (3a) enters into mitochondria and scavenges generated ROS and prevents MOS by attenuating mitochondrial protein oxidation, lipid peroxidation, and depletion of thiol. Iron is released from the Fe-S cluster of aconitase due to ROS-mediated damage and further aggravates oxidative stress by producing hydroxyl radical. SEGA (3a) also scavenges the released intramitochondrial iron. SEGA (3a) prevents the mitochondrial pathway of apoptosis by preventing indomethacin-induced activation of caspase-9 and caspase-3 and down-regulation of bcl-2 (antiapoptotic gene) and up-regulation of bax (proapoptotic gene). In conclusion, SEGA (3a) is a novel small molecule that protects gastric mucosa against NSAID-induced MOS-mediated gastric injury.