Design, synthesis and biological evaluation of indoline derivatives as multifunctional agents for the treatment of ischemic stroke

In this work, a series of indoline derivatives as multifunctional neuroprotective agents for battling ischemic stroke were designed, synthesized, and biologically evaluated. In antioxidant assay, all compounds showed significant protective effects against H2O2-induced death of RAW 264.7 cells. In oxygen glucose deprivation/reperfusion (OGD/R)-induced neuronal damage, some compounds significantly elevated the cell survival rate. Among them, 7i, 7j and 7r exerted comparable neuroprotective effects to ifenprodil, and exhibited binding affinity to N-methyl-D-aspartic acid receptors 2B (NMDA-GluN2B). At the concentrations of 0.1, 1 and 10 μM, 7i, 7j and 7r dose-dependently lowered the LPS-induced secretion of inflammatory cytokines, including TNF-α, IL-6 and NO, by BV-2 cells. Importantly, 7i and 7j can dramatically reduce the cerebral infarction rate and improve neurological deficit scores in middle cerebral artery occlusion (MCAO) rat model. As demonstrated by the above results, 7i and 7j are potential neuroprotective agents for the treatment of ischemic stroke. Graphical abstract Graphical abstract


Introduction
Ischemic stroke (IS) refers to focal cerebral tissue necrosis or brain injury caused by the transient or permanent decrease of cerebral aortic blood flow, causing brain tissue ischemia and hypoxia. Cerebral ischemia triggers the pathological pathway of the ischemic cascade and eventually causes irreversible neuronal damage to the ischemic core within a few minutes after onset [1]. Ischemicstroke, has emerged as the major cause of human death and long-term disability worldwide [2,3]. In recent years, with the increase of social pressure and the aging of the population, the number of stroke diseases continues to increase and there is a trend of younger generation [4].
At present, treatment of cerebral ischemic stroke mainly focuses on recovering cerebral circulation and protecting brain tissue, emphasizing the necessity of a combination of both thrombolytic drugs and neuroprotective drugs [5]. Although thrombolysis and thrombectomy are the main treatment strategies in the early stage of clinical ischemic stroke, due to the narrow treatment time window, the strategies are biased [6]. Other than vascular recanalization, the application of neuroprotective agents represents the most promising strategy for treating this disease [7]. It was found that the time window of thrombolytic therapy can be significantly prolonged by post-ischemic neuroprotective treatments. Therefore, the etiology and pathology of ischemic stroke should be simultaneously considered in order to find suitable therapeutic drugs and explore potential therapeutic molecular targets for neuroprotective sdrug development [8].
Mounting studies have revealed that ischemic stroke is initiated and promoted by a complex cascade of pathological events involving the release of numerous cytokines, and the activation of multiple signal transductions [9]. The main pathogenesis is associated with the intermediate blockage of the cerebral blood vessels, culminating in the severe lack of oxygen and energy, excessive release of excitatory amino acids, and inflow of a large amount of calcium ions [10]. Due to the hypoxia, a large number of active oxygen species are produced, thereby inducing cell damage, facilitating the peripheral leukocyte infiltration, activating endogenous microglia, and consequently mediating inflammation [11,12]. Furthermore, the cross-talk and interaction among these pathological processes initiates the neuronal apoptosis [13]. In light of the complex pathogenesis, the application of therapeutic agents with multi-faceted biological activ-ities, including anti-inflammation, antioxidant, and glutamate receptor antagonism, is beneficial to improving the efficacy for treating ischemic stroke.
Recently, a class of indoline derivatives has been reported to display more significant antioxidant activity than corresponding indole derivatives (Fig. 1). In vitro, they also exerted protective efficacy against reactive oxygen species (ROS)-inducedcytotoxicity. In addition, these compounds were capable to reduce the release of TNF-α and IL-6 from LPS-stimulated macrophages [14].
During cerebral ischemia, excessive stimulation of NMDA-GluN2B receptors caused by glutamate has long been thought to promote neuronal death [15][16][17]. Therefore, NMDA-GluN2B receptor antagonists have the potential to treat ischemic stroke. Classical selective GluN2B receptor inhibitors, including ifenprodil (Fig. 2), eliprodil and traxoprodil, can effectively inhibit glutamate-induced hippocampal neurotoxicity. However, the clinical investigation of traxoprodil has been terminated due to cardiovascular toxicity [18][19][20]. These classic NMDA-GluN2B receptor antagonists are structurally characterized by of the phenylethanolamine scaffold and a hydrophobic group tethered to the 4-position of the piperidine [21]. As confirmed by the antagonist/ receptor complex, the phenylethanolamine was vital for the binding affinity [22].
Considering the role of ROS, inflammation, and NMDA-GluN2B receptor in the pathology of ischemic stroke, indoline derivative DX, that exerts the anti-oxidative and anti-inflammatory activities, was structurally merged with ifenprodil for achieving a multi-target intervention. Upon this strategy, a novel series of indoline derivatives, incorporating the structural features of both DX and ifenprodil, were designed, synthesized, and biologically evaluated (Fig.  3). Among them, 7i and 7j were screened as potential multifunctional neuroprotective agents for the treatment of ischemic stroke.

Chemistry
The synthetic route for target compounds is displayed in Scheme 1. The commercially available substituted indoles were reduced by sodium cyanoborohydride to afford corresponding indolines [23]. Meanwhile, Grignard reaction between aromatic aldehydes 3a-r and vinylmagnesium bromide furnished the secondary alcohol, which was coped with chromium trioxide to generate the α, β-unsaturated ketones 5a-r [24,25]. The following Michael addition with the indolines and the reduction of the carbonyl functionality led to the formation of 7a-r as the target compounds (Table  1) [26].

Antioxidant assays
All the target compounds were evaluated for their protective effects against RAW 264.7 cell death induced by H 2 O 2 in vitro. As demonstrated by the biological data in Table 2, all compounds exhibited significant cytoprotective activity against H 2 O 2 induced oxidative damage compared with the model group. Among them, the antioxidant activity of compounds 7a, 7d, 7f, 7l, 7q, and 7i is better than the lead compound ifenprodil. Unfortunately, the antioxidant activity of these synthesized compounds is poorer than the lead compound DX at a concentration of 1~100 nM. Exploring the antioxidation results of the compound in vitro, We found that when R 2 is an electron donating group and the substituents of R 2 are the same, the antioxidant activity of R 1 containing H is better than that of R 1 containing OCH 3 . Among them, the antioxidant activity of compounds containing 4-CH 3 or 4-OCH 3 in R 2 decreases more significantly. Furthermore, when the phenyl group contains two electron-donating substituents, it is beneficial to increase the antioxidant act vity. However, when R 2 is an electron withdrawing group and R 2 is the same, the change of R 1 substituent has no obvious effect on the activity. Through the analysis of compounds 7d, 7h and 7p, it was found that when R 2 is Cl, the antioxidant activity of para position is better than that of meta and ortho position, but when R 1 is OCH 3 , its antioxidant activity will be reduced. In addition, the antioxidant activity of chlorine substitution is better than that of bromine substitution. Analyzing the structure-activity relationship of 7h, 7l, 6i and 6h, it is found that the activity of carbonyl group is better than that of hydroxyl group when R 1 is H. Interestingly, the activity of hydroxyl group is better than that of carbonyl group when Ar is thiophene. Overall, there is no order for magnitude difference in the antioxidant activity of these compounds compared with the lead compounds DX and ifenprodil.

Neuroprotection against injury by OGD/R
The neuronal damage induced by OGD/R is a frequently applied model for the in vitro evaluation of the protective effects against cerebral ischemia-reperfusion (I/R) injury. As shown in Fig. 4, in comparison with the control groups, a sharp decrease in the cell survival rate was observed as for the model groups treated by OGD/R. On the contrary, the survival rate of neuronal cells is in a dose-dependent manner upon the pretreatment with 7a, 7i, 7j, 7k, 7m, 7q, 7r and ifenprodil. Among the tested target compounds, 7i, 7j and 7r exhibited better protective effects compared with other compounds. Hence, they were selected for further functional investigation.
Competitive binding of 7i, 7j and 7r to GluN2B receptor The ifenprodil binding site, named after the negative allosteric modulator ifenprodil, is found exclusively on the GluN2B subunit. Therefore, compounds interacting with the ifenprodil binding site can only modulate the  [27][28][29]. However, the over activation of GluN2B receptor, is involved in the occurrence and development of ischemic stroke [30,31]. On the basis of the protective effect of OGD/R injury and in order to investigate the mechanism of action we evaluated the ability of someselected compounds to interact with the NMDA/NR2B receptorby testing [ 3 H]ifenprodil binding inhibition (Table 3) in Wistar rat cerebral cortex following a reported procedure and using ifenprodil as reference compounds [29]. Data presented in the Table 3 show that a 80 nM concentration of three of the  (100 μM), and then the cell viability was detected by MTT method. Survival rate(%) = (OD value of experimental group/OD value of control group) × 100%. All date were shown as the mean ± SD (n = 5 per group) of triplicate in three independent experiments, *P < 0.05, **P < 0.01 vs. H 2 O 2 group; # P < 0.05, ## P < 0.01 vs. control group Fig. 4 Neuroprotection of some compounds on injury by OGD/ R. Effects of some compounds on neuronal cultures exposed to OGD. Survival rate(%) = (OD value of experimental group/OD value of control group) × 100%. All data are expressed in mean ± SD. *p < 0.05, **p < 0.01 vs the OGD/R group, repeat five times for each concentration.
Potency to inhibit the secretion of TNF-α, IL-6 and NO by LPS induced BV-2 cells Considering the inflammatory response is a major pathological event in the occurrence and development of ischemic stroke, we then investigated the effect of 7i, 7j and 7r on the release of inflammatory cytokines. As shown in Fig. 5, when BV-2 cells were pretreated with the tested compound, the release content of TNF-α, IL-6 and NO was lowered in a dose-dependent manner. Among them, compound 7i showed the same inhibitory activity as the lead compound DX.
Efficacy of compounds 7i and 7j on a middle cerebral artery occlusion (MCAO) rat model To assess the in vivo therapeutic efficacy of compound 7i and 7j, a transient focal MCAO rat model was employed to simulate the temporary regional ischemia. In this experiment, infarct volume, neurological deficiency and  water maze experiment were used to systematically investigate the process of the MCAO in the rat model. As shown in Fig. 6, a significant neurological deficit score and extensive infarct volume were all presented by 90 min MCAO-induced ischemia and 24 h I/R as compared with the sham operated group. Pretreatment with Nbutylphthalide (NBP) (20 mg/kg), 7i (20 mg/kg), 7j (20 mg/kg), ifenprodil (20 mg/kg) and DX (4 mg/kg and 20 mg/kg) markedly ameliorated this increment symptom and improved the neurological score as compared with the model group (p < 0.01). Furthermore, NBP and 7i, 7j, ifenprodi, DX at each dosage can significantly reduce the cerebral infarction rate after MCAO (p < 0.01, p < 0.05), and the efficacy of 7i is slightly better than the 7j at a dose of 20 mg/kg, and is better than the ifenprodi at a dose of 20 mg/kg.
Morris water maze experiment of compound 7i, 7j and DX Furthermore, the Morris watered maze experiment, reflecting the learning and memory ability of experimental animals, was performed to evaluate the neuroprotective effect of the compound. As illustrated in Table 4, the I/R (ischemia-reperfusion) group had significant changes in the latency period and the times they crossed the platform (P < 0.01) compared with the sham group. Upon the treatment for three days, the NBP group, compound 7i (20 mg/kg), compound 7j and DX (20 mg/kg) can significantly improve the latency crossing the water maze and the times of crossing platforms (P < 0.01, P < 0.05) at the dosage of 20 mg/kg compared with the I/R group. There was no significant difference in NBP, 7i, 7j and DX-treated groups at the same dosage. Fig. 6 The neuroprotective effect of compound 7i, 7j, ifenprodi and DX were evaluated 24 h after cerebral ischemia and reperfusion in rats. a Representatives of TTC-stained brain slices. The white region shows the infarct size, while the red region shows the viable tissue. b Cerebral infarction rate(%). c Neurological deficit scores. All data are expressed as mean ± SD (n = 8 per group). # P < 0.05, ## P < 0.01 vs the sham group; △ P < 0.05, △△ P < 0.01 vs the I/R group; ▲ P < 0.05 , ▲▲ P < 0.01 vs positive group, respectively

Conclusions
In view of the complex pathology of ischemic stroke, a series of indoline derivatives with novel structures have been designed and synthesized. As a consequence of the successive biological screening, 7i and 7j were identified as potential neuroprotective agents for the treatment of ischemic stroke. In vitro, they displayed attractive protective activity against H 2 O 2 -induced cell death in RAW 264.7 cells, as well as comparable neuroprotective effects to ifenprodil against OGD/R-induced neuronal damage. Moreover, both compounds exerted binding affinity to NMDA-GluN2B, and were capable of lowering the LPSinduced secretion of inflammatory cytokines in a dosedependent manner. In vivo, 7i and 7j can significantly reduce the cerebral infarction rate and improve neurological deficit scores in MCAO rat model. In summary, compound 7i and 7j are potential neuroprotective agents for the treatment of ischemic stroke.

Chemistry
The reagents and solvents for reaction were purchased from common commercial suppliers or purified by standard techniques. All chemical reactions were tested by TLC (using thin layer board with 254 nm fluorescent indicator). 1  General method for preparation of compounds 7a-7r Compound 1 (13.3 mmol) was dissolved in AcOH (20 mL), NaBH 3 CN (15.9 mmol) was added in portions, and the reaction mixture was stirred at 25°C for 1 h. The reaction mixture was alkalized with 20% NaOH to pH = 9-10. The solution was poured into the 100 mL water and extracted with EtOAc (50 mL × 3). The organic layers were pooled and washed with saturated NaCl and dried over Na 2 SO 4 . After concentration, the crude product was purified using column chromatography (EtOAc: PE = 10:1) to afford Compound 2, yield 91-95%. A solution of vinylmagnesium bromide (32.0 mmol) in THF (32 mL) was added dropwise into a solution of 3a-3r (16 mmol) in THF (5 mL) at 0°C under N 2 (with stirring). After ending to drop, the reaction mixture was stirred at room temperature for 3 h. The reaction mixture was poured slowly into the solution of saturated NH 4 Cl (250 mL) and extracted with EtOAc (60 mL × 3). The organic layers were pooled and washed with saturated NaCl and dried over Na 2 SO 4 . The organic layer was filtered and concentrated under reduced pressure to give target compounds 4a-4r, yield 20-97%.
Jones reagent (15 mL) was added drop-wise into a solution of 4a~4r (15 mmol) in acetone (20 mL) at 0°C. Upon completing the addition, the reaction was stirred at ice bath for 2 h. The reaction mixture was poured into isopropanol (20 mL) and suction. Then the filtrate was diluted with wate (150 mL) and extracted with EtOAc (50 mL × 3). The organic phase was separated and concentrated under reduced pressure to afford oil. The crude product was purified by column chromatography (PE:EtOAc = 80:1) to give 5a-5r, yield 50-70%.
Compound 2 (4.7 mmol) was added dropwise to a solution of 5a-5r (4.8 mmol) in CH 3 CN (10 mL) and was stirred for 1 h, keeping the reaction between 20-25°C. After the reaction solution was concentrated under reduced pressure, the residue was purified by column chromatography (PE:EtOAc = 80:1) to give 6a~6r, yield 80-90%. NaBH 4 (4.2 mmol) was added into a solution of 6a-6r (3.5 mmol) in CH 3 OH (15 mL) in portions at 0°C and then the reaction mixture was stirred at room temperature for All data are expressed as mean ± SD (n = 8 per group). ## P < 0.01 vs the sham group; △ P < 0.05, △△ P < 0.01 vs the I/R group another 1 h. The reaction solution pour into water (120 mL) and extracted with EtOAc (50 mL × 3). The combined organic layers were washed with water and dried over Na 2 SO 4 . After the organic was filtrated, the filtrate was concentrated to afford 7a-7r, yield 90-95%.

Biological activity
Protective activity against oxidative stress The RAW264.7 cells in logarithmic growth phase were seeded at a density of 1 × 10 4 cells/well in a 96-well plate, and cultured at 37°C with 5% CO 2 , until the cells were 90% confluent, then cultured in serum-free base incubation for 2 h to synchronize the cells. Subsequently, the supernatant was discarded, and the EMEM medium containing each compound (0.1 nM, 10 nM, 100 nM) was added and incubated for 2 h, and then the medium was replaced with H 2 O 2 (100 μM) and the concentration was (0, 1 nM, 10 nM, 100 nM) drug medium continued to incubate with the cells for 24 h, 4 h before the end of the incubation, add 20 μL MTT solution (5 mg/mL) to each well. After the incubation, discard the supernatant of each well, add 150 μL DMSO to each well, place on a cell shaker for 10 min, and measure the OD 570 with a microplate reader after the crystals are fully dissolved [32].
Protective Activity against OGD/R-induced damage in rat cortical neuron cells The cells that grew to about seven days were selected for in vitro OGD/R model. Adjust the cell density and seed it in a 12-well culture plate with a cell density of 4×10 4 . Using DMEM medium containing 1% FBS and the corresponding concentration of medicine, the OGD/R model was prepared after 24 h of routine culture. First change the medium to DMEM medium without glucose and serum, and at the same time place the cells in a tri-gas incubator containing 5% CO 2 and 95% N 2 for 2 h to complete the hypoxia process; then change the cell medium to complete DMEM medium, and place the cells in an incubator containing 5% CO 2 for 24 h to complete the reoxygenation process. 4 h before the end of the incubation, add 20 μL of MTT solution (5 mg/mL) to each well. After the incubation, discard the supernatant of each well, add 150 μL DMSO to each well, shake on a cell shaker for 10 minutes, and measure OD 570 with a microplate reader after the crystals are fully dissolved [33].

Anti-inflammatory activity in vitro
BV2 cells in the logarithmic growth phase were seeded in a 96-well plate at 5 × 10 3 cells/well, cultured at 37°C and 5% CO 2 until the cells were 90% confluent, and then incubated with serum-free DMEM medium 2 h to synchronize the cells. Subsequently, the supernatant was discarded and the BV2 cells were incubated for 2 h with compounds at concentrations (0.1 μM, 1 μM, 10 μM), and then the medium was replaced with a medium containing LPS (200 ng/ml) and incubated for another 8 h. The cells treated with no LPS and drugs and no drugs only added to LPS were used as the blank control group and the LPS control group. After the incubation, collect the cell supernatant and freeze it at −80°C for later use. The levels of TNF-α, IL-6 and NO in the cell supernatant were determined according to the ELISA kit instructions. Each test is repeated three times [35].

Animal model of focal cerebral ischemia
These male rats (250-300 g in weight) were subjected to a transient MCAO model using an intraluminal thread.
Specifically, rats were anesthetized with 10% chloral hydrate (0.2 ml/10 g) intraperitoneally (i.p.). The left common carotid artery (CCA), external carotid artery (ECA), and internal carotid artery (ICA) were exposed and carefully isolated from the vagus nerve. The pterygopalatine branch of the ICA was sealed with an electric coagulator. The distal end of the ECA was ligated with a 4-0 silk suture. A 25-mm long nylon filament (0.26 mm), with its tip dipped in heparin before use, was inserted into the ECA lumen through a small puncture, and gently advanced into the lumen to a distance of 20 mm until the tip of the filament blocked the origin of the middle cerebral artery (MCA). After 90 min of ischemia, the nylon filament was withdrawn from the internal carotid artery lumen to establish reperfusion. The same procedure was performed on sham-operated rats, but the middle cerebral artery was not blocked. The neurological deficits were scored after 10 min. These rats with obvious neurological deficits were considered successful. The body temperature of the rats was maintained at 37 ± 0.5°C during the operation [36,37].

Morris water maze test
Positioning navigation experiment: Rats were put into water from four different quadrants at nine o'clock every morning and were trained four times. Recording the time that the rat seeks and climbs on the columnar platform, that is the latency of rats escape. This experimental process was continued for three days and the detection time is set to 120 s [36,37].
Measurement of neurological deficit score The neurological deficit of each rat, according to Zea-Longa method using a 5-point scales (no neuro-logical deficit = 0; failure to extend right paw fully = 1; circling to right = 2; falling to right = 3; being unable to walk spontaneously and depression of consciousness = 4) was evaluated by the same experimenter, who was blinded to the different treatments in the experiment, 24 h after reperfusion.
Infarct rate evaluation The size of infarction was evaluated at 24 h after MCAO using 2,3,5-tetrazolium chloride (TTC) staining. Mice were sacrificed with 10% chloral hydrate. Brains were removed and cut into 2 mm-thick coronal sections and stained with 2% of TTC for 30 min at 37°C. Then, we observed and taken pictures after the brains were fixed with 4% formaldehyde solution for 30 min. The normal brain tissue is bright red and the infarcted brain tissue is pale. The infarct areas were measured using the Image-J software. The percentage of cerebral infarction area is calculated by the following formula: infarct rate% = ischemic area/total area × 100%.
Experiment grouping The rats were randomly divided into nine groups with eight rats in each group, namely the sham operation group, the model group, low-dose group (4 mg/kg) and high-dose group (20 mg/kg) of the compound 7i, low-dose group (4 mg/kg) and high-dose group (20 mg/kg) of the compound 7j, low-dose group (4 mg/kg) and high-dose group (20 mg/kg) of lead compound and positive control butylphthalide dose group (20 mg/kg).
Drugs supply and dosage Compounds 7i, 7j and DX were synthesized as previously described. 3-N-Butylphthalide, a positive drug, was purchased from Sigma-Aldrich. In order to improve the water solubility of the compounds, we prepared the tested compounds in the form of the hydrochloride salt. After that, we formulated the compound into a solution of 2 mg/mL and 10 mg/mL with physiological saline. Finally, the administration was injected through the tail vein of the rat.
Dosing regimen Three days before the model was made, the rats were given the tail vein once a day. The sham operation group and the model group were given the same amount of normal saline. Two hours after the third administration, the model of cerebral ischemia was started to make. The nylon silk thread was drawn out carefully from the lumen of the internal carotid artery and ligate the broken end to restore the blood flow of the internal carotid artery in the rat when ischemic for about 90 min, so that the blood flow of the internal carotid artery of the rat can be reperfused. 24 h after the third administration, the compound group and the positive control group were administered once again through the tail vein. The sham operation group and the model group were injected with the same amount of solvent, the administration volume was 2 mL/kg body weight, for a total of 4 administrations.
Statistical analysis The SPSS 22.0 statistical software was used to statistically process the data, and the results are expressed as mean ± SD. Analysis of variance was performed between groups, and Tukey HSD test was used for pairwise comparison. P < 0.05 is statistically significant.