Adamantaniline Derivatives Target ATP5B to Inhibit Translation of Hypoxia Inducible Factor‐1α

Abstract Hypoxia inducible factor‐1α (HIF‐1α) plays a critical role in cellular adaptation to hypoxia and it is a potential therapeutic target for anti‐cancer drugs. Applying high‐throughput screening, here it is found that HI‐101, a small molecule containing an adamantaniline moiety, effectively reduces HIF‐1α protein expression. With the compound as a hit, a probe (HI‐102) is developed for target identification by affinity‐based protein profiling. The catalytic β subunit of mitochondrial FOF1‐ATP synthase, ATP5B, is identified as the binding protein of HI‐derivatives. Mechanistically, HI‐101 promotes the binding of HIF‐1α mRNA to ATP5B, thus inhibiting HIF‐1α translation and the following transcriptional activity. Further modifications of HI‐101 lead to HI‐104, a compound with good pharmacokinetic properties, exhibiting antitumor activity in MHCC97‐L mice xenograft model, and HI‐105, the most potent compound with an IC50 of 26 nm. The findings provide a new strategy for further developing HIF‐1α inhibitors by translational inhibition through ATP5B.


Introduction
Cancer is the second deadliest disease in the world, accounting for nearly 10 million deaths in 2020. [1] A striking characteristic feature of tumor cells is that they can escape senescence and proliferate indefinitely, which results in the formation of a hypoxic microenvironment. [2][3][4] Cells usually undergo a series of changes to adapt to the hypoxia microenvironment, such as the upregulation of vascular endothelial growth factor (VEGF), pyruvate dehydrogenase kinase isoenzyme 1 (PDK1) and glucose transporter 1 DOI: 10.1002/advs.202301071 (GLUT1). [5,6] These detrimental events might promote tumor angiogenesis, growth, and metastatic, enhancing resistance to the anti-tumor immune response at the same time in many solid malignancies. [7][8][9] A growing number of studies show that targeting the hypoxic microenvironment of malignant tumors is a promising therapeutic strategy to inhibit tumor growth. [10][11][12] Maintaining oxygen homeostasis is extremely important for tumor cell growth. Hypoxia inducible factor-1 (HIF-1), consisting of the hypoxic response factor-HIF-1 and the constitutively expressed aryl hydrocarbon receptor nuclear translocator-HIF-1 , is a transcription factor that is highly responsive to the hypoxic environment and is widely distributed throughout the human body. [13] Hydroxylated HIF-1 protein at P402 and P564 by prolyl-4-hydroxylases (PHDs) is recognized and ubiquitinated by the von Hippel−Lindau (VHL) E3-ubiquitin ligase complex and then degraded by the 26S proteasome subsequently under normoxia condition. However, inactivated PHDs fail to hydroxylate HIF-1 under hypoxia condition, which results in the stabilization and accumulation of HIF-1 . Stabilized HIF-1 can translocate to nuclear and dimerize with HIF-1 , which induces the transcription of a series of downstream target genes, including VEGF and PDK1, to promote tumor growth and metastasis. [14][15][16][17] Recent studies suggest that HIF-1 protein is highly overexpressed in various types of tumors, especially in solid tumors and the dramatic overexpression of HIF-1 is closely associated with poor prognosis. [18] A large number of clinical studies have shown correlations between HIF-1 and the metastasis, recurrence, vascular proliferation, and prognosis in cancer patients. [19] A previous report showed the expression level of HIF-1 in liver cancer tissues is higher than in corresponding adjacent tissues and the overexpression of HIF-1 is also associated with poor prognosis in liver cancer patients. [20] Thus, HIF-l is considered to be a promising target for liver cancer treatment and multiple HIF-1 inhibitors have been developed and are undergoing preclinical or clinical trials to date. [21,22] Mechanistically, HIF-1 inhibitors can be mainly classified into two broad categories: inhibition of HIF-1 transcriptional activity in either direct or indirect ways and induction of HIF-1 degradation. In recent years, a number of novel small-molecule inhibitors have been identified to inhibit HIF-1 transcriptional activity or induce HIF-1 degradation [21,23] (Figure 1A), including (aryloxyacetylamino) benzoic acid derivatives (LW6), reducing HIF-1 accumulation by inhibiting malate dehydrogenase 2 (MDH2) activity; [24,25] indazole/benzimidazole derivatives (YC-1), downregulating HIF-1 translational initiation by suppressing the PI3K/Akt/mTOR/4E-BP pathway; [26,27] benzofuran derivatives (moracin O), inhibiting the initiation of HIF-1 translation by binding to heterogeneous nuclear ribonucleoprotein A2B1(hnRNPA2B1); [28,29] Manassantin A derivatives (LXY7824), reducing HIF-1 expression level via VHL-proteasome degradation system [30,31] and so forth. [32][33][34][35] Unfortunately, there is no HIF-1 inhibitor approved by Food and Drug Administration (FDA) for the cancer treatment to date due to unsatisfied therapeutic effects or unexpected side effects. Thus, it is still an urgent need to discover novel HIF-1 inhibitors with brand-new mechanism, potential potency, excellent druggability as well as low toxicity.
Herein, we report HI-101, a hit compound with an adamantaniline moiety screened by a cell-based HIF-1 transcriptional activity assay ( Figure 1B). We discover that HI-101 inhibits HIF-1 transcriptional activity by decreasing HIF-1 protein level as a translation inhibitor. With modified probes HI-102 and HI-103, we identify ATP synthase subunit beta (ATP5B) as a potential target which HI-derivatives might interact with. Mechanism of action study reveals that HI-derivatives interfere with HIF-1 mRNA binding to ribosome by enhancing ATP5B binding to HIF-1 mRNA. Further modification led to HI-104, a compound with better pharmacokinetic properties, showing therapeutic effects on MHCC97-L mice xenograft model. This study provides new insight for better understanding HIF-1 translation related signaling pathways and demonstrates that ATP5B might be a promising target for the development of HIF-1 inhibitors as anticancer agents.

Identification of Hit Compounds by High-Throughput Screening
To screen HIF-1 inhibitors in a robust cell-based assay, we optimized the dual-luciferase assay with pLenti-HIF-1 P2A (P402A and P564A) instead of pLenti-HIF-1 in order to ensure the expression of HIF-1 under normoxia condition. In the cellbased assay, HEK293T cells were co-transfected with pLenti-HIF-1 P2A, renilla, and hypoxia-responsive element (HRE)-Firefly Luciferase ( Figure S1, Supporting Information). Inhibition of HIF-1 transcriptional activity by compounds decreased the expression of firefly luciferase resulting in attenuating firefly luminescence, and renilla luminescence was an internal reference for cell numbers. The inhibition rates (IRs) of screened compounds were initially measured at a compound concentration of 10 μm after incubation for 24 h. Three compounds were selected as the positive controls including two inhibitors (PX-478 2HCl and BAY 87-2243) and an activator (BAY 85-3934) in the screening assay.
A collection of 101 254 compounds in the National Compound Library of the Shanghai Institute of Materia Medica, Chinese Academy of Sciences, were screened based on dual-luciferase assay. Fifteen inhibitors were yielded to decrease significantly the firefly/renilla luciferase signal in HEK293T at a concentration of 10 μm ( Figure S2, Supporting Information).

Hit Compound Confirmation
To validate the results of the primary screening and exclude false positives for further studies, we reperformed luciferase assays and confirmed that these fifteen hit compounds decreased the luciferase activity in a concentration-dependent manner ( Figure  S3, Supporting Information). Next, we examined the effects of these compounds on the expressions of HIF-1 targeted genes by quantitative RT-PCR. As shown in Figure S4, Supporting Information, all these fifteen hit compounds significantly reduced HIF-1 target genes (VEGF and PDK1) mRNA expression. Besides, we examined their effects on HIF-1 protein level and the Effect of HI-101 on HIF-1 transcriptional activity and protein level. A) Activity measurement of HI-101 and positive controls on dual-luciferase assay. B) Effects of HI-101 and positive controls on HIF-1 target genes expression, including VEGF and PDK1. C) Expressions of HIF-1 were examined in HEK293T and MHCC97-L cells treated with HI-101 at indicated concentrations. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001(two-tailed Student's t-test for unpair wise comparisons). Data are mean ± SD. results showed that compounds 3,5,8,9,11, and 14 all significantly reduced HIF-1 protein level while the other compounds had almost no effect on HIF-1 protein level ( Figure S5, Supporting Information). The same effects of these fifteen hit compounds on HIF-1 transcriptional activity and target genes expressions were observed in HEK293T cells under hypoxia condition ( Figure S6, Supporting Information). We selected compounds with the same core structure (8 and 9) for further study after excluding pan-assay interference compounds. [36] Results showed that the hit compound 8 (named as HI-101) significantly inhibited HIF-1 transcriptional activity and the expressions of HIF-1 target genes (Figure 2A,B). In addition, HI-101 specifically reduced the protein level of HIF-1 in a dosedependent manner under hypoxic condition in both HEK293T and MHCC97-L cells ( Figure 2C).

Hit Compound Optimization and Probe Design
In order to obtain a potent bio-active probe for target identification, we performed the structure activity relationship (SAR) study. HI-101 was further studied in two regions, the phenyl ring (region A) and the adamantane phenyl group (region B). First, the phenyl ring in region A was modified with small hydrophobic and hydrophilic groups as well as long chain substituents. As shown in Tables 1 and 2, compounds containing hydrophilic groups, such as amino (1-3s-1-3u) and hydroxyl (1-3m-1-3o) were much more potent than compounds with lipophilic groups, such as hexyloxyl (1-3ai and 1-3aj) and octyl (1-3ak), suggesting that the substituents at this position might lie in a hydrophilic pocket. In addition, substitutions at the ortho position led to greatly reduced or loss of activity while para-and metasubstitutions were tolerable and the activity of compounds with long hydrophilic chains was sustained. Besides, the activity was mainly reserved when the benzene ring was replaced by various aromatic heterocyclic rings (1-4a-1-4r) ( Table 3). Second, we explored if the adamantane phenyl group (region B) was necessary for inhibiting HIF-1 transcriptional activity. Compounds either with substitutions (methyl or hydroxyl) at the adamantane group or linkers between adamantane and phenyl group were designed and synthesized. As shown in Tables 4 and 5, both substitutions at the adamantane group and introduction of linkers remarkably decreased activities which suggested the adamantane group was very important. Finally, a biotin group was introduced at the paraposition to give probe HI-102 for target identification and a bodipy group was introduced at the same position to obtain probe HI-103 for mechanism study. assays were carried out as described in the experimental procedures. The on-beads samples were performed with silver staining after SDS-PAGE and a clear protein band with molecular weights between 50 and 70 kDa was detected in HI-102 treatment group ( Figure 3A). The band was cut and sent for protein analysis by mass spectrometry in 2 replicates. ATP5B was found in both samples and ATP synthase alpha (ATP5A) was found for once from the mass spectrometry analysis ( Figure S7, Supporting Information). Competitive ABPP experiments in vitro using desulfurization biotin and HI-101 at indicated concentrations were performed and western blot analysis supported that ATP5A and ATP5B were the possible targets of HI-102 ( Figure 3B). Confocal immunofluorescence experiment was also performed and revealed that HI-102 co-localized with ATP5A and ATP5B respectively ( Figure 3D). ATP5A and ATP5B are important components of F O F 1 -ATP synthase, also known as ATP synthase, which is a protein complex that catalyzes ATP synthesis by utilizing energy from the translocation of protons across biological membranes. [40] F O F 1 -ATP synthase consists of two domains: the membrane-spanning domain, F O , which acts as a proton channel, and the soluble catalytic domain, F 1 , which synthesizes ATP and is composed of nine subunits( 3 3 ɛ). [41][42][43][44] Many studies have indicated that F O F 1 -ATP synthase is closely associated with the proliferation, www.advancedsciencenews.com www.advancedscience.com invasion, and metastasis of tumor cells. [45][46][47] To verify whether HI-101 targeted ATP5A and ATP5B, we knocked down the subunits of F 1 domains separately by shRNA. HI-101 failed to decrease HIF-1 protein expression only when ATP5A or ATP5B was knocked down, even at a relatively high concentration (30 μm) ( Figures S8 and S9, Supporting Information). Given the high sequence identity of ATP5A or ATP5B between humans and pigs with 98.7% and 97.4% respectively, we extracted and purified F 1 -ATP synthase complex from pig hearts as described by J. E. Walker [48] and fluorescence polarization binding assay showed that HI-103 had a high affinity for F 1 -ATP synthase complex with a K d of 2.6 nm ( Figure 3C). Competitive fluorescence polarization binding assay suggested that HI-101 bound to F 1 -ATP synthase complex in a similar way as HI-103 but different from reported F 1 -ATP synthase inhibitors Oligomycin and Aurovertin B ( Figure S10, Supporting Information). Taken together, all the above experiments demonstrated that HI compounds bound ATP5A/ATP5B to inhibit HIF-1 transcriptional activity.

HI-101 Inhibits HIF-1 Translation
HI-101 treatment exhibited a significant reduction in HIF-1 protein level in a dose-dependent manner which indicated HI-101 either enhanced the degradation of mature HIF-1 protein or impaired the synthesis of nascent HIF-1 protein. We first examined whether the degradation rate of HIF-1 protein was enhanced by HI-101. We used cycloheximide (CHX) to block protein synthesis and determined the HIF-1 protein decay curve by  collecting and analyzing protein lysates at different time points after CHX treatment. Cells were pre-incubated under hypoxia for 6-8 h to achieve a steady HIF-1 protein level. Then, the cells were co-incubated with 100 μg mL −1 CHX as well as 10 μm HI-101 or DMSO at the indicated time. The half-life of HIF-1 was examined by western blot analysis. As shown in Figure 4A, HI-101 decreased HIF-1 protein level while did not have a significant effect on the rate of HIF-1 protein degradation. Therefore, we speculated that HI-101 might reduce the synthesis of nascent HIF-1 protein.
Lysosome and ubiquitin/proteasome systems are two main pathways for cellular protein degradation. [49,50] MG-132 and PS-341 are potent, selective, and reversible proteasome inhibitors, chloroquine (CQ) and NH 4 Cl are autophagy inhibitors that inhibit lysosomal degradation by increasing lysosomal pH. MG-132, PS-341, CQ, NH 4 Cl, and DMSO at indicated concentrations were added to HEK293T cells for 6 h followed by treatment with DMSO or 10 μm HI-101 for another 12 h under hypoxia condition. Cells were then collected, lysed, and immunoblotted to detect HIF-1 protein level. The results showed that HIF-1 protein level was decreased with HI-101 treatment regardless of whether protein degradation inhibitors were present or not, while the control proteins level (HIF-1 and -actin) were unaffected ( Figure 4B), which suggested us that HI-101 decreased HIF-1 protein level might by inhibiting the production of nascent HIF-1 protein. In order to verify whether HI-101 inhibited the syn-www.advancedsciencenews.com www.advancedscience.com Table 5. HIF-1 inhibitory activity of compounds 3-2a-3-2i a) .

Compound
Structure Inhibition rate [%] @ 10 μm* thesis of HIF-1 protein, l-azidohomoalanine (AHA) labeling assay [51] was carried out. The nascent protein was labeled by AHA incorporation, enriched by click reaction, and detected by antibiotin antibody. Experimental results showed that HIF-1 protein synthesis was significantly inhibited by HI-101 ( Figure 4C). The whole protein synthesis process can be divided into four major steps: transcription, translation, posttranslational modifications and folding to a mature protein, and protein secretion. [52] We then investigated the potential effects of HI-101 on HIF-1 transcription and translation. Real time-qPCR (RT-qPCR) experiments were performed and demonstrated HI-101 did not affect HIF-1 and HIF-1 mRNA expression ( Figure  S11, Supporting Information), while it significantly decreased the mRNA expression of HIF-1 target gene VEGF. On the basis of the existing experimental results, we draw a conclusion that HI-101 inhibited the synthesis of nascent HIF-1 protein without affecting its transcription.
Given that HI-101 affected neither the transcription of HIF-1 nor its degradation, we next tested whether HI-101 regulated HIF-1 translation. Polysome profile analysis is a commonly used method for studying translation process by separating polysomes and ribosomal subunits using a sucrose density gradient (SDG). [53] After treating with HI-101 or DMSO for 12 h co-incubated with MG-132 under hypoxia condition, HEK293T cells were collected and polysome-bound and polysome-free mR-NAs were isolated by using 10-50% sucrose gradient fractionation. As evidenced in Figure 4D, ribosome-free subunits were upregulated and heavy polysomes were downregulated with HI-101 treatment, which indicated HI-101 efficiently inhibited HIF-1 translation. Overall, these results suggested that HI-101 decreased HIF-1 expression by inhibiting its translation.

HI-101 Inhibits HIF-1 Translation by Enhancing ATP5B and HIF-1 mRNA Interaction
Although we revealed that HI-101 bound to ATP5A/ATP5B and decreased HIF-1 protein expression by inhibiting HIF-1 translation, the exact mechanism still remained unclear. It has been reported recently that KUSC-5037 inhibited HIF-1 transcription by suppressing FoF 1 -ATP synthase activity. [35] Thus, we wondered whether HI-101 inhibited HIF-1 translation in a similar way. Several ATP synthase inhibitors with various mechanisms of action have been reported. [54] Among them, Oligomycin inhibits the ATP synthase activity by binding the F O domain; Bedaquiline, an FDA-approved drug, inhibits mitochondrial ATP production by targeting the gamma subunit of the ATP synthase; [55] Aurovertin B decreases ATP synthase activity by targeting the subunits of F 1 domain. We first measured HIF-1 transcriptional activity and protein level with the presence of reported ATP synthase inhibitors. Only Oligomycin decreased HIF-1 transcriptional activity and protein level in a dose-dependent manner (Figure 5A), which indicated that there was a relationship between HIF-1 transcriptional activity and FoF 1 -ATP synthase activity. However, HI-101 showed no effect on ATP synthase activity even at a relatively high concentration (50 μm) ( Figure 5B), which indicated that HI-101 inhibited HIF-1 translation in a FoF 1 -ATP synthase enzyme activity-independent manner. A recent study reported that ATP5B shut down mPTP by binding circRNA SCAR in an enzymatic-independent way, [56] which reminded us whether ATP5B played a similar role in blocking HIF-1 translation. RNA-binding protein immunoprecipitation was then performed using the Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore, Burlington, MA) to detect the interaction between ATP5B and HIF-1 mRNA. As shown in Figure 5C, HI-101 affected neither the protein expression of ATP5A/ATP5B nor their interaction. However, HI-101 improved HIF-1 mRNA binding to ATP5B only in the immunoprecipitation sample using anti-ATP5B antibody, indicating that this interaction was independent of ATP5A/ATP5B complex. To explain why ATP5A knock-down showed a similar phenotype under HI-101 treatment, we then checked the protein level of ATP5B in ATP5A knock-down cell line because both ATP5A and ATP5B are subunits of mitochondrial ATP synthase. The western blot result ( Figure  S12, Supporting Information) showed that ATP5A knock-down downregulated the protein level of ATP5B, which attenuated the HIF-1 -reducing effect of HI-101 in ATP5A knock-down cell lines.
In order to better understand the mode of the action of HI compounds with ATP5B, we tried to co-crystallize HI compounds and F 1 -ATP synthase. However, we only solved the crystal structure of pig F 1 -ATP synthase at a resolution of 3.1 Å and failed to obtain the co-crystal structure, probably due to poor aqueous solubility of HI compounds. Thus, we utilized molecular docking to reveal how HI compounds bind to ATP5B. The results ( Figure S16, Supporting Information) showed that HI-105 bound to ATP5B by forming two key hydrogen bonds with Leu342 and Arg412 in the hinge region. The adamantyl group fully occupied the hydrophobic pocket formed by Pro350, Leu351, and Leu378, demonstrating that modifications on the adamantyl group would lead to clashes and impaired affinity. In addition, the benzene ring extended into the solvent-exposed region and substituents on the benzene ring were tolerated, which correlated well with structure-activity relationship. Among them, HI-105 exhibits excellent inhibitory activity for an extra hydrogen bond with Leu342 via the hydroxyl group. Although the binding mode of HI compounds and ATP5B was proposed, the structure mechanism of how HI compound promotes ATP5B binding to HIF-1 mRNA still remains unclear.
To sum up, the above experimental results showed that HI-101 inhibited HIF-1 translation by enhancing the interaction of HIF-1 mRNA and ATP5B in an FoF 1 -ATP synthase activityindependent manner.

Pharmacokinetic Properties of Compound HI-104 In Vivo
First, we roughly screened the concentrations in the plasma of seven selected compounds (1-3n, 1-3au, 1-4a, 1-4k, 1-4l, 1-4m,   and 1-4p) at indicated times after intraperitoneal administration in the Institute of Cancer Research (ICR) mice at a dose of 50 mg kg −1 with two mice per compound, the blood at five different times was collected and the plasma concentration was analyzed finally. The results ( Figure S13, Supporting Information) showed that 1-3n, 1-4a, 1-4k, 1-4l, 1-4m, and 1-4p were barely detectable in the plasma and 1-3au (HI-104) had a high concentration in the plasma. Therefore, we further investigated the pharmacokinetic properties of HI-104 in detail. Compound HI-104 had a halflife of 5.66 h, a high maximum concentration (C max ) of 91.68 μg mL −1 , and an area under the curve (AUC 0-24 h ) of 860.56 μg h mL −1 , which showed a very good drug exposure in the blood. Besides, we examined the toxicity of HI-104 in ICR mice by administering for 1 week at a dose of 100 mg kg −1 by intraperitoneal injection. HI-104 was not overtly toxic, as there was no significant change in body weight and mental status of mice ( Figure S14, Supporting Information). The above experiments proved that HI-104 can be used for further efficiency studies in vivo.

HI-104 Attenuates Tumor Growth in MHCC97-L Xenograft Model
HIF-1 overexpression is implicated in human hepatocellular carcinoma (HCC). [57] To investigate the biological significance of HI-104 in the progression of HCC by targeting ATP5A or ATP5B, we performed xenograft models in nude mice with ATP5A or ATP5B stable knock-out MHCC97-L cell lines (Figure 6A). The cells were inoculated to 6-8 weeks old nude mice under the armpits, and daily intraperitoneal injection of HI-104 was started after the tumor volume reached 100 mm 3 . Tumor volume and body weight were monitored during the treatment. Knockout of ATP5B significantly reduced tumor growth in MHCC97-L xenograft model, which is similar to those previously reported in the literature. [58] Besides, we observed that HI-104 reduced the tumor volume and weight in wildtype MHCC97-L xenograft tumor model with a tolerant weight loss, while almost had no effect in ATP5B KO models, which indicated that HI-104 possibly inhibited tumor growth through ATP5B ( Figure 6A-C, Figure S15, Supporting Information).

Conclusion
In summary, we conducted a high throughput screening based on a dual luciferase-reporter assay in order to search for novel HIF-1 transcriptional inhibitors. Hit compound HI-101 with an adamantaniline group was identified, and then probe HI-102 was developed for target identification. Through a series of experiments, ATP5B was identified as the potential target. Further mechanism study suggested HI-101 inhibited HIF-1 translational process and decreased protein expression by promoting the binding of ATP5B with HIF-1 mRNA. Besides, extensive structural optimization was performed to improve potency and compound HI-105 exhibited the most potency in HIF-1 transcriptional activity with an IC 50 of 26 nm, HI-104 showed a mild degree of antitumor effect in xenograft mouse model. Taken together, we revealed that adamantaniline derivatives inhibited HIF-1 transcriptional activity by blocking its translational process via ATP5B, which provided a new strategy for antitumor agent development.
Class B compounds with a linker between adamantane and benzene ring were synthesized as Scheme 2. As shown in Table 3, since 1-4a demonstrated particularly good activity against HIF-1 transcriptional activity, we designed and synthesized compounds with the 2-pyridylderivatives or bioisosteres. 2-1 was synthesized through amide condensation using 2-picolinic acid and methyl 4-aminobenzoate as starting materials, followed by hydrolysis to give intermediates 2-2. 2-3a-2-3e were obtained by a similar procedure with corresponding amines (Scheme 2). Besides, substituents in adamantane phenyl group were introduced into the hit compound and these derivatives (3- following a similar synthetic route as shown in Scheme 1. Besides, Probes HI-102 and HI-103 were synthesized as shown in Scheme 3.

Experimental Section
Chemistry: All reagents and solvents were purchased commercially and used without further purification. Reaction processes were monitored by thin-layer chromatography (TLC) and visualized under UV light at 254 and 365 nm, or color reagents. Column chromatography was conducted on silica gel (300-400 mesh) using automatic purification apparatus. 1 H NMR and 13 C NMR spectra were recorded on Bruker AC400 and Bruker AC600NMR spectrometer respectively in Chloroform-d or DMSO-d 6 with tetramethylsilane (TMS) as an internal reference. High-resolution mass spectra were recorded on triple TOF 5600+ MS/MS system (AB Sciex, Concord, Ontario, Canada) in negative or positive ESI mode. The purity of target compounds was determined by high-performance liquid chromatography with Promosil C18 column from Agela Technologies (4.6 mm × 150 mm, 5 μm particle size). Mobile phase A was double distilled water containing 0.1% trifluoroacetic acid and mobile phase B was methanol containing 0.1% trifluoroacetic acid. Flow rate was 1 mL min −1 using linear gradients as follows: 0-1 min was 40%B, 1-5 min was from 40%B to 95%B, 5-7 min was 95%B, 7-8 min was from 95%B to 40%B, 8-10 min was 40%B. All the biologically tested compounds confirmed at least 95% purity.
General Procedure for the Synthesis of Compounds 1-3a-1-3r, 1-3aa-1-3af, 1-3ai-1-3ak, 1-3ar-1-3au,1-4a-1-4r: Step1. To a mixture of 1bromoadamantane (10.7 g, 0.05 mol) and acetanilide (108 g, 0.80 mol) in 500 mL three-necked flask equipped with a mechanical stirrer, a nitrogen inlet with a thermometer and a condenser. Anhydrous aluminum chloride (1.33 g, 0.01 mol) as the catalyst was added to the mixture until the reactants were melted at 120°C. The reaction was stirred vigorously for 36 h at 145°C. The mixture was poured into ice water containing 4 m hydrochloric acid then purified by precipitation in refluxed water for 5-8 times and recrystallization in toluene ultimately. The precipitants were dried in a vacuum oven.
Step2. The crude product was dissolved in a mixture of methanol and 4 m hydrochloric acid and refluxed for 24 h, anhydrous potassium hydroxide was added slowly into the solution with ice bath, and the suspension was extracted with ethyl acetate. The organic phase was washed with brine, dried over anhydrous Na 2 SO 4 , and filtered. The residue was concentrated in vacuo and purified by silica chromatography to get intermediate 1-2. www.advancedsciencenews.com www.advancedscience.com Step3. Corresponding acids (1.2 mmol) and HATU (570 mg, 1.5 mmol) were added to a solution of 5 mL dry DMF and stirred for 30 min, then the 1-2 (227 mg, 1 mmol) and DIPEA (870 μL, 5 mmol) were added slowly into the reaction mixture respectively and stirred vigorously overnight. The reaction was stopped by adding water and extracted with ethyl acetate, the organic phase was washed with brine twice, dried over anhydrous Na 2 SO 4 , and filtered. The residue was concentrated in vacuo and purified by silica chromatography to get target compounds 1-3a-1-3r, 1-3aa-1-3af, 1-3ai-1-3ak, 1-4a-1-4r.
Step4. Trifluoroacetate (382 μL, 5 mmol) was slowly pipetted drop by drop to corresponding aminos (compounds 1-3p-1-3r, 0.5 mmol) dissolved in dry dichloromethane under argon atmosphere in an ice bath and the mixture was stirred at rt overnight. Then saturated sodium bicarbonate solution (10-20 mL) was added to the reaction after it was completed followed by extracting with ethyl acetate and then the organic phase was washed with brine twice, dried with saturated anhydrous Na 2 SO 4 , and filtered, then the residue was purified by silica chromatography after removal of ethyl acetate to afford compounds 1-3s-1-3u.