key: cord-0027326-8dnhxl8s
authors: Du, Guoxin; Du, Weiwei; An, Yuanlong; Wang, Minnan; Hao, Feifei; Tong, Xiaochu; Gong, Qi; He, Xiangdong; Jiang, Hualiang; He, Wei; Zheng, Mingyue; Zhang, Donglei
title: Design, synthesis, and LFA-1/ICAM-1 antagonist activity evaluation of Lifitegrast analogues
date: 2022-02-17
journal: Med Chem Res
DOI: 10.1007/s00044-022-02851-9
sha: 6f8a0cc40fddbe9e1798133b47a03d2c8ca466ef
doc_id: 27326
cord_uid: 8dnhxl8s

The interaction between Lymphocyte function-associated antigen 1 (LFA-1) and intercellular-adhesion molecule-1 (ICAM-1) plays important roles in the cell-mediated immune response and inflammation associated with dry eye disease. LFA-1/ICAM-1 antagonists can be used for the treatment of dry eye disease, such as Lifitegrast which has been approved by the FDA in 2016 as a new drug for the treatment of dry eye disease. In this study, we designed and synthesized some new structure compounds that are analogues to Lifitegrast, and their biological activities were evaluated by in vitro cell-based assay and also by in vivo mouse dry eye model. Our results demonstrated that one of these analogues of Lifitegrast (compound 1b) showed good LFA-1/ICAM-1 antagonist activity in in vitro assay; meanwhile, it also significantly reduced ocular surface epithelial cells damage, increased goblet cell density in dry eye mouse and highly improved the symptoms of dry eye mouse. [Figure: see text]

Dry eye disease (DED) is a complex, multifactorial condition characterized by inflammation of the ocular surface and lacrimal glands, as well as the reduction in the quality and/or quantity of tears [1] . DED can be categorized as "dry eye with reduced tear production (aqueous deficient) and dry eye with increased evaporation of the tear film known as the hyperevaporative type" [2] . Dry eye patients suffer from ocular-related symptoms, such as stinging, burning, itching, light sensitivity, and blurry vision which limit the quality of their life, as well as their ability to work [3] . The prevalence of DED is between 5% and 50% in the different countries, and this prevalence can be up to 75% in adults over age 40, with mostly female patients. In adults aged 18-40, 2.7% of these people have had an experience of DED [4] . Lymphocyte function-associated antigen 1 (LFA-1) is a leukocyte cell surface glycoprotein widely expressed on cells of a hematopoietic lineage [5] . Previous functional studies indicated that intercellular-adhesion molecule-1 functions as a ligand for LFA-1-dependent adhesion by a variety of leukocytes [6] . LFA-1 functions are extremely varied but play a critical role in facilitating effective immune responses. It has been reported that LFA-1 was implicated in numerous autoimmune and inflammatory conditions including inflammatory bowel disease, psoriasis, diabetes, and arthritis [7] . The previous study suggested that LFA-1/ICAM-1 interaction may play important roles in the cell-mediated immune response and inflammation associated with DED. Notably, inhibition of LFA-1/ICAM-1 binding offers a novel and possible approach to reducing ocular surface inflammation in this condition [8] .

The immunosuppressive mechanism of Lifitegrast is different from the first-generation immunosuppressor (cyclosporine A) for dry eye treatment. Lifitegrast (Fig. 1) is a small molecule that inhibits T-cell-mediated inflammation by blocking the binding of LFA-1 and ICAM-1, thus lessening overall inflammatory responses [9] . In vitro, Lifitegrast potently inhibited the attachment of LFA-1 to ICAM-1 in a concentration-dependent manner, with a half-maximal inhibitory concentration (IC 50 ) of 2.98 nmol/L [10] . The results from the phase 3 clinic trial of Lifitegrast have shown its ability to improve ocular surface epithelial health and associated symptoms of dry eye patients, thereby demonstrating its potential as a relatively fast-acting second-generation agent for the treatment of DED [11] . Lifitegrast became the second topical anti-inflammatory agent which has been approved by the FDA in 2016 for the treatment of DED [12] .

Lifitegrast ophthalmic solution (Xiidra) is labeled for the treatment of DED in concentrations as high as 5.0%. Such high concentrations result in many side effects, such as ocular irritation, dysgeusia and reduced visual acuity, especially an unusual taste sensation in the mouth. The LFA-1/ICAM-1 antagonist activity of Lifitegrast is not high enough to overcome these side effects. Therefore, it is very interesting to find novel higher LFA-1/ICAM-1 antagonist activity compounds. The discovery process and structure-activity relationships (SARs) of Lifitegrast were reported in a previous report [13] . So far, no other medicinal chemistry studies about Lifitegrast analogues have been reported. In this study, a novel series of Lifitegrast analogues with systematic modifications were further designed, synthesized, and their activities as LFA-1/ICAM-1 antagonists were evaluated by in vitro and in vivo assays.

Previous SAR studies have shown that an aromatic group in the "left-wing" residue is necessary for LFA-1/ICAM-1 antagonist activity [15] . Therefore, we attempted to link tetrahydroisoquinoline and benzofuran with different groups (Fig.  2, 1a-1c) to investigate the LFA-1/ICAM-1 antagonist activity. In order to investigate the effect of X groups with different electron-absorbing properties and different sizes on the activity, we selected sulfone group, N-methylene carbamoyl and methylene carbonyl as substituents. Meanwhile, the effects of substitution of benzofuran with different aromatic groups (Fig. 2, 1d-1e ) on the activity were investigated. Although benzotriazole and benzofuran have different atomic compositions, they are both benzo five-membered rings. From the activity results of benzotriazole, it can be seen whether there is a more suitable Ar 1 group. Ethynylbenzene was selected to investigate whether Ar 1 ring opening would have a significant effect on activity. At the same time, we incorporated different aromatic groups into "right-wing" amino acids (Fig. 2, 1f-1j ) to investigate the LFA-1/ICAM-1 antagonist activity. We cyclized methylsulfonyl with phenyl and converted methylsulfonylphenyl to benzene six-membered rings such as quinoline and 2-oxo-1,2,3,4-tetrahydroquinolin to investigate the effect of cyclization and different group size on the activity. According to the results of previous studies, Lifitegrast is located in the I-domain allosteric site of LFA-1/ICAM-1. Ar 2 group of Lifitegrast interact with the protein mainly through hydrogen-bonds with residues such as Tyr257 and Lys287 [16] . Therefore, methylsulfonyl group was converted to acetyl group and hydroxyl group to change the interaction mode of the hydrogen bond between Ar 2 with the protein to investigate the effect on the activity.

Amide intermediate (4) was synthesized by condensation of carboxylic acid (2) with primary amine (3) in DMF in the presence of condensating agent HATU, which was deprotected Boc group to yield amide intermediate (5) (Scheme 1) [17] . The yield of condensation reaction is low due to steric size of compound 2 and 3.

Analogues 1a-1e were generated through the synthetic route outlined in Scheme 2. Amide intermediate (5) was treated with benzofuran-6-sulfonyl chloride (10), benzofuran-6-ylmethanamine (14) , 2-(benzofuran-5-yl) acetic acid (16) [18] , [1, 2, 4] triazolo [1,5-a] pyridine-6-carboxylic acid (18) [19] and 3-ethynylbenzoic acid 20 [20] in DCM in the presence of condensating agent HATU or CDI to yield benzyl group analogues 11, 15, 17, 19 and 21, which provided analogues 1a-1e in the presence of debenzylation agent H 2 or TMSOK in MeOH or THF through removing the benzyl group [21] . The condensation activity of carboxyl with amino is stronger than that of amine with amino group, so the yield of compound 11, 17, 19, 21 was higher than that of compound 15.

Williamson reaction of 2-bromo-1,1-diethoxyethane (6) with 3-bromophenol (7) yielded 1-bromo-3-(2,2-diethoxyethoxy)benzene (8) , which was cyclized to the corresponding 6-bromobenzofuran (9) by heating in toluene containing cyclizing agent PPA. Sulfonylation of intermediate 9 by treatment with sulfuryl chloride in THF catalyzed by halogen-lithium exchange reagent n-BuLi gave the respective benzofuran-6-sulfonyl chloride (10) (Scheme 3) [22] . The sulfonation of intermediate 9 first involved the exchange of halogen atom to lithium atom, and then lithium atom was replaced by sulfonyl group. The complexity of the reaction mechanism leaded to the low yield of this reaction.

6-bromobenzofuran (12) was treated with zinc cyanide and cyanation catalyst tetrakis(triphenylphosphine)palladium to give benzofuran-6-carbonitrile (13) , which was reduced by reducing agent lithium aluminum hydride yield benzofuran-6-ylmethanamine (14) (Scheme 4) [23] . The cyanidation reaction (13) and reduction reaction (14) were both classical reactions, the yields of this two reaction were normal.

Condensation reaction of benzofuran-6-carboxylic acid (22) with methyl 5,7-dichloro-1,2,3,4-tetrahydroisoquinoline-6-carboxylate (23) by condensating agent EDCI yielded carboxylic acid methyl ester intermediate (24) , which was hydrolyzed to the corresponding carboxylic acid intermediate (25) by heating in pyridine containing hydrolysis reagent lithium iodide (Scheme 5) [24] . The hydrolysis of methyl carboxylate (24) involved the formation of methyl iodide and carboxylic acid lithium, then the reaction yield was slightly lower.

Analogues 1f-1h were generated through the synthetic route outlined in Scheme 6. Carboxylic acid intermediate (25) was treated with ethyl 2-amino-3-(quinolin-6-yl)propanoate (30), ethyl 2-amino-3-(1,1-dioxido-2,3-dihydrobenzo[b]thiophen-6yl)propanoate (41), ethyl 2-amino-3-(2-oxo-1,2,3,4-tetrahydroquinolin-7-yl)propanoate (50) in DCM in the presence of condensating agent HATU to yield carboxylic acid ethyl ester analogues 31, 42, 51 which provided analogues 1f-1h in the presence of hydrolysis reagent sodium hydroxide in EtOH, H 2 O and THF mixed solvent. The hydrolysis of ethyl carboxylate (31, 42, 51 ) is more difficult than that of methyl carboxylate (24) , the yield is lower than that of methyl carboxylate.

Quinoline-6-carbaldehyde (26) was reduced in methanol in the presence of reducing agent sodium borohydride to (27) , which provided 6-(bromomethyl)quinoline (28) in acetic acid in the presence of hydrogen bromide. The activity of hydrogen bromide as a halogenated agent generally resulted in a slightly lower yield of halogenated reaction (28) . The intermediate (28) was treated with ethyl N-(diphenylmethylene)glycinate in THF in the presence of alkylation catalyst sodium hydride to yield ethyl 2-((diphenylmethylene)amino)-3-(quinolin-6yl)propanoate (29) , which was hydrolyzed in ethyl acetate in the presence of hydrolysis reagent hydrochloric acid to yield ethyl 2-amino-3-(quinolin-6-yl)propanoate (30) (33) was obtained by treating compound (32) with nitric and sulfuric acid. This nitrification reaction (33) was a classic reaction, and the yield is in normal range. The nitro group is reduced to an amine group to give 6-amino-2,3-dihydrobenzo[b]thiophene 1,1-dioxide (34) , which is replaced by the bromine atom to form 6bromo-2,3-dihydrobenzo[b]thiophene 1,1-dioxide (35). The activity of substituting amino group (34) with hydrogen bromide as halogenated reagent was higher than that of substituting hydroxyl group (27) . Zinc cyanide reacted with intermediate (35) to produce 2,3-dihydrobenzo[b]thiophene-6-carbonitrile 1,1-dioxide (36), which was reduced to give 2,3-dihydrobenzo[b]thiophene-6-carbaldehyde 1,1dioxide (37). 2-amino-3-(1,1-dioxido-2,3-dihydrobenzo[b] thiophen-6-yl)ethyl propanoate (41) was synthesized from intermediate (37) through reduction reaction, halogenating reaction, hydrocarbylation reaction and hydrolysis reaction [26] . These reactions belonged to the classical reactions, the yields were in the normal range.

2-amino-3-(2-oxo-1,2,3,4-tetrahydroquinolin-7-yl) ethyl propanoate (50) was generated from 7-hydroxy-3,4dihydroquinolin-2(1H)-one (43) in Scheme 9. The hydroxyl group of compound (43) was esterified by trifluoroacetic anhydride to give 2-oxo-1,2,3,4-tetrahydroquinolin-7-yl trifluoromethanesulfonate (44). Zinc cyanide reacted with intermediate (44) to produce 2-oxo-1,2,3,4-tetrahydroquinoline-7-carbonitrile (45), which was reduced to give 2-oxo-1,2,3,4-tetrahydroquinoline-7carbaldehyde (46). Since trifluoroacetate was an excellent leaving group, the cyanidation reaction yield was over 50%. 2-amino-3-(2-oxo-1,2,3,4-tetrahydroquinolin-7-yl) ethyl propanoate (50) was synthesized from intermediate (46) through reduction reaction, halogenating reaction, hydrocarbylation reaction and hydrolysis reaction [27] . These reactions belonged to the classical reactions, the yields were in the normal range.

Analogues 1i-1j were generated through the synthetic route outlined in Scheme 10. Acyl chloride intermediate (52) was obtained by treating carboxylic acid intermediate (25) with dichlorosulfoxide. Compound 52 was directly used in the next reaction without purification. Acyl chloride intermediate (52) was acylated with compound 3-(3-acetyl-4-hydroxyphenyl)-2-aminopropanoic acid (53) and 2amino-3-(3,4-dihydroxyphenyl)propanoic acid (54) in the presence of TEA in DMF to yield analogues 1i and 1j, respectively. The yields of condensations (1i, 1j) were low because the carboxyl group at the α position reduces the nucleophilicity of the amino group (53, 54).

A total of 5 target "left-wing" analogues, 5 target "rightwing" analogues and more than 30 intermediates were successfully synthesized. All the designed target analogues were confirmed correct structure by 1 H NMR, 13 C NMR, LCMS and HRMS, and the LCMS content reached more than 95%. The structures of analogues are shown in Table 1 and the detailed synthesis process of each compound can be found in Supplementary Materials of this article.

The IC 50 value activity test results were shown in Fig. 3 Fig. 3 ) measured by our assay, this IC 50 value was very close to the previously reported IC 50 value of Lifitegrast (IC 50 = 2.98 nM and 9 nM in Jurkat or in HUT 78T cells, respectively) [9] . The activity test results showed that the IC 50 value of compound 1b reached the level of 262 nM, and had a good LFA-1 inhibitory activity. The IC 50 of compounds 1c, 1f and 1h also reached several micromol levels. Among them, 1f and 1h are racemes, and the activity of their S configuration will be higher.

From the results of Jurkat T-cell adhesion assay, we may summarize some rules about the relationship between the structures of the designed target compounds and LFA-1/ ICAM-1 antagonist activities. With Lifitegrast as a lead compound, in the "left-wing" residue, changing carbonyl group in Lifitegrast to sulfone group (compound 1a), or changing furyl to triazole (compound 1d) or acetenyl (compound 1e) decreased LFA-1/ICAM-1 antagonist activity obviously when compared with Lifitegrast. The activity of compound 1a showed that the carbonyl group could not be replaced by the more electron-absorbing sulfone group between benzofuran and tetrahydroisoquinoline, probably because the charge distribution of the whole molecule was affected. The charge distribution of triazole is quite different from that of furyl, which may contribute to the poor activity of compound 1d. The ring structure of benzofuran was opened is the reason for the poor activity of compound 1e, indicating that the ring structure plays an important role in the activity. While the carbonyl was changed to the N-methylene carbamoyl (compound 1b) and methylene carbonyl (compound 1c), the analogues show certain LFA-1/ICAM-1 antagonist activity. This result indicated that the length of the carbonyl between benzofuran and tetrahydroisoquinoline can be appropriately extended without adversely affecting the binding of the molecule to the target protein. The IC 50 value of compound 1b reached the level of 262 nM, this result suggested that the LFA-1/ICAM-1 antagonist activity can be maintained or even improved by only changing the X group to the other groups. In the "right-wing" Ar 2 residue of the Lifitegrast, replacing methylsulfonyl benzene with quinoline (compound 1f), 1-dioxido-2,3-dihydrobenzothiophen (compound 1g), 2-oxo-1,2,3,4-tetrahydroquinolin (compound 1h), 3acetyl-4-hydroxyphenyl (compound 1i) or 3,4-dihydroxyphenyl (compound 1j) respectively decreased LFA-1/ ICAM-1 antagonist activity. The cyclization of the sulfone group with benzene (compound 1g) resulted in a loss of activity because of the influence on the binding mode of the sulfone group to the target protein. Meanwhile, the IC 50 value of compounds 1f and 1h also reached several micromol levels. These results showed that replacing the Ar 2 residue with benzene six-membered ring could not improve the LFA-1/ICAM-1 antagonist activity, this may be related to increasing the size of the AR 2 group. Conversion of methylsulfonyl group to acetyl or hydroxyl groups on benzene ring also could not produce good activity. This result indicated that adding hydroxyl groups to the benzene ring did not increase the hydrogen bond interaction between the analogues and the target protein.

Compound 1b reduced the damage of corneal epithelium in dry eye mice model

After the 14 days of desiccating stress and drug-induction, we established a dry eye model that causes dry eye symptoms closer to the symptoms in patients with chronic DED [28] . Then we evaluated if compound 1b could improve the permeability barrier integrity of the ocular surface in this animal model.

Fluorescein staining in Fig. 4 showed punctate staining of green on the corneal surfaces in the experimental and control groups at day 14. We can see that the fluorescence dyeing areas on the ocular surface were significantly increased in the dry eye model group (B, EDE group) compared with the blank group (A, control group). Treatment with 5% or 7% compound 1b eye drops (E, F groups) improved the damage of corneal epithelium with less fluorescence dyeing areas, and treatment with 5% Lifitegrast (C group) was used as the positive control, which reveals the prominent therapeutically effects on DED.

Compound 1b enhanced the down-regulated goblet cells in dry eye mice model

The conjunctival goblet cells were distributed mainly among the fornix epithelial cells. Goblet cells could secrete the mucin and keep the tear film stable. The symptoms of dry eye are usually accompanied by the loss of goblet cells [29] . Therefore, we assessed the effects of compound 1b on goblet cells in dry eye mice model. As shown in Fig. 5 , The number of goblet cells from the superior and inferior conjunctiva were markedly decreased in dry eye model group (B) compared with blank group (A, control group). Treatment with 5% or 7% compound 1b (E and F groups) significantly increased the density of goblet cells compared with the dry eye model group (B). It is obvious that compound 1b prevented the missing of the goblet cells in dry eye mice.

In order to investigate if compound 1b has anti-apoptotic effects to protect the ocular surface damage in dry eye mice, we detected the signals of apoptosis on corneal epithelium by using TUNEL staining method in dry eye mice model with or without compound 1b treatment. TUNEL immunofluorescence staining is a method to analyze apoptosis at a late stage by detecting DNA fragmentation through labeling the terminal end of nucleic acids in the cells. The results of TUNEL immunofluorescence staining in Fig. 6 demonstrated that compared to control group, mean numbers of corneal epithelium cells which had the signals of TUNEL staining with Cy3 were significantly increased in the dry eye model groups (EDE, and EDE + Vehicle groups). In compound 1b treatment group (EDE + 5% 1b, or EDE + 7% 1b), the amount of cells labeled with Cy3 were markedly decreased in cornea tissues when compared with that of dry eye model group, and the quantitative statistical analysis data of TUNEL positive cells in each group was shown in Fig. 6G . But the anti-apoptotic effect of compound 1b was weaker than that of Lifitegrast treatment group. These results in Fig. 6 suggested that treatment with compound 1b could protect corneal and conjunctival epithelium cells from apoptosis, maintained the integrity of ocular epithelium in dry eye mice.

The Lifitegrast eye drops showed a statistically significant reduction in ocular surface injury compared to the model control group, and the effect was stronger than compound 1b. As shown in Fig. 4 , 5% and 7% compound 1b significantly reduced ocular surface epithelial cell damage in dry eye mice. 5% and 7% compound 1b reduced the damage of corneal epithelium in dry eye mice model. The conjunctiva goblet cells secrete the mucin to maintain the stability of the tear film. The symptoms of dry eye are usually accompanied by a decreasing of the number or density of goblet cells. We evaluated the effect of 5% and 7% compound 1b on goblet cells in dry eye mice model. A statistical histogram of goblet cell density is shown in Fig. 5 . 5% and 7% compound 1b significantly increased goblet cell density and improved dry eye symptoms in mice with dry eye. As shown in Fig. 6 , 5% and 7% compound 1b has a protective effect on apoptosis of ocular surface cells in dry-eye mice, but the effect is weaker than that of Lifitegrast. The results from our in vivo animal model experiments showed that topical administration of 5% and 7% (W/ V) compound 1b improved the dry eye symptoms through reducing the ocular surface injuries, increasing the density of goblet cells on conjunctival epithelium and decreasing the apoptosis of cornea and conjunctiva. There was no statistical difference in the effect between 5% and 7% compound 1b. There was no statistical difference in the effect between 5% and 7% compound 1b. The in vitro assay showed the IC 50 value of compound 1b reached the level of 262 nM which is lower than that of Lifitegrast (IC 50 value is about 11 nM). This result suggested that the LFA-1/ICAM- Fig. 4 Effect of compound 1b on ocular surface damage in dry eye mice. After 14 days treatment on dry eye mice model, ocular surface of mice were stained with fluorescein sodium, the green fluorescence staining points represent the damaged area of ocular surface epithelium (A-F), and the corneal fluorescence staining score in graph A-F were calculated and shown in a bar graph (G). Data were shown as mean ± SEM (n = 10, eyes/ group). *P < 0.05, **P < 0.01, ***P < 0.001. Control: nontreated normal mice group; EDE: experimental dry eye mice group; EDE + Lifitgrast: experimental dry eye mice treated with Lifitgrast; EDE + Vehicle: experimental dry eye mice treated with control solvent; EDE + 5%1b: experimental dry eye mice treated with 5% compound 1b; EDE + 7%1b: experimental dry eye mice treated with 7% compound 1b 1 antagonist activity of compound 1b may be weaker than that of Lifitegrast also in vivo, therefore, the anti-dry eye effects of compound 1b is less as effective as Lifitegrast.

Although the effect of compound 1b is weaker than Lifitegrast in vitro and in vivo, the structure of compound 1b is different from Lifitegrast and circumvented the scope of Lifitegrast original patent. These results may serve as important milestone references for future structure optimization.

In summary, in this study, 10 analogues with different groups in the "left-wing" or "right-wing" residues in Lifitegrast were designed according to molecular dynamics simulations and were synthesized through a variety of chemical methods. The Jurkat cell attachment assay was used to assess the LFA-1/ ICAM-1 antagonist activity of these 10 analogues of Lifitegrast. Among them, the IC 50 value of compound 1b reached 262 nM. To further verify the biological activity of compound 1b in vivo, dry eye mice model was developed to investigate if compound 1b could improve the symptoms of dry eye. The results of animal model demonstrated that 5%(W/V) compound 1b significantly improved the symptoms of dry eye mice through reducing ocular surface epithelial cells damage, increasing goblet cell density and decreasing the apoptosis of cornea and conjunctiva in dry eye mice. The results of this study encourage us to further optimize the structure of compound 1b in our future work, looking for a compound with a better LFA-1/ICAM-1 antagonist activity.

All solvents and reagents were purchased from WuXi AppTec (Shanghai, China), Accela ChemBio Co., Ltd. (Shanghai, China), Titan Scientific Co.,Ltd. (Shanghai, China), and Hao Hong Pharma (Shanghai, China). Solvents and reagents were generally the best quality commercial-grade products and were used without further purification. 1 H and 13 C NMR spectra The results of histopathological evaluation of the goblet cells staining by periodic acid-schiff were shown in A-F images (size, ×400). The deep purple dyeing in conjunctiva represents goblet cells. The quantification of goblet cells densities in graph A-F was shown in G graph, data was shown as mean ± SEM (n = 10 eyes/group). *P < 0.05, **P < 0.01, ***P < 0.001. Control: non-treated normal mice group; EDE: experimental dry eye mice group; EDE + Lifitgrast: experimental dry eye mice treated with Lifitgrast; EDE + Vehicle: experimental dry eye mice treated with control solvent; EDE + 5%1b: experimental dry eye mice treated with 5% compound 1b; EDE + 7%1b: experimental dry eye mice treated with 7% compound 1b were taken on Bruker Avance III 500 or 400 NMR spectrometers operating at 400 MHz for 1 H NMR, and 100 MHz for 13 C NMR, with tetramethylsilane (TMS) as the internal standard and CDCl 3 , MeOD and DMSO-d6 as the solvent. Chemical shift values (δ) are reported in ppm relative to that of the internal TMS. The following abbreviations were used: s = singlet, d = doublet, t = triplet, q = quadruplet, m = multiplet, and b = broad. 13 C NMR spectra were recorded with complete proton decoupling. ESIMS data were recorded on Finnigan LCQ/DECA instruments. HRMS data were collected on Micromass QTOF Ultima (ESI) spectrometers. Silica gel F 254 was used in analytical thin-layer chromatography (TLC), and silica gel was used in column chromatography; visualizations were accomplished with UV light (254 nm).

A mixture of compound 2 (5.61 g, 16.2 mmol), compound 3 (4.50 g, 13.5 mmol), HATU ( 6 Compound 1b attenuated the apoptosis of corneal epithelium in dry eye mice model. The red fluorescence stain (TUNEL) represents apoptotic cells and the blue stain (DAPI) represents cell nucleus. The quantification of TUNEL positive staining cells in corneal epithelium were shown in G graphs, respectively. Data were shown as mean ± SEM (n = 10 eyes/group).*P < 0.05, **P < 0.01, ***P < 0. (11) To a solution of compound 5 (0.300 g, 501 µmol, HCl) in DCM (3.00 mL), TEA (152 mg, 1.51 mmol) and HATU (286 mg, 752 µmol) were added. The solution was stirred at 20°C for 3 h, then compound 10 (108 mg, 501 µmol) was added via several portions. The reaction mixture was stirred at 20°C for 2 h under N 2 atmosphere. LCMS showed starting material was consumed completely. The reaction was quenched with water (3.00 mL), and the organic was extracted with DCM (3.00 mL × 3), then the organic layer was concentrated under reduced pressure. Compound 11 (0.260 g, 350 µmol, 69.8% yield) was obtained as a brown solid. 1 

To a solution of compound 11 (0.260 g, 350 µmol) in MeOH (1.00 mL), Raney-Ni (30.0 mg, 350 µmol) was added under N 2 . The suspension was degassed under vacuum and purged with H 2 several times. Then the mixture was stirred under H 2 (15 psi) at 25°C for 3 h. TLC indicated starting material was consumed completely. After that, the suspension was filtered through a pad of Celite and the pad was washed with EtOH (2.00 mL × 3). The combined filtrates were concentrated to dryness to give product, and the crude product was purified by prep-HPLC (HCl condition). Compound 1a (0.022 g, 33.0 µmol, 9.42% yield, 97.7% purity) was obtained as a white solid. 1 (17) To a solution of compound 5 (0.200 g, 334 µmol) in DCM (2.00 mL), TEA (101 mg, 1.00 mmol, 139 µL) and HATU (191 mg, 502 µmol) were added, the solution was stirred at 20°C for 1 h, then compound 16 (58.9 mg, 334 µmol) was added via several portions. The reaction mixture was stirred at 20°C for 2 h under N 2 atmosphere. LCMS showed starting material was consumed completely and one major new peak with desired mass was detected. The solution was quenched with water (2.00 mL) and the organic was extracted with DCM (2.00 mL × 3), then organic was concentrated under reduced pressure. (S)-2-(2-([1,2,4]Triazolo[1,5-a]pyridine-6-carbonyl)-5,7-dichloro-1,2,3,4-tetrahydroisoquinoline-6carboxamido)-3-(3-(methylsulfonyl)phenyl)benzyl propanoate (19) To a solution of compound 5 (0.200 g, 334 µmol) in DCM (2.00 mL), TEA (101 mg, 1.00 mmol) and HATU (190 mg, 502 µmol) were added, the solution was stirred at 20°C for 1 h, then compound 18 (54.6 mg, 334 µmol) was added via several portions. The reaction mixture was stirred at 20°C for 2 h under N 2 atmosphere. LCMS showed starting material was consumed completely and one major new peak with desired mass was detected. The solution was quenched with water (2.00 mL) and the organic was extracted with DCM (2.00 mL × 3), then the organic was concentrated under reduced pressure. To a solution of compound 19 (0.200 g, 283 µmol) in THF (2.00 mL), TMSOK (109 mg, 849 µmol) was added, then the mixture was stirred at 70°C for 1 h. LCMS indicated starting material was consumed completely and one major new peak with desired mass was detected. The mixture was quenched with water (2.00 mL) and acidified with 1 N HCl to pH = 3, the mixture was extracted with EtOAc (2.00 mL × 3), then the organic layer was concentrated under reduced pressure. The crude product was purified by prep-HPLC (HCl condition). Compound 1d (0.04 g, 62.4 µmol, 22.0% yield) was obtained as a white solid. 

To a solution of compound 5 (0.200 g, 334 µmol) in DCM (2.00 mL), TEA (101 mg, 1.00 mmol) and HATU (190 mg, 501 µmol) were added, the solution was stirred at 20°C for 1 h, then compound 20 (48.9 mg, 334 µmol) was added via several portions. The reaction mixture was stirred at 20°C for 2 h under N 2 atmosphere. LCMS showed starting material was consumed completely and one major new peak with desired mass was detected. The reaction was quenched with water (2.00 mL) and the organic was extracted DCM 

To a solution of compound 21 (0.200 g, 290 µmol) in THF (2.00 mL), TMSOK (111 mg, 870 µmol) was added, then the mixture was stirred at 70°C for 1 h. LCMS indicated starting material was consumed completely, and new peak with desired mass was detected. The mixture was quenched with water (2.00 mL) and acidified with 1 N HCl to pH = 3, the mixture was extracted with EtOAc (2.00 mL × 3), then the organic layer was concentrated under reduced pressure. The crude product was purified by prep-HPLC (HCl condition). Compound 1e (0.040 g, 64.4 µmol, 22.2% yield) was obtained as a white solid. 1 

To a solution of compound 6 (10.0 g, 57.8 mmol) and 2bromo-1,1-diethoxy-ethane 7 (22.8 g, 115 mmol) in DMF (60.0 mL), K 2 CO 3 (7.99 g, 57.8 mmol) was added, then air in the reaction system was replaced with nitrogen three times, and the mixture was stirred at 100°C for 10 h. TLC indicated starting material was consumed completely. After that, the reaction solution was diluted with water (200 mL) and then extracted with ethyl acetate (60 mL × 3). The extract was dried over anhydrous sodium sulfate and concentrated. Then the oil was purified by silica gel column chromatography. Compound 8 (8.00 g, 27.6 mmol, 47.8% yield) was obtained as a colorless oil. 1 

PPA (2.27 g, 83.0 mmol) was added to a solution of compound 8 (8.00 g, 27.6 mmol) in toluene (40.0 mL), the mixture was stirred at 80°C for 4 h. TLC indicated starting material was consumed completely. The solution was cooled to room temperature, quenched with water (60.0 mL), extracted with EtOAc (40 mL × 3), the organic layer was washed with brine (40.0 mL), dried (Na 2 SO 4 ), filter. Then the organic layer was concentrated under reduced pressure to give a residue. The residue was purified by column chromatography. Compound 9 (4.40 g, 11.1 mmol, 40.3% yield,~50% purity) was obtained as brown oil. 1 

To a solution of compound 9 (0.400 g, 2.03 mmol) in THF (4.00 mL), n-BuLi (0.130 g, 812 µL) was added at −78°C with atmosphere of nitrogen, the solution was stirred at −78°C for 0.5 h. Then to the solution was added SO 2 (1.30 g, 20.3 mmol) in THF (2.00 mL), the solution was stirred at 20°C for another 1.5 h. The solution was concentrated under vacuum. The residue was dissolved in hexane, then the mixture was added sulfuryl chloride (548 mg, 4.06 mmol) at 0°C. The resulting solution was stirred for 10 h at 20°C. TLC indicated starting material was consumed completely. After that, the mixture reaction was filter, and the organic layer was concentrated under reduced pressure to left a residue. The residue was purified by prep-TLC. Compound 10 (0.200 g, 923 µmol, 22.7% yield) was obtained as a white solid. 1 Benzofuran-6-carbonitrile (13) To a solution of compound 12 (2.00 g, 10.2 mmol) in DMF (12.0 mL), Zn(CN) 2 (19.6 g, 167 mmol) and Pd(PPh 3 ) 4 (1.41 g, 1.22 mmol) were added under a nitrogen atmosphere, the reaction mixture was stirred at 80°C for 16 h. TLC indicated starting material was consumed completely, and two new spots were detected. The reaction mixture was diluted with toluene (10.0 mL) and the phases were separated. The aqueous phase was extracted twice with toluene (10.0 mL × 2). The combined organic was washed with brine (100 mL) and saturated aqueous ammonium hydroxide (20 mL), then dried (Na 2 SO 4 ) and concentrated to give a residue. The residue was purified by column chromatography (SiO 2 , Petroleum ether/Ethyl acetate = 100/1 to 10/1). Compound 13 (1.00 g, 6.99 mmol 68.8% yield) was obtained as a colorless oil. 1 

To a mixture of LiAlH 4 (265 mg, 6.99 mmol) in THF (10.0 mL), compound 13 (1.00 g, 6.99 mmol) was added in one portion at 20°C. The mixture was stirred at 20°C for 1 h. TLC indicated one new spot was detected. 1 mL of saturated Na 2 SO 4 was added to the reaction mixture carefully, and then filtered. The filtration was concentrated to give compound 14 (0.5 g, 3.40 mmol, 48.60% yield) as a light yellow oil. 1 2-(Benzofuran-6-carbonyl)-5,7-dichloro-1,2,3,4tetrahydroisoquinoline-6-methyl carboxylate (24) To a mixture of compound 23 (5.00 g, 16.8 mmol), ECDI (3.56 g, 18.5 mmol), DMAP (164 mg, 1.35 mmol) and TEA (5.12 g, 50.58 mmol) in DMF (30 mL), compound 22 (2.73 g, 16.8 mmol) was added. Then the mixture was stirred at 25°C for 16 h. TLC indicated reactant 23 was consumed completely, and the reaction was clean according to TLC. The reaction mixture was poured into H 2 O (120 mL). The mixture was diluted with DCM 10.0 mL and washed with H 2 O (50.0 mL), aq.citric acid (50.0 mL), aq. NaHCO 3 (50.0 mL). The combined organic layers were washed with brine 100 mL dried over Na 2 SO 4 , filtered and concentrated under reduced pressure to give a residue. Compound 24 (4.00 g, 9.90 mmol, 58.6% yield) was obtained as yellow solid. 1 2-(Benzofuran-6-carbonyl)-5,7-dichloro-1,2,3,4tetrahydroisoquinoline-6-carboxylic acid (25) To a solution of compound 24 (4.00 g, 9.90 mmol) in pyridine (24.0 mL), LiI (5.30 g, 39.5 mmol) was added. Then the mixture was stirred at 100°C for 16 h. TLC indicated reactant 24 was consumed completely and the reaction was clean according to TLC. The reaction mixture was concentrated under reduced pressure to remove pyridine. The reaction mixture was adjusted the pH to 2 by adding HCl (3 M, 40.0 mL), solid was precipitate out, then the mixture was filtered and the filter cake was washed with H 2 O (50.0 mL), dried in vacuum to give residue. The crude product was purified by re-crystallization from Petroleum ether: Ethyl acetate = 1: 1 (20.0 mL) at 15°C. Compound 25 (2.00 g, 4.77 mmol, 48.1% yield) was obtained as brown solid. 1 To a solution of compound 31 (200 mg, 324 µmol) in EtOH (2.00 mL) and THF (1.00 mL), NaOH (38.9 mg, 973 µmol) in H 2 O (1.00 mL) were added, the mixture was stirred at 20°C for 2 h. TLC indicated reactant 31 was consumed completely. EtOH and THF was removed, the mixture was adjusted pH = 4, then solid was formed, filtered and the filter cake concentrated to give crude product. The crude product was triturated with Petroleum ether: Ethyl acetate To a solution of compound 42 (0.200 g, 305 µmol) in EtOH (2.00 mL) and THF (1.00 mL), NaOH (36.6 mg, 915 µmol) in H 2 O (1.00 mL) was added. The mixture was stirred at 20°C for 20 h. TLC indicated reactant 42 was consumed completely. EtOH and THF was removed, the mixture was adjusted pH = 4, then solid was formed, filtered and the filter cake concentrated to give crude product. The crude product was triturated with Petroleum ether: Ethyl acetate = 10: 1 (20 mL) at 30°C for 30 min. Compound 1g (24.0 mg, 36.6 µmol, 12.0% yield) was obtained as a white solid. 2-(2-(Benzofuran-6-carbonyl)-5,7-dichloro-1,2,3,4tetrahydroisoquinoline-6-carboxamido)-3-(2-oxo-1,2,3,4-tetrahydroquinolin-7-yl)propanoic acid (1h)

To a solution of compound 51 (0.200 g, 315 µmol) in MeOH (2.00 mL) and THF (1.00 mL), NaOH (37.8 mg, 945 µmol) in H 2 O (1.00 mL) were added, the mixture was stirred at 20°C for 2 h. TLC indicated reactant 51 was consumed completely. EtOH and THF was removed, the mixture was adjusted pH = 4, then solid was formed, filtered and the filter cake concentrated to give crude product. The crude product was triturated with Petroleum ether: Ethyl acetate = 10: 1 (20 mL) at

To a solution of compound 26 (3.00 g, 19.0 mmol) in MeOH (15.0 mL), NaBH 4 (2.20 g, 58.1 mmol) was added in portions at 0°C. Then the mixture was stirred at 15°C for 3 h. TLC showed the starting material was consumed completely. The reaction mixture was partitioned between H 2 O (100 mL) and EtOAc (60.0 mL, 40.0 mL, 20.0 mL). The organic phase was separated, washed with brine 50.0 mL, dried over Na 2 SO 4 , filtered and concentrated under reduced pressure to give a residue. Compound 27 (2.00 g, 12.5 mmol, 65.8% yield) was obtained as yellow oil. 1 

To a solution of compound 27 (0.700 g, 4.40 mmol) in HBr/ AcOH (5.00 mL), the mixture was stirred at 20°C for 3 h. TLC indicated reactant 27 was consumed completely. The mixture was concentrated to give product. The crude product was triturated with Petroleum ether: Ethyl acetate = 5: 1 (10 mL) at 20°C for 30 min. Compound 28 (0.500 g, 1.65 mmol, 37.5% yield) was obtained as a yellow solid. 1 2-((Diphenylmethylene)amino)-3-(quinolin-6-yl) ethyl propanoate (29) To a solution of ethyl N-(diphenylmethylene)glycinate (882 mg, 3.30 mmol) in THF (10 mL), NaH (132 mg, 3.30 mmol) was added at 5°C, the mixture was stirred for 0.5 h, then compound 28 (0.500 g, 1.65 mmol) was added, the mixture was stirred at 20°C for 2 h. TLC indicated reactant 28 was consumed completely. The mixture was poured into NH 4 Cl (10 mL), extracted with EtOAc (20 mL), brine (10 mL), dried over Na 2 SO 4 , concentrated to give crude product. The residue was purified by column chromatography. Compound 29 (0.500 g, 1.16 mmol, 70.4% yield, 95.0% purity) was obtained as yellow oil. 1 

To a solution of compound 29 (0.500 g, 1.22 mmol) in EtOAc (3 mL) and HCl (1.00 M, 2 mL), the mixture was stirred at 20°C for 0.5 h. TLC indicated reactant 29 was consumed completely. The mixture was extracted with EtOAc (10 mL), the aqueous phase adjusted pH = 8 with aq.NaCO 3 , and extracted with EtOAc (10 mL × 2), brine (10 mL), dried over Na 2 SO 4 , concentrated to give product. 

To a solution of HNO 3 (35.0 g, 333 mmol) in H 2 SO 4 (46.0 g, 460 mmol) at 0°C, then compound 32 (10.0 g, 60.2 mmol) was added at 0°C for 1 h. TLC showed the starting material was consumed completely. The mixture was poured into ice water (300 mL) and the mixture was partitioned with EtOAc (200 mL) and the organic layer was wished with aq. NaHCO 3 (200 mL), dried over Na 2 SO 4 and evaporated to dryness. The residue was purified by column chromatography. Compound 33 (9.00 g, 42.6 mmol, 70.8% yield) was obtained as light yellow solid. 1 To a solution of compound 34 (5.00 g, 27.3 mmol) in HBr (50.0 mL, 40 % purity) at 0°C, NaNO 2 (2.45 g, 35.5 mmol) was added portion wise. The temperature was kept BELOW 8°C during this addition. The resulting mixture was then added to a suspension of CuBr (2.74 g, 19.1 mmol) in HBr (10.0 mL) at 70°C. The resulting mixture was stirred at 70°C for 1 h. TLC showed the starting material was consumed completely. The mixture was poured into ice water (200 mL) and then was partitioned with EtOAc (100 ml × 3), the organic layer was washed with H 2 O (150 mL) dried over Na 2 SO 4 and evaporated to dryness. The residue was purified by column chromatography. Compound 35 (4.00 g, 16.2 mmol, 59.1% yield) was obtained as light yellow solid. 

To a solution of compound 37 (2.00 g, 10.1 mmol) in MeOH (10.0 mL), NaBH 4 (1.16 g, 30.5 mmol)was added in portions at 0°C. Then the mixture was stirred at 15°C for 3 h. TLC showed the starting material was consumed completely. The reaction mixture was partitioned between H 2 O (100 mL) and EtOAc (60.0 mL, 40.0 mL, 20.0 mL). The organic phase was separated, washed with brine 50.0 mL, dried over Na 2 SO 4 , filtered and concentrated under reduced pressure to give a residue. 

A mixture of compound 38 (500 mg, 2.52 mmol) in HBr (3.00 mL) was stirred at 20°C for 2 h. TLC showed new spots were formed. The residue was poured into H 2 O (10 mL) and extracted with EtOAc (10 mL × 2). The organic layer was isolated, dried (MgSO 4 ) and concentrated in vacuo. Compound 39 (500 mg, 1.53 mmol, 60.7% yield) was obtained as yellow oil. 1 To a solution of compound 43 (5.00 g, 30.6 mmol) in CHCl 3 (50.0 mL), pyridine (5.28 g, 66.8 mmol, 5.39 mL) was added. The mixture was stirred at 25°C for 10 min and then cooled to 0°C. (Tf) 2 O (8.60 g, 30.6 mmol) was slowly added drop wise under 0°C followed by stirring for 10 min. The reaction was stirred at 25°C for 20 min. TLC showed the reaction was complete. The solution was poured into water (20.0 mL) and DCM (10.0 mL) was added. The organic phase was separated and the aqueous was extracted with DCM (2 × 10.0 mL). The organic phase were dried over magnesium sulfate, filtered and concentrated in vacuum. The crude product was purified by column chromatography. Compound 44 (5.00 g, 16.9 mmol, 55.2% yield) was obtained as white solid. 1 To a solution of compound 44 (5.00 g, 16.9 mmol) in DMF (30.0 mL), Zn(CN) 2 (1.59 g, 13.5 mmol) and Pd(PPh 3 ) 4 (1.96 g, 1.69 mmol) were added, then the reaction was stirred at 80°C for 12 h. TLC showed the reaction was complete. The solution was poured into water (20.0 mL) and DCM (10.0 mL) was added. The organic phase was separated and the aqueous was extracted with DCM (2 × 10.0 mL). The organic phase were dried over magnesium sulfate, filtered and concentrated in vacuum. Compound 45 (2.00 g, 9.35 mmol, 55.2% yield) was obtained as off-white solid. 1 Jurkat T-cell adhesion assay Cells and Cell Culture: Lifitegrast (Selleck, S3714, Shanghai, China), Human leukemia T lymphocytes (Jurkat, Clone E6-1, the catalog number is SPSC-513, was obtained from the cell bank of Chinese Academy of Sciences, Shanghai, China) were cultured in RPMI 1640 Medium (Invitrogen, 11875-093) supplemented with 5% fetal bovine serum at 37°C with 5% CO 2 in a humidifified atmosphere.

To evaluate the ability of compounds 1a-1j to inhibit the attachment of Jurkat cells to intercellular-adhesion molecule ICAM-1 in vitro, 100 mM stock solutions of the compounds and a positive control (Lifitegrast) were prepared in an aqueous solution of dimethylsulfoxide (1:1 (v/v)), respectively, and then diluted by adding assay medium to achieve and maintain the desired concentration throughout the assay. The Jurkat cells were labeled with an 8 μM solution of 2′, 7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein in growth medium at room temperature for 15 min. Labeled cells were incubated in 70 μL of assay medium in each well of a 96-well plate at 5 × 10 5 cells per well with 70 μL of compound or the positive control in assay medium at 37°C for 30 min. A 100 μL aliquot of this fluorescence-labeled Jurkat cell suspension was allowed to settle in the presence of compound or the positive control in wells of a 96-well plate coated with recombinant human ICAM-1 expressed as an Fc chimera (R&D Systems, Minneapolis, MN) at 37°C for 1 h. A dose titration of each test compound was run (10 −10 -10 −4 M) in three separate lanes on a single plate, nonadherent cells were removed by washing with the medium. The adherent cells were quantitated by measuring the intensity of fluorescence from adherent cells, and the IC 50 values were calculated by using a standard four-parameter logistic nonlinear regression analysis of the data (GraphPad Software Inc., San Diego, CA) [13] .

Dry eye mice model Drugs and instruments: Scopolamine (0.5 mg/0.2 mL, Sigma-Aldrich, S-098-1g, St. Louis, MO); 5% (W/V) or 7% (W/V) compound 1b eye drops (containing 2.6% glycerinum and 0.01% benzalkonium chloride 0.01%, pH range from 8.0 to 8.5). Lifitegrast (5%, Shire, 07548, Columbia, USA); OCT (VWR, 8551-00, Suwanee, USA); TUNEL(Beyotime Institute of Biotechnology, C1090, Haimen, China); (Nikon Eclipse Ni-U, Japan); Stereographic fluorescence microscope (Nikon SMZ1500, Japan); Frozen slicing machine (LEIKA CM1860UV, Germany); Intelligent drying oven (ZD-890C; Ouyi Company, Hangzhou, China). In this study, the animal experimental procedures were approved by the Guide for the Care and Use of Laboratory Animals (Ministry of Science and Technology of China, 2006) , and adhered to the standards in the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.

All experimental procedures related to animals' welfare in this research were performed according to the Guide for the Care and Use of Laboratory Animals (Ministry of Science and Technology of China, 2006) , and the standards in the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. All animal experimental protocols in this study were also approved by the animal ethics committee of He University. Thirty female C57/BL6 mice (age range, 6-8 weeks) were purchased from Liaoning Changsheng Biotechnology Co., LTD. The animals were divided into 6 groups (each group, n = 5): blank control group, model control group, two compound 1b treatment groups (concentrations: 5% and 7% W/V), vehicle control group treated with the vehicle of compound 1b solution and positive control group treated with 5% Lifitegrast. The dry eye model of mice was created by intelligent environment control system combined with drug induction. Two-step dehumidification method was used to control the environmental humidity, The relative humidity in the room was reduced to 40 ± 5% using the temperature dehumidifier, and the intelligent dehumidifier (dehumidification range RH 10-80%) is used to further reduce the humidity of the cabinet. At the same time, the air flow in the cabinet were maintained at 2.1 ± 0.2 m/s by the ventilator located 20 cm from the mouse cages in the cabinet. Through the above measures, humidity, temperature and air flow are effectively regulated. C57/BL6 mice were exposed to dry conditions (RH = 15; WV = 2 m/s; T = 21-23°C) for 14 days, subcutaneously were injected to hindquarters alternately with scopolamine hydrobrominde (0.5 mg/0.2 ml) three times a day (9:00 am, 1:00 PM, 4:00 pm) for 14 days to induce dry eye symptoms [30] . At the same time of dry eye development, the treatment groups mice were administrated on the surface of the both eyeball with 5 µL compound 1b eye drops, or treated with Lifitegrast (5%, USA) individually, and treatment with the vehicle of compound 1b solution was used as vehicle control group. There were 5 mice in each group, and twice a day for topical administration of drugs (at 9:00 am and 3:00 pm, respectively).

Periodic acid-Schiff (PAS) and TUNEL staining were performed on frozen section in this study. The animals were sacrificed, and the eyeballs and appendages (n = 10 eyes/ group) were surgically excised and then embedded by OCT (VWR, Suwanee, GA, USA). The central sagittal plane of the eyeball was sectioned with 6 µm thickness and stored at −80°C [31] .

The mice were in an unanesthetized state, 2 µL of 1% fluorescein sodium solution was dropped into the conjunctival sac of mice using a micropipette. The cornea was examined using slit-lamp microscopy in cobalt blue light 2 min after fluorescein instillation. The damaged areas of corneal epithelium were stained in green, the stained areas were assessed according to a standardized grading system ranging from 0 to 3, with the corneal surface divided into five regions, 0 score: no dyeing; 1 score: few punctate dyeing; 2 score: diffuse punctate dyeing; 3 score: flake dyeing [32] . Then the total score from the five regions was analyzed.

Ocular sections were cut at the center of the eye, where the lens has its maximum diameter. Sections (6 µm) were stained by periodic-Schiff (PAS) reagent for measuring goblet cell density. The sections were observed and taken for counting with a digital camera (Eclipse Ni-U; Nikon). Goblet cell density was measured in the superior and inferior conjunctival of the eyes of each group. And using image-analysis software (NIS Elements Software, Nikon) and expressed as number of goblet cells per mm [33] .

Frozen slices of 6 μm were used for the terminal deoxynucleotidyl transferase-mediated FITC-dUTP nick end labeling (TUNEL) apoptosis Assay kit (C1089, Beyotime Institute of Biotechnology, Haimen, China) detection. According to the manufacture's protocol, sections were fixed in 4% paraformaldehyde for 15 min at room temperature, permeabilized in 0.3% Triton X-100 for 3 min at 4°C. Then the frozen slices were incubated with TUNEL assay at room temperature for 60 min, and the slices were then stained with DAPI (0.5 μg/ml, staining the nucleus) for 5 min. After washing with PBS, the sections were visualized, and the images were captured on an inverted fluorescent microscope (×400 amplification, Nikon, ECLIPSE Ni, Japan). The central corneal epithelial field of 200 µm was selected, and the number of TUNEL positive cells was counted and recorded [34] .

Data are expressed as the mean ± standard error of the mean (SEM) of at least three independent experiments. The data were analyzed by one-way analysis of variance (ANOVA) with SPSS 20.0 (IBM). The histograms were plotted with Graphpad prism 8. Image J was used to show the image and calculate the gray value of protein bands in western blot.

m, 1H, Ar-H), 7.70-7.65 (m, 1H, Ar-H), 7.50-7.36 (m, 1H, Ar-H), 7.36-7.26 (m, 1H, Ar-H), 7.14-7.10 (m, 1H, Ar-H

739 (1C, Ar), 135.114 (1C, Ar), 132.110 (1C, Ar)

00 mL), then the mixture was stirred at 25°C for 0.5 h. TLC showed the reaction was complete. The solution was poured into water (10.0 mL) and DCM (10.0 mL) was added. The organic phase was separated and the aqueous was extracted with DCM (2 × 10.0 mL). The organic phase were dried over magnesium sulfate, filtered and concentrated in vacuum. The crude product was triturated with EtOAc

mmol, 33.9% yield) was obtained as a white solid. 1 HNMR (DMSO-d 6 , 400 MHz

90 (s, 1H, H-6)

Diphenylmethylene)amino)-3-(2-oxo-1,2,3,4-tetrahydroquinolin-7-yl)ethyl propanoate (49)

NaH (166 mg, 4.16 mmol, 60% purity) was added at 0°C, then the mixture was strried at 25°C for 5 h. TLC showed the reaction was complete. The solution was poured into water (5.00 mL) and DCM (5.00 mL) was added. The organic phase was separated and the aqueous was extracted with DCM (2 × 5.00 mL). The organic phase were dried over magnesium sulfate, filtered and concentrated in vacuum. The crude product was purified by column chromatography. Compound 49 (0.550 g, 1.29 mmol, 61.9% yield. 1 HNMR (DMSO-d 6

The mixture extracted with EtOAc (5.00 mL). The organic phase was adjusted to pH = 7-8 with NaHCO 3 , extracted with DCM (10 × 5.00 mL). The combined organic phase was washed with brine (20.0 mL), dried over Na 2 SO 4 , filtered and concentrated in vacuum

00 mL), was stirred at 70°C for 2 h, until the mixture was concentrated. A solution of 3-(3-acetyl-4-hydroxyphenyl)-2-aminopropanoic acid 53 (188 mg, 845 µmol) in DMF (3 mL) and TEA (233 mg, 2.31 mmol) were added, and the reaction mixture was stirred at 20°C for 2 h. TLC indicated reactant 25 was consumed completely. The mixture was poured into citric acid (10 mL), extracted with EtOAc (20 mL), concentrated to give crude product. The residue was purified by prep-HPLC

Ar-OH), 3.76 (s, 2H, CH 2 N), 3.16 (d, J = 4.8 Hz, 1H, * CH 2 CH 2 N), 2.95 (dd, J = 9.6, 13.8 Hz, 1H, * CH 2 CH 2 N), 2.81 (m, 2H, CH 2 CH), 2.61 (s, 3H, CH 3 CO). 13 C NMR (DMSO-d 6 , 101 MHz

Benzofuran-6-carbonyl)-5,7-dichloro-1,2,3,4-tetrahydroisoquinoline-6-carboxamido)-3-(3,4-dihydroxyphenyl)propanoic acid (1j)

mL), was stirred at 70°C for 2 h, until the mixture was concentrated. A solution of 2-amino-3-(3,4-dihydroxyphenyl)propanoic acid 54 (166 mg, 845 μmol) in DMF (3 mL) and TEA (233 mg, 2.31 mmol) were added, and the reaction mixture was stirred at 20°C for 2 h. TLC indicated reactant 25 was consumed completely. The mixture was poured into citric acid (10 mL), extracted with EtOAc (20 mL), concentrated to give crude product. The residue was purified by prep-HPLC

34(s, 1H, Ar-H) 7.02 (s, 1H, Ar-H), 6.68 (s, 1H, * CH=CHO), 6.62 (d, J = 7.6 Hz, 1H, Ar-H), 6.54 (d, J = 7.6 Hz, 1H, Ar-H), 4.74 (s, 2H, CH 2 N), 4.65-4.55 (m, 1H, CHCO), 3.76 (s, 2H, CH 2 N), 2.95 (dd, J = 5.6, 14.0 Hz, 1H, * CH 2 CH 2 N), 2.89-2.82 (m, 1H, * CH 2 CH 2 N), 2.82-2.78 (m, 2H, CH 2 CH). 13 C NMR (DMSO-d 6

Associations between signs and symptoms of dry eye disease: a systematic review

Dry eye disease: prevalence, assessment, and management. Home Healthc Now

The relationship between clinical signs and dry eye symptoms

Tear film osmolarity and dry eye disease: a review of the literature

Purified intercellular adhesion molecule-1 (ICAM-1) is a ligand for lymphocyte function-associated antigen 1 (LFA-1)

ICAM-1 a ligand for LFA-1-dependent adhesion of B, T and myeloid cells

LFA-1 in T Cell Migration and Differentiation

LFA-1/ICAM-1 Interaction as a Therapeutic Target in Dry Eye Disease

Lifitegrast, a Novel Integrin Antagonist for Treatment of Dry Eye Disease

Lifitegrast Ophthalmic Solution 5%: A Review in Dry Eye Disease

Lifitegrast Ophthalmic Solution 5.0% for Treatment of Dry Eye Disease: Results of the OPUS-1 Phase 3 Study

Dry eye syndrome: developments and Lifitegrast in perspective

Discovery and Development of Potent LFA-1/ICAM-1 Antagonist SAR 1118 as an Ophthalmic Solution for Treating Dry Eye

Discovery of tetrahydroisoquinoline (THIQ) derivatives as potent and orally bioavailable LFA-1/ICAM-1 antagonists

Structure-activity relationship (SAR) of the α-amino acid residue of potent tetrahydroisoquinoline (THIQ)-derived LFA-1/ICAM-1 antagonists

Allosteric inhibition abrogates dysregulated LFA-1 activation: Structural insight into mechanisms of diminished immunologic disease

Synthesis, characterization and cytotoxicity of surface amino-functionalized water-dispersible multi-walled carbon nanotubes

Structure−Activity Studies for a Novel Series of N-(Arylethyl)-N-(1,2,3,4-tetrahydronaphthalen-1-ylmethyl)-Nmethylamines Possessing Dual 5-HT Uptake Inhibiting and α2-Antagonistic Activities

Comparison of inhibitory activity of isomeric triazolopyridine derivatives towards adenosine receptor subtypes or do similar structures reveal similar bioactivities?

Rapid Discovery of Highly Potent and Selective Inhibitors of Histone Deacetylase 8 Using Click Chemistry to Generate Candidate Libraries

Chemoselective deprotection of benzyl esters in the presence of benzyl ethers, benzyloxymethyl ethers and n-benzyl groups by catalytic transfer hydrogenation

Discovery of Potent Benzofuran Derived Diapophytoene Desaturase (CrtN) Inhibitors with Enhanced Oral Bioavailability for the Treatment of Methicillin-Resistant S. aureus (MRSA) Infections

-furanyl)pyrimidin-2-amines as Janus kinase 2 inhibitors

A Convenient Synthesis of 2-Alkylated 1,4-Benzenediols

Synthesis of DL-[3-11C] phenylalanine

Design and synthesis of diarylamines and diarylethers as cytotoxic antitumor agents

Palladium-catalyzed cyanation of hindered, electron-rich aryl triflates by zinc cyanide

Esculetin protects human corneal epithelial cells from oxidative stress through Nrf-2 signaling pathway

Goblet cells of the conjunctiva: A review of recent findings

Interferon-γ Exacerbates Dry Eye-Induced Apoptosis in Conjunctiva through Dual Apoptotic Pathways

Expression profiling of ascorbic Acid-Related transporters in human and mouse eyes

Report of the National Eye Institute/Industry workshop on Clinical Trials in Dry Eyes

A Murine Model of Dry Eye Induced by an Intelligently Controlled Environmental System

EGFR Inhibition Blocks Palmitic Acid-induced inflammation in cardiomyocytes and Prevents Hyperlipidemia-induced Cardiac Injury in Mice

Acknowledgements This research was funded by Technology Pro-