Green Light‐Triggered Intraocular Drug Release for Intravenous Chemotherapy of Retinoblastoma

Abstract Retinoblastoma is one of the most severe ocular diseases, of which current chemotherapy is limited to the repetitive intravitreal injections of chemotherapeutics. Systemic drug administration is a less invasive route; however, it is also less efficient for ocular drug delivery because of the existence of blood‐retinal barrier and systemic side effects. Here, a photoresponsive drug release system is reported, which is self‐assembled from photocleavable trigonal small molecules, to achieve light‐triggered intraocular drug accumulation. After intravenous injection of drug‐loaded nanocarriers, green light can trigger the disassembly of the nanocarriers in retinal blood vessels, which leads to intraocular drug release and accumulation to suppress retinoblastoma growth. This proof‐of‐concept study would advance the development of light‐triggered drug release systems for the intravenous treatment of eye diseases.


Cell Culture
Human retinoblastoma cells (WERI-Rb-1) and human umbilical vein endothelial cells (HUVECs) were purchased from Stem Cell Bank, Chinese Academy of Sciences.
WERI-Rb-1 were cultured in RPMI-1640 (Gibco) supplemented with 10% FBS (Gibco) and 100 unitsmL -1 antibiotics (Penicillin-Streptomycin, Gibco) at 37 °C in a 5% CO2 humidified atmosphere. HUVECs were cultivated in DMEM (Gibco) supplemented with 10% FBS and 100 unitsmL -1 antibiotics at 37 °C in a 5% CO2 4 humidified atmosphere. The WERI-Rb-1-GFP-luc cells were transfected based on previously reported method without modification. [1] Animals BALB/c nude mice (male, 4 weeks, 20-22 g) were obtained from the Experimental Animal Center of Fudan Universityand maintained in SPF condition with access to food and water ad libitum.All the animal experiments were performed in compliance with the criteria of the National Regulation of China for Care and Use of Laboratory Animals.

Synthesis of (DEAdcCM)3-TAEA (DTAEA)
Compound 2: DEACM (compound 1, 300 mg, 1.2 mmol) was dissolved in dry dichloromethane (DCM, 20 mL) in a duplex flask. Then acetic acid (83 µL, 1.44 mmol, 1.2 eq) and 4-(dimethylamino) pyridine (DMAP, 180 mg, 1.44 mmol, 1.2 eq) were added into the solution of DEACM. The mixture was cooled to 0 o C and protected with nitrogen gas. 1,3-Dicyclohexylcarbodiimide (DCC, 300 mg, 1.44 mmol, 1.2 eq) was added slowly into the mixture. After stirring for 10 min at 0 o C, the mixture was warmed up to room temperature and stirred for 12 h in the dark. The mixture was then ten-fold diluted by DCM and washed with 1.2 M HCl and saturated solution ofNaHCO3separately for three times. The organic layer was collected and dried over Na2SO4and concentrated under vacuum. The residue was purified on a chromatography column by using 20:1 DCM/MeOH (v/v) to give compound 1 as yellow powder (Yield: 88.6%). Compound 3: Compound 2 (311 mg, 1.1 mmol) and Lawesson's reagent (285 mg, 0.68 mmol, 0.62 eq) were dissolved in dry toluene (40 mL) and protected by nitrogen gas in the dark. The mixture was heated to 115 o C and refluxed for 12 h. The solvent was removed by rotary evaporation and the residue was loaded into silica column directly.
The product was eluted by DCM to give orange-yellow powder as the product (Yield: 77.6%).
Compound 4: Compound 3 (175mg, 0.57 mmol) and malononitrile (52 mg, 0.91 mmol) were dissolved in 4 mL ACN. The mixture was added into TEA (0.3 mL) and stirred for 2 h in the dark at room temperature. Thin layer chromatography was used to confirm the complete consumption of compound 3. Then AgNO3 (221.8 mg, 1.3 mmol) was added and stirred for 2 h. After filtration, the solvent was removed by rotary evaporation. The residue was purified on a chromatography column by using 1:1 Hexene/DCM (v/v) to give compound 4 as orange-red powder (Yield: 72.4%).
Compound 5: Compound 4 (140mg, 0.41 mmol) was dissolved in absolute ethanol (50 mL). Aqueous HCl (37%, 3.36 mL, 0.04 mol) was added slowly and the mixture was refluxed at 85 o C for 16 h in the dark under nitrogen gas. The solvent was removed under reduced pressure and purified on a chromatography column by using DCM to give compound 5 as orange powder (Yield: 88.2%). 6 Compound 6: Compound 5 (108 mg, 0.37 mmol) was dissolved in 10 mL dry DCM. DIPEA (0.71 mL, 4.1 mmol) was added, and the mixture was cooled to 0 o C in the dark.
After stirring for 15 min, the solution of 4-nitrophenyl chloroformate (0.83 g, 4.1 mmol) in 5 mL dry DCM was dropwise added into the mixture. The resulting mixture was allowed to warm to room temperature and stirred for 6 h. The mixture was washed by 0.01 M HCl solution (100 mL x 2). The organic layer was collected and evaporated under reduced pressure. The residue was purified on a chromatography column by using 20:1 DCM/ethyl acetate (v/v) to give compound 6 as red powder (Yield: 90.8%).
Compound 7 (DTAEA): Compound 6 (153 mg, 0.33 mmol) was dissolved in 1.5 mL dry DCM under nitrogen gas and cooled to 0 o C. DIPEA (105 μL, 0.6 mmol) was added and stirred for 15 min. The solution of TAEA (15 μL, 0.1 mmol) in 1 mL dry DCM was slowly added into the above mixture at 0 o C. The resulting mixture was allowed to warm to room temperature. After stirring for about 1 h, a small amount of precipitation can be observed. Then, more DIPEA (105 μL, 0.6 mmol) was added and the mixture was stirred overnight. Thin layer chromatography was used to confirm the complete consumption of compound 6. Then the residue was evaporated under reduced pressure and loaded on a chromatography column. DCM/MeOH (0% to 4%) was used to elute the final product as orange-yellow powder (Yield: 63.4%).

Fabrication and characterization of DTNPs
The nanoparticles were fabricated by flash nanoprecipitation method that was reported previouslywith modifications [19,20] . Briefly, DTAEA (10 mgmL -1 ) and DSPE-mPEG2000 (20 mgmL -1 ) were dissolved in DMSO separately and then mixed at the weight ratio of 10:1 to form a stock solution. The stork solution (5 μL) was then added into 200 μL of filtrated water with vortexing. The resulting orange solution was further sonicated for 5 min at room temperature to give uniform nanoparticles. Then the solution was purified via a differential centrifugation method by ultrahigh-speed low-temperature centrifugate (ST 8R, Thermo Fisher Scientific, Waltham, MA, USA).
The solution of nanoparticles was first centrifuged at 3000×g for 10 min and the precipitation was removed, the process was repeated for three times or more until no precipitate was observed. The supernatant was then collected and further centrifuged at 30000×g for 20 min. The precipitated nanoparticles were washed for three times or more with water. The nanoparticles were finally collected as precipitates and dispersed into water or PBS. The concentration of DTAEA in DTNPs was determined by HPLC.
DTNPs were imaged by Philips CM100 transmission electron microscope. The size distribution and zeta potential of DTNPs were measured by dynamic light scattering (ZetasizerNano ZS90, Malven Instrument, southborough, MA, USA).

Drug loading in DTNPs
DOX can be encapsulated into DTNPs via co-assembly method during the flash nanoprecipitation of nanoparticles. [2,3] Hydrophobic DOX was obtained by the 8 conversion of commercial DOX hydrochloride with addition of TEA. Briefly, 10 mg DOX hydrochloride was dissolved in 1 mL DMSO and 2-fold molar of TEA was added.
The solution was kept stirring overnight. Subsequently, the solution was mixed with the stock solution of DTAEA and DSPE-mPEG2000 at different weight ratios (from 5% to 120%, w/w of DOX/DTAEA), in which the amount of DSPE-mPEG was 10% (weight ratio) of DTAEA. Drug-loaded nanoparticles (DOX/DTNPs) were fabricated and purified based on the fabrication protocol of DTNPs. The drug loading at different drug-to-material ratios were evaluated. To quantitatively determine the drug loading, the purified DOX/DTNPs was diluted and analysed by HPLC. The concentration of DTAEA and DOX in the nanoparticle solution was calculated based on a concentration-peak area standard curve. Drug loading capacity and encapsulation efficiency were calculated as follow:
The solution was added into a 1.5 mL vial and irradiated by green light (505 nm, 50 mWcm -2 , 0-5 min). Samples at different time points (irradiated for 0 min, 1 min, 3 min, and 5 min) were collected and directly analyzed by HPLC (C18 column, Poroshell 120, EC-C18, 2.7 μm). The mobile phase consisted of ACN and H2O with 0.1% TFA. The percentage of ACN was increased from 30% to 100% within 10 min with a flow rate at 1.5 mLmin -1 . For WERI-Rb-1 cells, cell viabilities were evaluated by CCK8 assay. WERI-Rb-1 cells were plated in 24-well plates at a density of 50000 cells per well. Various formulations were added into the culture medium and continued to incubate for 24 h. Cells were then collected by centrifugation (1000 g, 5 min) and washed with PBS for 3 times. The collected cells were re-dispersed in a new 24-well plate with fresh culture medium.

Real
CCK8 solutions (100 μL per well) were added and OD450 values were measured after 4 h of incubation for the calculation of cell viability.

Extravasation of DOX across endothelial monolayer
The in vitro inner BRB model was constructed based on the previously reported Transwell® method. [4] HUVECs were seeded onto the upper side of the 6.5 mm

In vivo biodistribution
Orthotopic retinoblastoma-bearing mice were established as the reported method. [1] Briefly, male BALB/c nude mice were anesthetized with 40 mgkg -1 pentobarbital sodium by intraperitoneal injection. Ocular local anesthesia was conducted by topical application of 0.4% oxybuprocaine hydrochloride, followed by 0.5% tropicamide to dilate the pupil. The retinoblastoma cells were suspended in PBS at a density of 10000 cells perμL. Two μL of the cell solution was loaded into the micro syringe (33 G, Hamilton TM , Thermo Fisher Scientific Inc., Waltham, MA, USA), which was slowly injected into the bottom of the vitreous cavity of the right eye. After removing the micro syringe, 0.5% chloramphenicol was topically administrated. The mice were further kept in SPF condition for 5-7 days until vitreous turbidities were observed.
Three groups (n = 3) of the tumor-bearing mice were intravenously injected with free DOX andDOX/DTNPs (5 mgkg -1 on the DOX basis) with or without the light irradiation (505 nm, 50 mWcm -2 , 5 min), separately.The light irradiation was performed post-injection at the tumor-bearing right eyes. The combined fluorescence of DEAdcCM and DOX from DOX/DTNPs was measured by IVIS (CailperPerkinElemer, United States). The mice were then euthanized. Organs including heart, lung, liver, spleen, kidney, as well as the eyes, were excised and washed with PBS for 3 times before the ex vivo fluorescence imaging by IVIS.

Eye irritation and retinal reactions
To evaluate the eye irritation of the formulations and light irradiation, BALB/c nude mice were intravenously injected with DOX/DTNPs or DOX/DTNPs + hv, separately. 13 The dose of nanoparticles was set as 5 mgkg -1 on the basis of DOX, and the light irradiation was set at 50mWcm -2 , 5 min. The whole-eye photographsand fundus images of retinal blood vessels of the eyes were collected 1 h post-treatment with a Phoenix Micron IV retinal imaging microscope (Phoenix Technology Group, LLC.CA, USA).
The mice were then euthanized, and the cornea and retina were excised and stained by H&E assay for histological analysis.

Anti-tumor effects in orthotopic retinoblastoma tumor model
The bioluminescence monitorable and orthotopic retinoblastoma-bearing mice were established by using WERI-Rb-1-GFP-luc cells during tumor implantation. The intraocular bioluminescence at the right eyes of the mice was detected 1 week aftertumor implantation. To induce bioluminescence, tumor-bearing mice were intraperitoneally injected with 150 mgkg -1 D-luciferin and anesthetized with isoflurane before IVIS imaging. Sixteen tumor-bearing mice with approximative bioluminescence intensities were selected and randomly divided into four groups (n = 4) and intravenously injected with saline, free DOX, DOX/DTNPs and DOX/DTNPs+ hv, separately. The dosage of drug administration was precisely controlled based on the body weight of mice and set as 5 mgkg -1 of body weight. The injection of the formulations as well as the light irradiation after the injection were applied every three daysfrom day 0 to day 12 for 5 times . The mice were continued to be fed in a dark room and monitored until day 25. In vivo imaging of bioluminescence was carried out on day 0, 7, 15, 20, 25. After the treatment, the mice were euthanized on day 25. The treated eyes were fixed in 4% paraformaldehyde and stained with hematoxylin-eosin (H&E) for further observation and histology analysis.

Statistical analysis
All experiments were conducted three times or more independently. Data were presented as the mean ± standard deviation (S.D.). The one-way ANOVA method and Independent-samples t-test were adopted to determine the statistical significance of differences using GraphPad Prism 8.0.2 software. The criterion for statistical significance was taken as*p < 0.05, **p < 0.01 and ***p < 0.001. Figure            Data were shown as means ± SD (n = 5).