Nanomicelle‐Assisted Targeted Ocular Delivery with Enhanced Antiinflammatory Efficacy In Vivo

Abstract Ocular inflammations are common diseases that may lead to serious vision‐threatening obstacles. Eye drops for antiinflammation therapy need to be administered multiple times daily at a high dosage due to the rapid precorneal removal and low bioavailability of drugs. To overcome these problems, a cRGD‐functionalized DSPE‐PEG2000 nanomicelle (DSPE‐PEG2000‐cRGD) encapsulated with flurbiprofen is proposed. The tailored nanomicelles trigger specific binding to integrin receptors on the ocular surface, which leads to rapid and robust mucoadhesion, superior ocular surface retention, and transcorneal penetration behaviors of nanomicelles. Due to the enhanced drug delivery on ocular surface and in aqueous humor, the functionalized nanoformulation significantly improves ocular antiinflammation efficacy at a low dosage by blocking the synthesis of inflammatory mediators and cytokines. The present study demonstrates a promising strategy that uses a functional peptide combined with nanomicelles for targeted delivery to the eye in ophthalmologic applications.

Free drugs were excluded by dialyzing the micelle solution against water through dialysis membrane tubes (molecular weight cut-off 6-8 KDa, Millipore) (Millipore, USA). The preparation process for the cornea-targeting peptide-functionalized nanomicelles (CTFM-FBP) was the same as that for M-FBP except that the DSPE-PEG 2000 was replaced with a mixture of DSPE-PEG 2000 and DSPE-PEG-cRGD (4:1, wt/wt). For the in vitro confocal microscopy and in vivo fluorescence microscopy, the fluorescence probe coumarin 6 was used as a model drug and was loaded into micelles using the above procedure.

Characterization of the nanomicelles
The amount of FBP in the nanomicelles was quantified using a high-performance liquid chromatograph (HPLC) system (CBM-20A, Shimazu, Japan) equipped with a UV detector. A C 18 column with a guard column was utilized. The mobile phase was acetonitrile and KH 2 PO 4 (pH 4.5, 0.01 M) buffer (55:45, vol/vol) at a flow rate of 1 mL/min and a 250 nm detection wavelength. The drug-loading capacity and drug encapsulation efficiency were calculated using the following formula: drug-loading capacity (DLC, %) =W 1 /W 2 ×100 % drug encapsulation efficiency (EE, %) = W 1 /W 0 ×100 % where W 1 is the amount of FBP that was loaded into nanomicelles, W 2 is the total amount of drug loaded nanomicelles, and W 0 is the initial fed amount of FBP. Morphology and size characterizations of the nanomicelles were conducted using low-acceleration-voltage TEM (HT7700, Hitachi). 5 L of each micelle solution (micelle concentration of 0.5 mg/mL) was dropped onto a carbon film-supported copper grid. After the samples were half-air dried, 1 % (wt/vol) uranyl acetate was used to stain the samples for 5-10 minutes. The excess staining reagent was removed and followed air-drying before TEM observation. The zeta potential and size distribution of the nanomicelles were determined using dynamic light scattering (DLS, Nano-ZS, Malvern).

Stability studies
The physical stability of M-FBP and CTFM-FBP nanomicelle solution (water solution) was evaluated after storage for 3 months under different temperature. Exact volumes of each solution was stored in glass bottles and placed at 4 and 25 ℃. After stored for 1 and 3 months, the particle size, morphology, zeta potential and drug encapsulation efficiency were determined.
For the stability study of nanomicelles in artificial tears, both M-FBP and CTFM-FBP nanomicelles were firstly dispersed in artificial tears (FBP concentration: 0.3 mg/mL). The size, morphology, zeta potential and drug encapsulation properties of the nanomicelles were analyzed after different storage time (1, 4, 7 days) at 4 and 25 .

In vitro drug release in artificial tears
The artificial tears were prepared with PBS buffer (0.01 M), 0.03 mM CaCl 2 , 1 mg/mL lysozyme and 0.1 mg/mL bovine serum albumin. 1 The release of FBP from nanomicelles was performed with dialysis membrane tubes (molecular weight cut-off 6-8 KDa, Millipore). First, the nanomicelle solution was dispersed in artificial tears. Then, 150 L nanomicelle solution was transferred to each dialysis membrane tube. The tubes were put in a container filled with 2 L artificial tears and stirred at 37 for 12 hours. At 0.25, 0.5, 1, 2, 4, 6, 9 and 12 hour time points, three tubes were collected and the residual FBP in each tube was measured using the HPLC method mentioned above. The released FBP was calculated using the initial fed FBP minus the residual FBP in each tube.

Cell culture
The human corneal epithelial cell line (HCEC) HCE-2 was purchased from the American Type Culture Collection and was tested negative for mycoplasma before use (Mycoplasma Detection Kit, Lonza). The primary rabbit corneal epithelial cells (RCECs) were obtained according to a modified Chan and Haschke's method. 2 Briefly, eyes from New Zealand albino rabbits (1.5 Kg) were enucleated and placed in sterile PBS buffer with 400 units/mL penicillin and 400g/mL streptomycin for 30 min. Then, the corneal epithelium was peeled from the stroma and along the limbus with an iris restorer and fine forceps. The dissected corneal epithelium was cut into small pieces and incubated with 0.05 % trypsin (Gibco, USA) for 30 min at 37 . The cells were harvested and washed by centrifugation at 1000 rpm for 5 min. All steps were performed under sterile conditions. The culture media for the HCEC and RCEC cells were the same, including a 1:1 mixture of Dulbecco's modified eagle medium (DMEM) and Ham's F12 medium (Gibco), 5 µg/ml insulin (Sigma), 100 ng/mL hydrocortisone (J&K Co.), 5 ng/mL epidermal growth factor (Invitrogen, CA), 100 unit/mL penicillin, 100 g/mL streptomycin, and 10 % fetal bovine serum (Gibco). All cells were cultured in an incubator at 37 with 5 % CO 2 .

Immunofluorescence
The immunofluorescence methodology was used to determine the expression of integrin  1 on the HCECs and RCECs (n = 3). First, HCECs and passage 3 RCECs were cultured in complete medium to 70 % confluence. Then, the cells were fixed with 4 % paraformaldehyde for 10 min, washed with PBS buffer 3 times, and blocked with 10 % BSA for 1 hour at room temperature to block nonspecific conjugation sites. The cells were incubated with primary monoclonal mouse anti-integrin 1 antibody (1:50, ab30394, Abcam, USA) at 4 overnight. After washing with PBS 3 times, the cells were incubated with secondary FITC-conjugated goat anti-mouse antibody (1:500, ZSBG-bio, China) for 1 hour at room temperature. Cells treated with secondary antibody only were used as controls. The cell nuclei were stained with 5 g/mL DAPI (Invitrogen) for 30 min at 37 and observation was conducted using confocal laser scanning microscopy (LSM700, Carl Zeiss). For flow cytometry, cells without nuclear staining were harvested using trypsin digestion and analyzed using an Attune® acoustic focusing cytometer (Life Technologies, CA) with 488 nm excitation. All of the experiments were carried out in triplicate.

Western blot assay
The quantitative analysis of integrin  1 expression was conducted using a western blot assay (n = 3). The whole cell proteins were extracted from HCECs and RCECs using a RIPA lysis buffer (Beyotime, China) with protease inhibitors (Roche). The protein concentration was determined using a BCA protein assay kit (Lot#23225, Thermo Fisher). A total of 50 g protein was used for SDS-PAGE with 10 % gel and was transferred to a nitrocellulose filter membrane (PALL) for blotting. The membrane was blocked with 5 % defatted milk for 1 hour at room temperature, incubated with primary mouse monoclonal anti-integrin  1 antibody (1:200, ab30394, Abcam, USA) overnight at 4 , and incubated with secondary horseradish peroxidase-conjugated antibody (1:5000, ZSBG-bio, China) for 1 hour at room temperature. The results were recorded and analyzed using a chemiluminescence imaging system (ChemiDoc MP, Biorad). All samples were assayed in triplicate.

In vitro nanomicelle binding study
Human corneal epithelial cells (HCECs) were planted in glass microscopy dishes and incubated in complete medium. To facilitate observation, coumarin 6 was loaded into the micelles instead of FBP. After the cells grew to 70 % confluence, the cells were washed with PBS twice and pre-incubated with different endocytosis inhibitors to reduce the cell internalization of coumarin 6.
Four endocytosis inhibitors were added into serum-free medium to pre-incubate the cells for 30 min at 37 : 5 M chlorpromazine (clathrin-mediated endocytosis inhibited), 8 M dynasore hydrate (dynamin-dependent endocytosis inhibited), 10 M cytochalasin D (macropinocytosis-mediated endocytosis inhibited), and 100 M genistein (caveolae-mediated endocytosis inhibited). Then, different formulations of nanomicelles were incubated with cells in complete medium with 1 g/mL coumarin 6 for 2 min. The nanomicelle solution was removed and the cells were fixed. The cell membranes were stained with 5 M Dil for 30 min. For the cRGD peptide competition study, cells were pre-incubated with excess cRGD peptide in 20 g/mL for 30 min at 37 before treatment with endocytosis inhibitors. The experiment was assayed in triplicate and the nanomicelle binding properties were observed using confocal microscopy.

Trans-corneal penetration study using 3D-cultured multilayer HCEC spheroids
Multilayer corneal epithelium spheroids were constructed with HCECs using methods described in the literature. 3 Briefly, a 0.8 % agarose solution was freshly prepared and heated. The 96-well plates were coated with 50 L of the agarose solution per well and sterilized for 30 min using UV irradiation. 100L of cell suspension (500 cells) were planted into each agarose-coated well and cultured in complete medium. The medium was changed every 2 days until the corneal epithelial spheroids grew to 200-250 m diameter. To evaluate the trans-corneal penetration ability of different formulations, HCEC spheroids were incubated with free coumarin 6 (free C6), coumarin 6 nanomicelles (M-C6) and functional coumarin 6 nanomicelles (CTFM-C6) at a final concentration of 2 g/mL coumarin 6 for 4 h (n = 5 spheroids). HCEC spheroids cultured with complete medium were used as the control. After treatment, the spheroids were rinsed with PBS and fixed with 4 % (wt/vol) formaldehyde solution for 20 min at room temperature. Then, spheroids were transferred into chambered coverslips and observed using two-photon confocal microscopy. Z-stack images were obtained by scanning the spheroids from top to bottom with 10 m thickness and a total 60 m depth (~15 images in each spheroid). To measure the penetration distance of each formulation, the spheroid image with the biggest diameter was selected and the penetration distance from spheroid edge to interior was measured using Adobe Photoshop CS 6 software. The relative penetration distance was defined as the ratio of nanomicelle penetration distances to free C6 penetration distances.

Animal experiments
All animal experiments were in compliance with guidelines for the care and use of laboratory animals and were approved by the institutional animal care and use committee of Wenzhou Medical University and the National Center for Nanoscience and Technology of China. The New Zealand male albino rabbits and Sprague Dawley○ R rats (Charles River, China) were housed under standard conditions with free access to food and water.

Ocular surface retention study
The ocular surface retention study was conducted to evaluate the retention of nanomicelles after a single instillation. Sprague Dawley○ R rats (500-800 g) were randomly divided into three groups (n=5). In each group, 1.5 L of either free C6, M-C6 or CFTM-C6 with a final C6 concentration of 2g/mL was instilled into the conjunctival sac of the left eye. The contralateral eye was used as untreated control. We used fluorescence microscopy for eye imaging at the initial instillation (0 h), 0.5 h, 1h, 2h, 3h and 4 h after instillation. The mean fluorescence intensities of areas inside the eyelid in each image were quantified by Image J software.

In vivo anti-inflammatory study
To study the anti-inflammatory effect of the FBP formulations, ocular inflammation was induced by topical administration of sodium arachidonate (SA). 4-6 New Zealand albino male rabbits As an important ocular inflammatory feature, the polymorphonuclear (PMN) leukocytes in tears and aqueous humor were counted using a hemocytometer. For tear collection, the ocular surface of the rabbit eye was washed with 50 L PBS from the upper eyelid to the conjunctiva three times and the tears and PBS were collected immediately. For aqueous humor collection, the rabbits were anesthetized with 2 % xylazine hydrochloride (wt/vol, 0.2 mL/Kg) and 3 % pentobarbital sodium (wt/vol, 1 mL/Kg) followed by topical anesthetic proparacaine hydrochloride eye drops (0.5 %, wt/vol). 150 L of aqueous humor from each rabbit was withdrawn by anterior chamber paracentesis with an insulin syringe. The PMN leukocytes were quantified immediately after sample collection.
All samples were assayed in triplicate and stored at -80 for ELISA measurement (n = 6).

ELISA
The changes in prostaglandin E 2 (PGE 2 ) and inflammatory cytokine IL-6 were measured using enzyme-linked immunoassay (ELISA). The mouse PGE 2 ELISA kit (R&D Systems, USA) and rabbit IL-6 ELISA kit (BD Biosciences, USA) were operated under manufacturers' instructions.

Ocular tolerance study
The potential ocular irritancy and toxicity of the FBP nanomicelle formulation were evaluated using a modified Draize test. 9 Briefly, the New Zealand male albino rabbits were randomly divided into five groups (n = 3). 50 L of PBS (irritant negative control), 1 % SDS (wt/vol, irritant positive control), and three FBP formulations (FBP concentration of 0.3 mg/mL) were instilled into the left eye of rabbits every 30 min for 6 h. At 10 min, 6 h and 24 h after the last instillation, all eyes were examined using a slit lamp microscope and scored according to the scoring system above. In the last examination at 24 h, all experimental eyes were evaluated using optical coherence tomography (OCT) and H&E staining for histological examination. All experiments were carried out in triplicate.

OCT examination
A custom-built ultra-high resolution OCT (UHR-OCT) instrument provided by Wenzhou Medical University was used to assess changes of topographic thickness of the cornea and retina.
Male New Zealand albino rabbits were anaesthetized prior to OCT examination.

Histological examination
To evaluate the influence of different formulations on normal eye tissues, the New Zealand albino rabbits were sacrificed by intravenous air injection after the ocular tolerance experiment and the eyes were enucleated immediately. The eye balls were cleaned and washed with PBS and fixed in 10 % formalin solution (vol/vol) for 2 days. Then, the eye balls were dehydrated in a gradient alcohol series and embedded in paraffin. 5 m cross sections were cut and stained with hematoxylin and eosin (H&E) for histopathological evaluation (~5 slices for each eye).

Statistical analysis
All data are presented as mean ± s.e.m. in this work. Statistical analysis was conducted using the Graphpad Prism software. Statistical significance was determined by Student's t-test, one-way ANOVA and two-way ANOVA test set at P < 0.05.

Data availability
All data in this manuscript are available from corresponding authors upon request.

Table S2
Clinical scores of ocular inflammation in the iris of rabbit eyes, examined using a slit lamp (n = 6).