Sunitinib-loaded MPEG-PCL Micelles for the Treatment of Age-Related Macular Degeneration

Age-related macular degeneration (AMD) will be responsible for the vision impairment of more than five million late-aged adults in the next 30 years. Current treatment includes frequent intravitreal injections of anti-vascular endothelial growth factor (VEGF) agents. However, there are methods of drug delivery that can decrease the frequency of intravitreal injections by sustaining drug release. MPEG-PCL ((methoxypoly(ethylene glycol) poly(caprolactone)) has been reported as biocompatible and biodegradable. Polymeric micelles of MPEG-PCL can be useful in efficiently delivering anti-VEGF drugs such as sunitinib to the posterior segment of the eye. In this study, the novel micellar formulation exhibited an average dynamic light scattering (DLS) particle size of 134.2 ± 2.3 nm with a zeta potential of −0.159 ± 0.07 mV. TEM imaging further confirmed the nanoscopic size of the micelles. A sunitinib malate (SM)-MPEG-PCL formulation exhibited a sustained release profile for up to seven days with an overall release percentage of 95.56 ± 2.7%. In addition to their miniscule size, the SM-MPEG-PCL formulation showed minimal cytotoxicity onto the ARPE-19 human retinal pigment epithelial cell line, reporting a percent viability of more than 88% for all concentrations tested at time intervals of 24 h. The SM-MPEG-PCL micelles also exhibited exceptional performance during an anti-VEGF ELISA that decreased the overall VEGF protein expression in the cells across a 24–72 h period. Furthermore, it can be concluded that this type of polymeric vehicle is a promising solution to symptoms caused by AMD and improving the management of those suffering from AMD.

ii iii

Introduction
Age-Related Macular Degeneration, or AMD is an eye disease that gradually deteriorates the macula and is the leading causes of vision failure among late-aged adults (2,17). The macula is located in the posterior segment of the eye near the retinal center and it is a key component for central vision processing using its photoreceptor cells (1). Macular degradation as a result of AMD disrupts the process of normal vision processing in the optic nerve causing blurry and distorted vision. It is projected that by 2050, more than 5 million Americans will have some form of AMD (2).
Vascular Endothelial Growth Factor, or VEGF is a protein that causes the formation of blood vessels within the eye as a result of AMD. Anti-VEGF drugs inhibit the expression of VEGF at the angiogenic site to prevent central vision damage induced by hyperoxic conditions (22). Generally, anti-VEGF drugs such as Axitinib, Pazopanib, and Sorafenib are available for use in cancer treatments, but recently have been propitious agents for treating AMD-induced neovascularization (11)(12). Despite the success of these angiogenic inhibitors, a newer drug named Sunitinib has proved to be a notable contender for the treatment of ocular diseases (14)(15).
Sunitinib malate is a tyrosine kinase inhibitor that exhibits anti-VEGF and anti-HIF properties (16). Originally approved for treatment of renal cell carcinoma, Sunitinib has displayed usefulness in the treatment of ocular diseases like AMD that depend on VEGF or PDGFR receptors (3). Sunitinib can specifically inhibit all three types of VEGF receptors (VEGFR1-VEGFR3) along with seven other important components of cell proliferation (3).
Notably, VEGFR1, VEGFR2, and platelet-derived growth factor beta, or PDGFR(beta) are the main integrants of angiogenesis in the body. Additionally, Sunitinib can inhibit eight total tyrosine kinase receptor proteins, such as Kit, Flt-3, and CSF-1R (28). Because Sunitinib is able to terminate tyrosine kinase binding, it can potentially inhibit ocular angiogenesis, a common symptom of AMD (4). Nonetheless, it is noted that the use of sunitinib is restricted due to its toxicity among higher dosages (14)(15).
However, there are significant physical barriers including the blood-retinal barrier to reaching the posterior segment of the eye where the retina and macula are housed (5,(23)(24).
Notably, Poly(ethylene glycol)-block-poly (-coprolactone), or PEG-PCL copolymers have been recognized for their low toxicity, biocompatibility, and ability to aid in sustained drug release (6). PEG, along with PEG derivatives like PEG-PCL have been approved by the FDA for medical uses (7). Optimal biodistribution of PEG-PCL nanoparticles was observed in vivo and ex vivo studies using parameters between 20-200nm in size and molecules with a negative zeta potential indicating that circulation time of drugs carried by PEG-PCL copolymers can be greatly extended reducing the frequency of administration (8). PEG-PCL micelles for delivery of anti-VEGF drug Axitinib also expressed its usefulness in inhibiting angiogenesis while displaying exceptional cytocompatibility (9,27). These advancements are promising in fulfilling the need to decrease frequent intravitreal injections and significantly improve patient compliance (18,(25)(26).
Consequently, MPEG-PCL micelles in conjunction with Sunitinib appears to be a viable option considering its ability to inhibit eight receptors in the body responsible for cell proliferation and angiogenesis (10,27). This study will examine the effectiveness of sunitinibloaded MPEG-PCL micelles by evaluating their morphological characterization, in vitro drug release, cytotoxicity and VEGF expression.

Cell Culture
Cell line studies were carried out using ARPE-19 (ATCC© CRL2302™) cells derived from human retinal epithelial cell line. Nourishment for the cells were provided from complete media, a combination of DMEM F12 medium, 10% v/v Fetal Bovine Serum (FBS), 1% 10000 u/ml penicillin antibiotic. Additionally, cells were stored in a moisture-controlled CO 2 incubator set at 37℃.

Formulation of SM-MPEG-PCL Micelles
A mass of 2 mg of Sunitinib malate and 200l of DMSO were dissolved into a 5 ml centrifuge tube. This mixture is sonicated for eight minutes to achieve full dissolution of the drug into the organic solvent. Afterwards, 1.8 ml of acetone along with 20 mg of Methoxy poly(ethylene glycol)-block-poly (-coprolactone), or MPEG-PCL polymer is added to the DMSO/sunitinib mixture and sonicated again to fully dissolve the drug. Mixture is then vortexed to mix the two solvents as this will be the 2 ml organic solvent component. The accompanying aqueous phase is 4 ml of filtered distilled water. Under magnetic stirring, the 2 ml organic phase is added dropwise using a 23G syringe to the aqueous phase. The mixture is then stirred magnetically for 24 hours at 700 rpm to allow for adequate precipitation of organic solvents.
Following 24 hours, the formulation is then centrifuged at 5000 rpm for 15 minutes to extrapolate any free drug or polymer as seen in Figure 1C.  (32). Similar to size characterization, zeta potential data is also measured in triplicates.

Differential Scanning Calorimetry
Differential scanning calorimetry analysis, or DSC was executed using a DSC Q 20 instrument (TA Instruments, New Castle, DE, USA, Q series Q20-2288-DSC software). Four samples were analyzed beginning with polymer, pure sunitinib drug, physical mixture of sunitinib and MPEG-PCL polymer, and SM-MPEG-PCL formulation. Five milligrams of each sample was taken and placed inside of an aluminum pan specially made for DSC analysis (29).
Samples were then heated to approximately 30-300°C at a rate of 10°C while simultaneously under a nitrogen purge at a rate of 50 mL/min (30).

Transmission Electron Microscopy
Morphology of the SM-MPEG-PCL micelles were measured using a Transmission Electron Microscope, or TEM. To prepare the formulation for imaging, the sample was inserted onto a 200 Cu film square grid. The sample was then allowed to settle for approximately 10 minutes. After the sample was air dried, it is prepared for negative staining using 2% w/v phosphotungstic acid (PTA). Furthermore, images were obtained at 80,000X magnification at an accelerating voltage of 120kV.

Entrapment Efficiency
The SM-MPEG-PCL formulation was centrifuged at 5000 rpm for 15 minutes to separate any excess drug or polymer. Furthermore, the supernatant was collected and centrifuged once more at 18000 rpm for approximately 20 minutes to transform the nanoparticles into pellet form for further analysis. In order to prepare the formulation for UV spectrophotometer analysis, the pellet was washed three times with filtered distilled water to further ensure the removal of unentrapped drug. The nanoformulation was paired with methanol as a solvent and analyzed using a UV Spectrophotometer at a wavelength of 432 nm (λmax). From this data, the amount of drug within the sample was calculated and used in the equations below to determine the entrapment efficiency and loading efficiency percentages. Samples were diluted with methanol in preparation to be read using a UV Spectrophotometer at a wavelength of 432 nm.

Cytotoxicity Study
Cytotoxicity to dissolve the formazan crystals that indicate cell viability by displaying a distinct purple color.
The plate was then read using a UV spectrophotometer at a wavelength of 595 nm to obtain the absorbances of each well. For comparison, cells treated with DMEM-F12 were the negative control and 0.1% Triton X100 was the positive control respectively.

Anti-VEGF ELISA
In a sterile 96-well plate, 48 wells were seeded with a density of 5,000 cells per well and treated with complete media to induce confluency. After 24 hours, complete medium was removed and cells received treatment. Two treatments were used in the ELISA beginning with free drug solution and the SM-MPEG-PCL formulation. Complete and incomplete medium was used as negative controls and two treatments each at 10μM were treated into quadruplicates for 24, 48, and 72 hours (36). After 72 hours of incubation, VEGF expression was assessed using an Invitrogen Human VEGF-A Platinum ELISA Kit. Quantification of cell protein content was measured using a Pierce BCA Protein Assay Kit following the collection of cell lysate.
Quantitative analysis of samples were read using an ELISA plate reader at absorbances of 450 nm. The differences between the two absorbances were noted and used to determine the inhibitory effects of VEGF secretion using a standard curve.

Statistical Analysis
In order to quantize all data reported in this study, an extensive statistical analysis was done using the mean, standard deviation, and t-test.

Differential Scanning Calorimetry
The physiochemical properties of pure SM drug, pure MPEG-PCL polymer, physical mixture of polymer and SM, and SM-MPEG-PCL micelle formulation were analyzed using DSC analysis. Figure  indicative of its melting point ( Figure 2). In contrast, the DSC spectra for the SM-MPEG-PCL formulation did not exhibit a peak at this melting point, rather at a peak near 105.26℃ affirming the entrapment of SM in MPEG-PCL micelles and SM in an amorphous form (13). Furthermore, an endothermic peak shown at ~201℃ is seen in the physical mixture of SM and MPEG-PCL.

Size and Zeta Potential
Using a 1:10 drug to polymer ratio, DLS analysis reported the mean size of the 2k-2k sunitinib micelles were 167.8 nm with an average PDI of 0.211 (Table 1). Whereas, the zeta potential for 2k-2k sunitinib micelles was +4.34 mV. 5k-2k SM-MPEG-PCL micelles with a 1:10 drug to polymer ratio were seen to have an average size of 134.2 nm with an average PDI of 0.160. The zeta potential reported for this ratio was -0.159 mV ( Figure 3A). Because the 5k-2k MPEG-PCL polymer provided a more desirable size, a 1:7 ratio was also tested in hopes of a smaller particle size. The 1:7 ratio reported a mean particle size of 170.8 nm, ZP of +0.842 mV and a PDI of 0.104. Furthermore, a 1:15 ratio of the 5k-2k polymer was tested and showed an average size of 120.9 nm, ZP of +10.5 mV and a PDI of 0.149 (Table 1). The low PDI values indicate the homogeneity and monotonous distribution of the nanoparticles within the formulation (19). It is concluded that the 1:10 ratio of the 5k-2k polymer was the optimal molecular ratio to use for this experiment as the positive zeta potential values indicate the possibility of aggregation within the body (20). Table 2 displays the size difference between the 2k-2k & 5k-2k MPEG-PCL blank micelles for size comparison to compare the weightiness of SM drug onto the micelles.

Transmission Electron Microscopy
TEM images reported particle sizes ranging from 86.76 nm to 114.51 nm in contrast to the mean particle size of 134.2 nm reported from DLS ( Figure 3B). The discrepancy between the difference in particle sizes among both DLS and TEM are due to the differences in the way the formulation is processed. TEM analyzes the nanoparticles within a fixed copper grid, whereas DLS utilizes the hydrodynamic diameter of the micelles to report a mean particle size (21). TEM imaging of SM-MPEG-PCL micelles affirmed the size uniformity of the nanoparticles coinciding with the low PDI of the 5k-2k formulation. Figure 3B also exhibits the physical characteristics of the nanoparticles having a spherical structure and monotonous surface.
It is also accurate to infer from TEM imaging that the particles are in agreement with the negative zeta potential value and less likely to agglomerate. use for further studies despite its smaller particle size.

In Vitro Drug Release
The in vitro drug release study using the dialysis method compared the release capacity   Figure 5A). The viability of the drug solution was 68% for 24h and 61% inferring that the solution is toxic to the cells in higher concentrations.
Results for 48 hours also showed that the SM-MPEG-PCL formulation, along with the blank micelles are viable with percentages above 95% for all six concentrations ( Figure 5B). The MTT assay analysis reaffirmed the SM-MPEG-PCL and blank micellar formulations ability to show no significant cytotoxicity within the ARPE-19 cell line indicating that the formulation is suitable for further comprehensive studies.