*Article* **Retinal Delivery of the Protein Kinase C-**β **Inhibitor Ruboxistaurin Using Non-Invasive Nanoparticles of Polyamidoamine Dendrimers**

**Rehab A. Alshammari 1,2 , Fadilah S. Aleanizy 2,\*, Amal Aldarwesh <sup>3</sup> , Fulwah Y. Alqahtani <sup>2</sup> , Wael A. Mahdi <sup>2</sup> , Bushra Alquadeib <sup>2</sup> , Qamraa H. Alqahtani <sup>4</sup> , Nazrul Haq <sup>2</sup> , Faiyaz Shakeel <sup>2</sup> , Hosam G. Abdelhady <sup>5</sup> and Ibrahim A. Alsarra <sup>2</sup>**


**Abstract:** Ruboxistaurin (RBX) is an anti-vascular endothelial growth factor (anti-VEGF) agent that is used in the treatment of diabetic retinopathy and is mainly given intravitreally. To provide a safe and effective method for RBX administration, this study was designed to develop RBX nanoparticles using polyamidoamine (PAMAM) dendrimer generation 5 for the treatment of diabetic retinopathy. Drug loading efficiency, and in vitro release of proposed complexes of RBX: PAMAM dendrimers were determined and the complexation ratio that showed the highest possible loading efficiency was selected. The drug loading efficiency (%) of 1:1, 2.5:1, and 5:1 complexes was 89.2%, 96.4%, and 97.6%, respectively. Loading capacities of 1:1, 2.5:1, and 5:1 complexes were 1.6%, 4.0%, and 7.2% respectively. In comparison, the 5:1 complex showed the best results in the aforementioned measurements. The in vitro release studies showed that in 8 h, the RBX release from 1:1, 2.5:1, and 5:1 complexes was 37.5%, 35.9%, and 77.0%, respectively. In particular, 5:1 complex showed the highest drug release. In addition, particle size measurements showed that the diameter of empty PAMAM dendrimers was 214.9 ± 8.5 nm, whereas the diameters of loaded PAMAM dendrimers in 1:1, 2.5:1, 5:1 complexes were found to be 461.0 ± 6.4, 482.4 ± 12.5, and 420.0 ± 7.1 nm, respectively. Polydispersity index (PDI) showed that there were no significant changes in the PDI between the free and loaded PAMAM dendrimers. The zeta potential measurements showed that the free and loaded nanoparticles possessed neutral charges due to the presence of anionic and cationic terminal structures. Furthermore, the safety of this formulation was apparent on the viability of the MIO-M1 cell lines. This nanoformulation will improve the therapeutic outcomes of anti-VEGF therapy and the bioavailability of RBX to prevent vision loss in patients with diabetic retinopathy.

**Keywords:** polyamidoamine dendrimers; diabetic retinopathy; protein kinase C-β inhibitor; nanoparticles; ruboxistaurin

### **1. Introduction**

Diabetic retinopathy (DR) is a well-known consequence of poorly controlled hyperglycemia. It is responsible for significant blindness in diabetes mellitus (DM) patients [1]. When DR progresses to a vision-threatening complication, vitreous surgery and laser

Aleanizy, F.S.; Aldarwesh, A.; Alqahtani, F.Y.; Mahdi, W.A.; Alquadeib, B.; Alqahtani, Q.H.; Haq, N.; Shakeel, F.; Abdelhady, H.G.; et al. Retinal Delivery of the Protein Kinase C-β Inhibitor Ruboxistaurin Using Non-Invasive Nanoparticles of Polyamidoamine Dendrimers. *Pharmaceutics* **2022**, *14*, 1444. https://doi.org/10.3390/ pharmaceutics14071444

**Citation:** Alshammari, R.A.;

Academic Editors: Vibhuti Agrahari and Prashant Kumar

Received: 2 June 2022 Accepted: 6 July 2022 Published: 11 July 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

photocoagulation are the only effective procedures in restoring or maintaining the vision. Nevertheless, the laser-treated retina usually loses its function and develops scarring tissue [2]. The main drug classes used for the management of DR and other DM related complications are anti-vascular endothelial growth factor (anti-VEGF) drugs, protein kinase C (PKC) inhibitors, corticosteroids, and somatostatin analogues [1]. The development of anti-VEGF therapy as a new approach for treating proliferative DR (PDR) has recently been adopted [3]. Anti-VEGF agents have been identified to reduce or prevent neovascularization and vessel drainage through antagonization of VEGF's effects as they bind to VEGF, and prevent its cellular action [3–6]. Ranibizumab, bevacizumab, and pegaptanib intravitreal (IVT) injections are currently being used in the treatment of DR [1].

A modern anti-VEGF therapy which relies on using PKC ß-inhibitor has been developed for the treatment of DR. Ruboxistaurin (RBX) is an investigational drug that has recently attracted a great deal of attention as a potent drug for treating DR. It acts as a selective PKC ß-inhibitor that is still pending for the United States Food and Drug Administration (FDA) and European Medicines Agency (EMA) approval. In diabetic rats, RBX has been found to inhibit PKCβ activity when given orally or intravitreally [7,8]. The findings reported by Aldeweesh (2016) demonstrated that RBX is very potent at lowering VEGF release from human retinal macroglial Müller cells (MIO-M1) in the first 24 h under simulated conditions of hyperglycemia. Furthermore, prolonged exposure to RBX resulted in a continued reduction which could indicate the breakdown of VEGF over time [9]. This medicine could slow down or prevent DR progression and which will lead to better quality of life for DM patients. Therefore, this potential candidate needs to be formulated in an appropriate dosage form to be available upon approval by the FDA and EMA [10]. The ideal route for RBX administration can be attained through reaching the retina at a constant rate. In recent years, invasive drug delivery systems (DDSs) have been chosen to deliver anti-VEGFs to the retina, to reduce the risk of systemic side effects and provide better drug targeting for oral and systemic drug delivery [11]. Delivering RBX to the retina via invasive methods can successfully slow down the progression of DR by exposing the retina to a high level of RBX, but unfortunately, will lead to the appearance of unwanted side effects. As a consequence, these side effects will not be tolerated which affects the compliance of patients and thus minimizes the efficacy of the therapy. In this case, microvascular complications of DR remain advanced over time, causing vision to be distorted and quality of life to be reduced. Non-invasive dosage forms, however, are self-administered by the patient and therefore well tolerated as they do not involve injections or invasive equipment. Hopefully, they would increase the patient compliance toward the treatment without experiencing the ocular discomfort and complications associated with IVT injections [12]. If the non-invasive delivery of retinal drugs is achievable and efficient, RBX eye drops could be used as prophylactic approach to lessen the formation of new fragile vessels in the retina [13]. In recent years, researchers have been studying topical drug delivery for their potential in the treatment of retinal diseases [13,14]. The distribution of dexamethasone (DEX) in the ocular tissue was studied by Bessonova et al. (2011). In fact, DEX was found in the eyelid, cornea, aqueous humor, iris, lens, vitreous cavity, retina, and choroid at a range of concentrations following topical application. However, the highest levels of DEX were found in the anterior structures of the eye. This study has shown that DEX cannot be transported to the posterior chamber of the eye via topical application, resulting in a lack of therapeutic responses to treat retinal diseases [14].

Likewise, delivery of oral drugs has been investigated as a non-invasive alternative; however, it has largely been ineffective [15]. Hence, the ideal method is to develop a DDS that can passively deliver RBX from the ocular surface, through the layers of the eye and reside in the retinal tissues [11–13]. This can be achieved by the production of nanocarriers that are capable of encasing and delivering therapeutic agents to their site of action [11–13]. In the past years, a number of nanocarriers have been investigated for ocular drug delivery, especially degradable nanoparticles (NPs) made with polymers, such as dendrimers, liposomes, polylactic-co-glycolic acid (PLGA) nanoparticles, and chitosan nanoparticles [11–13]. These studies suggested that the properties of nanocarriers could influence their application in both

segments of the eye [11–13]. In fact, dendrimers are exceptional polymers, which offer various advantages over conventional linear or branched polymers [1]. These advantages are better solubility in common solvents, monodisperse distribution, controlled particle size, and higher area/volume ratio [1,12]. These nanocarriers have proved themselves to be very effective to be used as bioavailability, permeability, and solubility enhancers for different routes of administration [1]. For that reason, the goal of this study was to formulate and characterize RBX nanoparticles by encapsulating RBX within the internal cavities of polyamidoamine (PAMAM) dendrimer generation 4.5 and 5, respectively. Out of different generation PAMAM dendrimers, only intermediate generation 3.5–5 dendrimers are suitable for drug delivery carriers. Due to the application of generation 4.5 and 5 as drug delivery carriers, these carriers were developed in this study [12]. Generation 4.5 dendrimers are anionic in nature; however, generation 5 dendrimers are neutral in nature [1,13].

#### **2. Materials and Methods**

#### *2.1. Materials*

RBX (LY-333531 hydrochloride, 1 mg vial), dimethyl sulfoxide (DMSO), dialysis cellulose membrane (MWCO 14,000 Da), reusable plastic sample cuvette, folded capillary ζ-cells, and human vascular endothelial growth factor (Hu VEGF) ELISA were obtained from Sigma Aldrich (St. Louis, MO, USA). Polyamidoamine dendrimer generation 4.5 (PA-MAM dendrimers G4.5, 1 g vial) and polyamidoamine dendrimer generation 5 (PAMAM dendrimers G5, 1 g vial) were purchased from Nanosynthons (Mt. Pleasant, MI, USA). All chemicals and reagents were of analytical grade.

#### *2.2. Formulation of Anionic G4.5 and Neutral G5 Complexes*

To achieve loading of RBX into anionic G4.5 and neutral G5 PAMAM dendrimers, RBX (1 mg) was dispersed into 1 mL of dimethyl sulfoxide (DMSO) making a concentration of (0.1% *w*/*v*). The obtained PAMAM dendrimers G4.5 or G5 (1 g) were further dispersed into 10 mL of purified water to achieve a concentration of (10% *w*/*v*). A number of formulations were prepared for the optimization of the best one. In this study, a molar ratio of 1:1, 2.5:1, 5:1, and 25:1 PAMAM dendrimers G4.5: RBX and PAMAM dendrimers G5: RBX were prepared and coded as G4.5 1:1, G4.5 2.5:1, G4.5 5:1, and G4.5 25:1, respectively for G4.5 complexes and G5 1:1, G5 2.5:1, G5 5:1, and G5 25:1, respectively for G5 counterparts. The amounts of these complexes were calculated and added to the reaction mixture. In each one of these mixtures, the volume was completed with deionized water to 1 mL. Briefly, RBX and amounts for each proposed complex are shown in Table 1. After preparation, the reaction mixtures were stirred for 24 h in the dark at 4 ◦C [16]. The formulated complexes were extracted by removing the excess amounts of the undissolved drug by cold dialysis using cellulose membrane (MWCO 14,000 Da) against distilled water for 24 h [17,18]. The complexes were then lyophilized for 72 h using Labcono Free Zone 6 Liter Benchtop Freeze Dry System and stored at −30 ◦C for further use [19].


**Table 1.** Calculated amounts of G4.5 and G5 complexes.

### *2.3. Measurement of Particle Size (PS), Polydispersity Index (PDI), and ζ-Potential*

PS and PDI of G4.5 and G5 complexes and the empty PAMAM dendrimers G4.5 and G5 were measured by Zetasizer (Malvern Instruments Ltd., Malvern, UK) [20], which was based on dynamic light scattering (DLS). DLS analysis was performed in triplicate at 25 ◦C and scattering angle of 90◦ . The volume-average PS was determined for each sample. The surface charge of G4.5 and G5 complexes and the empty PAMAM dendrimers G4.5 and G5 was determined using Zetasizer (Malvern Instruments Ltd., Malvern, UK) [20]. The ζ-potential of these nanoparticles was determined in an electric field. The velocity and direction of the nanoparticle movement was measured and was demonstrated to be proportional to their ζ-potential. Analysis was performed in triplicate at 25 ◦C [21].

#### *2.4. Determination of Drug Loading Efficiency (DE%)*

The loading of different amounts of the drug into the dendritic structure of PAMAM dendrimers G4.5 and G5 was studied to estimate the maximum drug loading efficiency of each complex. To achieve this, DE% of G4.5 and G5 complexes was determined by using the dialysis method. G4.5 and G5 complexes were placed into a cellulose membrane (MWCO 14,000 Da), then immersed in deionized water and stirred for 24 h at 4 ◦C [22]. After that, a sample from the surrounding media (i.e., deionized water) was withdrawn and measured by Nanodrop UV spectrophotometer at 254 nm. This was carried out to indirectly determine the amount of the drug loaded in PAMAM dendrimers. The measured amount of free drug from the surrounding media was referred as amount of free drug, and the original amount of drug added in the mixture was called amount of total drug. The DE% was calculated by the following equation:

$$\text{DE\%} = \frac{\text{Amount of total drug} - \text{amount of free drug}}{\text{Amount of total drug}} \times 100\tag{1}$$

#### *2.5. Transmission Electron Microscopy*

The loading visual pattern as well as the morphology of G4.5 and G5 complexes and the empty PAMAM dendrimers, G4.5 and G5, was determined by transmission electron microscopy (TEM) (JEM1230EX; Tokyo, Japan). The image of the nanoparticles was created by using scattered electrons. These electrons were transmitted through the sample, and then the image was formed by detecting the reflected electrons [21].

#### *2.6. In Vitro Drug Release of G4.5 and G5 Complexes*

In vitro drug release study was carried out to determine the amount of drug released from G4.5 and G5 complexes. To conduct this study, a phosphate buffer solution (PBS) media (pH 7.4) was prepared and used as a dissolution media [23,24]. A total of 200 µL of each studied complex containing 10 µg of RBX was placed into a cellulose membrane with a molecular weight cut-off (MWCO) 14,000 Da, then was immersed in 20 mL of PBS media (pH 7.4). In vitro drug release of G4.5 and G5 complexes was tested at 37 ± 0.5 ◦C with slow magnetic stirring for a period of 8 h. Aliquots of 500 µL were withdrawn from the solution and replaced with the same volume of fresh PBS at predetermined time points (i.e., 1, 2, 3, 4, 5, 6, 7, and 8 h). After that, RBX amounts in the withdrawn samples was determined by Nanodrop UV spectrophotometer at 254 nm. Throughout the experiments, sink conditions were maintained [1,16].

#### *2.7. Stability Studies*

According to the manufacturer, PAMAM dendrimers should be stored in the dark at 2–8 ◦C, while RBX is a light sensitive drug and should be kept at −20 ◦C. In this work, the effect of light and temperature on the stability of the formulated nanoparticles was investigated. Samples of G4.5 and G5 complexes were stored in the dark and daylight and kept at different temperatures ranging from (4–50) ◦C as shown in Supplementary Table S1. In addition, the stability of G4.5 and G5 complexes was studied at periods of 1, 3, and 6 months.

#### 2.7.1. Physical Stability

At the end of the 6-month study period, the nanoparticles were observed and signs of color changes, precipitation, or turbidity were recorded [25]. The most physically stable complex was then visualized using scanning electron microscopy (SEM) (JSM 7610F; Tokyo, Japan) to confirm that the complex still preserves its shape and size under certain storage conditions.

#### 2.7.2. Chemical Stability

In addition to the physical stability of G4.5 and G5 complexes, drug content was also assessed under storage conditions mentioned in Table S1. This was carried out by calculating the percent content of RBX remaining in each tested complex. Drug content was quantified by ultra-performance liquid chromatography coupled to the mass spectrometry (UPLC-MS/MS) method as reported in the literature [26].

#### *2.8. Cell Culture*

#### 2.8.1. Moorfield's/Institute of Ophthalmology-Müller-1 (MIO-M1) Cells

In this work, cell culture studies were carried out to assess the viability of macroglial Müller cells when exposed to RBX, PAMAM dendrimers, and the proposed complexes as well. The viability of these cells was assessed in controlled and high glucose treatment with and without the tested complexes. The cell line used in this study is MIO-M1; a spontaneously immortalized human cell line. These cell lines were obtained from UCL institute of Ophthalmology, London, UK. These cells were named after the institution where they were isolated, MIO-M1 [27].

#### 2.8.2. Culture of the MIO-M1 Glial Cell Line

MIO-M1 cells were cultured in Dulbecco's Minimal Essential Medium (DMEM) containing GLUTAMAX and physiological glucose levels of 5.55 mM. The medium was supplemented with 10% fetal bovine serum (FBS) and 50 mg/L penicillin/streptomycin. Prior to experimental work, cells were seeded at 5000 cells per well in 96 well plates (100 µL medium). After that, cells were grown for three days to achieve >80% confluence and were starved with serum-free DMEM for at least 24 h before the experiment [9].

#### 2.8.3. Cell Viability Studies

Cell viability was assessed using (Cell Titer 96 Aqueous Proliferation Assay, Promega, Southampton, UK). This assay was carried out according to the manufacturer's protocol. This test is a colorimetric assay used to determine the number of viable cells. It is based on the conversion of a 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2- (4-sulfophenyl)-2H-tetrazolium (MTS) into a brown formazan product when reduced by active cells [28]. The tetrazolium reduction takes place in the mitochondria and measures mitochondrial metabolic rate as a measure of cell viability [29]. Briefly, after exposing cells to experimental conditions, the medium was removed from the 96 well plate and replaced with 100 µL MTS solution. At the end of each tested period (i.e., 24, 48, and 72 h), the fluorescence of formazan was measured at 490 nm with a micro-plate reader [30].

#### 2.8.4. High Glucose Experiments

In this work, MIO-M1 cells were exposed to simulated conditions of hyperglycemia to test the cells' behavior in the presence of RBX, PAMAM dendrimers, and the proposed complexes. This was achieved by incubating the cells in controlled DMEM (5.55 mM glucose) and high glucose DMEM of (25 mM glucose) for 24, 48, and 72 h with and without the tested compounds. Then, the medium was collected for subsequent cell viability studies [9].

#### 2.8.5. High Glucose Treatment of PAMAM Dendrimers and Complexes

In this study, a drug-dose–response study of RBX, PAMAM dendrimers, and the proposed complexes was carried out in MIO-M1 cells cultured in controlled and high glucose medium. In hyperglycemic conditions, the viability of MIO-M1 cells was assessed using an

MTS assay at the end of each tested period (i.e., 24, 48, and 72 h). Two concentrations (200 and 500 nM) of pure RBX were tested. In addition, two concentrations (200 and 500 nM) of RBX formulated as nanoparticles (G4.5 and G5 complexes 25:1) were tested as well. Moreover, PAMAM dendrimers G4.5 and G5 were tested in both cell mediums in the same amount available in G4.5 and G5 complex respectively. This was carried out to examine whether PAMAM dendrimers in the presented concentrations could cause any cytotoxicity to the cells. Furthermore, this step was performed to investigate the safety of the tested compounds in simulated conditions of hyperglycemia and also, to determine the most tolerable doses by MIO-M1 cells.

#### 2.8.6. In Vitro Permeability Study

The integrity of the monolayer cell membrane of MIO-M1 was evaluated via an in vitro permeability study on Transwell® and transendothelial electrical resistance (TEER). The in vitro studies were performed in a 24-well plate format on a polyester membrane (PET) Transwell® (0.33 cm<sup>2</sup> , 0.4 µm pore size). MIO-M1 cells were seeded at a density of 0.5 <sup>×</sup> <sup>10</sup><sup>6</sup> cells/insert in the apical (upper) compartment with 300 µL media, and the basolateral (lower) compartment was filled with 900 µL media. The cells were grown in Dulbecco's Modified Eagle's medium (DMEM) containing glutamine supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin. MIO-M1 cells were incubated at 37 ◦C at 5% CO2, along with DMEM media of well and insert replenished every two days [31,32].

It is important that MIO-M1 make confluent monolayers so that they require monolayer integrity and will function as a barrier. To evaluate monolayer integrity, the measurement of TEER was performed using an endothelial volt-ohmmeter (EVOM) (World Precision Instruments, Sarasota, FL, USA) according to the manufacturer's instructions. The resistance value (R Blank) in units of (Ω) measured for each insert (no cells) was subtracted from the resistance (R Total) across the cell layer on the membrane to obtain the cell specific resistance (R MIO-M1):

$$\mathbb{R}\text{ }\mathbb{R}\text{ }\text{MIO-M1}\text{ (}\Omega\text{)}=\mathbb{R}\text{ }\text{Total}-\mathbb{R}\text{ }\text{Blank}\tag{2}$$

TEER resistance (Ω·cm<sup>2</sup> ) = R MIO-M1 (Ω) <sup>×</sup> SA semipermeable membrane surface Area (cm<sup>2</sup> ) (3)

R MIO-M1 value was then multiplied by the surface area of the insert (0.33 cm<sup>2</sup> ) to obtain the value of TEER resistance of MIO-M1 (Ω·cm<sup>2</sup> ). Each filter was measured three times, and each culture condition was carried out at least with triplicate filters [33–36].

At the time of the experiments, MIO-M1 was cultured for at least 14 days to reach the highest and constant TEER value of the membrane. Transwells® were washed twice with sterile PBS without calcium and magnesium and transferred to a serum-free medium for the duration of the experiments. At 37 ◦C in 5% CO<sup>2</sup> in a humidified incubator, pure drug (RBX), or dendrimers (with or without RBX), at 500 nM of RBX concentration, was incubated for 24 h. The monolayer membrane's integrity was evaluated by quantifying tracer permeability of solution containing fluorescein isothiocyanate conjugate (FITC)-dextran at an initial concentration of 100 µg/mL, which was added to the apical (upper) compartment. At 1, 3, 6, 12, and 24 h, samples were drawn from the basolateral compartment, which was replaced with an equal volume of fresh medium each time. The fluorescence intensities were measured using a Synergy H1 plate reader (BioTek) at emission/excitation wavelengths of 495/520 nm and were converted to the concentrations of FITC-dextran with the calibration curve [31,32,37–39]. The apparent permeability coefficient (Papp) was calculated based on the following equation [38,40]:

$$\text{Papp} \text{ (cm/s)} = \text{J/(AC}\_{\text{D}}) \tag{4}$$

where, J is the rate of appearance of the compounds in the receiver compartment, A is the surface area of the filter membrane, and C<sup>D</sup> is the initial concentration in the donor compartment.

#### *2.9. Statistical Analysis*

Values were expressed as the means and standard error of the mean (SEM). Differences among treated groups and control (MIO-M1 only) were evaluated by one-way analysis of variance (ANOVA) followed by *post-hoc* comparisons using Tukey's multiple comparisons using Graph Pad Prism (version 9.0, GraphPad Inc., La Jolla, San Diego, CA, USA). A *p* value of 0.05 or less was considered statistically significant.

#### **3. Results**

#### *3.1. Characterization of G4.5 Complexes*
