Hybrid Formulations of Liposomes and Bioadhesive Polymers Improve the Hypotensive Effect of the Melatonin Analogue 5-MCA-NAT in Rabbit Eyes

For the treatment of chronic ocular diseases such as glaucoma, continuous instillations of eye drops are needed. However, frequent administrations of hypotensive topical formulations can produce adverse ocular surface effects due to the active substance or other components of the formulation, such as preservatives or other excipients. Thus the development of unpreserved formulations that are well tolerated after frequent instillations is an important challenge to improve ophthalmic chronic topical therapies. Furthermore, several components can improve the properties of the formulation in terms of efficacy. In order to achieve the mentioned objectives, we have developed formulations of liposomes (150–200 nm) containing components similar to those in the tear film and loaded with the hypotensive melatonin analog 5-methoxycarbonylamino-N-acetyltryptamine (5-MCA-NAT, 100 µM). These formulations were combined with mucoadhesive (sodium hyaluronate or carboxymethylcellulose) or amphiphilic block thermosensitive (poloxamer) polymers to prolong the hypotensive efficacy of the drug. In rabbit eyes, the decrease of intraocular pressure with 5-MCA-NAT-loaded liposomes that were dispersed with 0.2% sodium hyaluronate, 39.1±2.2%, was remarkably higher compared to other liposomes formulated without or with other bioadhesive polymers, and the effect lasted more than 8 hours. According to the results obtained in the present work, these technological strategies could provide an improved modality for delivering therapeutic agents in patients with glaucoma.


Introduction
Ophthalmic drug delivery is one of the most interesting and challenging endeavors facing the pharmaceutical sciences. The anatomy, physiology and biochemistry of the eye render this organ exquisitely impervious to foreign substances [1]. So to get the desired therapeutic effect, repeated and frequent applications of topical ophthalmic formulations are usually required.
Glaucoma is a multifactorial, progressive and neurodegenerative disease characterized by atrophy of the optic nerve and loss of retinal ganglion cells that can eventually lead to loss of visual acuity and visual field [2,3]. High intraocular pressure (IOP) is considered the greatest risk factor for the development of glaucoma, so most treatments consist of the chronic application of eye drops containing hypotensive agents. Despite the usefulness of topical administration, glaucoma treatments are usually associated with adverse reactions generated by the frequent exposure to drugs and excipients undergone by the eye. Among the excipients, preservatives can induce ocular surface alterations [4,5,6,7] that contribute to the development of secondary ophthalmic diseases such as the dry eye syndrome [8,9]. This in turn can compromise patient compliance. However, the elimination of preservatives from ophthalmic formulations is not always enough to avoid side effects on the ocular surface. For that reason, the incorporation of new components in formulations with beneficial properties for the eye and able, at the same time, to increase the bioavailability of the drug, results of great interest in this field.
Melatonin receptors have been identified in the cornea, ciliary body, lens, choroid, and sclera and play a role in aqueous humor dynamics [10,11]. In recent years, the role of melatonin and its analogues in the control of IOP has been investigated. The melatonin analogue 5-methoxycarbonylamino-N-acetyltryptamine (5-MCA-NAT) induces an IOP reduction after topical administration in normotensive rabbits [12,13] and glaucomatous monkeys [14]. In a previous work, the hypotensive effect of 5-MCA-NAT was enhanced up to 30% (maximum IOP reduction) with the use of mucoadhesive polymers, although the effect could not be prolonged longer than 6 hours [15].
Ophthalmic formulations of liposomes have been proposed to increase the efficacy of topical ophthalmic formulations and, depending on composition, to resemble the tear film lipid layer. Liposomes are spherical vesicles composed of an aqueous core enclosed by concentric phospholipids bilayers [16]. On the ocular surface, these vesicles establish a very tight contact with tissues, thus increasing the residence time of the formulation and, subsequently, the corneal penetration of the drug [17]. When active substances are included in these systems they are protected from the medium and the half-life is usually increased.
Our research group recently developed a novel artificial tear based on liposomes that improve the quality and stability of an unstable tear film [18]. In this formulation, liposomes were dispersed in aqueous solutions of polymers with mucoadhesive properties. The addition of these polymers had positive effects attributed to their rheological and biophysical properties, which are similar to those of mucins involved in the maintenance of the tear film on the ocular surface. Bioadhesive polymers also provide long-lasting hydration and lubrication of the ocular surface while minimizing the loss of formulation through the nasolacrimal drainage pathway [19].
Other materials, such as thermosensitive amphiphilic block copolymers, namely poly (ethylene oxide)-poly (propylene oxide)poly (ethylene oxide) (PEO-PPO-PEO, poloxamers), have been recently investigated in the design of novel drug delivery systems. When poloxamers are in solution, they form micelles and, depending on the concentration and temperature, are able to self-organize and transform into a viscous gel, allowing a controlled release of the drug [20].
In the present study we have used hybrid combinations of liposomes, mucoadhesive or thermosensitive polymers, and the intraocular hypotensive agent 5-MCA-NAT. These pharmaceutical formulations were characterized in terms of physicochemical properties, in vitro release of 5-MCA-NAT, in vivo tolerance and in vivo intraocular pressure reduction, after topical ophthalmic application in nonsedated normotensive rabbits.
2. Sample preparation. Five formulations (Table 1)  Formulation F1, the aqueous non-liposomal solution of 5-MCA-NAT, was prepared by dissolution of 5-MCA-NAT in PG (10 mg/ mL) and further dilution with an aqueous solution of NaCl. The final formulation contained 100 mM 5-MCA-NAT, 0.275% PG, and 0.788% NaCl. 5-MCA-NAT-loaded liposomes (F2, F3, F4, and F5) were prepared following the procedure described by Bangham [21] and modified by Vicario-de-la-Torre [22]. According to the procedure PC, Cht and vitE (8:1:0.08) were dissolved in chloroform in a round-bottom flask. Then, the organic solvent was slowly removed at 33uC with a rotator evaporator to produce a thin film of dry lipids on the inner surface of the flask. The dry film was then hydrated with an aqueous solution of 200 mM 5-MCA-NAT to produce the 5-MCA-NAT-loaded liposomes. The liposomes were extruded ten times through 0.22 mm pore polycarbonate membranes (Nucleopore Lipex Biomembrane, Vancouver, Canada). To ensure full lipid hydration, vesicles were allowed to mature overnight under refrigeration. After that, stock formulations were diluted (1:2) with different aqueous solutions to achieve the final liposomal formulations (Table 1): F2, 0.788% NaCl and 0.275% PG; F3, 0.4% SH in 0.788% NaCl and 0.275% PG; F4, 1% CMC in 0.788% NaCl and 0.275% PG); F5, PX407 and PX188 (24/16, w/w) in 0.17% NaCl and 0.275% PG. The concentration of NaCl was adjusted to attain osmolarity values in the final formulations within an acceptable range for ophthalmic administration [23]. Due to the high influence of PX in the osmolarity of the formulations [24], a lower concentration of NaCl was used to adjust F4. Final liposomal formulations were sterilized by filtration through 0.22 mm pore membranes.
3. Determination of the encapsulation efficiency of liposomes. The encapsulation efficiency of 5-MCA-NAT in the liposomes was determined upon separation by centrifugation (18,000 rpm, 4uC, 20 min) from the dispersing medium containing non-encapsulated 5-MCA-NAT. The amount of free 5-MCA-NAT in the supernatant was determined in triplicate by HPLC. The determination of the percentage of drug loading (PDL) in liposomes, expressed as mean 6 standard deviation, was calculated using the following equation [25]: Osmolarity was analyzed by a vapor pressure osmometer (model K-7000 Knauer, Berlin, Germany). Before performing the analyses, the osmometer was calibrated with a solution of NaCl (400 mOsm). The determinations were made in triplicate at 33uC (ocular surface temperature) [26].
Viscosity of liposomal samples was assessed in triplicate at 33uC with a thermostatically controlled rheometer (HaakeRheostress R1, Düsseldorf, Germany) using a parallel plate geometry (diameter 60 mm, gap 0.5 mm). The viscosity was calculated at a shear rate of 100 s 21 , which was within the linear viscoelastic range for all of the formulations.

5.
Determination of particle size and zeta potential. Measurements of liposome particle size and zeta potential were carried out by photon correlation spectroscopy (PCS, Zetatrac, Largo, FL, USA). For the analyses, formulations were diluted 1/20 (v/v) in an aqueous medium. All determinations were performed in triplicate at room temperature (25uC).

6.
Hypotensive efficacy studies in vivo: IOP determinations. In this work, experiments were performed in both eyes of nonsedated normotensive male New Zealand white rabbits (2-2.5 kg). Each formulation was evaluated in 10 animals (n = 20 eyes) and each control in 5 (n = 10 eyes). In order to reduce the number of rabbits, the same animals were used multiple times as experimental and control. The animals were kept in individual cages with free access to food and water and maintained in a controlled 12/12 h light/dark cycle. All of the protocols herein were approved by the Ethics Committee for Animal Research of Complutense University of Madrid. Also, animal manipulations followed institutional guidelines, European Union regulations for the use of animals in research, and the ARVO (Association for Research in Vision and Ophthalmology) statement for the use of animals in ophthalmic and vision research. IOP was measured with a Tonovet rebound tonometer (Tiolat, Helsinki, Finland). With this technique IOP is assessed without the need of topical anesthesia. For each eye, IOP was set at 100% with two basal readings taken 30 min before and immediately before the instillation. Then a single dose of the formulation (25 mL) was applied to both eyes. IOP determinations were performed once every hour over the next 8 hours. As a control, rabbits received formulations without the hypotensive agent. The administration protocol included a washout period of at least 48 hours between experiments.
7. In vitro drug release. The release of 5-MCA-NAT from the different formulations was studied using a dialysis method. To do this, a dialisys membrane (Spectra/Por Float-A-Lyzer G2; 20,000 MW cut off; Iberlabo, Madrid, Spain) was employed (0.8 mL of sample). Three membranes were prepared for each formulation. A conventional solution of 100 mM 5-MCA-NAT without liposomes was used as reference. Dialysis membranes were placed inside a flask with 50 mL of a phosphate buffered solution isotonized with NaCl (pH 7.4). The flask was kept on a magnetic stirrer and stirring was maintained at 100 rpm at 33uC. 1 mL of release sample was withdrawn at pre-set times (5 min, 15 min, 30 min, once every hour over a period of 8 and 24 hours). The release of 5MCANAT was analyzed by HPLC following the above mentioned method.
8. In vivo tear osmolarity measurements. This study was restricted to formulations F3 (5-MCA-NAT-loaded liposomes dispersed in 0.2% sodium hyaluronate) and F4 (5-MCA-NATloaded liposomes dispersed in 0.5% carboxymethylcellulose), that provided the highest in vivo hypotensive effect among all the tested formulations. An isotonic saline solution was used as control. Each formulation (control included) was tested in both eyes of three nonsedated male New Zealand white rabbits (n = 6 eyes).
Tear osmolarity was measured with the TearLab Osmolarity System (TearLab Corporation, San Diego, CA, USA). The device, stored in a temperature and humidity controlled room, was recalibrated using electronic check cards at the beginning of each test session following the manufacturer's guidelines. Tear samples of 50 nL were collected from the lower lateral meniscus of the eyes. To avoid fluctuation of tear osmolarity, all the measurements were made between 9 and 11 AM. Both eyes of rabbits were used for the evaluations. Measurements were taken before the administration of 25 mL of the formulation and 5, 15, 30 and 60 min after instillation.
9. In vivo short-term tolerance study. This study was restricted to formulations F3 (5-MCA-NAT-loaded liposomes dispersed in 0.2% sodium hyaluronate) and F4 (5-MCA-NATloaded liposomes dispersed in 0.5% carboxymethylcellulose), that provided the highest in vivo hypotensive effect among all the tested formulations.
The experiment was carried out on both eyes of six male New Zealand rabbits. A single administration of 25 mL of the formulation were administered in the right eye of the animal. The contralateral eye was used as control and being instilled with the same volume of an isotonic solution of NaCl (pH 7.4). Clinical symptoms and signs were evaluated in accordance with a protocol described previously [27]. Ocular surface evaluation was made before the instillation of the formulation or control, and then 1, 8 and 24 h after their administration. The short-term tolerance was evaluated by macroscopic examination of the eye surface of the animal and graded from 0-2, indicating the absence or presence of the following clinical signs: loss of corneal transparency, conjunctival signs (hyperemia, edema), eyelid swelling, and intense blinking (what would show a lack of tolerance).
10. Statistical analysis. Intraocular hypotensive reduction was expressed as means 6 standard error of the means (SEM). Other parameters as means 6 standard deviation (SD) were also evaluated. Statistical differences between two mean values were evaluated by two-tailed Students t-test. If necessary, an analysis of variance (ANOVA) was employed. Results were taken as significantly different at p-values less than 0.05.

5-MCA-NAT dose determination
The HPLC method to quantify 5-MCA-NAT was validated with respect to linearity, accuracy, and reliability in the range of concentrations between 5 and 50 mg/mL. In all cases, the method allowed sufficient separation of the drug from the vehicle. The retention time of 5-MCA-NAT was 10.760.5 minutes. There were non-significant differences (p = 0.57) between the theoretical concentration of 5-MCA-NAT (100 mM, ,27.53 mg/mL) and the experimental values obtained (28.561.07 mg/mL). The percentage of 5-MCA-NAT loaded in the liposomes was 7.963.5%, relative to the entire amount of drug in the liposomal formulation. Taking into consideration that the volume of the instilled formulation in rabbit eyes was 25 mL, the dose of 5-MCA-NAT administered was 0.7160.04 mg.

pH, osmolarity, and viscosity
The pH and osmolarity were measured for all formulations, and the viscosity was measured for liposomal formulations F2-F5 (Table 2). Formulation F1, without liposomes, had an acidic pH of 5.6. The incorporation of lipid vesicles in formulation F2 produced an increase of pH to 6.9, a nearly neutral value. The incorporation of polymer in formulations F3, F4, and F5 produced a minor decrease in pH, with values of 6.4, 6.5, and 6.5 respectively. The osmolarity of formulations F1, F2, F3, and F4 were within the range of isotonicity (Table 2). Formulation F5, with an osmolarity value of 23763.1 mOsm, was hypotonic. The lowest viscosity, 1.2 mPa?s, was obtained in F2 (Table 2), the liposomal formulation without polymers. The addition of sodium hyaluronate 0.2% (F3) and carboxymethylcellulose 0.5% (F4) increased the viscosity of the formulations to 2.0 mPa?s and 7.3 mPa?s, respectively. The liposomal formulation with poloxamer (F5) had the highest viscosity, 28.3 mPa?s.

Mean particle size and zeta potential of liposomal formulations
All of the liposomal formulations showed unimodal size distributions (Figure 1), and the average particle sizes were less than 200 nm in all cases ( Table 3). The average zeta potential was neutral (range between 210 to 10 mV) in all cases (Table 3).

Effect of 5-MCA-NAT formulations on IOP in rabbits
All formulations reduced IOP in normotensive rabbits though the maximal effects were different for each formulation (Figure 2). The maximum percentage of IOP reduction, the area under the DIOP (%) time curve from 0 to 8 hours (estimated by the trapezoidal rule) and the duration of the hypotensive effect (h), were calculated for all the formulations (Table 4). In addition, several other IOP parameters were considered: mean of the minimum and maximum IOP, and mean difference between maximum and minimum IOP ( Table 5). The IOP values measured were within the range of 8 to 16 mm Hg (Table 5)  The release rate of the drug was higher for the solution of 5-MCA-NAT (F1), while the rate of release from 5-MCA-NATloaded liposomes dispersed in SH (F3) or CMC (F4) was lower than that of conventional liposomes (F2). Statistically significant differences between the solution of 5-MCA-NAT (F1) and liposomal formulations (F2-F4) were found after 1 hour of the beginning of the study (p,0.05 at all the time points studied). Comparing liposomal formulations, non-significant differences were found between conventional liposomes (F2) and liposomes dispersed in SH (p.0.25 in all cases). In the case of F4, 5-MCA-NAT released from liposomes resulted significantly lower to the ones that were found for the solution of 5-MCA-NAT (F1) or for conventional liposomes (F2) after 1 hour of the beginning of the study and up to 8 hours (p,0.003 in all cases). From 6 to 8 hours, a significant different release rate can be observed between F3 and F4, being F4 (5-MCA-NAT-liposomes dispersed with CMC) the formulation that showed a slower release profile (p,0.05).

In vivo tolerance study
Before testing, all animals had a normal ocular surface and corneal transparency. None had any conjunctival disorders including hyperemia or edema, eyelid swelling, or intense blinking (grade 0). Animals showed no discomfort or irritation during the test or within 24 h after the administration of formulations F3 and F4, which contained 5-MCA-NAT-loaded liposomes dispersed in 0.2% SH and 0.5% CMC, respectively. The cornea remained transparent (no vessels) throughout the assay (grade 0) and the coloration of the conjunctiva remained normal (grade 0). No animal presented signs of mucus secretion.

Discussion
In general, topical ophthalmic anti-glaucoma therapies combine the use of one or more hypotensive drugs with several instillations per day in chronic treatments. In these patients, the eye surface is continuously exposed to drugs and preservatives that produce, in most cases, ocular surface alterations and generally lead to a therapeutic failure [28]. Eye drops for chronic treatments should include components that are well tolerated for the ocular surface and, at the same time, enable enhanced drug bioavailability Table 2. pH, osmolarity, and viscosity data of the 5-MCA-NAT formulations.   Data are expressed as means 6 SD (n = 3). *Significant differences with formulation F2 (p-value,0.05). doi:10.1371/journal.pone.0110344.t003  [29,30]. In general, toxicity of formulations can be reduced by removing preservatives from the composition. However, in some cases the toxic effect of the drug still remains [31,32,33]. The inclusion of bioadhesive polymers have demonstrated to increase the tolerance of formulations while extending the effect of hypotensive drugs such as timolol maleate [34]. Nevertheless, the benefits produced by these polymers in terms of efficacy are limited, so it becomes necessary to use other technological approaches, such as combination of the drugs with other colloidal systems. Liposomes have gained considerable attention for ocular drug delivery. They have been primarily investigated as a modality to enhance corneal drug absorption through the ability to come into intimate contact with the corneal and conjunctival surfaces, thereby increasing the ocular drug penetration [17].
In this study, we designed novel liposomal formulations loaded with 5-MCA-NAT, a melatonin derivative that reduces IOP [12,15]. The liposomes that we employed (150-200 nm, 6-10 mV zeta potential) simulate the lipid composition of the pre-ocular tear film and have demonstrated to be effective in the treatment of the dry eye syndrome [18]. The anti-glaucoma formulations of the current study were designed to (a) control the delivery of 5-MCA-NAT, thus prolonging the hypotensive effect, and (b) replenish the tear film, which may improve an ocular surface damaged by chronic exposure to topical treatments. To achieve the objectives, we prepared liposomes with the biocompatible components PC, Cht, and vitE -all of which are present in the natural tear filmthat were loaded with 100 mM 5-MCA-NAT. Once prepared, the liposomes were dispersed in an isotonic aqueous solution alone or with the bioadhesive biopolymers SH or CMC, or a thermosensitive polymer, PX, to determine whether or not the use of these combinations would enhance the hypotensive effect of 5-MCA-NAT. The bioadhesive polymers and concentrations were selected among the available existing commercial artificial tear products [35]. A conventional solution of 100 mM 5-MCA-NAT without liposomes was used as the reference.
All the formulations developed satisfied the given requirements of pH and osmolarity for ophthalmic solutions [36]. Regarding pH values, the presence of lipid vesicles in formulation F2 produced a slight increase of pH to a nearly neutral value, probably due to the presence of PC in the formulation. The incorporation of polymers in formulations F3, F4, and F5 produced minor decreases in pH, presumably due to the acidic nature of SH, CMC and PX. For the formulations F2-F5, the pH values were nearly neutral, thus helping to maintain the optical properties of the eye surface, epithelial cell functions, and cellular homeostasis. With respect to the osmolarity, all of the formulations were within an acceptable range for ophthalmic administration [23].
To obtain a homogeneous film after instillation, the viscosity of a formulation should be similar to that of natural tears, 1-9 mPa?s [37,38]. Previous studies showed that the most significant relative improvement in ocular bioavailability occurred for vehicles in the viscosity range from 1 to 15 mPa?s [39,40]. In our study, the viscosities of formulations with bioadhesive polymers were within this range, and the most effective formulations for reducing IOP, F3 and F4, had viscosity values of 2.0 mPa?s and 7.3 mPa?s, respectively. The formulation F5 had the highest viscosity, 28.3 mPa?s, but it was no more effective in reducing IOP than were the non-liposomal F1 or the liposomal F2 without a bioadhesive polymer. This anomalous behavior of F5 (liposomes dispersed with Poloxamer) could be associated with its high viscosity and sol-gel phase transition phenomena [41].
The bioadhesive properties and rheological behavior of SH and CMC are widely described [42,43]. Mucins present in the preocular tear film can interact with the bioadhesive polymers and  In vitro release studies also corroborate these results. Formulation F4 (5-MCA-NAT-liposomes dispersed with CMC) shows significant differences in the release rate from 6 h to 8 h with the rest of formulations and a tendency to extend the hypotensive effect in animals. Taking into account that hypertensive animals show higher IOP decrease after the use of hypertensive drugs (in comparison to normotensive animals), an induced hypertensive  animal models might be useful to appreciate differences in the extension of the hypotensive effect [44,45]. In addition, these polymers hydrate and protect the ocular surface by forming a film that covers the eye. So, the liposomes dispersed with bioadhesive polymers contribute to the modulation of drug delivery by adequately mixing with the natural tears that cover the ocular surface. The liposomes also reduce water evaporation from tears and help improve symptoms such as dryness, redness, and visual impairment associated with dry eyes. The mucoadhesive behavior of these hybrid formulations was exhibited when rabbit eyes' tear osmolarity decreased to reach osmolarity values similar to the formulations F3 and F4 that were maintained for at least 60 minutes after a single instillation. These results must be confirmed in humans due to the physiological differences between both species that can affect the residence time of the formulations on the eye surface (i.e. lower blink rate than humans). In any case, further studies are necessary to demonstrate the efficacy of these formulations, in terms of ocular surface damage, that require experimental animal models.
The goal of the study was to enhance the effect of the hypotensive agent 5-MCA-NAT by using an unpreserved drug delivery system that could increase its bioavailability while being soft with the ocular surface. The novel hybrid formulations simulate tear film composition and include biocompatible components that might relieve symptoms of secondary diseases affecting the ocular surface derived from the chronic use of antiglaucoma topical medications. With these improvements, patient adherence to topical glaucoma therapy could be enhanced.

Conclusions
A number of applications of liposomes in ophthalmic drug delivery have been extensively studied [46,47,48,49,50]. The improvement in the precorneal retention, transcorneal permeation, and therapeutic efficacy has been explored in detail, providing information about the interaction between liposomes and ocular tissues. In this work we prepared novel liposomal formulations dispersed in aqueous solutions of bioadhesive or amphiphilic block copolymers for the delivery of the ocular hypotensive agent 5-MCA-NAT. In vivo efficacy studies in rabbits showed that the hypotensive effect of the drug was remarkably increased with the combination of liposomes and bioadhesive polymers and also tear osmolarity decreased significantly for 60 min. This improvement could be attributed to the influence of liposomes as drug carriers, the increased residence time of the formulation on the eye surface derived from the mucoadhesive properties of the polymers and the ability of these hybrid formulations to simulate and replenish the tear film. Addition of the thermosensitive polymer poloxamer to the drug-loaded liposomes was no more effective than the non-liposomal 5-MCA-NAT. In this work, the application of liposomes combined with bioadhesive polymers for the improvement of the precorneal retention of hypotensive drugs has shown potential for further investigation. In conclusion, hybrid nanosystems composed by liposomes combined with bioadhesive polymers might serve as potential ocular drug carriers that prolong drug retention and improve biocompatibility of the formulations on the ocular surface.