Most ocular diseases are treated with topical drug applications formulated as solutions, suspensions or ointments. Ophthalmic topical formulations bear some drawbacks related to poor ocular bioavailability due to the many anatomical and physiological barriers existing in the eye [1]. The oral and intravenous routes are seldom used in ophthalmology, because the blood ocular barrier impairs the achievement of satisfactory drug concentration in intraocular tissues [2,3].

The blood ocular barrier is composed of the blood-aqueous barrier and the blood-retinal barrier, protecting the eye from entry of toxic substances and maintaining the homeostatic control that underpins the ocular physiology [4,5]. The blood-aqueous barrier is formed by the nonpigmented epithelium of the ciliary body, the posterior iris epithelium, the endothelium of the iris vessels with tight junctions of the leaky type, and the endothelium of Schlemm’s canal. The blood-retinal barrier consists of the retinal pigment epithelium (outer barrier) and the endothelial membrane of the retinal vessels (inner barrier), both with tight junctions of the nonleaky type. The two functional barriers restrict the movement of blood elements to the intraocular chambers [6] and explain why drugs administered orally or intravenously can hardly reach therapeutic levels in intraocular tissues (Figure 1).

The barrier properties of the blood-brain barrier are similar to the blood-retinal barrier. Small protein tracers like microperoxidase (19 kDa) are capable of entering the perioxonal space but not the retina [6], indicating that the intercellular junctions of the retinal endothelium that are sealed with overabundant amounts of zonulae occludentes contribute to form the bulk of the blood-retinal barrier [7]. The ciliary epithelium of the ciliary body of the rabbit eye does not have as effective tight junctions as those found in the retina and in the brain capillaries [8]. Thus, the blood-aqueous barrier is not as active as the blood-retinal barrier in respect to limiting molecular diffusion. After intravenous injection, a lot of test substances, such as inulin, chloride, sucrose, phosphate, potassium, sodium, urea, proteins and some antibiotics could be found on the anterior side of the vitreous humor resulting from the ciliary circulation, but could not reach the retina due to the blood-retinal barrier [9].

There are some situations when the breakdown of the blood-ocular barrier can occur, such as ocular injuries and ocular hypotonia. Many types of ocular injuries, such as surgical and non-surgical traumas, intraocular inflammation (e.g., uveitis, scleritis), vascular disorders (e.g., M. Coats, M. Eales), systemic disorders (e.g., diabetes) and intraocular tumors (e.g., retinoblastoma, uveal melanoma) can disturb the blood-ocular barrier, resulting in a variable inward movement of inflammatory cells and blood plasma constituents such as proteins, cytokines and growth factors [10] (Scheme 1).

Ocular hypotony can promote the breakdown of the blood-aqueous barrier, permitting the leakage of serum proteins into the anterior and posterior chambers through the opening of the non-fenestrated endothelial layer of the iris capillaries and the zonulae occludentes of the ciliary body [11] (Figure 1). Ocular hypotony may also cause a breakdown of the blood-ocular barrier when intraocular pressure is lower than the episcleral venous pressure, leading to reversion of the flow direction of the ocular fluids, with plasma proteins entering the anterior chamber from the blood-filled Schlemm’s canal [12]. The blood retinal barrier is concomitantly broken down with the lowering of intraocular pressure, which affects the zonulae occludentes of the retinal pigmented epithelium (Verhoeff’s membrane) and the retinal capillaries [13,14,15] (Scheme 2).

The three commonly used ocular routes of drug administration are topical, local and systemic, but they bear limitations and may lead to complications. The topical route is frequently accompanied by toxicity in the cornea, conjunctiva and nasal mucosal [16,17,18,19]. Furthermore, daily topical instillation frequently leads to a lack of compliance [20,21]. Local toxicity can also develop with the use of subconjunctival and sub-Tenon routes. These procedures can cause muscle trauma [22], and inadvertent globe perforation [23,24]. Finally, retinal toxicity and infections may occur after intravitreal injections [25,26,27]. The systemic application of drugs is another method of reaching the posterior segment of the eye, but it is still a challenge due to the blood-ocular barrier. Despite the presence of an extensive vascular network in the choroid, the entry of drugs into the posterior segment of the eye is limited by the outer and inner blood-retinal barriers. In addition to the limitations imposed by the blood-ocular barrier, other facts limit the systemic application, since dilution and degradation of the drug before reaching the target site may occur. However, the systemic route could be a useful alternative, since this route is less invasive and the patients prefer it to the local route in the treatment of some diseases. There is the additional benefit of facilitating out-patient treatment [28].

Innovative techniques of drug delivery are required to overcome the limitations of systemic treatment of ophthalmic disorders. One of these strategies is the use of nanoparticles in drug delivery systems, that can be intravenously administrated and that overcome the physiological barriers.

Any physical system that is engineered at nanoscale is defined as a nanosystem. Nanosystems include nanoparticles (nanospheres and nanocapsules), liposomes, dendrimers, niosomes, and soluble macromolecules, amongst others. Some examples of these systems are shown in Figure 2.

Nanoparticles are spherical particles varying in size from 10 nm to 1 µm, containing a drug molecule or not. They can be prepared using materials of different chemical classes, for example proteins, lipids, polymers or polysaccharides [30]. The active agent may be encapsulated/entrapped, dissolved or distributed to a nanoparticle matrix, and then can be broadly classified into nanospheres, nanoparticles or nanocapsules.

Nanoparticles can have the ability to overcome physiological barriers, passing even through capillaries, and transporting the drug to specific cells or intracellular compartments by several targeting mechanisms [31,32,33,34]. Singh et al. used arginine-glycide-aspartic acid peptide or dual-functionalized poly-(lactide-co-glycolide) nanoparticles to target delivery of anti-vascular endothelial growth factor (VEGF) intraceptor plasmid to choroidal neovascularization lesions in rats. They demonstrated that nanoparticles enable targeted gene delivery to the neovascular rat eye but not the control eye [35].

Liposomes are membrane-like vesicles that consist of one or more concentric phospholipid or cholesterol bilayers designed to carry a drug either into the core or into the bilayer. They can have a positive, negative or neutral surface charge, depending on their chemical composition, and they present different properties, such as binding affinity [36].

Light-targeted liposomal delivery was employed after systemic administration of thermosensitive liposomes for the diagnosis and therapy of chorioretinal neovascularization in age-related macular degeneration, showing the possibility of targeting a particular tissue [37]. In this approach, a liposomal formulation of a light-sensitive delivery system was selectively withheld in neovascular areas of the eye for the treatment of age-related macular degeneration, which demonstrated targeted therapy, even though this therapy was not completely successful in its purpose of destroying abnormal blood vessels without causing damage to normal surrounding tissues.

Dendrimers, macromolecular compounds from which a number of highly branched, tree-like arms originate in a symmetric fashion [38], are considered attractive for biomedical applications due to their unique physicochemical properties. As the dendrimer’s size can be controlled based on the stepwise chemical synthetic processes of their generation, they can become similar in size to a number of biological structures, such as G5 polyamidoamine dendrimers with the size of a hemoglobin molecule [39]

Niosomes are bilayered nonionic surfactant vesicles chemically stable with the ability to carry both lipophilic and hydrophilic drugs. Niosomes are nonimmunogenetic, biodegradable and biocompatible, and show low toxicity due to their nonionic structure [40].

Micelles are aggregates of amphiphilic molecules dispersed in a liquid colloid. Some micelar formulations for topical applications seem to be promising, but one of the major disadvantages of micellar formation is the demicellization that happens as a result of dilution upon injection in vivo [29].

Macromolecules in nanometer dimensions, when present in solution, including virtually any drug, can provide special properties to the drug molecule. Macugen^®, a marketed treatment for neovascular age-related macular degeneration, is a nanoconjugate of an oligonucleotide (aptamer) with two polyethylene glycols (PEG) that prolongs the half-life of the drug in the vitreous humor [41].

Our laboratory has pioneered the use of artificially made, solid nanoparticles directed to therapeutic targets, called LDE [42]. This nanoemulsion has a structure similar to that of a low-density lipoprotein (LDL), with a core constituted of cholesteryl esters with trace amounts of triglycerides, surrounded by a phospholipid monolayer; at the surface monolayer there is a small proportion of free cholesterol. When the LDE is injected into the bloodstream, it acquires apo E from the native lipoproteins and binds to LDL receptors that recognize both apo B and apo E [42]. The mechanism behind this drug delivery system lies in the overexpression of cell surface receptors that bind and internalize LDL, as well as the artificial lipid nanoemulsions in tissues undergoing cell proliferation [43,44,45]. This drug-targeting nanoemulsion system can be used not only for cancer treatment, but also to carry drugs directed against atherosclerotic lesions in rabbits [46,47,48]. We also showed that the intense immunogenic-inflammatory processes in rabbits following heart transplantation could be attenuated when using nanoemulsion associated with anti-proliferative drugs [49].

Despite the limitations of the systemic route to ocular drug delivery, researchers have made continuous efforts to develop drug delivery systems to administer drugs systemically in order to clear these barriers.

Among the challenges of effective systemic nanodelivery into tissues, the limited penetration across the vascular endothelium and the uptake by the reticuloendothelial system are considerable factors. The current nanodelivery systems depend on transvascular exchange and tissue accumulation, which require high dosages to create large concentration gradients to drive nanoparticles passively across the blood-tissue interface [66]. Nevertheless, passive accumulation allows only a fractional dosage of nanoparticles penetrating into target tissues, which diminishes therapeutic efficacy and aggravates potential side effects. Therefore, active delivery of targeted nanoparticles across the vascular endothelium could increase the desired therapeutic index with fewer side effects [66]. In addition, other facts limit the systemic application, since dilution and degradation of the drug before reaching the target site can occur. Furthermore, drug-drug interactions in patients being treated for coexisting diseases may also influence the administration of systemic drugs [66].

The use of nanoparticles that have a higher selectivity for the target tissue is a challenge to be achieved to improve systemic administration. Hence, specific targeting systems are needed to transport molecules to deeper layers of the eye, overcoming the blood-ocular barrier. Recent studies have shown promising results with systemic administration of nanoparticles. Kim et al. have demonstrated that 20 nm gold nanoparticles, administered intravenously, were able to pass through the blood-retinal barrier and were distributed into all retinal layers [67]. The absence of toxicity was noted in retinal endothelial cells, astrocytes and retinoblastoma cells [67]. Another recent study demonstrates that transferrin, arginine-glycine-aspartic acid peptide, or dual-functionalized poly-(lactide-co-glycolide) nanoparticles carrying anti-VEGF intraceptor plasmid, administrated intravenously, were able to reach choroidal neovascularization lesions, probably due to the broken blood-retinal barrier as a result of choroidal neovascularization in the laser-treated rat eye [39].

The similarities between the blood-brain and the blood-retinal barriers allow the proposal of the hypothesis that there is the possibility of systemically administered nanoparticles also gaining the ability to pass through the ocular barrier. Dalargin, a hexapeptide analog of leucine-enkephalin containing D-alanine, which produces central nervous system analgesia, can cross the blood-brain barrier when conjugated with poly(butylcyanoacrylate) nanoparticles and accumulate in the brain of mice, but not when it administrated without nanoparticles [68]. Furthermore, these studies on blood-brain barriers suggest the future application of systemically administered nanosystems for ophthalmic purposes.

The evaluation of systemic nanosystem toxicity is a complex task, since the drug-delivery systems have to take into account the toxicity of the polymer used, the drug, and entire system after administration in vivo. Moreover, systemic use of nanoparticles can induce changes in every organ of the human body, which may cause alterations in blood clinical chemistry, liver function, kidney function, blood cell counts, and other unknown side effects.

The establishment of a complete risk-benefit ratio of the use of nanoparticles by systemic routes depends on the system configuration, specific materials used, dosage, and frequency of administration, besides the individual patient factors. Any change in these elements can affect the immunological response, cellular uptake, excretion, and, by consequence, the toxic effects elicited by the nanoparticle.

New methods and biological models are necessary to increase the research in nanotoxicology. Recent toxicology studies seem to be promising to reach new evaluation standards. Kong et al. have shown the most important factors affecting nanoparticle cytotoxicity assays due to the current lack of standardized protocols, in order to improve the method that could evaluate the toxicity without false-negative or false-positive misinterpretations [76].

Almeida et al. have recently published an extensive review about nanoparticle biodistribution and toxicity [72]. The authors pointed out that nanoparticle modifications can be made to prolong blood circulation and enhance treatment efficacy, but simultaneous nonspecific distribution is unavoidable, as well as the need for long-term in vivo studies to properly investigate the innate and adaptive immune response, excretion over time and toxicity of nanoparticles. Another important point of this study is the need for more systematic in vitro approaches to improve the correlation with in vivo work, and to clarify the real nanoparticle therapeutic action and side effects [72].

Stensberg et al. have shown the activity, transport and fate of silver nanoparticles at the cellular and organismic level, in conjunction with the available methods of nanoparticle characterization [77]. They proposed several mechanisms of cytotoxicity based on such studies, as well new opportunities for investigating the uptake and fate of silver nanoparticles in living systems [77].

Maranhão et al. have shown that the association of chemotherapeutic agents, such as carmustine, etoposide, and paclitaxel, with artificial lipid nanoemulsion (LDE) pronouncedly reduces the toxicity of those drugs while preserving or even increasing the antineoplastic activity, as shown in tumor-implanted rats and mice [78,79] and in trials enrolling patients with advanced cancers [43,44,45,74].

There are recent reports regarding the toxicity concerns of the use of nanoparticles for ophthalmic purposes. Escobar et al. have shown that budesonide-PLA microparticles administered periocularly in rabbits remained at the periocular site of injection for a long time without causing local side effects, such as elevation in intraocular pressure, lens opacity or blood chemistry changes [80]. In addition, Amrite et al. have demonstrated that periocular administered celecoxib-PLGA particles were useful sustained drug delivery systems for inhibiting diabetes-induced elevation in PGE2 and VEGF without inducing fibrotic reactions at the site of administration or damage to the retina [81].

The systemic use of nanoparticle drug delivery systems for ophthalmic treatments is rare, equally scarce are the corresponding toxicological studies regarding this route of administration.

The breakdown of the blood-ocular barrier is an attractive manner to allow the penetration of systemically applied drugs into the eye. We hypothesize that the breakdown of the blood-ocular barrier—that may occur in acute inflammation following intraocular surgeries, induced ocular hypotony and controlled use of inflammatory mediators—could allow the use of drugs carried in targeting nanosystems administered by the systemic route for ophthalmic purposes.

Blood-ocular barrier breakdown can occur in some situation, such as acute inflammation caused by intraocular surgery, induced ocular hypotony, and the use of inflammatory mediators.

The proposition presented in this article refers to the possibility of taking advantage of the breakdown of the blood-ocular barrier to use the systemic route as an alternative to access intraocular tissues. Monitoring this breakdown is currently possible, indicating the correct time to use intravascular medication. Besides the natural enhanced permeability of the blood-ocular barrier in ophthalmic surgeries, we suggest two possible ways to induce a transient, controlled breakdown of the blood-ocular barrier, both with limitations that need to be overcame.

The induced breakdown of the blood-ocular barrier would be an interesting and acceptable option to facilitate the ocular access of the systemically administered nanosystem, if a topical route is chosen to administer pharmacological agents that induce it, and likewise, other topical agents are used to reverse it after the therapeutic effect is achieved. Apart from the invasive fact of the intravenous administration of the nanosystem, only non-invasive additional topical methods would be plausible to be applied concurrently. Moreover, the reversion of the breakdown of the blood-ocular barrier, when induced, would be achieved with the use of topical anti-inflammatory or anti-hypotensive drugs. The utilization of short half-life inductive drugs would be essential to carry out a self-limiting process. Scheme 4 summarizes the hypothetical pathways that are taken advantage of the breakdown of the blood-ocular barrier in order to allow the systemically administered nanosystems to reach the ocular tissues.

The pharmacological evaluation of drugs injected into the bloodstream aiming to permeate the intraocular tissues is necessary to establish their safety, especially if formulated with nanocarriers that still require additional pharmacodynamic and pharmacokinetic studies [66,133]. The future development of methods enabling a better understanding of the complete physicochemical properties of the nanomaterials and the availability of new nanotoxicological standards will allow the real evolution in this field, making it feasible to use nanosystems via systemic route for ocular drug delivery.

In summary, the availability of new effective and safe ocular delivery systems remains a major challenge for pharmaceutical researchers. The development of methods to induce a controlled, safe and transient breakdown of the blood-ocular barrier could be a novel approach to allow the access of systemically injected drugs to the internal parts of the human eye. The ideal nanoparticle system for systemic use in order to reach the intraocular tissues is not presently available, nor is there any evidence of the facilitation of access of these nanosystems with the transient breakdown of blood-ocular barriers. However, previous studies and pathophysiological events mentioned in this article encourage us to propose this hypothesis with the aim to conceive the basis for the possible development of a viable system, allying nanoparticles properties with the facilitated permeation to intraocular tissues by taking advantage of a natural or induced breakdown of the blood-ocular barrier.

The intravenous route could be reconsidered as a good alternative to complement the ophthalmic therapeutics currently in use. As the ability to deliver drugs and other compounds across the blood-ocular barrier via the systemic route for therapeutic purposes is not reached, further studies using nanoengineered technologies and focusing on the development of nanosystems with minimal toxicity and enhanced efficacy are necessary in order to expand the knowledge of the ways to selectively and efficiently overcome the blood-ocular barrier. Moreover, more detailed studies on the normal physiological and pathological conditions of the eye are essential to yield a consistent understanding of nanosystem characteristics under different circumstances in the target organ. In addition, methods to take advantage of the natural or induced breakdown of blood-ocular barriers can complement the nanotechnological properties to get easier access to nanomedicines administered systemically and targeted to the hitherto almost impervious areas of the human eye.