You are on page 1of 14

Progress in Retinal and Eye Research 29 (2010) 596 e 609

Progress in Retinal and Eye Research 29 (2010) 596 e 609 Contents lists available at ScienceDirect

Contents lists available at ScienceDirect

Progress in Retinal and Eye Research

jou rn al hom ep ag e: www. el s evi e r. com/lo c a t e/p r e r

rn al hom ep ag e: www. el s evi e r. com/lo c a t

Applications of nanoparticles in ophthalmology

Yolanda Diebold a , b , * , Margarita Calonge a , b

a Ocular Surface Research Group, Edi cio IOBA, Campus Miguel Delibes, Instituto de Oftalmobiología Aplicada (IOBA), Universidad de Valladolid, Paseo de Belén, 17, E-47011 Valladolid, Spain b CIBER de Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Spain

a b s t r a c t

Keywords:

Drug delivery Gene delivery Nanomedicine Nanoparticles Ocular drug administration Toxicity

Contents

Nanocarriers, such as nanoparticles, have the capacity to deliver ocular drugs to speci c target sites and hold promise to revolutionize the therapy of many eye diseases. Results to date strongly suggest that ocular medicine will bene t enormously from the use of this nanometric scale technology. One of the most important handicaps of the eye as a target organ for drugs is the presence of several barriers that impede direct and systemic drug access to the speci c site of action. Super cial barriers include the ocular surface epithelium and the tear lm, and internal barriers include the blood e aqueous and blood e retina barriers. Topical application is the preferred route for most drugs, even when the target tissues are at the back part of the eye where intraocular injections are currently the most common route of administration. Direct administration using any of these two routes faces many problems related to drug bioavailability, including side effects and repeated uncomfortable treatments to achieve therapeutic drug levels. In this regard, the advantages of using nanoparticles include improved topical passage of large, poorly water-soluble molecules such as glucocorticoid drugs or cyclosporine for immune-related, vision-threatening diseases. Other large and unstable molecules, such as nucleic acids, delivered using nanoparticles offer promising results for gene transfer therapy in severe retinal diseases. Also, nanoparticle-mediated drug delivery increases the contact time of the administered drug with its target tissue, such as in the case of brimonidine, one of the standard treatments for glaucoma, or corticosteroids used to treat autoimmune uveitis, a severe intraocular in ammatory process. In addition, nanocarriers permit the non-steroidal anti-in ammatory drug indomethacin to reach inner eye structures using the transmucosal route. Finally, nanoparticles allow the possibility of targeted delivery to reach speci c types of cancer, such as melanoma, leaving normal cells untouched. This review summarizes experimental results from our group and others since the beginnings of nanocarrier technology to deliver drugs to different locations in the eye. Also, it explores the future possibilities of nanoparticles not only as drug delivery systems but also as aides for diagnostic purposes. 2010 Elsevier Ltd. All rights reserved.

1. Introduction: what is nanomedicine?

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.597

2. Drug delivery systems

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.597

3. The eye as a target organ for drug delivery systems

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.597

4. Types of

NPs for ocular delivery

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

599

5. NPs

and the

anterior segment of

the

eye

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

600

6. posterior segment of

NPs

and the

 

the

eye

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

603

Abbreviations: AS-ODNs, Antisense oligonucleotides; CS, Chitosan; CSO, Chitosan oligomers; CsA, Cyclosporine A; ESF, European Science Foundation; HA, Hyaluronic acid; HA-CS NPs, Hyaluronic acid and chitosan-based nanoparticles; HA-PECL NPs, Hyaluronic acid-coated poly- e-caprolactone nanoparticles; IOP, Intraocular pressure; LCS-NPs, Liposome e chitosan nanoparticle complexes; NIH, National Institutes of Health; NPs, Nanoparticles; OIR, Oxygen-induced retinopathy; PBCA, Poly(butyl-cyanoacrylate); PECL, Poly- e-caprolactone; PEG, Polyethyleneglycol; pGFP, plasmid green uorescent protein; PIBCA, Poly(isobutyl-cyanoacrylate); PLA, Poly- D -lactic acid; PLGA, Poly- D - lactic-co-glycolide; RNAi, RNA interference; rAAV, adeno-associated virus vectors; RPE, Retina pigment epithelium; siRNA, small interfering RNA. * Corresponding author. Ocular Surface Research Group, Edi cio IOBA, Campus Miguel Delibes, Instituto de Oftalmobiología Aplicada (IOBA), Universidad de Valladolid, Paseo de Belén, 17, E-47011 Valladolid, Spain. Tel.: þ 34 983 184750; fax: þ 34 983 184762. E-mail address: yol@ioba.med.uva.es (Y. Diebold).

1350-9462/$ e see front matter 2010 Elsevier Ltd. All rights reserved. doi: 10.1016/j.preteyeres.2010.08.002

Y. Diebold, M. Calonge / Progress in Retinal and Eye Research 29 (2010) 596 e609

597

7.

NPs and gene delivery/therapy

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

604

8.

Nanoparticle safety: toxicity and interaction with the immune system

 

605

9.

Future directions

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

607

10.

Summary and

 

conclusions

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

607

11.

Author disclosure statement

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

607

Acknowledgements

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

. 607

References

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

. 607

1. Introduction: what is nanomedicine?

In 2003, the European Science Foundation (ESF) initiated a project aimed at gathering European experts from academia and industry to prepare what was called ESF Forward Look on Nanomedicine, published in 2004 ( http://www.esf.org/publications/forward-looks. html). Several workshops were conducted to (i) de ne the eld, (ii) discuss the future impact of nanomedicine on healthcare practice

and society, (iii) review the state-of-the-art of research, (iv) identify Europes strengths and weaknesses, and (v) deliver recommenda- tions on research trends and organization. These experts formally de ned the eld of nanomedicine as the science and technology of diagnosing, treating and preventing disease and traumatic injury, of relieving pain, and of preserving and improving human health, using molecular tools and molecular knowledge of the human body .

It is noteworthy that this concept is based in complex systems of

nanometre-scale size, i.e., from one nanometre to hundreds of nanometres, with the ultimate goal of using them to achieve medical bene ts. It is important to bear in mind that the nanometre scale is the scale at which molecules and compounds operate inside living cells. Soon afterward the ESF s publication, the National Institutes of

Health (NIH) in the U.S.A. developed a Nanomedicine Roadmap Initiative ( http://nihroadmap.nih.gov/nanomedicine/ ). As a centre- piece of this initiative, a national network of eight collaborative Nanomedicine Development Centers was established in 2006. That multidisciplinary research initiative was primarily directed towards

gathering extensive information about nanoscale intracellular biological structures. That information was to be used in the application of newly developed nanomedical therapies to treat speci c diseases. As an example of the interest that this topic has awakened in the vision research community, an education course Nanotechnology and Nanomedicine: Applications for Vision Research , was organized in 2005, co-sponsored by the Association for Research in Vision and Ophthalmology and the NIH s National Eye Institute. As anyone can envision, nanomedicine has a relevant position in the global agenda for future development of medical research in the 21st Century. One of the main topics in nanomedicine research is the pharmaceutical development of drug delivery systems. Its goal is the development of improved nano-sized drug carriers consisting of at least two components, one of which is the active therapeutic ingredient. These drug-loaded carriers can be termed nanopharmaceuticals or nanomedicines in a broad sense. This review will focus on the ocular applications of nanoparticles (NPs),

a

particular type of these drug delivery systems.

2.

Drug delivery systems

Among the different approaches that have been taken to develop more ef cient treatments to ght against human and animal life-threatening or debilitating diseases, the development of drug delivery systems is noteworthy. The purpose of a drug delivery system is to act as a carrier or vehicle for an entrapped or bound

therapeutic agent to reach precisely and effectively the desired site of action. Here we focus on delivery systems that target ocular structures. This concept is particularly interesting when one takes into account the physicochemical features of the frequently mar- keted biotechnological macromolecules, such as peptides, protein, antibodies, and nucleic acids ( Conti et al., 2000; Degim and Celebi, 2007; Levy-Nissenbaum et al., 2008 ). However, a drug delivery system is more than a simple (nano) carrier. All of the science and technology behind the design of drug delivery systems intends to achieve solutions for key aspects of modern treatments: (i) to control the release of the active agent so that a therapeutic concentration is maintained over a prolonged period of time, (ii) to develop organ or site-speci c or even disease- speci c targeting, and (iii) to provide new or more convenient routes of administration for drugs able to reach those locations in the body that are dif cult to access. The ultimate goals are to better manage relevant drug-related parameters, such as pharmacoki- netics, pharmacodynamics, non-speci c toxicity, immunogenicity, biorecognition, and to improve therapeutic ef cacy ( Vanderwoot and Ludwig, 2007; Sahoo et al., 2008 ). There are different kinds of drug delivery technologies that are designed to serve as drug delivery systems ( Medina et al., 2007; Sahoo et al., 2008; Gaudana et al., 2009 ). These include, among others, transdermal patches, implants, nanodevices, and cell encapsulation devices. Among nanoparticulate-based drug delivery systems (or nanosystems) ( Table 1 ) one can nd different polymeric formulations made of non-degradable polymers and biodegradable polymers that are either hydrophilic or hydrophobic. Examples of nanosystems that differ in composition include the following:

(i) Nanoparticles (NPs) consist of 1 mm or smaller particles composed of various polymers or materials. These are described in detail below. (ii) Liposomes are lipidic membranes, similar to plasma membranes, and surround an aqueous core. A variant of liposomes are niosomes, consisting of non-ionic surfactants. (iii) Emulsions consist of stabilised oil-in-water or water-in-oil mixtures. Others include (iv) nanosuspensions, (v) dendrimers, (vi) nanoparticle- loaded contact lenses, (vii) nanotubes and fullerenes composed of carbon-based nanomaterials, and (viii) quantum dots made of semiconductor materials with uorescent properties and covered with other materials. However, drug delivery systems other than those collectively named as nanoparticles are out of the scope of this review, and therefore we will not comment on them.

3. The eye as a target organ for drug delivery systems

There are a plethora of ocular disorders that may be vision- threatening. The responsiveness towards classically developed drugs is limited and most fail to correct the underlying problem. Thus, there is a scarcity of truly curative treatments for most eye diseases. The main reasons for these limitations are bio- pharmaceutical problems related to the special characteristic of the eye that restricts drug bioavailability. The eye is partially isolated from the remainder of the body by several types of barriers that impede the effective passage of many drugs ( Fig. 1 ), leading to

598

Y. Diebold, M. Calonge / Progress in Retinal and Eye Research 29 (2010) 596 e609

Table 1 Nanoparticulate drug delivery systems (or nanosystems) used as carriers for drug administration.

Nanosystem

Composition

Potential application in the eye

Nanoparticles Liposomes Niosomes Emulsions Nanosuspensions Dendrimers Nanoparticle-loaded contact lenses

Natural or synthetic polymers, metals, lipids, phospholipids Phospholipids Non-ionic surfactants Oil-in-water and water-in-oil mixtures that require surfactants Inert polymer resins Synthetic polymers Different hydrogel-based lenses with nanoparticulate-based drugs incorporated in the lens matrix

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Nanotubes and fullerenes

Carbon-based nanomaterials

Not tested yet

Quantum dots

Semiconductor materials covered with other materials

Yes (diagnostic)

According to Sahoo et al. (2007) and Gaudana et al. (2009) .

a minimal dose absorption ( Urtti, 2006 ). These barriers consist of the (i) muco-aqueous layer of the tear lm that protects the ante- rior surface of the eye, (ii) corneal epithelium with abundant tight junctions and desmosomes, (iii) iris blood vessels that lack fenes- trations, (iv) non-pigmented layer of the ciliary epithelium that constitutes the blood e aqueous barrier and limits the passage of molecules from the blood to the inner part of the eye, and (v) retinal pigment epithelium (RPE), along with the endothelium of the retinal vessels, constitute the inner and the outer blood e retina barriers, respectively, that limit the passage of molecules from the blood to the retina and vitreous cavity. Additionally, certain physiological processes contribute to the poor ef cacy of conventional drug formulations. For instance, blinking and tear drainage through the lachrymal drainage system act to reduce the residence time of topically applied molecules. Eye drops placed onto the ocular surface are washed away in less than

placed onto the ocular surface are washed away in less than Fig. 1. Schematic presentation of

Fig. 1. Schematic presentation of the ocular structure showing a summary of ocular pharmacokinetics. The numbers refer to following processes: 1) transcorneal perme- ation from the lachrymal uid into the anterior chamber, 2) non-corneal drug permeation across the conjunctiva and sclera into the anterior uvea, 3) drug distri- bution from the bloodstream via the blood eaqueous barrier into the anterior chamber, 4) elimination of drug from the anterior chamber by aqueous humour passage into the trabecular meshwork and Sclemm s canal, 5) drug elimination from the aqueous humor into the systemic circulation across the bloode aqueous barrier, 6) drug distribution from the blood into the posterior eye across the blood eretina barrier, 7) intravitreal drug administration, 8) drug elimination from the vitreous via the poste- rior route across the blood eretina barrier, and 9) drug elimination from the vitreous via the anterior route to the posterior chamber. (Taken from Urtti A., Adv Drug Deliv Revs (2006), with permission of Elsevier)

30 s ( Kaur and Kanwar, 2002 ). The maintenance of corneal trans- parency is based upon several strategies, one of them being the sealing of the corneal epithelium by means of specialized struc- tures, such as tight junctions and desmosomes. The corneal epithelium is therefore almost impermeable to any substance larger than 500 Da ( Hämäläinen et al., 1997 ). Most of the commonly used topical drugs are larger than that and, consequently, do not cross the cornea. Instead, they permeate throughout the conjunc- tiva and the underlying sclera in what is known as non-productive passage . Altogether, less than 5% of topically administered drugs reach intraocular tissues ( Keister et al., 1991 ). Therefore, the exis- tence of several ocular tissue and cell barriers along with the physiological processes impede the effective passage of many drugs, leading to a minimum dose absorption into the eye. Drug delivery systems hold promise to be an important part of the new therapeutic armamentarium in ophthalmology. Since the early study of Wood et al. (1985) showing the intrinsic capacity of NPs to adhere to the ocular surface and interact with the epithe- lium, applications of nanotechnologies to solve eye problems have been sought. Knowledge derived from drug delivery systems using non-ocular routes of administration has stimulated researchers to nd applications for them in ophthalmology. What makes them attractive to treat eye diseases is the possibility of the controlled release of drugs, especially poorly water-soluble ones, that surpasses the ocular barriers and effectively reaches the target. Nanoparticulate systems improve the delivery of poorly water- soluble drugs while signi cantly reducing toxicity compared to the free drug. Increasing attention has been focused particularly on this aspect due to the clear therapeutic implications. Also, the micro- or nano-size of such drug carriers is appealing. To envision the potential of nanoparticles as drug carriers that can treat ocular disorders, it is necessary to understand the peculiarities of the drug administration routes in the eye. First, local delivery of the drug is preferred over systemic delivery. There are several different modalities for ocular drug administration ( Table 2 ). The most common include liquids topically applied onto the front of the eye in the form of eye-drops, sub-conjunctival or sub-Tenon s injection in the conjunctival tissue or below the Tenon s capsule, and intravitreal injection. Also, intraocular implants made of either biodegradable or non-biodegradable polymers loaded with different drugs are used to provide long- term drug presence at the implantation site. Topical administra- tion of drugs is used to alleviate the symptoms and signs caused by ocular surface in ammatory disorders, such as dry eye syndrome and allergic diseases that affect millions of people worldwide. They are also used to treat infections and complex, vision-threatening diseases, such as glaucoma or intraocular in ammation (uveitis). In most cases, the patients themselves

Y. Diebold, M. Calonge / Progress in Retinal and Eye Research 29 (2010) 596 e609

599

Table 2 Modalities for ocular drug administration.

Modality

Anatomical location

Kind of

 

administered drug

Topical eye-drops

Ocular surface (onto the tear lm and corneal and conjunctival epithelia)

Arti cial tears Anti-infectives Anti-allergics Anti-in ammatories Anti-hypertensives Anaesthetics Mydriatics, miotics, cyclopegics

Periocular injection

Sub-conjunctival

Sub-conjunctval space

Anti-infectives

Sub-Tenon

Mydriatics

Corticoids

Transeptal

Anterior orbit

Anaesthetics

 

Corticoids

Retro-orbital

Posterior orbit

Anaesthetics

 

Corticoids

Intraocular injection

Intracameral

Anterior chamber

Anti-infectives

Intravitreal

Vitreous

Anti-infectives

 

Anti-angiogenics

Corticoids

instil the drops many times a day, and the treatment is not usually curative but just palliative. To summarize the necessity of using a carrier for ocular drugs, the key is to achieve adequate bioavailability. There are many challenges for ocular drug delivery systems. In some cases , the goal is to stop or reverse problems such as degenerations in the retina and neovascularisation in the cornea and in the re tina. In other cases, the drugs must contribute to the success of refractive corneal surgery and the healing process, tissue transplantation and growth factor delivery for retinal and corneal regenerative medicine, and gene therapy for heredi tary retinal disorders.

4. Types of NPs for ocular delivery

There has been an evolution in the design of nanoparticles (NPs), and Sánchez and Alonso (2006) extensively reviewed the features of polymers used to prepare the carriers. We analyze in this review the progress that has been made from a practical point of view in terms of therapeutic and diagnostic applications of NPs

in the eld of ophthalmology. The general term nanoparticle will be used in this review for the sake of simpli cation. However, this term can refer to nanospheres or nanocapsules, all of which are more properly designated as colloidal systems ( Fig. 2 ). Nano- spheres are matrical-type nanostructures that entrap or adsorb the biologically active molecule onto the surface. Nanocapsules are reservoir-type nanostructures within a surrounding polymeric wall containing an oil core where the active molecule is dissolved. These nanostructures can be coated with a hydrophilic polymer or even functionalized with antibodies attached to the coating. There are different methods to prepare NPs and load them with therapeutic molecules (for review, please see Pinto-Reis et al., 2006 ). NPs can be made of a great variety of materials including organic carbon-based biopolymers and inorganic ceramic, metallic, and semiconductor materials. Some of the most commonly used biomaterials include polyacrylates, polyalkylcyanoacrylates, poly- lactide (PLA), polylactide e polyglycolide (PLGA), polycaprolactones, dextran, albumin, gelatin, alginate, collagen, hyaluronic acid, and chitosan. The possibilities for their design are almost in nite. The intended application in uences the kind of material used for their preparation. The different chemical ways in which bioactive molecules can be associated with polymers and give rise to drug- loaded NPs include entrapment, encapsulation, adsorption, or attachment to a polymer. Also, NPs can be prepared in different sizes, charge, and other physicochemical features. This confers great versatility upon them. The physicochemical characteristics of NPs not only confer versa- tility in terms of the kind of drug to be loaded, but they also in u- ence the cellular uptake and intracellular trafcking. Additionally, the physicochemical characteristics are critical for other properties, such as interaction with plasma proteins, other blood components ( Dobrovolskaia et al., 2008 ), and with immune cells ( Dobrovolskaia and McNeil, 2007 ), all of which are relevant to the organ distribu- tion. For instance, opsonisation of NPs covers the surface by opso- nins present in the blood and creates a molecular signature ( Aggarwal et al., 2009 ) that determines the route of NP internali- zation in phagocytic cells and eventually their fate. However, if the NP surface is covered by a hydrophilic coat of poly(ethylene glycol) (PEG), opsonisation is prevented. This has been termed a stealth effect ( Gref et al., 1994 ), and it reduces the rate of cellular uptake. Consequently, PEG-NPs stay longer in the bloodstream. In addition, the rate of clearance of NPs is an important consideration with regard to potential toxicity caused by their permanence in the target tissue. Therapeutic targets for drugs can be located extra- or intra- cellularly. With regard to classical drugs with sizes greater than

With regard to classical drugs with sizes greater than Fig. 2. Evolution of colloidal systems: The

Fig. 2. Evolution of colloidal systems: The rst generation of colloidal systems consisted of polymerized matrices known as nanoparticles and nanocapsules composed of a polymeric wall containing an oil core. Second generation systems were similar to nanocapsules, but with improved hydrophilicity associated with a coating polymer. Third generation systems are functionalized by the attachment of antibodies or lectins, among other targeting moieties to the surface structure. (Adapted from Sánchez A. and Alonso M.J.,

2006 ).

600

Y. Diebold, M. Calonge / Progress in Retinal and Eye Research 29 (2010) 596 e609

1 kDa, stability conditions or hydrophilicity makes them unable to cross plasma membrane. NPs, as well as other drug delivery systems, enable the loaded bioactive molecule to cross cell membranes and epithelial barriers by using different internaliza- tion pathways. This is one of the most interesting aspects of this technology. Reports showing that submicrometer delivery systems such as NPs can enter cells and become concentrated into non- diffusible molecules appeared in the 1970s ( Couvreur et al., 1977 ). Much research has been done since then and much knowledge has been gained related to both carrier design and mechanisms of biological action. It is now known that the physicochemical char- acteristics of nanocarriers, such as size, shape, surface charge, surface coating, and surface functionalization with targeting ligands, along with the target cell type, determine the internali- zation pathway and the intracellular fate. Increasing numbers of studies about cell targeting and internali- zation pathways are being accomplished, mainly using nano- medicines to treat infectious diseases and cancers. Recent interesting reviews deal with these topics (Bareford and Swaan, 2007; Frenkel, 2008; Hillaireau and Couvreur, 2009). The main internalization pathways for NPs discussed in those reviews are summarized in Table 3. Phagocytosis is the typical pathway for NPs intravenously admin- istered. It implies NP opsonisation as mentioned above and the involvement of specic cells with phagocytic capacity that will ingest the NPs. In the cytoplasm, NPs and their payload will become accessible to lysosomes and experience enzymatic degradation. This step is important to ensure drug release. Endocytosis occurs in all mammalian cells by means of pitted or invaginated membrane regions that are coated by either of two speci c proteins. Pitted membranes coated by clathrin can be formed by receptor-mediated and receptor-independent processes with different internalization rates. Invaginated membranes coated by another protein, caveolin, can also conduct receptor-mediated endocytosis. NPs can be targeted to interact with cell receptors of interest to facilitate internalization by either of these receptor- mediated endocytic mechanisms. Loaded NPs internalized by cla- thrin-mediated endocytosis are directed to lysosomal degradation. Those internalized by caveolin-mediated endocytosis accumulate in the endosomes (caveosomes) and are delivered to other subcellular compartments different from lysosomes. This last mechanism has been used, for instance, to deliver chemothera- peutics to cancer cells. Also, clathrin- and caveolin-independent endocytosis mechanisms have been described. Finally, macropynocytosis also occurs in all kind of cells. It is directed by actin-driven membrane protrusions that form large ( > 1 mm) endocytic vesicles (macropinosomes). Eventually, those vesicles fuse with lysosomes. This pathway is the least speci c of all mentioned. Notably, several endocytic mechanisms can take place

simultaneously. The above mentioned review articles present abundant examples of all these mechanisms for different kinds of NPs tested in a variety of cells and tissues. Surprisingly, there is a scarcity of studies about internalization pathways and intracel- lular traf cking of NPs intended for ocular application (see Table 1 ). Intracellular metabolism of delivered drugs differs according to the internalization pathway. It implies a physical separation from the NP transporter, and their physicochemical characteristics will determine the degradation ratio. These are key aspects for the pharmacological activity of the carried molecule. Different kinetic processes are described to control the release pro les of drugs from NPs, including (i) desorption of the surface bound or adsorbed drug, (ii) diffusion through the NP matrix or the polymer wall, (iii) NP wall erosion, and (iv) a combination of erosion and diffusion processes ( Soppimath et al., 2001; Harush-Frenkel et al., 2008 ). For drugs that uniformly distribute or dissolve through the NP matrix, diffusion and biodegradation govern the release process. If drug diffusion is faster than polymeric degradation, drug release mainly occurs through diffusion; if not, then drug release will occurs through degradation of the polymer. Finally, there are other important factors involved in the design of a nanomedicine, including the modality of administration such as injection or topical application, and the features of the entrapped drug itself. This complex picture is completed with the requirement of minimal potential toxicity (see speci c chapter) and the contin- uous need for improved ef ciency. Different kinds of NPs prepared with some of the materials described above have been studied in the search for alternatives in ophthalmic treatments. We present in this review those that have been tested either in vitro or in vivo.

5. NPs and the anterior segment of the eye

The bioavailability of an instilled conventional drug onto the ocular surface is usually low. Most of it is lost due to physiological mechanisms, such as tear drainage and blinking, a few seconds after instillation ( Bayens and Gurny, 1997 ). Thus, there is a short pre-corneal residence time, usually less than 2 min, and a non- productive absorption thorough the well vascularised conjunctiva and the nasolachrymal drainage system ( Jarvinen et al., 1995 ). Therefore, the picture we face includes a very limited absorption of drug, a potential although limited access to intraocular tissues through the conjunctival e scleral pathway, and the risk of systemic side effects. For those reasons, intensive research in recent decades has focused on increasing the corneal penetration of topically applied drugs ( Schoenwald, 1990; Sasaki et al., 1999; Edwards and Prausnitz, 2001; Mannermaa et al., 2006 ). NPs were considered to offer the possibility of a more facile delivery and transport across tissues, and consequently their potential started to be studied.

Table 3 Main pathways for nanoparticle (NP) internalization and preferential degradation route.

Pathway

Target cells

Endocytic

Degradation route

Examples of NPs studied for ocular delivery

 

vesicles size

Phagocytosis Clathrin-mediated endocytosis

Professional phagocytes All mammalian cells

0.25 e 10 mm z 120 nm

Early endosomes and lysosomes Early endosomes and lysosomes

e SLNs loaded with pCMS-EGFP ( del Pozo-Rodríguez et al., 2008 ) HAS NPs loaded with SOD1 gene ( Mo et al., 2007 ) HA-CSO NPs loaded with pSEAP (Contreras-Ruiz et al., Submitted for publication) PLGA NPs loaded with 6-cumarin ( Qaddoumi et al., 2003 ) e

Caveolin-mediated endocytosis

All mammalian cells

z 60 nm

Endosomal accumulation;

 

nondegradative pathway

Clathrin- and caveolin-independent endocytosis Micropynocytosis

All mammalian cells

z 90 nm

Early endosomes

All mammalian cells

> 1 mm

Macropinosomes and lysosomes

SLNs, solid lipid NPs; HSA NPs, human serum albumin NPs; SOD1, Cu, Zn superoxide dismutase gene; pCMS-EGFP, plasmid DNA encoding enhanced green uorescent protein; HA-CSO, hyaluronic acid-chitosan oligomers; pSEAP, plasmid DNA encoding secreted alkaline phosphatise; PLGA, Poly- D -lactic-co-glycolide.

Y. Diebold, M. Calonge / Progress in Retinal and Eye Research 29 (2010) 596 e609

601

In 1986, an early attempt to use NPs was done with poly(butyl- cyanoacrylate) (PBCA) NPs loaded with progesterone, a highly lipophilic molecule ( Li et al., 1986 ). The NPs were topically applied to rabbit eyes and concentration e time pro les in different ocular tissues were assessed. The authors found a decreased progesterone concentration in tissues when compared to control solutions, suggesting a greater af nity of progesterone for the NP polymer than expected. As a consequence, progesterone was less available for absorption during its residence time in the pre-corneal area. The lesson learned from these experiments was that it is important to

optimize the physicochemical relations between the polymers and drug to obtain an ef cient carrier. Later, Losa et al. (1993) , used poly (isobutyl-cyanoacrylate) (PIBCA) and poly- e-caprolactone (PECL) nanocapsules for the ocular delivery of metipranolol, a beta-blocker used to treat certain types of glaucoma. In that work, the polymer coating was not responsible for the differences observed regarding the controlled release of the encapsulated drug. Instead, the drug release pro le was mainly in uenced by the type of oil core in the nanocapsule. However, a certain contribution of the polymer coating on emulsion stability was acknowledged. Our group later studied PECL nanocapsules loaded with 1% cyclosporine (CsA) in

a rat model of corneal transplantation rejection ( Juberías et al., 1998 ). This formulation had been developed to improve the

corneal penetration of CsA applied topically ( Calvo et al., 1996 ). This well-known immunosuppressive drug, widely used in trans- plantation patients, has severe systemic side effects such as neph- rotoxicity, hepatotoxicity, and hypertension. These side effects have limited its use in the local management of immune rejection of corneal grafts. By using PECL nanocapsules, toxic effects of systemic administration of CsA were avoided along with improved ocular penetration. While corneal graft rejection was not prevented by the CsA-loaded PECL NPs, the failure was not considered to be

a consequence of negative interactions between the polymer and the drug. Improved formulations using polymers with known biocom-

patibility and biodegradability, such as poly- D -lactic acid (PLA) and its copolymer glycolic acid (PLGA), were developed later. Flurbi- profen-loaded PLGA NPs successfully reduced in ammation in an

in vivo rabbit model of ocular in ammation ( Vega et al., 2006 ). In

comparison with commercial urbiprofen formulations, urbipro- fen-loaded NPs were more effective in reducing in ammation as evaluated by direct observation of clinical signs. More recently, we and others have developed applications using chitosan (CS)-based nanosystems. CS is a natural polysaccharide with interesting features, such as biocompatibility and biodegrad- ability, mucoadhesiveness, and the ability to transiently enhance the permeability of mucosal barriers. These features have made it quite useful in the development of drug delivery systems for transmucosal administration (for review see Alonso and Sánchez,

2003; Paolicelli et al., 2009 ). NPs made of CS and carbopol,

a cross-linked polymer of acrylic acid ( Kao et al., 2006 ), combined properties of both polymers, such as stability in aqueous solution, small size, improved biocompatibility, and positive charge to facilitate interaction with the negatively charged biological membranes. CS/carbopol NPs can be loaded with pilocarpine,

a parasympathomimetic drug used as treatment for open-angle

glaucoma. Patients with this form of glaucoma need to frequently instil pilocarpine eye-drops, which increases aqueous humour out ow for only 4 e 8 h Kao et al. (2006) compared the ef cacy of pilocarpine-loaded NPs against liposomes, gel, and eye-drops formulations of pilocarpine in normal rabbits. Pilocarpine induces a miotic response measured as a decrease in pupil diameter, and the effect of the CS/carbopol NPs lasted up to 24 h, much longer than the other formulations. The release pro le, previously studied in vitro, showed an initial burst release followed by a sustained

release for at least 24 h. These results are very promising, and further evaluation in a glaucoma animal model is now warranted. Recently, another CS-based nanosystem, CS/sodium alginate NPs, has been reported as a potential reservoir for topical delivery of gati oxacin, a potent fourth-generation uoroquinolone with low intraestromal penetration that is mainly used for microbial keratitis ( Motwani et al., 2008 ), but so far, the NP characterization has been only physicochemical, and no in vitro or in vivo experiments have been reported. The transport pathways by which NPs penetrate the ocular surface tissues are of great interest. Zimmer et al. (1991) studied the ocular transport pathway of uorescein-labeled PBCA NPs in rabbits. They found the uorescence signal localized inside conjunctival and corneal epithelial cells, and observed differences in depth of tissue penetration. They proposed a variety of pathways to explain their data, including NP endocytosis, lysis of cell membrane by NP metabolic degradation products, and a trans- cellular route. The transcellular route was also proposed for coated PECL nanocapsules ( de Campos et al., 2003 ). A critical question with regard to NP delivery of drugs is the concentration and duration of the drug in the target and surrounding tissues. Losa et al. (1991) tested PBCA NPs with different stabilizer agents to improve the binding of the antibiotic amikacin sulphate to the NPs. One of the formulations, using Dextran 70000 as a stabilizer, resulted in a signi cant increase of the amikacin concentration not only in the cornea but also in the aqueous humour. de Campos et al. (2001) studied the distribution of CsA-loaded CS NPs in different rabbit ocular tissues. In that study, therapeutic concentrations of this immunosuppressant drug adequate to modulate the local immune response were maintained in the cornea and conjunctiva for 48 h post-administration. However, those concentrations were not achieved using a formu- lation consisting on a CsA suspension in either a chitosan aqueous solution or a CsA suspension in water. The concentration of CsA in aqueous humour, iris, and ciliary body were extremely low. In addition, no detectable CsA levels were measured in plasma. Therefore, a prolonged local drug delivery was achieved using the CS NPs with no signi cant accumulation in intraocular tissues. The surface characteristics of the nanocarriers also have an in uence on the interaction with the ocular surface structures. For instance, CS-coated PECL nanocapsules enter the corneal epithe- lium in vivo more ef ciently than uncoated PECL or polyethylene glycol-coated PECL nanocapsules ( de Campos et al., 2003 ). We have already mentioned the exceptional biological features of CS, espe- cially mucoadhesiveness and the ability to transiently enhance the permeability of mucosal barriers. Our group is most interested in this approach using mucoadhesive NPs able to reach the anterior structures of the eye after topical administration. We started testing CS NPs because of the great potential envisioned for this polymer in the ophthalmology eld ( Alonso and Sánchez, 2003 ). We reported that CS NPs did not cause toxicity-related alterations in several cell lines derived from human conjunctiva epithelium ( de Campos et al., 2004; Enríquez-de-Salamanca et al., 2006 ). We were able to identify albumin-loaded CS NPs inside the cells by uorescence microscopy in a time-dependent manner. With in vivo experiments using albino rabbits, we con rmed that CS NPs were well-tolerated by ocular surface structures, causing no harm or in ammation, as determined by histopathological analysis. The corneal and conjunctival tissues that took up the CS NPs showed interesting tissue-related distribution patterns ( Fig. 3 ). Both cornea and conjunctiva incorporated the NPs; however, the conjunctiva was more permeable to the nanocarrier as shown by the deeper pene- tration into the epithelium and underlying stroma. Other authors have also explored similar strategies. For instance, Yenice et al. (2008) used hyaluronic acid (HA)-coated PECL nanospheres for

602

Y. Diebold, M. Calonge / Progress in Retinal and Eye Research 29 (2010) 596 e609

/ Progress in Retinal and Eye Research 29 (2010) 596 e 609 Fig. 3. CS NP

Fig. 3. CS NP in vivo uptake. Fluorescence microscopy of ocular surface structures of sham-treated (A, D), CS NP-treated (B, E) and contralateral control (C, F) rabbit eyes. Representative corneal (A, B, C) and conjunctival (D, E, F) sections are shown. No uorescence was detected in sham control corneas (A) or conjunctivas (D). (B) Corneal epithelial cells of the CS NP-treated rabbits were uniformly uorescent. Inset in (B): zoom area showing details of the corneal epithelium uorescence pattern. (E) Fluorescence in conjunctival epithelial cells was intense in the apical cell membranes and positive along the basolateral cell membrane. Inset in (E): zoom area showing the basolateral membrane uorescence staining in goblet and non-goblet cells. (C, F) Some uorescence was detected in corneal and conjunctival epithelial cells from contralateral control eye (OS), although much less intense than in the treated (OD) eye. Scale bar (A e F) ¼ 50 mm; Inset scale bar ¼ 10 mm. (Taken from Enríquez de Salamanca et al., Invest. Ophthalmol. Vis. Sci., 2006, with permission of the Association for Research in Vision and Ophthalmology).

corneal CsA delivery. The concentrations in rabbit cornea were 10 e15-fold higher than that achieved when CsA was administered as solution in castor oil. These results moved us to explore different biomaterial combinations intended speci cally for the ocular surface tissues. Consequently we joined liposomes and CS NPs to form a new nanosystem that we termed liposome e chitosan nanoparticles (LCS-NPs) ( Diebold et al., 2007 ). The theoretical advantages of these complexes are the combination liposome biocompatibility with biological membranes and the demonstrated properties of CS NPs. We tested three different complex formulations, showing again their potential for ocular administration. Using mucus-producing primary cultures of conjunctival epithelium, we observed that the three nanosystems were rst retained in the in vitro mucus layer

and then entered the epithelial cells, depending on the particular NP composition ( Fig. 4 ). We consider this feature a potential advantage that can be used to modulate the retention time of the nanocarrier in the tear lm. Different in vivo retention times would be required depending on the encapsulated drug. Use of different lipid carriers has had renewed interest. An example is the recent paper by Attama et al. (2009) that presents phospholipid nanoparticles made in theobroma oil. These were designed to incorporate timolol hydrogen maleate, a water-soluble drug used as a rst-choice treatment for glaucoma. Using a modi- ed Franz diffusion cell apparatus, they did drug permeation experiments using a bio-engineered cornea construct, composed of human immortalized corneal endothelial cells, stromal broblasts, and epithelial cells. This in vitro study, although showing promising

cells. This in vitro study, although showing promising Fig. 4. Confocal microscopy images of primary cultures

Fig. 4. Confocal microscopy images of primary cultures of human conjunctival epithelium. Control cultures had normal morphology when viewed with transmitted light (TL). During 30 min of incubation, LCS-NP complexes (green) passed through the mucus layer and were present in deeper cell layers. They formed aggregates with different patterns for each type of LCS-NP tested. Z-series prole images (lowest panels) are projections of stacked image proles from optical sections captured along the Z -axis. These con rmed the intracellular presence of LCS-NPs (green), which were localized among the actin bres in the cytoskeleton (red) stained with phalloidin. Scale bar ¼ 25 mm. Images are repre- sentative of at least three independent experiments. (Taken from Diebold et al., 2007 (Biomaterials), with permission of Elsevier).

Y. Diebold, M. Calonge / Progress in Retinal and Eye Research 29 (2010) 596 e609

603

results in terms of sustained release and timolol permeation, lacks a complementary toxicity study. We have recently started working with NPs composed of hya- luronic acid and chitosan (HA-CS NPs) ( de la Fuente et al., 2008a ). This new development is capable of encapsulating macromolecules of both hydrophilic nature, such as the protein bovine serum albumin, and of hydrophobic nature, such as the polypeptide CsA. Also, HA-CS NPs can carry plasmid DNA and may be suitable for gene delivery as shown with ocular surface-derived cell lines ( de la Fuente et al., 2008b ). The most important aspect is the good in vivo tolerance of these nanosystems ( Contreras-Ruiz et al., 2009 ), which opens the possibility of testing them in actual disease models. By exploiting the known properties of HA-CS NPs, our group is currently working to design speci c nanomedicines that utilize gene therapy to target certain ocular surface in ammatory diseases. It is also worth mentioning that primary open-angle glaucoma is the most widespread neuropathy, affecting 60.5 million people in the world ( Quigley and Broman, 2006 ). Aside from surgery, the only partially effective treatment for this disease is IOP-reducing drugs applied topically onto the cornea. The necessity of numerous daily applications of eye-drops for life is a serious issue for patients, besides the risk of side effects in the anterior structures of the eye. While there was an early interest in drug delivery systems, such as the Ocusert in 1974 for pilocarpine sustained release, there are surprisingly few reports testing NPs intended for glaucoma treat- ment. Zimmer et al. (1994) tested different pharmaceutical aspects of pilocarpine-loaded PBCA NPs versus a standard pilocarpine solution in an elevated IOP rabbit model. The NPs were better in reducing IOP and maintaining miosis, especially at lower drug concentrations, than the drug solution. The NPs induced maximum reduction of IOP at 2 e 3 h, whereas the drug solution maximum response was at 1 e2 h. Epinephrine-loaded poly-N-isopropylacrylamide NPs were also tested in rabbit to evaluate IOP-lowering effect ( Hsiue et al., 2002 ). The polymer in this nanosystem is thermosensitive and undergoes a phase transition when the temperature increases to about 32 C. This allows the progressive release of epinephrine after being topically administered. The NPs had six times more long- lasting effect compared to conventional eye-drops. Finally, Wadhawa et al. (2009) reported a signicant IOP reduction in rabbit eyes exposed to CS-HA NPs loaded with timolol compared to stan- dard timolol eye-drops or blank NPs. There were no irritant effects. Although not properly NPs but a nanodevice, it is worth mentioning here the development of a prototype for a nanodrainage system to be implanted in the sclera as a bypass route for humour aqueous outow ( Pan et al., 2006 ). This new concept might revolutionize glaucoma treatment in the coming years.

6. NPs and the posterior segment of the eye

Even though the cornea constitutes one of the most selective barriers to foreign molecules for the eye, transcorneal penetration of topically administered ophthalmic medicines intended for the posterior segment is persistently sought. The reason is that currently, the best way to treat intraocular in ammation, either infectious or non-infectious, is by injecting drugs into the vitreous. The vitreous is a gelatinous, cell-free structure that is capable of retaining molecules and also delivering them to nearby structures, such as the ciliary body or the RPE, a vital component of the retina. Frequent intraocular injections are needed to treat retinal disor- ders. With these injections come potential undesired side effects, higher risk of infections, and poor acceptance by the patient. More frequent side effects associated with repeated intravitreal injec- tions include increased risk of cataract development, vitreous hemorrhage, retinal detachment, and endophthalmitis.

The prospect of frequent intravitreal injections to treat serious intraocular disorders affecting the choroid and retina has moved researchers to look for better solutions derived from the use of NPs as drug carriers. However, the scenario is quite challenging for NPs. Typically, the cornea is penetrated by less than 5% of drugs applied as liquid eye-drops ( Keister et al., 1991 ) because of the limitations mostly imposed by tear turnover associated with blinking and the nasolacrymal drainage system. In addition, the drugs must diffuse

a great distance between the ocular surface and the intraocular

targets. Therefore, usually eye-drops do not provide suf cient drug

concentration in the posterior ocular tissues. On the other hand, systemic drug administration delivered through the blood vascular system is not very effective because of the uveal blood e aqueous and blood e retina barriers. As a consequence, a poor dos-

e e response pro le for vitreoretinal diseases is generally achieved, and a large amount of the drug is needed to maintain therapeutic levels, usually for insuf cient amounts of time ( Geroski and Edelhauser, 2001 ). Additionally, the high concentration of drugs needed to penetrate the blood e ocular barriers is often associated with systemic side effects. Thus, much effort has been invested in the last decade to opti- mize drug delivery systems for intraocular treatment. Alternatives to intravitreal or periocular injections, including scleral implants and devices, transdermal patches, and different iontophoretic devices including hydrogel reservoirs, have been explored with variable results (for review, see del Amo and Urtti, 2008 ). However, the ability to achieve long-term drug delivery in the retina and nearby tissues while reducing the number of intraocular injections

to just one seems feasible at this time. Several kinds of NPs carrying

diverse active molecules, including genetic material, are currently in pre-clinical studies using the above mentioned approaches

( Bourges et al., 2003; Bejjani et al., 2005; Normand et al., 2005; Irache et al., 2005; Farjo et al., 2006; Paasonen et al., 2007 ). The underlying idea is to take advantage of the vitreous capacity for retaining and delivering molecules to tissues with which it is in direct contact and to use it as a biological reservoir once the NPs are placed inside. If therapeutic levels of a drug can be maintained for months after a single intravitreal injection, that would be consid- ered an enormous improvement for the quality of life of many patients. There are promising studies reported in the recent literature on the use of intravitreally injected NPs. Ganciclovir-loaded albumin NPs are an interesting example. Ganciclovir is one of the standard treatments for cytomegalovirus retinitis, a prevalent infectious retinal disease in immunosuppressed patients, such as those with AIDS. In vitro experiments demonstrated that albumin NPs released ganciclovir in a sustained way ( Merodio et al., 2001 ), with

a signi cant improvement of drug uptake by cytomegalovirus-

infected human cells ( Merodio et al., 2002 ). For single intravitreal

injections in rats, these NPs were safe, well-tolerated carriers not only for ganciclovir, but also for the anti-cytomegaloviral oligonu- cleotide analog formivirsen. They were present in the vitreous and ciliary body for at least two weeks ( Irache et al., 2005 ). RPE cells have the capacity to take up different kinds of NPs ( Bourges et al., 2003; Bejjani et al., 2005; Normand et al., 2005 ) opening the possibility of using them to treat retinal disorders associated with ageing or photoreceptor dystrophies. The purpose will be to target these cells with speci c molecules or genetic material capable of reversing or stopping the processes leading to these diseases. Bourges et al. (2003) tested in rats intravitreal PLA NPs loaded with uorochromes and showed a preferential locali- zation at the RPE cells after 24 h. The most interesting achievement was that RPE cells retained the NPs, which continuously delivered the uorochrome for months after the single injection. Fluores- cence diffusing from the NPs was observed in distant parts of

604

Y. Diebold, M. Calonge / Progress in Retinal and Eye Research 29 (2010) 596 e609

retinal tissue, ganglion cells, and rod outer segments for up to four months after the injection. In contrast, detection after injection of the free uorochrome lasted barely one week. Later, Bejjani et al. (2005) studied in vitro and in vivo PLA and PLGA NPs loaded not only with uorochromes but also with model plasmids. NPs encapsulating a plasmid encoding red nuclear uorescent protein were localized in the RPE cells 24 h after intravitreal injection in rats. Effective plasmid expression was achieved after four days of injection and expression-associated red uorescence remained detectable in RPE cells during the following three weeks, with no apparent tissue damage or toxicity. In all of these studies, the association of the delivered molecules with the NPs and kinetic release rate pro les were determined prior to the in vivo experiments.

From a therapeutic point of view, not only is the reduction in the number of intravitreal injections by the use of NPs a goal, but the improvement in intraocular availability of topically applied drugs is an important and remarkably challenging goal. An illustrative example is the case of steroidal and non-steroidal anti-in amma- tory drugs. There are many clinical situations in which these drugs are normally used. However, because they are almost completely insoluble in water and because they are effectively excluded from intraocular sites by the various blood e ocular barriers, the ability to reach intraocular targets is quite low. For instance, surgical traumas, such as cataract surgery, cause miosis that is treated with non-steroidal anti-in ammatory drugs. Pignatello et al. (2002) developed ibuprofen-loaded NPs made of inert polymeric resins (Eudragit RS100) optimized as a pharmaceutical preparation. The NPs showed a controlled ibuprofen release pro le in vitro, and they had a high ef cacy in reducing miosis with an in vivo model of ocular trauma. The therapeutic effect was achieved with lower drug concentration than in an eye-drop formulation and without any toxic effect on ocular tissues. Recently, Kassem et al. (2007) studied different nanosuspensions prepared by high-pressure homogeni- zation of three insoluble glucocorticoids: hydrocortisone, prednis- olone, and dexamethasone. Using normotensive rabbits, they determined if the glucocorticoid-associated NPs instilled into the lower cul-de-sac demonstrated enhanced absorption and improved intensity of drug action. Based on the measured increase in intra- ocular pressure (IOP), they reported not only improvements in both, but also a signi cant extension of the glucocorticoid action. Interestingly, intravenously injected PLA NPs encapsulating beta- methasone phosphate effectively controlled in ammation in a rat model of experimental autoimmune uveoretinitis ( Sakai et al.,

2006 ).

The studies described above have opened a new perspective for the treatment of retinal and uveal disorders. Nevertheless, to implement this kind of delivery system in a clinical setting, more functional studies are needed to exclude any impairment of the retinal function and vision and the development of accompanying chronic in ammatory processes. Surely, in coming years we shall see reports dealing with all of these topics.

7. NPs and gene delivery/therapy

The Roadmap for Nanomedicine ( http://nihroadmap.nih.gov/ nanomedicine/ ) released by the NIH presents NPs as a strategy to improve non-viral gene transfer. The eye is an excellent candidate for gene therapy for two main reasons: it has immune privilege, and it is affected by many well understood genetic-based diseases. Immune privilege means that the immune system is driven towards tolerance to foreign antigens rather than rejection and in ammation, the normal responses. Immune tolerance serves to protect vision by avoiding the collateral in ammation that is associate d with the immune response against any antigen. Also, it is

a small and closed organ, with very limited diffusion of locally

applied active molecules to the bloodstream because of the blood e tissue barriers. Hence, there is a growing interest in exploring the suitability of gene delivery strategies in ocular

therapy since the 90s ( Nussenblatt and Csaky, 1997; Tanelian et al.,

1997 ).

Genetic-based therapies can be developed using different nucleic acids such as DNA, antisense oligonucleotides (AS-ODNs), small interfering RNA (siRNA), and aptamers. AS-ODNs are synthetic molecules of short sequences, 13 e 25 nucleotides, that bind to speci c mRNAs. By binding to the mRNA molecules, AS-ODNs are capable of stopping translation of the mRNA and, consequently, block the protein synthesis of the targeted gene ( Loke et al., 1989 ). siRNAs share with AS-ODNs the capacity of blocking protein synthesis from a given mRNA. However, the gene silencing mech- anism by which it is performed, called RNA interference (RNAi), is different ( Leung and Whittaker, 2005; Bumcrot et al., 2006 ). RNAi is induced in mammalian cells by means of synthetic double-stranded siRNAs. These molecules have small sequences, 21 e23 nucleotides, are highly selective and sequence-speci c, and have better stability compared to that of AS-ODNs. Finally, aptamers are synthetic single- stranded DNA or RNA molecules with a unique 3-D structure. They are able to speci cally bind other molecules and are particularly prone to bind the functional domains. This feature makes them useful as modulators of the targeted molecule ( Proske et al., 2005; Nimjee et al., 2005 ). Therapeutic applications of all of these mole- cules in the eye have been extensively reviewed ( Borrás, 2003; Henry et al., 2004; Campochiaro, 2006; Fattal and Bochot, 2006, 2008; Levy-Nissenbaum et al., 2008 ). Delivery of genetic material is quite a challenge from a phar- maceutical point of view. It is unstable in biological uids and has poor cell penetration due to its size or charge. For instance, plasmid DNA is large; however, siRNA is quite small. These facts imply that suitable carriers to deliver it to ocular tissues are needed. A partially successful approach in recent years has been the use of viral vectors such as adenovirus, retrovirus, lentivirus, and mainly recombinant adeno-associated virus (rAAV), as gene carriers ( Snyder, 1999 ). rAVV vectors carry single-stranded DNA which is inserted into the genome of the targeted cell. In general, gene delivery using rAAV shows a lack of pathogenicity, good long-term transgene expres- sion, and no toxicity. However this technology has limitations such as lack of effective transduction in some cell types or presence of neutralizing antibodies for some rAVV serotypes ( Rabinowitz and

Samulski, 1998; Grimm and Kay, 2003 ). Also, progress has been made to deliver naked DNA to cells ( Herweijer and Wolff, 2003 ).

Different studies have established that some viral vectors can

ef ciently deliver transgenes to ocular tissues while others cannot.

For instance, several rAAV serotype vectors appear unsuitable for anterior segment delivery; however, rAAVs appear to have a selec- tive tropism for retinal ganglion cells ( Borrás et al., 2002 ). There are examples of effective rescue of genetic de cits in the eye using viral

and rAAV-mediated gene therapy in different in vitro and in vivo models ( Campochiaro, 2002; Martin et al., 2002; Borrás, 2003; Ziche et al., 2004; Ralph et al., 2006 ; Roy et al., 2010 ). One remarkable use of this technology is the recovery of visual function in RPE65-de cient dogs. Genetic de ciency of RPE65, a protein involved in retinoid metabolism in RPE cells, results in blindness similar to human Leber s Congenital Amaurosis. Recovery of visual function in the dogs was achieved after subretinal injection of rAAV encoding RPE65 ( Acland et al., 2001 ). These important results have led to a gene therapy clinical trial for the RPE65-de cient form of the human disease ( Bainbridge et al., 2008; Cideciyan et al., 2008 ). However, the use of viral vectors poses risks for the safety of patients. Additionally, the effectiveness in the eye may be limited due to several factors, including cell tropism, the size of the

Y. Diebold, M. Calonge / Progress in Retinal and Eye Research 29 (2010) 596 e609

605

sequence to be carried and expressed, and most importantly, host immunity ( Bennett, 2003 ). Despite the promising results of virally- mediated ocular gene therapy, non-viral carriers would be the preferred choice. Still, the issue of an optimal delivery system remains to be solved. Features that gene delivery systems should provide for optimal activity of the delivered nucleic acid molecule include improved stability, increased half-life, speci c tissue and cellular targeting, improved cellular penetration, and release of the molecule in the right intracellular compartment (i.e., DNA in the nucleus, however, siRNAs in the cytoplasm). Hence, gene delivery using non-viral carrier systems holds promise of being safer and more effective, as those limiting factors cited above do not in uence them so strongly. Several investiga- tions have started using different approaches, including naked DNA delivery by means of electroporation and the formation of lip- oplexes by DNA condensation with cationic lipids among others ( Kachi et al., 2005 ; Bloquel et al., 2006 ; Johnson et al., 2008 ). Regarding genetic material incorporation to NPs, there are exam- ples using compacted DNA NPs. Regarding the anterior segment of the eye, our group has obtained promising results using HA-chitosan oligomers (CSO) NPs loaded with a model plasmid encoding for alkaline phosphatase (manuscript submitted). We are currently studying the intracellular trafcking of those nanoparticles and contents in human epithelial cell lines derived from the ocular surface. Our preliminary results show an effective delivery of the plasmid to the cell interior and alkaline phosphatase expression 48 h after transfection using HA- CSO NPs. Another recent example is the work published by Klausner et al. (2010) in which ultrapure CSO (NOVAFECT) was used to complex with a model plasmid encoding for green uorescent protein (pGFP) to form NPs. Transgene expression of the pGFP was detected in cell cultures. In rat corneas pGFP was expressed 5.4 times that of control polyethylenimine-pGFP NPs. It is worth mentioning that in vivo transgene expression was identi ed in corneal stroma but not in the epithelium or the endothelium. Another approach is DNA condensed with polycationic poly- mers. Farjo et al. (2006) packaged compacted DNA in PEG- substituted lysine peptides that formed the NPs. The compacted DNA was a model plasmid encoding for enhanced green uores- cent protein (pEGFP), and its expression was studied in mice. Two days after an intravitreal injection, uorescence was present in the lens, cornea, trabecular meshwork, sclera, choroid, RPE, and other retinal cells. However, two days after a subretinal injection, uo- rescence was restricted to retina and RPE cells, choroid, and sclera, with a minimal presence in the lens. The authors reported no alteration in visual function, evaluated by electroretinography, and no evidence of in ammation in the histological analysis of the ocular tissues. In this study, the duration of DNA expression was not reported. Ding et al. (2009) recently reported the preparation of single molecule of pEGFP compacted with PEG-substituted polysysine (CK30PEG). The formed NPs were subretinally injected in mice and resulted in ef cient retinal cells transfection. This work was more focused on toxicity issues related with that tech- nology. There were no signs of local in ammatory response in terms of in ltration of in ammatory cells or chemokine marker expression. Even though this method of gene therapy is poten- tially applicable for multiple ocular diseases, a more directed tar- geting is desirable. A quite recent concept is the use of light-sensitive NPs made by combining a structural protein of the Herpes simplex virus, VP22, with AS-ODNs bound through the C-terminal end of the viral protein. This leads to the formation of spherical NPs of 0.3 e1 mm in diameter named vectosomes . AS-ODNs selectively modulate the expression of a given gene by displaying a base sequence that is complementary to a speci c mRNA ( Helene and Toulme, 1990 ). The

precise local delivery of bound ODNs in target cells is controlled by illuminating them ( Normand et al., 2001; Zavaglia et al., 2003 ). This technology, used for tumour cells, is particularly interesting in ophthalmology due to the frequent use of lasers for therapeutic purposes. Light-induced delivery of AS-ODNs from vectosomes has been studied in vitro, using human melanoma and RPE cell lines,

and in vivo with rats ( Normand et al., 2005 ). Interestingly, the vectosomes followed a rapid transretinal migration pattern after intravitreal injection in rats, and they were internalized by RPE cells and other cell types. Transscleral illumination of injected eyes induced disruption of the light-sensitive vectosomes and migration of released AS-ODNs to the cell nucleus, where they were localized in a light-dependent manner. Not only were the AS-ODNs localized in the RPE cells, they were also detected in ganglion cells, inner nuclear layer, and even in the choroid. Much work is needed to learn about the light wavelengths and energy produced as

a consequence of vectosome rupture, which can potentially harm

the retina. However, this approach offers great therapeutic potential. A nal example of potential gene therapy treatment of retinal diseases is the recent work of Park et al. (2009) . They encapsulated in PLGA NPs an expression plasmid for a natural angiogenic inhibitor, plasminogen kringle 5 (K5) that inhibits ischemia- induced neovascularisation in a rat model of oxygen-induced reti- nopathy (OIR). After administering K5-loaded NPs intravitreally to animals with OIR, the authors reported high K5 expression in the inner retina for four weeks. Also, retinal vascular leakage and retina neovascularisation were reduced in those K5-NP injected eyes when compared to fellow control eyes. Even though these possibilities make those of us who work in the development of new therapies in ophthalmology dream of the end of many handicapping disorders, many challenges remain. For instance, we must consider the potential capacity of the vitreous to

act as a barrier for gene delivery, mainly due to its composition and biological characteristics that affect diffusion of large molecules ( Xu et al., 2000; Peeters et al., 2005 ). We need a better under- standing of the biological processes affecting intraocular structures to help design more speci c solutions. The identi cation of more target genes involved in the development of each pathological condition is necessary. We must learn about potential immune responses derived from the use of different non-viral vectors. For pre-clinical studies, it is important to have available and select carefully appropriate animal models of the target disease. Finally, we will need to identify those elements that control the long-term expression of the delivered transgenes or the permanent shutting off of genes with delivered AS-ODNs.

8. Nanoparticle safety: toxicity and interaction with the immune system

Not just ocular, but any biomedical application of NPs as

a therapeutic agent requires biocompatibility. This means that NPs,

both the components and the assembled NP itself, need to be biologically compatible with living tissues by not producing toxic, injurious, or immunological responses in them. Key aspects in u- encing the biocompatibility of NPs are the physicochemical char- acteristics, such as size, shape, charge, solubility, and chemical groups on the surface that provide particle charge and lipo- or hydro-phobic features ( McNeil, 2005 ). However, the same proper- ties that make NPs attractive for biomedical applications may make them reactive in biological systems and develop toxicity. For instance, smaller size NPs are preferred for better interactions at the cellular level. However, smaller NPs have larger surface area per unit mass, which may mean higher reactivity and consequently, cell or tissue toxicity ( Kipen and Laskin, 2005 ).

606

Y. Diebold, M. Calonge / Progress in Retinal and Eye Research 29 (2010) 596 e609

Risks posed by both organic and inorganic NPs include aggre- gation, tissue accumulation, and adsorption of plasma proteins onto the surface. This latter consideration is particularly important when thinking about an intravenous NP administration or a potential access of the blood system using other non-systemic administration routes, e.g., intraocular administration to treat posterior segment diseases that involve blood e retina barrier impairment. NP aggre- gation may block cell metabolism or even impair tissue function. For instance, aggregation of topically applied NPs onto the ocular surface may block the lachrymal drainage punctum and impair tear lm recycling. Additionally, indiscriminate NP accumulation in ocular tissues may distort tissue architecture and consequently, alter function. Finally, there may be toxic effects due to the presence of high levels of the loaded drug in a non-target tissue. Potential cytotoxic activity of NPs may include alterations of cell membranes such as membrane disruption, as has been described for carbon nanotubes ( Panessa-Warren et al., 2009 ). However, sometimes this particular property may be sought, especially for gene delivery ( Kiang et al., 2004; Akagi et al., 2010 ). For instance, Kiang et al. (2004) used poly (propyl acrylic acid) to formulate chitosan-DNA NPs with enhanced in vitro transfection ef ciency. That polymer was speci cally designed to disrupt the lipid bilayer in cell membranes. By changes in the pH, it triggered the release of DNA from the endosomal compartment. Very recently, Akagi et al. (2010) evaluated the relationship of different physicochemical characteristics of 200 nm-size NPs composed of poly(gamma-glu- tamic acid). The protein-loaded NPs had signi cant haemolytic activity in erythrocytes, depending on NP hydrophobicity and pH, with the greatest activity present at pH 7 to 5.5 and absent at physiological pH. Noxious effects of different kinds of NPs have been reported in several organ systems (for review please see Medina et al., 2007 ). One of the main mechanisms described by which NPs may harm cells and tissues is oxidative stress generation, which in turn, may lead to the activation of different transcription factors ( Medina et al., 2007 ). Generally, NPs can be taken up by lymphatic nodes and distributed through the lymphatic system in parallel with the blood vascular system. The ocular mucosa possesses lymphoid

tissue that drains to different face and neck lymphatic ganglia. TW Prow (2009) recently published a very comprehensive review about the toxicity of nanomaterials in the eye. He nicely summa- rized in vitro and in vivo relevant studies accomplished since 1996 for about thirty different kinds of nanoparticles and other nano- carriers tested for ocular applications. The types of toxicity reported included cell morphology and viability, clinical signs evaluation, gross tissue examination, irritation test, histology and functional analyses, and in ammatory response ( Table 4 ). That review high- lights the importance of accomplishing toxicity testing of newly developed drug carrier candidates. Another consideration is the potential in ammatory, immu- nostimulatory, and immunosuppressive properties described for different kinds of NPs ( Dobrovolskaia and McNeil, 2007; Zolnik et al., 2010 ). There are limited data available about this topic ( Zolnik et al., 2010 ), but it is known that some NPs are antigenic themselves. The antigenicity depends on particle size, especially those of ultra-small size (25 nm or smaller) that improves lymphatic uptake, and surface charge ( Reddy et al., 2007; Manolova et al., 2008 ). Allergic or hypersensitivity reactions can be induced or aggravated in animal models and humans by dendrimers ( Toyama et al., 2008 ), carbon nanotubes ( Nygaard et al., 2009 ), lipid-based NPs ( Szebeni et al., 2007 ), titanium dioxide NPs ( Yanagisawa et al., 2009 ), and polystyrene NPs ( Yanagisawa et al., 2010 ). However, it is important to bear in mind that sometimes NPs are speci cally designed to target the immune system, and interactions with immune cells are considered bene cial. Such is the case for those NPs intended for vaccine development, in which the immunogenic properties are exploited. NPs can serve as adjuvants as they can be conjugated with antigens. Currently, a major issue related to the development of NP-based novel therapies is the rigorous evaluation of the potential immu- notoxic effects. Researchers in the eld of nanomedicine agree that the potential environmental and health-related risks should be carefully analyzed. Importantly, we, in the eye community, test NPs for toxic effects less than we should. The reasons for that are many, but the most important one is the absence of a common regulatory framework. Technological advancements develop faster than

Table 4 Summary of key questions to be addressed during early phase pre-clinical evaluation of nanoparticles (NPs) intended for use in biomedical applications.

Assay category

Questions to address

In vitro Hemolysis Platelet aggregation Coagulation time Complement activation Colony-forming unit granulocyte macrophage Leukocyte proliferation Uptake by macrophages Cytokine induction

Nitric oxide production Cytotoxicity of natural killer cells Endotoxin contamination Microbial/viral/mycoplasma contamination

Do NPs change integrity of red blood cells? Do NPs interfere with cellular components of the blood coagulation cascade? Do NPs cause changes in the function of the coagulation factors? Do NPs activate the complement system? Do NPs cause myelosuppression (toxicity to bone marrow precursors)? Do NPs have adverse effects on leukocyte proliferative responses? Are NPs internalized by specialized phagocytes? Do NPs activate immune cells to elicit cytokine production or interfere with that caused by known immunogens? Do NPs induce oxidative stress? Indirect test for potential endotoxin contamination Do NPs interfere with the ability of natural killer cells to recognize and kill tumour target cells? Pyrogen contamination test Sterility test

In vivo Single-dose toxicity study:

Standard toxicity tests (blood chemistry, hematology, histopathology, gross pathology)

T cell dependent antibody response (TDAR) Host resistance studies; evaluation of cell-mediated immunity

Repeated dose toxicity study; immunogenicity

These tests aim at answering the following questions:

Do NPs cause toxicity to immune cells and organs? Are there any indications for additional toxicity studies? Additional studies are conducted on a case-by-case basis using weight-of-evidence approach This test is recognized for its high predictability for human models These tests are recommended for 1) testing the potential effects that particles might have on host resistance towards pathogens and tumour cells, and 2) to check for contact sensitization and delayed type hypersensitivity reactions Do NPs elicit particle speci c immune response?

Y. Diebold, M. Calonge / Progress in Retinal and Eye Research 29 (2010) 596 e609

607

regulations in the nanomedicine eld. Standardization of proce- dures to study immunological properties and toxicology-related issues would also help. There certainly are regulations in Europe, U.S., and Japan intended to assess the immunotoxic potential of newly developed pharmaceuticals ( Putman et al., 2003; Snodin, 2004 ). However, there are no speci c protocols for those nano- technology-based tests because the properties of the NPs may interfere with the established tests ( Stone et al., 2009 ). A rst step towards that purpose has been taken by the National Characterization Laboratory, U.S. National Cancer Institute ( http:// ncl.cancer.gov/working_assay-cascade.asp ), whose mission is to perform and standardize the pre-clinical characterization of nanomaterials intended for cancer therapeutics and diagnostics. As an example, in a recent review Dobrovolskaia and McNeil (2007) suggested the most important parameters that need to be addressed during an initial evaluation of new nanotechnology- derived pharmaceuticals ( Table 4 ).

9. Future directions

A novel direction for nanomedical applications involves nano- ceria particles. This unique type of NP is made of nanocrystalline cerium oxide, CeO 2 , also known as ceria. It is a rare earth oxide from the lanthanide series of the periodic table. Properly speaking, nanoceria particles do not deliver any drug; rather they are the drugs. These NPs can scavenge free radicals and reactive oxygen species ( Heckert et al., 2008 ). Interest in them for biomedical applications derives from several in vitro studies in which they increased the lifespan of cultured brain cells (Rzigalinski et al., 2003 ), protected non-tumoral cells from radiation therapy effects ( Tarnuzzer et al., 2005 ), and protected in vitro rat spinal cord neurons from oxida- tive stress ( Das et al., 2007 ). For ocular applications, there are only a few studies with promising results. One of them, by Chen et al. (2006) , showed that nanoceria particles were effective in the inhi- bition of the reactive oxygen intermediate-induced photoreceptor cell death. As macular degeneration and retinitis pigmentosa, among other blinding diseases, are thought to generate reactive oxygen species, the use of nanoceria may be a useful therapeutic strategy. In another study, Pierscionek et al. (2010) showed the potential of antioxidant nanoceria particles in cataract treatments. It is certain that more studies will arise in the near future to explore the potential bene t of nanoceria particles in different eye diseases. Other inorganic NPs made of noble metals, such as gold ( Hayashi et al., 2009 ) or silver ( Gurunathan et al., 2009 ), are generating much interest due to their small size, about 20 nm, and the great potential of traversing the blood e retina barrier ( Kim et al., 2009 ). There are very few reports, and most of them show preliminary results. More and deeper studies using these kinds of small NPs are necessary to understand the pharmacokinetics and clearance mechanisms ( Amrite et al., 2008 ) and the actual therapeutic potential. Also, the use of inorganic NPs as novel contrast agents for molecular imaging in other tissues suggests that this technology can be applied to ocular imaging. A few interesting examples of NPs used as imaging agents are gadolinium-loaded nanoemulsions for brain imaging, gold NPs for optical coherence tomography imaging, and superparamagnetic iron oxide NPs with gold shells for magnetic resonance imaging. Regarding ocular tissues, Yamamoto et al. (2007) have shown that quantum dots can be used for imaging the vitreous. No doubt, the coming years will bring exciting developments in this eld.

10. Summary and conclusions

It is currently possible to design nanocarriers with speci c delivery requirements for ocular administration. Those carriers are

able to safely deliver the loaded therapeutic molecule while pre- venting damage or deactivation of it. The loaded agents can act more ef ciently and with fewer side effects when compared to the same agents administered without the nanocarrier. However, many questions still remain. It is an urgent matter to resolve them so that new and more ef cient drug formulations based on NP technology for ocular therapy can be made available for patient care.

Author disclosure statement

The authors report no nancial interest in any of the materials or nanosystems presented in this review.

Acknowledgements

This work was supported by a Spanish Ministry of Science and Technology Grant (MAT2007-64626-C02) and the Networking Research Centre on Bioengineering, Biomaterials and Nano- medicine (CIBER-BBN), Spain. CIBER-BBN is an initiative funded by the VI National R&D&I Plan 2008 e 2011, Iniciativa Ingenio 2010 , Consolider Program, CIBER Actions and nanced by the Instituto de Salud Carlos III with assistance from the European Regional Devel- opment Fund.

References

Acland, G.M., Aguirre, G.D., Ray, J., Zhang, Q., Aleman, T.S., Cideciyan, A.V., Pearce- Kelling, S.E., Anand, V., Zeng, Y., Maguire, A.M., Jacobson, S.G., Hauswirth, W.W., Bennet, J., 2001. Gene therapy restores vision in a canine model of childhood blindness. Nat. Genet. 28, 92 e 95. Aggarwal, P., Hall, J.B., McLeland, C.B., Dobrovolskaia, M.A., McNeil, S.E., 2009. Nanoparticle interaction with plasma proteins as it relates to particle bio- distribution, biocompatibility and therapeutic ef cacy. Adv. Drug Deliv. Rev. 61,