Gold nanoparticles (Au-NPs) in the inhibition of retinal neoangiogenesis

There are currently several drugs that have been developed against retinal neovascularisation and used in ophthalmology, but the excessive costs and the possible presence of medium- and long-term complications associated with their administration have prompted researchers to search for alternative routes for inhibiting neoangiogenesis. These include the gold nanoparticles (Au-NPs) certainly have good potential, by virtue of their low cytotoxicity, their ability to withstand surface modifications induced by thiol-containing molecules, to immobilise a wide range of biomolecules (amino acids, enzymes, DNA) and for their high optical quenching coefficient [7].
The characteristics of gold as a stable metal particle were probably already known in ancient Egypt and China in the 4th-5th centuries BC. To this period dates the well-known Cup of Lycurgus, a characteristic Roman glass artefact that shows an intense red colour when light shines through it and an opaque green colour when light is reflected, due to the presence of small amounts of gold and silver in the form of nanoparticles within the glass itself.
It is of Francisci Antonii the first book on gold particles [2] (od colloidal gold, as it was called) and dates back to 1618; this work contains information on the preparation of colloidal gold as well as interesting clinical cases.
Over time and up to the present day, various methods have been described for the preparation and synthesis of colloidal gold and what we now more correctly call gold nanoparticles (Au-NPs). Two different types of approach currently exist: the Top Down and those of Bottom Up. The former provide for the synthesis of nanostructures from the bulk material; the latter, on the other hand, exploit chemical reactions whereby nanostructures are prepared from simple molecules that are assembled to the desired size. Among the methods of Bottom Up there is also the possibility of obtaining Au-NPs using biological systems, i.e. small quantities of microorganisms such as fungi or bacteria, such as the Bacillus licheniformis or the Brevibacterium casei.
The aim of this discussion is precisely to highlight the characteristics of Au-NPs nanoparticles, their functions, possible mechanisms of action and their potential clinical use by analysing the most significant studies in this field.

Biological synthesis of gold nanoparticles from the Bacillus licheniformis
Beveridge and Murray were the first to describe in 1980 that the exposure of Bacillus subtilis with tetrachloroaurate (AuCl4^–) enabled the synthesis of gold nanoparticles [3].
The synthesis of these gold nanoparticles by biological methods undoubtedly has several advantages: it is tightly controlled, highly reproducible, offers the possibility of creating biocompatible particles, and does not involve the use of toxic surfactants or organic solvents. Moreover, the bacteria are easy to handle and can be genetically modified.
What Kalishwaralal et al. describe is the synthesis of stable gold nanocubes through the reduction of AuCl4^– in aqueous solution through the Bacillus licheniformisat room temperature [4]. The process requires a single step without the use of chemical toxins rather than very strict and rigorous solutions. Spectroscopic analysis and X-ray diffraction (XRD) demonstrate the crystalline nature of nanoparticles. On the other hand, the scanning electron microscope (SEM) confirms that the synthesis of such nanoparticles takes place through the Bacillus licheniformis.
The B. licheniformis is grown in a nitrate medium. The cultures are incubated at 200 rpm for 24 hours at room temperature and then centrifuged at 15,000 rpm for 15 minutes. The pellets are harvested to obtain approximately 1 g of wet cells that are then suspended in 100 ml of a 1 mM AuCl4 aqueous solution and incubated for 48 hours at 200 rpm, followed by sonication and centrifugation. The nanoparticles suspended in the supernatant are then separated from the solution by slow evaporation of the water at 50-55°. The aqueous chloride ions are reduced to metallic gold on exposure to the bacterial mass, which also makes the nanoparticles more stable: the colour of the reaction changes from pale yellow to dark purple, indicating the formation of the nanoparticles. Using the appropriate formulas, the diameter of these nanoparticles was determined to be between 10 and 100 nm, with an average absolute error of 3%. In this process, the bacterium functions as a cellular efflux pump and as a periplasmic protein that specifically binds the particles to the cell surface, resulting in altered solubility and toxicity through reduction, bioabsorption and bioaccumulation in the absence of specific metal transport systems.
From what has been said, biological synthesis seems to be undoubtedly advantageous, not only because it comprises a single step, but also because it turns out to be cost-effective and require a very general approach, without particularly restrictive criteria.
Kalishwaralal et al. used biologically synthesised Au-NPs to test their ability to inhibit retinal neoangiogenesis, using endothelial cells and retinal pigmented epithelium cells of bovine origin for their studies [5,6]. In particular, the role of Au-NPs in the inhibition of VEGF- and IL-1-induced cell proliferation and migration was investigated, as well as the possible underlying mechanism of modulation of the Src pathway.

Inhibition of VEGF-induced angiogenesis in retinal endothelial cells
Angiogenesis is a physiological process involving the growth of new vessels from existing ones. Pathological angiogenesis is responsible for the development of ocular complications (diabetic retinopathy, choroidal neovascularisation, age-related macular degeneration, retinopathy of the premature, etc.), in the pathogenesis of which the vascular endothelial growth factor (VEGF) is involved along with fibroblast growth factor (FGF). This neoangiogenesis is regulated by various growth factors and cytokines, which are responsible for the dysfunction, proliferation and migration of endothelial cells, as well as their permeability through adhesive contact with the extracellular matrix. Similarly, in tumour angiogenesis, the role of VEGF is to elicit endothelial cell proliferation and the growth of new vessels.
The influence of Au-NPs on angiogenetic phenomena was examined quantitatively by studying the proliferation, migration and tubule formation patterns in vitro [5]. In order to determine the degree of cytotoxicity of Au-NPs, bovine retinal endothelial cells (BRECs) at a density of 1 x 103 cells/ml were exposed to various concentrations of Au-NPs (from 100 nM to 600 nM) over 24 hours. Exposure to amounts above 500 nM causes significant cell death [5].
The effect of Au-NPs on VEGF-induced BRECs proliferation was then studied. Cells at a density of 1 x 103 cells/ml were treated with and without Au-NPs (500 nM) or with PP2 (Protein Phosphatase type 2 inhibitor of Src, the family of tyrosine kinase enzymes).
The addition of Au-NPs with VEGF significantly reduced the proliferation of BRECs compared to the control with VEGF alone [5]. The Src inhibitor PP2 was used as a positive control.
Src also plays a key role in regulating VEGF-induced cell migration and changes at the endothelial level. Cell migration is certainly important in neoangiogenesis and consequently in tumour growth and the development of metastases. The method used for this assay is wound-healing, in which the wound at the level of the BRECs was carried out using the tip of a micro-pipette in order to stimulate the healing. The maximum capacity of Au-NPs of inhibit cell migration was obtained for concentrations of 500 nM, the same concentration at which a significant reduction in VEGF-induced cell proliferation had been observed [5]: also in this case, however, the cells were exposed to higher concentrations of Au-NPs. The ability of Au-NPs to prevent migration and the ability of these nanoparticles to prevent cell migration. A stimulus for migration and complete healing within 24 hours was observed in the plates treated with VEGF; in contrast, a significant area of the wound remained uncovered in those treated with Au-NPs when compared to controls. The latter phenomenon was also observed in plates treated with a combination of VEGF and Au-NPs [5].
Another important step in the process of angiogenesis is the formation of tubules. Two-dimensional assays on Matrigel were used to examine the potential effects of Au-NPs on the formation of these structures. When BRECs were placed on VEGF-reduced Matrigel, long, robust tubule-like structures were formed after incubation in the presence of VEGF; their length was calculated using an inverted phase-contrast microscope. BRECs were inoculated onto the surface of Matrigel and treated with VEGF in the presence and/or absence of Au-NPs (500 nM) and PP2. It was revealed that both Au-NPs and PP2 inhibited VEGF-induced tubule formation of endothelial cells [8].

Fate of Au-NPs
Investigation using a transmission electron microscope (TEM) showed that, after 6 hours, in the Au-NPs+VEGF-treated BRECs samples, the nanoparticles remained mostly bound to the cell membrane and only a small amount was present inside the cell, in the endosomes. In the samples treated instead with Au-NPs, again after a time interval of 6 hours, the nanoparticles were found peripherally within endosomes or multi-vesicular bodies of the cell [5]. The results revealed that particles of a size of approximately 50 nm are internalised in endothelial cells during treatment with VEGF and Au-NPs.

Downregulation VEGF- and IL-1-induced cell proliferation in RPE
After studying the effects of Au-NPs on endothelial cells, the attention of Kalishwaralal et al. focused on the retinal pigmented epithelium, the outermost component of the BRB involved in certain pathological conditions such as age-related macular degeneration and vitreo-retinal proliferation. The latter was analysed by these authors in an interesting study [6].
Retinal vitreous proliferation (PVR) is one of the major ocular complications that can follow retinal damage and retinal detachment surgery. Cell development, migration and proliferation are considered to be the events behind the onset of ISR. This condition, we recall, is characterised by the formation of a membrane due to the proliferation and migration of retinal pigment epithelium (RPE) cells, fibroblasts and M?ller cells on both the retinal surface and the vitreous. Contraction of the epiretinal and subretinal membranes leads to tractional retinal detachment and, in the worst cases, to blindness.
Although there are many factors involved, a key role in the proliferation and migration of pigmented epithelium cells is played by VEGF and the chemotherapeutic effect of IL-1? In particular, VEGF is involved in BRB disruption and retinal neoangiogenesis; IL-1?, on the other hand, produced by RPE cells accumulates in ocular fluids during inflammatory phenomena, trauma and ISR. Recent studies have also shown that during ISR, the expression of not only VEGF but also its receptors VEGF-R1 and VEGF-R2 increases [7].
For the above-mentioned reasons, considering that the retinal pigment epithelium is a source of various cytokines and growth factors and that the disruption of the blood-retinal barrier can release a number of growth factors, ISR can therefore be explained as a condition that results in an uncontrolled interaction between growth factors and cytokines, on the one hand, and different cell types of the eye, on the other.
Au-NPs were biologically synthesised in a similar manner to the method described above, but using the Brevibacterium casei. The images obtained by transmission electron microscopy (TEM) revealed that almost monodisperse (i.e. approximately the same size and shape) gold nanoparticles measuring 50 nm had been synthesised. Again, RPE cells were isolated from bovine eyes (which is why we will now refer to them by the acronym BRPEs) suitably treated and subsequently cultured. Isolation was performed using a simple trypsinisation method in a relatively short time. This method proved to be quite reliable, economical and allowed a satisfactory number of cells to be harvested in a relatively short processing time [6].
To analyse the effect of Au-NPs, VEGF and IL-1? on their viability, BRPEs were treated with various concentrations of Au-NPs (1 to 1000 nM), VEGF and IL-1? and incubated for 24 hours. With regard to the Au-NPs, it was observed that the viability of the BRPEs was not impaired up to concentrations of 300 nM, but as this value increased, significant cell death was evident. This suggests the dose-dependent relationship of the reduction in cell viability treated with Au-NPs [6]. The same relationship was observed with VEGF and IL-1?, used to analyse the proliferation of BRPEs previously treated with fetal bovine serum (FBS) 0.5% for 8 hours. After various concentrations of VEGF (5, 10, 25, 50 and 100 ng/ml) and IL-1? (1, 5, 10, 25, 50 and 100 ng/ml) for 24 hours, results were obtained such that the maximum proliferative effect was obtained for VEGF contractions of 50 ng/ml and IL-1? concentrations of 25 ng/ml when compared to controls treated exclusively with FBS 0.5% [6].
We then went on to study the antiproliferative capacity of Au-NPs on the effect induced by VEGF and IL-1?: BRPEs were therefore treated with gold nanoparticles about 30 minutes before treatment with VEGF and/or IL-1?, and the viability of the cells was then assessed by an MTT assay. The addition of 300 nM Au-NPs to VEGF (50 ng/ml) and IL-1? (25 ng/ml) completely blocked the induced proliferation. Au-NPs were also equally adept at blocking the synergistic effect of VEGF and IL-1? [6]. Similarly, a considerable inhibition of cell-bond formation and cell development induced by VEGF and IL-1? and characterised by the formation of a clearly defined cytoplasmic halo around the nucleus was observed. However, by using quantities of Au-NPs up to 300 nM, only partial control could be exerted [6].
As we have already mentioned, the earliest events during ISR are cell migration and development. Therefore, in order to find out the effects of growth factors on BRPEs, an assay was used wound-stratch. In line with what was expected, both VEGF and IL-1? promote the migration of pigmented epithelium cells, whereas Au-NPs inhibit it. In particular, treatment with 50 ng/ml VEGF and treatment with 25 ng/ml IL-1? accelerates cell migration and allows complete wound closure within approximately 24 hours; in contrast, treatment with 300 nM AuNPs leaves a large area of the wound uncovered on the plate [6].

Role of Src and mechanism of action of Au-NPs
The analysis of the two studies conducted by Kalishwaralal et al. shows many similarities with regard to the role of Src in neoangiogenesis phenomena in both retinal endothelial cells and retinal pigmented epithelium, testifying that inhibition of the Src pathway underlies the effect of Au-NPs [9].
In the first case, with the aim of clarifying the role of Src in VEGF-induced endothelial cell proliferation, BRECs were transfected with DN Src (DNA Dominant Negative Src) and with CA Src (DNA Consitutive Active Src). The transfected BRECs were treated with Au-NPs and PP2 (a specific Src inhibitor) in the presence and absence of VEGF for 24 h at 37°C. The overexpression of Src DN (i.e. without active constituents, 'deficient HA") significantly blocks VEGF-induced cell proliferation, bringing it down to control levels, while CA Src overexpression has an additive effect after growth factor treatment. What is even more interesting is that CA Src confers resistance to the inhibitory effect of Au-NPs and PP2 on cell proliferation: treatment with VEGF+Au-NPs and VEGF+PP2 induces cell proliferation in cells transfected with CA Src when compared to controls [5].
In the second case, a similar approach was taken and the results obtained are in line with those of the previous study. Briefly, BRPEs were grown at a density of 2 x 103 cells per well, in 96 wells and placed in an IMDM (Iscove's Modified Dulbecco Medium) with 5% serum for 8 hours. The cells were then incubated with 10 ?M of PP2 for 30 min before treatment with VEGF (50 ng/ml) and Il-1? (25 ng/ml). Again, an MTT assay was used to determine Src kinase activity in cell proliferation. It was seen that pre-treatment with PP2 inhibited VEGF- and IL-1? induced proliferation, in agreement with the theory that basal Src activity is required for cells to proliferate. [6]
Again, transfection of CA Src increases the proliferation of pigmented epithelium cells, while transfection of DN Src causes a reduction. Overexpression of DN Src blocks VEGF- and IL-1-mediated proliferation, bringing it down to the levels of control cases. Interestingly, Au-NPs are also unable to block cell proliferation in cells transfected with CA Src, regardless of whether they are treated with or without VEGF or IL-1? These are undoubtedly proof of the AuNPs' mechanism of action, according to which the gold nanoparticles are able to inhibit the proliferating effect of BRPEs induced by VEGF and IL-1b by blocking the Src pathway. [6]
In order to support the hypothesis that the inhibitory effects of AuNPs and PP2 were specifically directed by the Src pathway, immunoassays were conducted that demonstrated that treatment with VEGF and IL-1? significantly increased the levels of phosphorylated Src (Y419). In VEGF-induced BRECs, treatment with AuNPs results in a clear reduction of this phosphorylation to levels comparable to controls [5]. In retinal pigmented pigment epithelium cells, on the other hand, phosphorylated Src levels are lowered with the use of 300 nM AuNPs prior to treatment with 50 ng/ml VEGF or 25 ng/ml IL-1?
These results clearly show how gold nanoparticles inhibit the VEGF- and IL-1-induced cell proliferation process by inhibiting the activation of the phosphorylated Src protein [6].

The administration of nanoparticles intravenously
Studies to date show that gold nanoparticles are indeed able to express, in vitrotheir functions at the level of both endothelial cells and retinal pigmented epithelium, thus suggesting excellent potential. Having reached this point, however, one wonders whether nanoparticles can actually be used in common clinical practice, that with which the doctor must interface on a daily basis, taking into account the pharmacokinetic and pharmacodynamic characteristics of the drug he decides to administer to his patient. Therefore, while it is only right to emphasise the importance of testing at vitro, from which one absolutely cannot and must not disregard, it is equally true that one must consider, in vivo, the ability of nanoparticles to cross the blood-retinal barrier, their volume of distribution, and their possible toxicity.
Recently, several authors have spoken of a size-dependent distribution of Au-NPs within tissues in vivo [8,9]: specifically, nanoparticles of the size of 10 nm were widely distributed in all the various tissues, including the retina, while those larger than 50 nm were only found in the blood. Although the potential toxic effect of the nanoparticles, or at least the possibility of interaction with biological molecules, is not investigated in these two papers, there is one fact that clearly emerges: Au-NPs are able to cross the blood-brain barrier (blood-brain barrier, BBB) and thus the blood-retinal barrier. Very interesting in this regard is the work produced by a Korean study group [10] that analyses the distribution of nanoparticles and their cytotoxic effect in the various retinal layers following the administration of intravenous Au-NPs in mice. Twenty-four hours after the administration of Au-NPs of sizes 20 nm and 100 nm in two separate groups of mice, the mice were sacrificed and their retinas analysed under a transmission electron microscope (TEM). The 100 nm particles were not found in the retina, whereas the 20 nm particles were able to cross the BRB and distribute in all retinal layers and especially in neurons, endothelial cells and peri-endothelial glial cells [10]. No cells in which Au-NPs were present showed any structural alteration compared to cells without them [10].
The evaluation of histological changes in the seven days following the intravenous administration of Au-NPs also yielded good results: compared to controls, no histological changes were observed in any retinal layer, nor were inflammatory cells present in the vitreous, retina or choroid [10]. A TUNEL assay showed that the number of apoptotic cells in the retina did not increase following intravenous administration of Au-NPs when compared with controls [10]. At the same time, there was also no alteration in the expression of two membrane components of retinal endothelial cells, namely ZO-1 at the tight-junctions and glut-1, which is the most important glucose transporter of BRB. The viability of retinoblastoma cells and astrocytes also remained largely unchanged after treatment with 20 nm Au-NPs [10].

Conclusions
Au-NPs have so far been shown to be potent anti-angiogenic substances, capable of acting on VEGF and IL-1? and capable of exerting their effect on both the retinal pigment epithelium and retinal endothelial cells. The possibility of interfering with those processes that are fundamental in the phenomenon of neoangiogenesis, such as proliferation, cell migration, and phosphorylation of VEGF- and IL-1?-induced Src, makes them unquestionably interesting from a therapeutic point of view, not only in the field of ophthalmology but also and above all in the field of oncology.
Reference must be made to the cytotoxicity of Au-NPs, as this represents a parameter of primary interest for any drug and in particular for a potential drug. Although concentrations above 500 nM were responsible for a quantitative increase in cell death at the endothelial level [5], it is equally true that no toxicity or increased apoptosis was recorded in any of the cells of the ten layers that make up the retina following the administration of nanoparticles of the size of 20 nm.
It is precisely the ability of 20 nm Au-NPs to cross the BRB that represents a very interesting aspect in view of their possible therapeutic application in clinical ophthalmological practice. Considering their ability not to induce inflammatory events in any of the retinal cells, some authors have gone so far as to suggest them as a possible alternative for the drug delivery to be applied in vivo [13].
Au-NPs therefore have great potential. Nowadays, several drugs are used in ophthalmology to inhibit retinal neovascularisation, such as bevacizumab, a humanised monoclonal antibody with high affinity for human VEGF administered by intravitreal injection, or pegaptanib sodium, an anti-VEGF aptamer that selectively inhibits isoform 165. These are extraordinary drugs of proven efficacy, but also very expensive by virtue of the methods used in their production and development. In addition, despite the considerable anti-VEGF effects of monoclonal antibodies, there are limitations, mainly due to the fact that several intravitreal injections per year are required, up to seven or eight, with the real risk of increasing injection-related complications.
Furthermore, direct inhibition by anti-VEGF monoclonal antibodies can aggravate retinal ischaemia, induce the destruction of mitochondria in photoreceptors or influence the behaviour of retinal neuronal cells in some way [11, 12, 13]. For this reason, Au-NPs of a size of 20 nm, by virtue of their ability to cross the BRB, could, for example, be thought of as a carrier to be used to increase the bioavailability of drugs in the retina and brain. In this case, however, the problem of toxicity linked to greater biodistribution of the drug in these sites could arise: new studies are therefore undoubtedly needed to establish exactly which types of particles must or can actually be administered, their size, concentrations, and the methods of administration that can minimise toxic or undesirable effects but also increase the bioavailability of the drug so as to avoid the drawbacks currently linked to the need for numerous intravitreal injections. The use of gold nanoparticles could therefore represent an excellent, interesting and innovative starting point for the future of ophthalmology.

Pietro Distante
Almo Collegio Borromeo of Pavia
E-mail: piero_distante@libero.it

BIBLIOGRAPHY
1. Farjo, K.M.; Ma, J.X.. The potential of nanomedicine therapies to treat neovascular disease in the retina. J. Angiogenes. Res. 2010, 2, 21.
2. Antonii, F. Panacea Aurea-Auro Potabile; Bibliopolio Frobeniano:Hamburg, 1618.
3. Beveridge, T.J., Murray, R.G.E., 1980. Site of metal deposition in the cell wall of Bacillus subtilis. J. Bacteriol. 141, 876-887.
4. Kalishwaralal K. et al. (2009). Biological synthesis of gold nanocubes from Bacillus licheniformis. Bioresour Technol 100:5356-5358.
5. Kalishwaralal K. et al. (2011). Gold nanoparticles inhibit vascular endothelial growth factor-induced angiogenesis and vascular permeability via Src dependent pathway in retinal endothelial cells. Angiogenesis. 14(3):409-10.
6. Kalishwaralal K. et al. (2010). Gold nanoparticles downregulate VEGF-and IL-1b-induced cell proliferation through Src kinase in retinal pigment epithelial cells. Exp Eye Res. 91(5):769-78.
7. Chen Y.S. et al. 1997. Localisation of vascular endothelial growth factor and its receptors to cells of vascular and avascular epiretinal membranes, Br.J. Ophthalmol. 81, 919e926.
8. De Jong W.H., Hagens W.I., Krystek P., Burger M.C., Sips A.J. and Geertsma R.E. (2008). Particle size-dependent organ distribution of gold nanoparticles after intravenous administration, Biomaterials 29 1912-9.
9. Sonavane G., Tomoda K. and Makino K. (2008). Biodistribution of colloidal gold nanoparticles after intravenous administration: effect of particle size Colloids Surf. B 66 274-80.
10. Kim J.H. et al. (2009). Intravenously administered gold nanoparticles pass through the blood-retinal barrier depending on the particle size, and induce no retinal toxicity, Nanotechnology, 20 (2009) 505101.
11. Saint-Geniez M. et al, Endogenous VEGF is required for visual function: Evidence for a survival role on muller cells and photoreceptors. PLoS One 2008, 3, e3554.
12. Lee S.J.; Koh H.J. Enlargement of the foveal avascular zone in diabetic retinopathy after adjunctive intravitreal bevacizumab (avastin) with pars plana vitrectomy. J. Ocul. Pharmacol. Ther. 2009, 25, 173-174.
13. Inan U.U., Preclinical safety evaluation of intravitreal injection of full-length humanized vascular endothelial growth factor antibody in rabbit eyes. Invest. Ophthalmol. Vis. Sci. 2007, 48, 1773-1781.