Effect of circulation on the disposition and ocular tissue distribution of 20 nm nanoparticles after periocular administration

Abstract

Purpose: Our previous studies indicated that while 20 nm particles are rapidly cleared from the periocular space of the rat following posterior subconjunctival injection, 200 nm particles persisted for at least two months. To understand faster clearance of 20 nm particles, the purpose of this study was to determine transscleral permeability and in vivo disposition in the presence and absence of circulation. Further, it was the purpose of this study to simulate sustained retinal drug delivery after periocular administration of rapidly cleared and slowly cleared nanoparticles.

Methods: The permeability of 20 and 200 nm particles over 24 h was examined across isolated bovine sclera and sclera-choroid-RPE with or without a surfactant (Tween 20, 0.1% w/v) added to the preparation. The in vivo disposition of nanoparticles was performed using Sprague Dawley rats. The rats, either dead or alive, were administered with 400 microg of the nanoparticles in the periocular space, and the particle disposition in the eye tissues was assessed 6 h later. To evaluate the role of the reticulo-endothelial system and lymphatic circulation, isolated liver, spleen, and cervical, axillary, and mesenteric lymph nodes were analyzed using confocal microscopy. Mathematical simulations with Berkeley Madonna were used to evaluate the effect of nanoparticle size on retinal drug levels following periocular administration. Celecoxib was used as the model drug and the finalized pharmacokinetic model from a previous study was used with some modifications for the simulation.

Results: Transport of 20 nm particles across sclera in the presence and absence of the surfactant were 0.1%+/-0.07% and 0.46%+/-0.06%, respectively. These particles did not permeate across the sclera-choroid-RPE in 24 h. There was no quantifiable transport for 200 nm particles across the sclera or the sclera-choroid-RPE. In live animals, the 20 nm particles were undetectable in any of the ocular tissues except in the sclera-choroid following periocular administration; however, in dead animals, the particle concentrations in the sclera-choroid were 19 fold higher than those in live animals, and particles were detectable in the retina as well as vitreous. The retention of 20 nm particles at the site of administration was two fold higher in the dead animals. In live animals, the particles were clearly detectable in the spleen and to a very low extent in the liver as well. The particles were also detected in the cervical, axillary, and mesenteric lymph nodes of the live animals. Simulations with two particles (20 nm and 200 nm) with different clearance rates demonstrated that the retinal drug levels were affected by particle clearance. Larger nanoparticles sustained retinal drug delivery better than smaller nanoparticles. With an increase in drug release rate from the particles, these differences diminish.

Conclusions: The 20 nm particles are transported across the sclera to a minor degree; however, there is no significant transport across the sclera-choroid-RPE. Periocular circulation (blood and lymphatic) plays an important role in the clearance of the 20 nm particles. The higher particle levels in the ocular tissues in the post-mortem studies indicate a dynamic physiologic barrier to the entry of particles into the ocular tissues after periocular administration. The particle size of the delivery system can play an important role in the observed retinal drug levels after periocular administration. Slow release nanoparticles with low clearance by blood and lymphatic circulations are suitable for prolonged transscleral drug delivery to the back of the eye.

Figure 1

Nanoparticle size distribution and permeability across sclera and sclera-choroid-RPE. A: Particle size distribution after 24 h storage in assay buffer used for transport studies. Majority of the particles are distributed around the 50 nm particle size with another small group of particles of higher size distribution. B: Particle size distribution after 24 h storage in assay buffer containing 0.1% tween-20. In the presence of the surfactant, the particle size distribution shifts towards greater particle size indicating probable particle aggregation. C: Transport of nanoparticles (20 nm; 100 µg/ml) across isolated bovine sclera and sclera-choroid-RPE. The 20 nm particles can cross the sclera but not the sclera-choroid-RPE combination to any quantifiable extent. D: Transport of nanoparticles (20 nm; 100 µg/ml) across isolated bovine sclera and sclera-choroid-RPE in the presence of 0.1% tween-20. The particle transport across sclera is reduced in the presence of surfactant probably due to the shift in particle size distribution. Data are expressed as mean ± s.d. for n = 5-6. No quantifiable transport was observed either across the sclera or the sclera-choroid-RPE with the 200 nm particles.

Figure 2

Confocal images of the sclera at the end of the 24 h nanoparticle (20 nm) transport study. A shows the control fluorescence and combined fluorescence while B shows the phase contrast images. C and D are nanoparticle exposed tissue fluorescence and combined fluorescence and phase contrast images, respectively. In each image, the scleral (donor) side is on the left and the vitreal (receiver) side is on the right. The particles are concentrated on the outer edge of the sclera. There are very few or no particles on the vitreal side of the tissue.

Figure 3

Confocal images of the sclera-choroid-RPE combination at the end of 24 h nanoparticle (20 nm) transport study. Panel A shows the fluorescence image and Panel B shows the combination (fluorescence plus phase contrast) image of the control sclera-choroid-RPE tissue. Panel C shows the fluorescence and Panel D shows the combination (fluorescence plus phase contrast) image of the sclera-choroid-RPE tissue that was exposed to nanoparticles during the transport study. In each panel, the scleral (donor) side is on the left and the vitreal (receiver) side is on the right. The particles are concentrated on the outer edge of the sclera. There are very few or no particles seen on the vitreal side of the tissue.

Figure 4

Pharmacokinetic modeling of the disposition of 20 nm particles in the periocular space. Nanoparticle (20 nm) elimination from the periocular tissue is biphasic. The solid line represents the model predicted data while the circles represent the observed data. T_1/2α and T_1/2β represent half-lives for elimination from the periocular space. R²: regression coefficient for the correlation between observed and predicted data.

Figure 5

Nanoparticles (20 nm) are cleared by ocular (blood and lymphatic) circulation. Following periocular administration of 400 µg dose of 20 nm particles to either live (blood and lymphatic circulation present) or dead rats (blood and lymphatic circulation absent), the percent dose remaining at the site of administration was determined 6 h post-dosing. The amount of particles remaining in the periocular tissue was more than 2 fold higher in the dead rats as compared to the live rats. The data are expressed as mean ± SEM for n=4. The asterisk indicates a statistically significant difference between live and dead animals (p<0.05).

Figure 6

Dynamic barriers prevent significant entry of 20 nm particles into ocular tissues in live animals. Following periocular administration of 400 µg dose of 20 nm particles to either live (blood and lymphatic circulation present) or dead rats (blood and lymphatic circulation absent) the particle levels in the ocular tissues were quantified. Higher levels of the particles are seen in the sclera-choroid, retina, vitreous, and the cornea of dead rats as compared to live rats. The data are expressed as mean ± SEM for n=4.

Figure 7

Representative confocal micrographs of various tissues 6 h after periocular administration of 20 nm particles. Following periocular administration of 400 µg dose of 20 nm particles to live rats, the nanoparticles can be found in the organs of the reticulo-endothelial system (liver and spleen). The various tissues including the eye, the periocular tissue, the liver and the spleen were removed and sectioned 6 h after administration. The figure shows the fluorescence images of sections of the: eye (Panels A and B); periocular tissue (Panels C and D); liver (Panels E and F); and spleen (Panels G and H). The left panels (A, C, E, and G) are fluorescence images from control rats that were not dosed with the nanoparticles whereas the right panels (B, D, F, and H) are images from the rats that were dosed with the nanoparticles. Nanoaprticles can be seen in the periocular tissue, spleen and to some extent in the liver of the dosed animals.

Figure 8

Representative confocal images of lymph nodes sections 6 h after periocular administration of 20 nm nanoparticles. Lymphatic circulation plays a role in the clearance of nanoparticles (20 nm) after periocular administration. Representative confocal images of lymph nodes sections, 6 h post periocular administration of 20 nm nanoparticles. Nanoparticles (20 nm; green) were administered to SD rats, live (Panels B, E, and H) and dead (Panels C, F, and I) by periocular injection. Lymph nodes, namely, cervical (Panels A-C), axillary (Panels D-F), and mesenteric (Panels G-I), were analyzed for the presence of nanoparticles by confocal microscopy. Lymph nodes of undosed SD rats were treated as controls (Panels A, D, and G). Green fluorescence associated with nanoparticles was observed in lymph node sections of live, but not dead, SD rats 6 h post periocular administration of nanoparticles. This suggests that in live animals lymphatic drainage delivered nanoparticles to various lymph nodes, however in dead rats, which are devoid viable lymphatic system, nanoparticles could not be detected in lymph nodes.

Figure 9

Sustained retinal delivery of a model drug (celecoxib) from nanoparticles with different clearance rates and drug release rates. The profiles were simulated for 20 nm and 200 nm particles for a period of 60 days. The elimination rate of the 20 nm formulation was obtained by curve fitting to the previously published data [12]. The estimated elimination half-life for 20 nm particles was 5.5 h. The elimination half-life for the 200 nm particles was assumed to be 180 days since they persisted almost completely for at least two months in the periocular space [12]. All other model parameters used in the model are shown in Table 1. The structural model is shown above the simulation. The panels depict profiles of 20 and 200 nm particles with a release rate constant of 0.016 min⁻¹ (A), profiles of 20 and 200 nm particles with a release rate constant of 0.0016 min⁻¹ (B), profiles of 20 and 200 nm particles with a release rate constant of 0.00016 min⁻¹ (C), and profiles of 20 and 200 nm particles with a release rate constant of 0.000016 min⁻¹ (D). The insets in each panel are the profiles for the first 24 h of drug release to better show the early differences between the retinal concentrations of celecoxib using 20 and 200 nm particles.