Application of an organotypic ocular perfusion model to assess intravitreal drug distribution in human and animal eyes.

Intravitreal (ITV) drug delivery is a new cornerstone for retinal therapeutics. Yet, predicting the disposition of formulations in the human eye remains a major translational hurdle. A prominent, but poorly understood, issue in pre-clinical ITV toxicity studies is unintended particle movements to the anterior chamber (AC). These particles can accumulate in the AC to dangerously raise intraocular pressure. Yet, anatomical differences, and the inability to obtain equivalent human data, make investigating this issue extremely challenging. We have developed an organotypic perfusion strategy to re-establish intraocular fluid flow, while maintaining homeostatic pressure and pH. Here, we used this approach with suitably sized microbeads to profile anterior and posterior ITV particle movements in live versus perfused porcine eyes, and in human donor eyes. Small-molecule suspensions were then tested with the system after exhibiting differing behaviours in vivo. Aggregate particle size is supported as an important determinant of particle movements in the human eye, and we note these data are consistent with a poroelastic model of bidirectional vitreous transport. Together, this approach uses ocular fluid dynamics to permit, to our knowledge, the first direct comparisons between particle behaviours from human ITV injections and animal models, with potential to speed pre-clinical development of retinal therapeutics.

Transpupillary collagen photocrosslinking for targeted modulation of ocular biomechanics.

The central vision-threatening event in glaucoma is dysfunction and loss of retinal ganglion cells (RGCs), thought to be promoted by local tissue deformations. Here, we sought to reduce tissue deformation near the optic nerve head by selectively stiffening the peripapillary sclera, i.e. the scleral region immediately adjacent to the optic nerve head. Previous scleral stiffening studies to treat glaucoma or myopia have used either pan-scleral stiffening (not regionally selective) or regionally selective stiffening with limited access to the posterior globe. We present a method for selectively stiffening the peripapillary sclera using a transpupillary annular light beam to activate methylene blue administered by retrobulbar injection. Unlike prior approaches to photocrosslinking in the eye, this approach avoids the damaging effects of ultraviolet light by employing red light. This targeted photocrosslinking approach successfully stiffened the peripapillary sclera at 6 weeks post-treatment, as measured by whole globe inflation testing. Specifically, strain was reduced by 47% when comparing treated vs. untreated sclera within the same eye (n = 7, p=0.0064) and by 54% when comparing the peripapillary sclera of treated vs. untreated eyes (n = 7, p<0.0001). Post-treatment characterization of RGCs (optic nerve axon counts/density, and grading), retinal function (electroretinography), and retinal histology revealed that photocrosslinking was associated with some ocular toxicity. We conclude that a transpupillary photocrosslinking approach enables selective scleral stiffening targeted to the peripapillary region that may be useful in future treatments of glaucoma.

Author Correction: Improving Stem Cell Delivery to the Trabecular Meshwork Using Magnetic Nanoparticles.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Improving Stem Cell Delivery to the Trabecular Meshwork Using Magnetic Nanoparticles.

Glaucoma is a major cause of blindness and is frequently associated with elevated intraocular pressure. The trabecular meshwork (TM), the tissue that primarily regulates intraocular pressure, is known to have reduced cellularity in glaucoma. Thus, stem cells, if properly delivered to the TM, may offer a novel therapeutic option for intraocular pressure control in glaucoma patients. For this purpose, targeted delivery of stem cells to the TM is desired. Here, we used magnetic nanoparticles (Prussian blue nanocubes [PBNCs]) to label mesenchymal stem cells and to magnetically steer them to the TM following injection into the eye's anterior chamber. PBNC-labeled stem cells showed increased delivery to the TM vs. unlabeled cells after only 15-minute exposure to a magnetic field. Further, PBNC-labeled mesenchymal stem cells could be delivered to the entire circumference of the TM, which was not possible without magnetic steering. PBNCs did not affect mesenchymal stem cell viability or multipotency. We conclude that this labeling approach allows for targeted, relatively high-efficiency delivery of stem cells to the TM in clinically translatable time-scales, which are necessary steps towards regenerative medicine therapies for control of ocular hypertension in glaucoma patients.

Differential effects of tyrosine-rich amelogenin peptide on chondrogenic and osteogenic differentiation of adult chondrocytes.

Current approaches to treat osteoarthritis (OA) are insufficient. Autologous chondrocyte implantation (ACI) has been used for the past decade to treat patients with OA or focal cartilage defects. However, a number of complications have been reported post-ACI, including athrofibrosis and symptomatic hypertrophy. Thus, a long-term ACI strategy should ideally incorporate methods to 'prime' autologous chondrocytes to form a cartilage-specific matrix and suppress hypertrophic mineralization. The objective of this study is to examine the effects of tyrosine-rich amelogenin peptide (TRAP; an isoform of the developmental protein amelogenin) on human articular cartilage cell (HAC) chondrogenic differentiation and hypertrophic mineralization in vitro. Effects of chemically synthesized TRAP on HAC chondrogenic differentiation were determined by assessing: (1) sGAG production; (2) Alcian blue staining for proteoglycans; (3) collagen type II immunostaining; and (4) expression of the chondrogenic genes SOX9, ACAN and COL2A1. Hypertrophic mineralization was assayed by: (1) ALP expression; (2) Alizarin red staining for Ca(+2)-rich bone nodules; (3) OC immunostaining; and (4) expression of the osteogenic/hypertrophic genes Ihh and BSP. Chemically synthesized TRAP was found to suppress terminal osteogenic differentiation of HACs cultured in hypertrophic mineralization-like conditions, an effect mediated via down-regulation of the Ihh gene. Moreover, TRAP was found to augment chondrogenic differentiation of HACs via induction of SOX9 gene expression when cells were cultured in pro-chondrogenic media. The results obtained from this proof-of-concept study motivate further studies on the use of TRAP as part of a preconditioning regimen in autologous chondrocyte implantation procedures for OA patients and patients suffering from focal cartilage defects.

Quantitative analysis of three-dimensional fibrillar collagen microstructure within the normal, aged and glaucomatous human optic nerve head.

The aim of this study was to quantify connective tissue fibre orientation and alignment in young, old and glaucomatous human optic nerve heads (ONH) to understand ONH microstructure and predisposition to glaucomatous optic neuropathy. Transverse (seven healthy, three glaucomatous) and longitudinal (14 healthy) human ONH cryosections were imaged by both second harmonic generation microscopy and small angle light scattering (SALS) in order to quantify preferred fibre orientation (PFO) and degree of fibre alignment (DOFA). DOFA was highest within the peripapillary sclera (ppsclera), with relatively low values in the lamina cribrosa (LC). Elderly ppsclera DOFA was higher than that in young ppsclera (p < 0.00007), and generally higher than in glaucoma ppsclera. In all LCs, a majority of fibres had preferential orientation horizontally across the nasal-temporal axis. In all glaucomatous LCs, PFO was significantly different from controls in a minimum of seven out of 12 LC regions (p < 0.05). Additionally, higher fibre alignment was observed in the glaucomatous inferior-temporal LC (p < 0.017). The differences between young and elderly ONH fibre alignment within regions suggest that age-related microstructural changes occur within the structure. The additional differences in fibre alignment observed within the glaucomatous LC may reflect an inherent susceptibility to glaucomatous optic neuropathy, or may be a consequence of ONH remodelling and/or collapse.

Protective ocular mechanisms in woodpeckers.

AIMS: Woodpeckers possess mechanisms protecting the eye from shaking/impact. Mechanisms available to woodpeckers but not humans may help explain some eye injuries in Shaken Baby syndrome (SBS). METHODS: Gross dissection and histologic examination of eyes and orbits of seven woodpeckers. RESULTS: All birds showed restricted axial globe movement due to the tight fit within the orbit and fascial connections between the orbital rim and sclera. The sclera was reinforced with cartilage and bone, the optic nerve lacked redundancy, and the vitreous lacked attachments to the posterior pole retina. CONCLUSIONS: Woodpecker eyes differ from human infants by an inability of the globe to move axially in the orbit, the sclera to deform, and the vitreous to shear the retina. These findings support current hypotheses that abusive acceleration-deceleration-induced ocular injury in human infants may be related to translation of vitreous within the globe and the globe within the orbit. The woodpecker presents a natural model resistant to mechanical forces that have some similarity to SBS.

Rate of change in central corneal thickness: a viability indicator for conventional drainage tissues in organ culture.

Organ culture of human anterior segments is a powerful tool for understanding trabecular meshwork biology. However, data from a significant percentage of cultured anterior segments are unusable because tissues fail to meet quality control requirements, such as having adequate trabecular meshwork histology. The purpose of the present study was to evaluate a novel, real time method for assessing the viability of conventional drainage tissues in the human anterior segment perfusion model. Twenty-two human anterior segments were perfusion cultured using standard techniques for one week while measuring outflow facility and central corneal thickness (CCT). After perfusion-fixation, toludine blue-stained histological sections of drainage tissues from all four quadrants of each anterior segment were graded and endothelial cell nuclei from cornea centers were stained with 4',6-diamidino-2-phenylindole and counted. We found that most anterior segments with a stable outflow facility had a CCT that decreased over time, while anterior segments with an unstable outflow facility had CCT measurements that failed to decrease over time (P<0.01). When comparing CCT measurements to histological appearance of outflow tissues, we found that in 11/11 cases, anterior segments with an acceptable histological score had a negative CCT slope (P<0.01). Conversely in 3/4 instances, anterior segments with an unacceptable histological score had a positive CCT slope. Lastly, we observed a significant relationship between CCT measurements and corneal endothelial density (P<0.01). Thus, the simple procedure of measuring CCT during anterior segment perfusion provides a second useful measure to assess the viability of the anterior segment during the perfusion process.

Computational modeling of arterial biomechanics: insights into pathogenesis and treatment of vascular disease.

We review how advances in computational techniques are improving our understanding of the biomechanical behavior of the healthy and diseased cardiovascular system. Numerical modeling of biomechanics is being used in a wide variety of ways, including assessment of effects of mural and hemodynamically induced stresses on atherogenesis, development of risk measures for aneurysm rupture, improvement in interpretation of medical images, and quantification of oxygen transport in diseased and healthy arteries. Although not amenable to routine clinical use, numerical modeling of cardiovascular biomechanics is a powerful research tool.

Cationic ferritin changes outflow facility in human eyes whereas anionic ferritin does not.

PURPOSE: To determine the effect of charged moieties within the outflow pathway on aqueous outflow facility in human eyes. METHODS: After baseline facility measurement in human eye bank eyes (n = 10 pairs), one eye of each pair received anterior chamber exchange and continued perfusion with medium containing 10 mg/ml cationic ferritin. Contralateral eyes were treated in a similar manner with anionic ferritin (10.0 or 102 mg/ml). Eyes were fixed by anterior chamber exchange and perfusion with universal fixative at 8 mm Hg (corresponding to a physiologic pressure of 15 mm Hg in vivo) and examined by transmission electron microscopy. In a second series of human eyes (n = 8 pairs), facility was measured before and after anterior chamber exchange, with a solution containing 0.1 U/ml neuraminidase. RESULTS: Perfusion of eyes with anionic ferritin at either 10.0 or 102 mg/ml caused a negligible 2% increase in facility, whereas cationic ferritin perfusion reduced facility by 66% (P < 0.00001). Perfusion with fixative reduced facility by approximately 60% in both cationic and anionic ferritin-perfused eyes, relative to facilities after perfusion with ferritin. Transmission electron microscopy showed that the distribution of ferritin was segmentally variable. Cationic ferritin consistently labeled the luminal surface of the inner wall of Schlemm's canal, and variably labeled the juxtacanalicular connective tissue (JCT) and trabecular beam surfaces. Anionic ferritin was more prominent in the JCT and intertrabecular spaces and less so on the luminal surface of Schlemm's canal. By scanning electron microscopy, cationic ferritin was seen to accumulate at intercellular margins of the inner wall. Neuraminidase perfusion had no significant effect on outflow facility. CONCLUSIONS: Cationic ferritin reduces outflow facility, presumably by binding to negatively charged sites in the outflow pathway. A possible mechanism is partial or complete blockage of intercellular clefts in the inner wall of Schlemm's canal by the ferritin that accumulates on the luminal surface of the inner wall. Although they are possible targets for ferritin binding, sialyl residues themselves seem to have little direct effect on outflow facility. Our data indicate that positively charged molecules, especially if they can interact with inner wall pores, have the potential to markedly alter outflow facility.

Requirements for mesh resolution in 3D computational hemodynamics.

Computational techniques are widely used for studying large artery hemodynamics. Current trends favor analyzing flow in more anatomically realistic arteries. A significant obstacle to such analyses is generation of computational meshes that accurately resolve both the complex geometry and the physiologically relevant flow features. Here we examine, for a single arterial geometry, how velocity and wall shear stress patterns depend on mesh characteristics. A well-validated Navier-Stokes solver was used to simulate flow in an anatomically realistic human right coronary artery (RCA) using unstructured high-order tetrahedral finite element meshes. Velocities, wall shear stresses (WSS), and wall shear stress gradients were computed on a conventional "high-resolution" mesh series (60,000 to 160,000 velocity nodes) generated with a commercial meshing package. Similar calculations were then performed in a series of meshes generated through an adaptive mesh refinement (AMR) methodology. Mesh-independent velocity fields were not very difficult to obtain for both the conventional and adaptive mesh series. However, wall shear stress fields, and, in particular, wall shear stress gradient fields, were much more difficult to accurately resolve. The conventional (nonadaptive) mesh series did not show a consistent trend towards mesh-independence of WSS results. For the adaptive series, it required approximately 190,000 velocity nodes to reach an r.m.s. error in normalized WSS of less than 10 percent. Achieving mesh-independence in computed WSS fields requires a surprisingly large number of nodes, and is best approached through a systematic solution-adaptive mesh refinement technique. Calculations of WSS, and particularly WSS gradients, show appreciable errors even on meshes that appear to produce mesh-independent velocity fields.

Factors influencing blood flow patterns in the human right coronary artery.

Evidence suggests that atherogenesis is linked to local hemodynamic factors such as wall shear stress. We investigated the velocity and wall shear stress patterns within a human right coronary artery (RCA), an important site of atherosclerotic lesion development. Emphasis was placed on evaluating the effect of flow waveform and inlet flow velocity profile on the hemodynamics in the proximal, medial, and distal arterial regions. Using the finite-element method, velocity and wall shear stress patterns in a rigid, anatomically realistic model of a human RCA were computed. Steady flow simulations (ReD=500) were performed with three different inlet velocity profiles; pulsatile flow simulations utilized two different flow waveforms (both with Womersley parameter=1.82, mean ReD=233), as well as two of the three inlet profiles. Velocity profiles showed Dean-like secondary flow features that were remarkably sensitive to the local curvature of the RCA model. Particularly noteworthy was the "rotation" of these Dean-like profiles, which produced large local variations in wall shear stress along the sidewalls of the RCA model. Changes in the inlet velocity profiles did not produce significant changes in the arterial velocity and wall shear stress patterns. Pulsatile flow simulations exhibited remarkably similar cycle-average wall shear stress distributions regardless of waveform and inlet velocity profile. The oscillatory shear index was very small and was attributed to flow reversal in the waveform, rather than separation. Cumulatively, these results illustrate that geometric effects (particularly local three-dimensional curvature) dominate RCA hemodynamics, implying that studies attempting to link hemodynamics with atherogenesis should replicate the patient-specific RCA geometry.

Mass transport in an anatomically realistic human right coronary artery.

The coronary arteries are a common site of atherosclerotic plaque formation, which has been putatively linked to hemodynamic and mass transport patterns. The purpose of this paper was to study mass transport patterns in a human right coronary artery (RCA) model, focusing on the effects of local geometric features on mass transfer from blood to artery walls. Using a previously developed characteristic/finite element scheme for solving advection-dominated transport problems, mass transfer calculations were performed in a rigid, anatomically realistic model of a human RCA. A qualitative and quantitative examination of the RCA geometry was also carried out. The concentration field within the RCA was seen to closely follow primary and secondary flow features. Local variations in mass transfer patterns due to geometric features were significant and much larger in magnitude than local variations in wall shear stress. We conclude that the complex secondary flows in a realistic arterial model can produce very substantial local variations in blood-wall mass transfer rates, and may be important in atherogenesis. Further, RCA mass transfer patterns are more sensitive to local geometric features than are wall shear stress patterns.

Effect of Healon and Viscoat on outflow facility in human cadaver eyes.

PURPOSE: To compare the acute effects of Healon (sodium hyaluronate) and Viscoat (sodium chondroitin sulfate-sodium hyaluronate) on outflow facility in human cadaver eyes and determine which viscoelastic agent is least likely to cause an intraocular pressure (IOP) spike after cataract surgery. SETTING: The Glaucoma Research Lab, University of Toronto, Ontario, Canada. METHODS: In this prospective paired study, 15 pairs of human cadaver eyes were used. Following the construction of a 3.0 mm scleral tunnel, 0.25 cc of Healon was injected into the anterior chamber of 1 eye and 0.25 cc of Viscoat was injected into the contralateral eye. The viscoelastic agents were removed from both eyes in a standardized fashion and the scleral tunnels closed. The eyes were then perfused at a constant IOP of 8.0 mm Hg, corresponding to 16.0 mm Hg in vivo. Outflow facility (microL/minute [min]/mm Hg) was recorded every 15 minutes for 24 hours using standard methods. RESULTS: Outflow facility in the Viscoat-treated eyes decreased appreciably for the first 3 hours, then recovered somewhat after 12 hours; facility in the Healon-treated eyes showed less of an overall decrease. Over the 24 hour perfusion period, mean outflow facility was 0.037 microL/min/mm Hg +/- 0.015 (SD) in the Viscoat-treated eyes and 0.060 +/- 0.012 microL/min/mm Hg in the Healon-treated eyes. Healon reduced outflow facility significantly less than Viscoat between 3.25 and 10.50 hours postoperatively (P < .05, 2-tailed t test). CONCLUSIONS: Healon reduced outflow facility less than Viscoat between 3.25 and 10.50 hours postoperatively.

Computational blood flow modeling based on in vivo measurements.

Study of the relationship between hemodynamics and atherogenesis requires accurate three-dimensional descriptions of in vivo arterial geometries. Common methods for obtaining such geometries include in vivo medical imaging and postmortem preparations (vessel casts, pressure-fixed vessels). We sought to determine the relative accuracy of these methods. The aorto-iliac (A/I) region of six rabbits was imaged in vivo using contrast-enhanced magnetic resonance imaging (MRI). After sacrifice, the geometry of the A/I region was preserved via vascular casts in four animals, and ex situ pressure fixation (while preserving dimensions) in the remaining two animals. The MR images and postmortem preparations were used to build computer representations of the A/I bifurcations, which were then used as input for computational blood flow analyses. Substantial differences were seen between MRI-based models and postmortem preparations. Bifurcation angles were consistently larger in postmortem specimens, and vessel dimensions were consistently smaller in pressure-fixed specimens. In vivo MRI-based models underpredicted aortic dimensions immediately proximal to the bifurcation, causing appreciable variation in the aorto-iliac parent/child area ratio. This had an important effect on wall shear stress and separation patterns on the "hips" of the bifurcation, with mean wall shear stress differences ranging from 15% to 35%, depending on the model. The above results, as well as consideration of known and probable sources of error, suggests that in vivo MRI best replicates overall vessel geometry (vessel paths and bifurcation angle). However, vascular casting seems to better capture detailed vessel cross-sectional dimensions and shape. It is important to accurately characterize the local aorto-iliac area ratio when studying in vivo bifurcation hemodynamics.

Measurement of Gd-DTPA diffusion through PVA hydrogel using a novel magnetic resonance imaging method.

Polyvinyl alcohol-cryogel (PVA-C) is a hydrogel that is an excellent tissue mimic. In order to characterize mass transfer in this material, as well as to demonstrate in principle the ability to noninvasively measure solute diffusion in tissue, we measured the diffusion coefficient of the magnetic resonance (MR) contrast agent gadolinium diethylene triaminopentaacetic acid (Gd-DTPA) through PVA-C using a clinical MR imager. The method involved filling thick-walled rectangular PVA-C "cups" with known concentrations of Gd-DTPA solutions. Then by using a fast inversion recovery spin echo MR imaging protocol, a signal "null" contour was created in the MR image that corresponded to a second, known concentration of Gd-DTPA. By collecting a series of MR images through the PVA-C wall as a function of time, the displacement of this second known isoconcentration contour could be tracked. Application of Fick's second law of diffusion yielded the diffusion coefficient. Seven separate experiments were performed using various combinations of initial concentrations of Gd-DTPA within the PVA-C cups (3.2, 25.6, or 125 mM) and tracked isoconcentrations contours (0.096, 0.182, or 0.435 mM Gd-DTPA). The experimental results and the predictions of Fick's law were in excellent agreement. The diffusivity of Gd-DTPA through 10% PVA hydrogel was found to be (2.6 +/- 0.04) x 10(-10) m(2)/s (mean +/- s.e.m.). Separate permeability studies showed that the diffusion coefficient of Gd-DTPA through this hydrogel did not change with an applied pressure of up to 7.1 kPa. Accurate measurements could be made within 30 min if suitable Gd-DTPA concentrations were selected. Due to the excellent repeatability and fast data acquisition time, this technique is very promising for future in vivo studies of species transport in tissue.

A numerical study of blood flow patterns in anatomically realistic and simplified end-to-side anastomoses.

PURPOSE: Recently, some numerical and experimental studies of blood flow in large arteries have attempted to accurately replicate in vivo arterial geometries, while others have utilized simplified models. The objective of this study was to determine how much an anatomically realistic geometry can be simplified without the loss of significant hemodynamic information. METHOD: A human femoral-popliteal bypass graft was used to reconstruct an anatomically faithful finite element model of an end-to-side anastomosis. Nonideal geometric features of the model were removed in sequential steps to produce a series of successively simplified models. Blood flow patterns were numerically computed for each geometry, and the flow and wall shear stress fields were analyzed to determine the significance of each level of geometric simplification. RESULTS: The removal of small local surface features and out-of-plane curvature did not significantly change the flow and wall shear stress distributions in the end-to-side anastomosis. Local changes in arterial caliber played a more significant role, depending upon the location and extent of the change. The graft-to-host artery diameter ratio was found to be a strong determinant of wall shear stress patterns in regions that are typically associated with disease processes. CONCLUSIONS: For the specific case of an end-to-side anastomosis, simplified models provide sufficient information for comparing hemodynamics with qualitative or averaged disease locations, provided the "primary" geometric features are well replicated. The ratio of the graft-to-host artery diameter was shown to be the most important geometric feature. "Secondary" geometric features such as local arterial caliber changes, out-of-plane curvature, and small-scale surface topology are less important determinants of the wall shear stress patterns. However, if patient-specific disease information is available for the same arterial geometry, accurate replication of both primary and secondary geometric features is likely required.

Enthacrynic and acid effects on inner wall pores in living monkeys.

PURPOSE: The influence of the inner wall of Schlemm's canal on aqueous outflow facility remains poorly understood. We examined the relationship between inner wall pore characteristics and outflow facility in living primate eyes in which facility had been pharmacologically increased by ethacrynic acid (ECA) infusion and in contralateral control eyes. METHODS: Outflow facility (two-level constant pressure perfusion) was measured in eight pairs of living monkey eyes before and after administration of a bolus dose of either 0.125 mM ECA or vehicle. After exsanguination, eyes were fixed in situ under constant-pressure conditions (mean fixation pressure approximately 19 mm Hg). The density and diameter of inner wall pores and the number and area of platelet aggregates on the inner wall of Schlemm's canal were measured by scanning electron microscopy. RESULTS: In ECA-treated eyes, outflow facility increased 63% (P < 0.0001), intracellular pore density decreased 46% (P = 0.0094), intracellular pore size increased 27% (P = 0.049), platelet aggregate density increased 158% (P < 0.0001), and area covered by platelets increased 210% (P = 0.012) relative to contralateral controls. Although the average density and size of intercellular pores were essentially unaffected by ECA, an increased density of large (> or = 1.90 microm) intercellular pores was seen in ECA-treated eyes. The density of intracellular pores increased with the duration of fixative perfusion. Other than a weak negative correlation between outflow facility and intracellular pore density in ECA-treated eyes (P = 0.052), facility was not correlated with inner wall pore features. CONCLUSIONS: Our data are most consistent with a scenario in which ECA promotes formation of large intercellular pores in the inner wall of Schlemm's canal, which are then masked by platelet aggregates. Masking of intercellular pores, combined with fixation-induced alteration of inner wall pore density, greatly complicates attempts to relate facility to inner wall structure and suggests that in vivo pore density is smaller than in fixed tissue. Additionally, facility-influencing effects of ECA on the juxtacanalicular tissue cannot be excluded.