A versatile laser-induced porcine model of outer retinal and choroidal degeneration for preclinical testing

Abstract

Over 30 million people worldwide suffer from untreatable vision loss and blindness associated with childhood-onset and age-related eye diseases caused by photoreceptor (PR), retinal pigment epithelium (RPE), and choriocapillaris (CC) degeneration. Recent work suggests that RPE-based cell therapy may slow down vision loss in late stages of age-related macular degeneration (AMD), a polygenic disease induced by RPE atrophy. However, accelerated development of effective cell therapies is hampered by the lack of large-animal models that allow testing safety and efficacy of clinical doses covering the human macula (20 mm2). We developed a versatile pig model to mimic different types and stages of retinal degeneration. Using an adjustable power micropulse laser, we generated varying degrees of RPE, PR, and CC damage and confirmed the damage by longitudinal analysis of clinically relevant outcomes, including analyses by adaptive optics and optical coherence tomography/angiography, along with automated image analysis. By imparting a tunable yet targeted damage to the porcine CC and visual streak - with a structure similar to the human macula - this model is optimal for testing cell and gene therapies for outer retinal diseases including AMD, retinitis pigmentosa, Stargardt, and choroideremia. The amenability of this model to clinically relevant imaging outcomes will facilitate faster translation to patients.

Keywords: Gene therapy; Ophthalmology; Retinopathy; Stem cell transplantation; Stem cells.

Figure 1. Schematic of the effect of laser injury on RPE ablation followed longitudinally using clinically relevant imaging modalities.

A 532 nm micropulse laser, with a DC ranging in power from 1% to 3% was used to ablate RPE cells. Live imaging of laser-damaged eyes was performed using optical coherence tomography (OCT) to evaluate retina structure; fluorescein angiography (FA) and indocyanine green angiography (ICGA) to evaluate choriocapillaris and retinal vasculature leakage, atrophy, and proliferation; optical coherence tomography-angiography (OCTA) to evaluate choroidal vasculature confluency; and adaptive optics (AO) to image cone PRs at a single-cell resolution. Images were produced using BioRender software with publication license OT25AR5EPP.

Figure 2. Fluorescein angiography (FA) of laser injured pig retina.

(A–P) Early (A–H) and late (I–P) phase FA images of the full fundus (A–D and I–L) and higher magnification of the lasered area in the visual streak (E–H and M–P) for baseline (A, E, I, and M), 2 weeks (B, F, J, and N), 6 weeks (C, G, K, and O), and 12 weeks (D, H, L, and P) past 1% and 1.5% duty cycle (DC) laser treatment of the pig eye. Blue arrowheads mark large choroidal vessels visible through damaged RPE and choriocapillaris. Scale bars: 500 μm.

Figure 3. OCTA segmentation of choriocapillaris.

(A–C) Simultaneous evaluation of 2 weeks 1% DC laser lesion (A) by ICGA (B) and OCTA (C). (D and E) Serial OCTA scans are projected as pseudo-color yellow signal; blue arrowhead shows margin of choriocapillaris as a continuous yellow signal. (E) OCTA signal was segmented into the retinal capillaries, choriocapillaris (CC), and choroid vasculature (red dotted lines); IZ/RPE layer was used as a landmark to define retinal from CC vasculature. (F–K) The choriocapillaris signal was represented in C-scan (en face) OCTA images of 1% duty cycle laser lesion (F–H) and 1.5% duty cycle laser lesion (I–K) shown at 2-, 6-, and 12 weeks past the laser lesion. (L) The graph depicts the pixel gray value changes in percent in CC density from baseline to 12 weeks as seen in en face images. Box and whiskers represent minimum to maximum, 25th and 75th percentile, median, and single values. Data were analyzed with mixed-effect model (REML) and Bonferroni multiple-comparison test. n = 13, 7, 4, 4, 4, 4 eyes for 1% DC at baseline, 2 weeks, 4 weeks, 6 weeks, 8 weeks, and 12 weeks, respectively; n = 13, 10, 10, 4, 4, 4 for 1.5% DC at baseline, 2 weeks, 4 weeks, 8 weeks, and 12 weeks, respectively. P values are reported as *P < 0.05; **P < 0.005; ***P < 0.0005; ****P < 0.0001. Scale bars: 500 μm

Figure 4. Qualitative OCT evaluation of the lasered retina.

(A and B) Comparison between OCT (A) and histology (B) of the pig retina identifying the neuron fiber layer (NFL), ganglion cell layer (GCL)/inner plexiform layer (IPL), inner nuclear layer (INL), horizontal cells cytoplasm layer (HCC), outer plexiform layer (OPL), outer nuclear layer (ONL), ellipsoid zone (EZ), and interdigitation zone (IZ)/retinal pigment epithelium (RPE). (C and D) Fundus image and correspondent OCT scan at 2 weeks (C) and 12 weeks (D) after 1% and 1.5% DC laser treatment. (E–Q) Higher-magnification OCT images of laser lesions of the pig retina from 1% laser duty cycle (F–K) and 1.5 % laser duty cycle (M–Q); baseline (E and L), 2 weeks (F and M), 4 weeks (G and N), 6 weeks (H and O), 8 weeks (J and P), and 12 weeks (K and Q) after the laser injury. Scale bars: 100 μm (A and B), 1 mm (C and D), and 50 μm (E–Q). Arrowheads in G-J and L-P mark hyperreflective foci postlaser damage.

Figure 5. Segmentation and quantification of OCT data.

(A and B) Representative image of injured retina before (A) and after segmentation (B). (C–I) Quantification of OCT images from 1% (orange line) and 1.5% (black line) duty cycle laser–injured retina based on manual segmentation of different retinal layers. Percent change of the area of the damaged zone (DZ), the neuron fiber layer (NFL), ganglion cell layer (GCL)/inner plexiform layer (IPL), inner nuclear layer (INL), outer nuclear layer (ONL), ellipsoid zone (EZ), and interdigitation zone (IZ)/retinal pigment epithelium (RPE) in 1% and 1.5% duty cycle laser–treated retinas weeks 2–12 after laser injury compared with the baseline. OCT, optical coherence tomography; HCC, horizontal cells cytoplasm layer; OPL, outer plexiform layer. (D–I) Box and whiskers represent minimum to maximum, 25th and 75th percentile, median, and single values. Data were analyzed with mixed-effect model (REML) and Bonferroni multiple -comparison test. Three OCT scan per eye were analyzed. n = 7, 4, 3, 4, 4 eyes for 1% DC at 2 weeks, 4 weeks, 6 weeks, 8 weeks, and 12 weeks, respectively; n = 10, 10, 3, 4, 4 for 1.5% DC at 2 weeks, 4 weeks, 8 weeks, and 12 weeks, respectively (C–I). P values are reported as *P < 0.05; **P < 0.005; ***P < 0.0005; ****P < 0.0001.

Figure 6. Adaptive optics–based (AO-based) quantification of cone photoreceptor cell damage in laser-injured pig retinas.

(A) En face view of pig fundus with aligned AO montage images corresponding to 1%, 1.5%, and 2% DC laser damage. The cyan arrowhead points to the choroidal vessels visible through the damaged retina. (B) Higher magnification of an AO imaged area in 1.5% duty cycle laser–injured retina; cyan arrowheads are pointing at the margin of the lasered retina. Detail of pig RPE imaged outside the visual streak. (C and D) Cone diameter calculated on untreated pig retina. (E–J) Representative AO image of baseline (E) and 1% duty cycle laser injured retinal area (H); identification and segmentation of photoreceptors in baseline (F and G) and 1% duty cycle laser–injured retinal area (I and J); and red dots marking cone PRs used for automated quantification at baseline, with green dots indicating adjusted quantification in lasered areas. (K) Photoreceptor cell density per mm² compared with baseline images in 1% (orange) and 1.5% (gray) laser–injured pig retinas at 4, 8, and 12 weeks after injury. Scale bars: 500 μm (A, B, E, and H) and 100 μm (C, F, and I). (D and K) Box and whiskers represent minimum to maximum, 25th and 75th percentile, median, and single values. Data were analyzed with mixed-effect model (REML) and Bonferroni multiple-comparison test. Five regions of interest (ROI) in 3 images for each eye were analyzed. n =25 eyes (D); n = 6, 5, 3, 3 for 1% DC at baseline, 2 weeks, 4 weeks, 8 weeks, and 12 weeks, respectively; 6, 6, 3, 3 for 1.5% DC at baseline, 2 weeks, 4 weeks, 8 weeks, and 12 weeks, respectively (K). The 6-week time point was excluded from quantification; 1% and 1.5% DC were shown as n = 2 eyes. *P < 0.05; ***P < 0.0005; ****P < 0.0001.

Figure 7. Histological evaluation of the laser injured retina.

(A–H) representative images of healthy (A, C, E, and G) and laser-injured (B, D, F, and H) retina after 2 and 12 weeks of 1% and 1.5% DC laser treatments. (I–K) Nuclei number quantification and comparison between healthy and laser-treated area of the retina for INL (I), ONL (J), and RPE (K) at 2 and 12 weeks after 1% (orange) and 1.5% DC (gray) laser treatments. (L and M) Quantification of retina thickness for INL (L) and ONL (M) at 2 and 12 weeks after 1% and 1.5% DC laser treatments. (N) Quantification of migrating pigmented cells per mm² of retina identified at 2 and 12 weeks after 1% and 1.5% DC laser treatments. (I–N) Box and whiskers represent minimum to maximum, 25th and 75th percentile, median, and single values. Data were analyzed by 2-way ANOVA and Bonferroni multiple-comparison test. Five slides for each eye were analyzed. n = 3 eyes per condition (I–N). Arrow heads in B, D, F, and H mark migratory pigmented cells *P < 0.05; **P < 0.005; ***P < 0.0005. Scale bar: 100 μm.

Figure 8. IHC confirms damage to photoreceptors and RPE and shows immune cell activation in the lasered injured retina.

(A–F) Healthy retina (A and D) and 1% and 1.5% DC laser–treated retina at 2 weeks (B and E) and 12 weeks (C and F) after treatment. RPE65 labels the retinal pigment epithelium (magenta). PNA labels the cone photoreceptor outer segment (white). IBA1 labels immune cells (orange). DAPI labels cell nuclei (cyan). Scale bar: 50 μm. Two of the total eyes (1% DC n = 3 at 2 weeks, n = 5 at 12 weeks; 1.5% DC n = 7 at 2 weeks, n = 5 at 12 weeks) were fixed and processed for IHC.