Widespread subclinical cellular changes revealed across a neural-epithelial-vascular complex in choroideremia using adaptive optics

Abstract

Choroideremia is an X-linked, blinding retinal degeneration with progressive loss of photoreceptors, retinal pigment epithelial (RPE) cells, and choriocapillaris. To study the extent to which these layers are disrupted in affected males and female carriers, we performed multimodal adaptive optics imaging to better visualize the in vivo pathogenesis of choroideremia in the living human eye. We demonstrate the presence of subclinical, widespread enlarged RPE cells present in all subjects imaged. In the fovea, the last area to be affected in choroideremia, we found greater disruption to the RPE than to either the photoreceptor or choriocapillaris layers. The unexpected finding of patches of photoreceptors that were fluorescently-labeled, but structurally and functionally normal, suggests that the RPE blood barrier function may be altered in choroideremia. Finally, we introduce a strategy for detecting enlarged cells using conventional ophthalmic imaging instrumentation. These findings establish that there is subclinical polymegathism of RPE cells in choroideremia.

Fig. 1. Enlarged retinal pigment epithelial (RPE) cells observed in choroideremia.

a, b Color fundus images of an affected male and a female carrier showing areas of outer retinal atrophy and mild pigmentary changes, respectively. The white ‘x’ denotes the fovea (eccentricity = 0.0 mm). Subject codes are shown in brackets (L = left eye, R = right eye; A = affected male, C = female carrier). The number in the brackets refers to the subject number (Supplementary Table 1). White rectangles indicate areas where high resolution adaptive optics (AO) images were obtained. Scale bar, 1 mm. c, d Late phase adaptive optics enhanced indocyanine green (AO-ICG) images of the RPE mosaic obtained in the white rectangles from (a and b). e For comparison, an AO-ICG image obtained from a similar area in a healthy right eye is shown. Green hexagons denote the approximate size of single RPE cells in the immediate vicinity of the hexagon (barely visible in the healthy eyes). The nuclei of RPE cells are visible in some of the enlarged cells (arrows point to examples of nuclei visible as hypofluorescent spots). In the healthy eye, RPE cell size is nearly constant across the image; in contrast, the RPE cells in the carrier eye are variable in size, with many of the cells enlarged. The RPE cells in the affected eye are consistently enlarged across the entire image. Scale bar for (c–e), 100 µm. f Quantitative measurements of RPE cell density obtained at different locations (eccentricities) from the fovea out to ~5.0 mm in the temporal direction (female carriers, unfilled symbols; affected males, filled symbols; triangles, left eye; squares, right eye). Measurements of RPE cell density in the affected eyes were possible only up to the edge of atrophy, which was generally <3.0 mm. RPE density in all subjects was significantly lower than expected normal values (gray dots; the gray band represents 99.9% confidence interval).

Fig. 2. Multimodal imaging corroborates the finding that the RPE cells are enlarged in choroideremia.

Co-registered adaptive optics optical coherence tomography (AO-OCT) images and (AO-ICG) images from choroideremia are shown in comparison to healthy examples. a–h RPE cells from both affected males and female carriers are enlarged compared to (i–l) healthy RPE cells. Green circles are examples of individual RPE cells from each pair of images. RPE cell nuclei can be seen as the small dark spots in AO-OCT images (e.g., white arrows). The increased spacing between cell nuclei in choroideremia compared to healthy eyes can be seen in the AO-OCT images. The retinal locations (eccentricities) of the images are (a, b) 1.0 mm, (c, d) 1.5 mm, (e–h) 0.5 mm, (i, j) 0.5 mm, and (k, l) 1.5 mm. Scale bar, 50 µm.

Fig. 3. Visualization of the photoreceptor, RPE, choriocapillaris complex in choroideremia.

a AO images acquired at the fovea (eccentricity = 0.0 mm) showing foveal cone photoreceptors (PR), RPE cells, and the choriocapillaris (CC) microvasculature. Images from an affected male, a female carrier, and a healthy eye are shown. Note that subject A4L has a relatively well-preserved island of RPE cells at the fovea in which RPE cells are less enlarged compared to other affected males; outside of the fovea, RPE cells still form a contiguous monolayer but are dramatically enlarged (see Fig. 5a). Scale bars: PR, 10 µm; RPE, 50 µm; CC, 100 µm. b Box plots of photoreceptor spacing, RPE spacing, and choriocapillaris flow void diameters show that the RPE is the most severely affected layer of these three layers (center line: median; box limits: upper/lower quartiles; whiskers: 1.5× interquartile range; points beyond the whiskers: outliers). Data corresponding to the subjects shown in (a) can be determined using the legend. Measurements of photoreceptors, RPE, and choriocapillaris performed in choroideremia were compared to normative histologic data, normative in vivo RPE data, and normative in vivo choriocapillaris data, respectively. For subjects who had two visits, only data from the first visit was used for this analysis. c Longitudinal imaging acquired at the same location and co-registered across visits revealed the degree to which photoreceptors, RPE, and choriocapillaris changed from one visit to the next (time between visits varied between 2 and 12 months; Supplementary Table 1). The largest changes were observed in the RPE layer, further corroborating our finding that the RPE layer is the most disrupted layer out of these three layers.

Fig. 4. ICG labeled photoreceptors in mice and human.

a Following intraperitoneal injection of ICG in mice, ICG can be detected within the RPE in unfixed cryosections, as has been previously demonstrated. The overlying photoreceptors (PR) remain unlabeled under normal conditions due to the tight junction between RPE cells, which together establish the outer blood retinal barrier. The PR outer segments (OS) and outer nuclear layer (ONL) can be discerned using autofluorescence imaging (430 nm). A faint infrared autofluorescence can be observed in the choroid due to melanin. Scale bar, 50 µm. b No detectable ICG signal was observed in the photoreceptor layer after systemic injection of ICG. The presence of photoreceptor profiles was confirmed using differential interference contrast (DIC) microscopy. However, following an ex vivo incubation with ICG, the photoreceptors were readily labeled with ICG. These results were further corroborated using a custom-assembled adaptive optics microscope capable of simultaneous acquisition of en face confocal reflectance (displayed in log scale), non-confocal split detection, and AO-ICG images of the photoreceptor mosaic. Scale bar (all images in b), 10 µm. c ICG labeled photoreceptors observed in a female carrier. The heterogeneous RPE fluorescence pattern can be observed in the background. Zooms of the white box show that the fluorescently labeled photoreceptors are consistent with photoreceptors imaged using non-confocal split detection. Scale bars: 50 µm (C6L larger image), 10 µm (zooms, AO-ICG and Split Det). d The densities of cone photoreceptors in areas of ICG labeling are similar to normative histologic values. e Outer retinal length (ORL) measurements were performed using AO-OCT images acquired in areas of ICG labeling and in areas without ICG labeling. There was no apparent difference in ORL between labeled or non-labeled photoreceptors, and both were similar to ORL measurements performed in healthy eyes. f Retinal sensitivity measurements (microperimetry) performed within patches of ICG labeled photoreceptors were within normal limits and similar to measurements obtained in neighboring, non-labeled photoreceptors.

Fig. 5. Detection of RPE cell polymegathism using late phase ICG imaging.

a Late phase ICG image acquired using a commercially available Heidelberg scanning laser ophthalmoscope (SLO) showing widespread enlarged RPE cells in an affected male. b Zoom of the white square in (a), contrast adjusted to show cellular detail. c AO-ICG image of the same region as (b). Many of the cells observed in (c) can be detected in (b). In the lower-left corner of (b and c), the RPE cells are smaller than those in the upper-right corner, but individual RPE cells are still visible. d Heidelberg SLO image from a female carrier. Numerous hyperfluorescent enlarged RPE cells can be observed, especially near the left edges of the image. e, g Contrast-adjusted zoom of white rectangles. f, h AO-ICG images corresponding to (e and g) showing that the heterogeneous pattern of fluorescence captured using the Heidelberg SLO matches the pattern imaged using AO-ICG. These side-by-side comparisons illustrate that individual enlarged RPE cells can be detected using conventional imaging, even without AO. a, d The green ‘x’ denotes the fovea. Scale bar, 1 mm. b, c, e–h Scale bar, 500 µm.