Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Nov 1.
Published in final edited form as: Exp Eye Res. 2024 Oct 16;248:110128. doi: 10.1016/j.exer.2024.110128

Ghost vessels in the eye: cell free choriocapillaris domains in atrophic age-related macular degeneration

Robert F Mullins 1,2, Miles J Flamme-Wiese 1,2, Emma M Navratil 1,2,3, Erin A Boese 1,2, Katayoun Varzavand 1,2, Megan J Riker 1,2, Kai Wang 1,4, Edwin M Stone 1,2, Budd A Tucker 1,2
PMCID: PMC11532014  NIHMSID: NIHMS2030504  PMID: 39419369

Abstract

The choriocapillaris is a dense vascular bed in the inner choroid that supplies the photoreceptor cells and retinal pigment epithelium (RPE). While loss of choriocapillaris density has been described in association with age-related macular degeneration (AMD), whether these changes are primary or secondary to RPE degenerative changes in AMD has been debated. In this study we characterized choriocapillaris loss by quantifying “ghost” vessels in a series of 99 human donor maculae labeled with the UEA-I lectin, and found significant increases in early-intermediate AMD and a greater difference in geographic atrophy in areas with intact RPE. Eyes were genotyped at the CFH Tyr402His locus, and those homozygous for the His allele showed significantly more ghost vessels than those with other genotypes. When only non-AMD eyes were evaluated, His homozygotes had increased ghost vessel density but this trend did not reach statistical significance. These results support the notion that choriocapillaris death often precedes RPE degeneration in AMD and that this loss is an important therapeutic consideration for AMD.

Keywords: Choriocapillaris, aging, age-related macular degeneration

1. Introduction

Vascular endothelial cells perform a number of crucial functions in maintaining cell, tissue and organismal homeostasis. In addition to forming tubes that serve as the major interstate for circulating cells, oxygen, hormones, and other bioactive molecules and their metabolic byproducts, endothelial cells perform diverse activities such as maintaining hemostasis (maintaining the blood in a liquid state or facilitating its clotting as needed), and interconverting CO2 and bicarbonate ions. Moreover, new functions of the endothelium continue to be discovered including guidance of non-vascular tissue development and remodeling (Tamari et al., 2020; Tuckermann & Adams, 2021) and regulating adaptive immunity globally through the uptake and sialylation of circulating immunoglobulins (Glendenning et al., 2023). In the case of the choriocapillaris endothelium of the posterior pole, the large caliber fenestrated capillaries support the outer retina and RPE (Linsenmeier & Padnick-Silver, 2000), impact the developing retinal pigment epithelium through paracrine regulation of its adhesion molecules (Benedicto et al., 2017), provide circulating retinoids for the visual cycle, and impact the immune milieu by producing cytokines, adhesion molecules, and MHC antigens (Goverdhan et al., 2005; McLeod et al., 1995).

Apart from normal physiology, choroidal endothelial cells can exhibit altered structure and function during aging and disease. Structural and molecular changes in the aging choriocapillaris include loss of fenestrae (Grebe et al., 2019), thickening of the inner wall of the vessel, and loss of CD34 with increased ICAM1 (Sohn et al., 2014; Voigt et al., 2020). These cellular changes are accompanied by striking changes in the extracellular environment including altered composition of the extracellular matrix (Uno et al., 2006), deposition of esterified and unesterfied cholesterol (Curcio et al., 2005), advanced glycation end products (Handa et al., 1999), monomeric C-reactive protein (Bhutto et al., 2011; Chirco et al., 2016; Molins et al., 2016) and complement components including the cytolytic C5b-9 membrane attack complex (MAC). The last of these is elevated with increased age and further in eyes with early-intermediate age-related macular degeneration (Mullins et al., 2014).

Degenerative changes in the choriocapillaris in AMD have been appreciated for decades, although there has not been strong consensus as to whether they represent an early cause of disease, an epiphenomenon of RPE failure, or whether both pathways may exist in different individuals. Although evidence for vascular loss as a contributor to early AMD had been suggested by earlier studies of fluorescein angiography (Hayreh, 2010; Pauleikhoff et al., 1990) and laser Doppler flowmetry (Berenberg et al., 2012; Grunwald et al., 1998), due in part to the lack of clinical imaging of the choriocapillaris, uncertainty remained. Support for increased loss of molecular species derived from the vasculature, as compared to the RPE, has also come from biochemical studies of RNA (Whitmore et al., 2013). Likewise, proteomic studies, in which loss of choriocapillaris enriched protein species such as carbonic anhydrase 4, plasmalemma vesicle associated protein, and MHC class I molecules was observed during AMD progression (Yuan et al., 2010). In the current study, we evaluated a previously studied cohort of human donor eyes (Sohn et al., 2019). Instead of quantifying the density of healthy vessels, which could be altered by both loss and physiological regulation, we measured the number of ghost vessels of the choroid per unit length between eyes with healthy aging, early-intermediate atrophic AMD, and geographic atrophy. We describe a remarkable increase in choriocapillaris ghost vessels in eyes with anatomically intact RPE and geographic atrophy.

2. Methods

2.1. Human donor eyes

Human donor eyes were obtained following informed consent of the next of kin through the Iowa Lions Eye Bank (Iowa City, IA). Macular punches centered on the fovea centralis were collected and fixed for 2 hours in 4% paraformaldehyde in phosphate buffered saline (PBS, 1x from 10x stock, Gibco), followed by cryoprotection and freezing in OCT compound (Sakura) as described previously (Barthel & Raymond, 1990). The majority of samples analyzed were collected as part of a previous study (Sohn et al., 2019) but had not been analyzed for density of ghost vessels or in relationship to genotype.

2.2. Evaluation of morphometric data

We performed immunofluorescence and Ulex europaeus agglutin-I (UEA-I) lectin histochemistry on a series of formaldehyde fixed donor maculae. This set contained sections from 147 donors that were categorized as healthy controls (i.e., were negative for macular degeneration on chart review and/or histology), early-intermediate AMD, or geographic atrophy. We re-evaluated samples from this data set and quantified the number of ghost vessels per mm of Bruch’s membrane (i.e., measurable length of tissue section since quality and quantity of tissues varied donor-to-donor). All measurements were collected masked to donor genotype and affection status. Ghost vessels were defined as UEA-I negative cavities that were hypoautofluorescent in the UV channel compared to Bruch’s membrane and the choroidal stroma and that generally were bounded by intercapillary pillars or other autofluorescing extracellular matrix (Figure 1)(Mullins, Johnson, et al., 2011).

Figure 1.

Figure 1.

Open in a new tab

Appearance of ghost vessels in the human macula. A-C, images from a 79-year-old male donor clinically diagnosed with early atrophic AMD. (A), phase contrast image showing the intact RPE and healthy appearing choriocapillaris with numerous capillary segments in cross section; (B) same field as A labeled with the lectin UEA-I showing absence of endothelium in several of the capillary segments (asterisks); (C) same field with UEA-I and brightfield images overlaid as well as FITC channel autofluorescence and DAPI nuclear stain. (D) Stereo pair of the submacular RPE and choroid of an 88 year old female donor with histopathologic evidence of AMD labeled with UEA-I (red), anti-CD45 antibody green), and DAPI (blue). Scalebar: 25μm (A-C), 10μm (D).

During processing of frozen sections, RPE can be lost due to either artifactual separation or pathology. For this study, eyes without RPE were removed for further analysis. If the RPE layer was mostly or partially intact, measures of the vasculature were collected only beneath intact RPE. One hundred eyes from one hundred donors passed this filtering (68 control, 25 AMD, and 7 GA). One of the AMD eyes was found to possess a type 1 macular neovascular membrane and was not included with the AMD set; the rest were used for subsequent analysis (n=99).

The total number of ghost vessels was determined and divided by the measurable section length (specifically, a line drawn along the contours of Bruch’s membrane to accommodate section folds) ranging from 1,847 to 8,052μm. Masked data were reported and analyzed as number of ghost vessels per mm. For the images in Figure 1 A–C, the depicted field shows 5 ghost vessels under 111μm of measurable Bruch’s membrane length and with intact RPE.

2.3. Genotyping

Donor eyes were genotyped as described previously (Mullins, Dewald, et al., 2011) using DNA from either peripheral blood or, in some cases, a small piece of ciliary body collected at the time of tissue processing. Genotyping was performed with a TaqMan assay for the rs1061170 SNP which is responsible for a non-conservative histidine substitution for tyrosine at position 402 of the FH preprotein. Data were analyzed for 84 non-GA eyes for which DNA and genotypes were available.

2.4. Histochemistry

Immunofluorescence and histochemical staining of macular sections from eyes with high and low numbers of ghost vessels was performed. Immunohistochemistry was conducted as described previously (Mullins et al., 2014) using antibodies directed against the membrane attack complex (Dako, Cat. # M077701-5, 1:300 dilution) and CD45 (BD Biosciences, Cat. # 555480, 1:200 dilution). Antibodies were visualized using Alexa-488 conjugated anti-mouse secondary antibody (Thermo Fisher Scientific, Cat. #A21202, 10 μg/mL). Sections were also labeled with the fucose binding lectin UEA-I (Vector laboratories, 40 μg/mL (1:50 dilution) conjugated to rhodamine isothiocyanate, which was added at the secondary antibody step. Grayscale images of red, green, and blue channels were collected individually and merged, and levels were adjusted using Photoshop.

For some experiments, neutral lipids, cholesterol, and general classes of lipids were detected using Oil Red O (Millipore-Sigma, O1391), filipin (Millipore-Sigma, SAE0087-1ML), and Sudan black B (Fisher Scientific, AC419830100).

2.5. Choroidal whole mounts

Biopsy punches (4.0mm) of RPE-choroid tissue adjacent to OCT-embedded sections was obtained from two eyes and the RPE was removed by incubation with trypsin (Thermo Fisher Scientific, Cat. #15050-065, undiluted) for 1 hr at 37C followed by gentle debridement. Punches were then incubated in 500 μL of 40 μg/mL UEA-I-rhodamine overnight at 4C with agitation, washed, and coverslipped in Aquamount (Epredia) and were imaged RPE-side up on an upright epifluorescence microscope (Olympus BX41).

2.6. Confocal microscopy

Crysotat sections (14 μm thickness) from an 81-year-old male donor were collected and used for immunofluorescence with UEA-I, anti- CD45 (BD Biosciences # 555480, 1:200) and DAPI (Thermo-Fisher, 1:25,000 of 5mg/mL stock), and were viewed on a LSM 980 Airyscan2 confocal microscope (Zeiss). Z-series were obtained at 0.5μm steps and used to construct stereo pairs in ImageJ (ImageJ2 v2.14.0/1.54f). A set of 29 optical sections was projected using the 3D project method (Brightest Point, Y-axis rotation, 10 pixel spacing, initial angle 357°, total rotation 8°. A 2 panel montage was generated from the projection.

2.7. Transmission Electron Microscopy

A subset of donor eyes determined to have ghost vessel counts in the top and bottom quartiles (as determined above) was evaluated by transmission electron microscopy as described previously (Mullins et al., 2007). Sections were collected on formvar coated grids and images were of the RPE-Bruch’s membrane-choriocapillaris complex were obtained using either a JEOL JEM-1230 or Hitachi H-7800 transmission electron microscope in order to characterize ghost vessels at high resolution at different stages of evolution. Brightness and contrast levels were adjusted for each panel in Photoshop.

2.8. Data analysis

Ghost counts per unit length were evaluated related to disease status (control, early-intermediate AMD and GA); complement factor H (CFH) Y402H genotype for all eyes; and CFH Y402H genotype for control eyes only. Data were evaluated using either ANOVA analysis or a negative binomial regression using the glm.nb function in the R package MASS (version 7.3–60).

3. Results

3.1. Evaluation of ghost vessels

Ghost capillary segments were identified in dark field micrographs as hypofluorescent lumens lacking a surrounding UEA-I-labeled endothelial cell. Interestingly, eyes can contain multiple ghost vessels without a grossly abnormal appearance (Figure 1A–C). Figure 1 shows the challenge of identifying ghost vessels on routine histology, in which capillary segments lacking an endothelial cell can grossly appear normal (Figure 1A, early AMD) without the application of a marker of viable endothelial cells such as UEA-I (Figure 1B, C, asterisks). Only a shallow basal laminar deposit can be noted on the right-hand side of the image. Confocal microscopy (Figure 1D) shows the appearance of ghost vessels that can be occupied by immune cells, which is also deceptively suggestive of an intact, perfused capillary.

Whole mounts of choroids with low (Figure 2A) and high (Figure 2B) densities of ghost cells provide additional detail regarding the structure of the choriocapillaris in AMD.

Figure 2.

Figure 2.

Open in a new tab

Whole mount of a 97-year-old control donor (A) and a 96-year-old donor with geographic atrophy (B). Sections are labeled with UEA-I lectin (red). Note the preservation of autofluorescent intercapillary pillar stubs (green) that persist after degeneration of the endothelium. Some RPE lipofuscin yellow) remains in both samples (*).

Following the dropout of healthy capillaries, intercapillary pillars can be observed as hyperautofluorescent stubs in the FITC channel.

3.2. Quantification of ghost vessels-AMD affection status

The number of ghost vessels per mm of measurable Bruch’s membrane was performed on images only under an intact RPE monolayer in a masked fashion on submacular choriocapillaris from age-matched control eyes, eyes with atrophic early to intermediate AMD, and eyes with geographic atrophy (omitting areas of RPE degeneration). Compared to control eyes, eyes with early-intermediate AMD had significantly higher density of ghost vessels (p<0.05). Eyes with geographic atrophy had significantly higher density of ghost vessels beneath the intact RPE than control eyes (p<10−4) and the AMD group (p<0.01) (Figure 3A).

Figure 3.

Figure 3.

Open in a new tab

Distribution of ghost vessels. Density of ghost vessels was related to AMD affection status when comparing unaffected, early-intermediate AMD, and GA eyes (A). Panel B shows the relationship between CFH Y402H genotype and ghost vessel density in all samples (n=84) and panel C depicts AMD-depleted controls only (n=63). Values are depicted as average number of ghost vessels per mm. Only areas with intact RPE were evaluated. *, p<0.05; ** p<0.01; *** p<10−4.

3.3. Quantification of ghost vessels-CFH genotype

In addition to disease affection status, we sought to determine if genotype at the rs1061170 SNP, which encodes the Y402H polymorphism, was associated with altered density of ghost vessels. GA eyes were excluded from this study, and genotypes were available for 84 maculae. Of the measured samples, 11 were homozygous for the high risk (hereafter referred to as His/His), 45 were heterozygous (Tyr/His) and 28 were homozygous for the low-risk allele (Tyr/Tyr). When all eyes were pooled for the analysis, eyes with the His/His genotype had a significantly higher ghost vessel density than Tyr/His or Tyr/Tyr eyes (p<0.05). Heterozygous eyes did not have significantly more ghost vessels than Tyr/Tyr eyes (Figure 3B). When eyes diagnosed with AMD were removed from the analysis and only control eyes were evaluated, His/His eyes showed a trend toward a higher ghost vessel density (1.6-fold higher on average) but this trend was non-significant (Figure 3C).

3.4. Complement distribution in ghost vessels

Sections of eyes with ghost vessels were labeled with antibodies directed against the C5b-9 membrane attack complex (MAC). The MAC is normally distributed in human eyes in Bruch’s membrane in domains surrounding and often overlapping with the choriocapillaris as well as in some subRPE deposits, especially hard drusen (Mullins et al., 2014). Figure 4 depicts section from the maculae of two donors with geographic atrophy, with 4B corresponding to the same eye shown in Figure 2B. In fields with both intact capillaries and ghost vessels, labeling was stronger on intact capillaries (Figure 4, arrows) compared to ghosts (asterisks) indicating a level of complement turnover or replenishment that requires intact vasculature.

Figure 4.

Figure 4.

Open in a new tab

Distribution of the membrane attack complex in normal and ghost vessels. Immunofluorescence of anti-C5b-9 MAC (green) colabeled with UEA-I lectin in the maculae of two donors clinically diagnosed with geographic atrophy. Note the presence of green fluorescence associated with ghost vessels (asterisks) that is relatively attenuated compared to intact (red) vessels. Scalebar =50μm. The MAC is also present in some subRPE deposits including hard drusen.

3.5. Ultrastructure of ghost vessels and natural history

Eyes with high and low numbers of ghost vessels were evaluated using transmission electron microscopy. The normal choriocapillaris (Figure 5A) is fenestrated, especially along the RPE facing surface, and may exhibit considerable numbers of caveolae (Nakanishi et al., 2016). Adjacent endothelial cells may elaborate intricate junctional complexes and, like other capillaries, generate a basal lamina that is partially shared with pericytes (when the latter are present). In what are presumed to be stages of their development, ghost vessels (asterisk) initially appear as electron lucent cavities between intercapillary pillars with abundant electron dense profiles (Figure 5B). We did not observe residual basal lamina, which probably recedes rapidly following endothelial cell death. In both immunofluorescent (e.g., the right half of Figure 1C) and ultrastructural preparations (Figure 5C), nuclei are frequently observed within the space created by the loss of the endothelium, which can lead to the overestimation of intact, perfused vasculature in histological studies. Longstanding ghost vessels appear to undergo sedimentation with extracellular matrix that includes both fibrillar collagens (Figure 5D) and long-spacing collagen, which is reminiscent of basal laminar deposits (Figure 5E).

Figure 5.

Figure 5.

Open in a new tab

Natural history of choroidal ghost vessels. Intact capillary with a circulating leukocyte. Nuclei of choriocapillaris endothelial cell and RPE (top right) are apparent. After degeneration of the endothelium (B), ghost vessels have a lucent core (asterisk), while intercapillary pillars are still apparent. These spaces can become occupied by solitary cells (C) and eventually fill with extracellular matrix which can include loose fibrillar collagen (D)and long spacing collagen (E), ultimately filling with dense extracellular matrix (F). Intercapillary pillars (G) recede during “depillarization” (Li et al., 2018), often leaving nodular remains. In some cases, the loss of the choriocapillaris layer results in migration of Sattler’s layer vessels toward the RPE (H). Scalebars: 2μm (A-F, H), 500 nm (G).

Ghost vessels ultimately fill with a dense mixture of collagen and ground substance (Figure 5F). The original capillary space is obliterated and the intercapillary pillars undergo concomitant degeneration (Figure 5G). The loss of intercapillary pillars has recently been observed in thick (1μm) sections of donor eyes and referred to as depillarization (Li et al., 2018) and suggests that intact endothelial cells are involved in maintenance of Bruch’s membrane. During the loss of the choriocapillaris, the intermediate vascular layers of the choroid (referred to as Sattler’s layer vessels) can migrate internally (Figure 5H).

4. Discussion

Death of microvascular endothelial cells is a common event in numerous diseases of aging including Alzheimer disease, vascular dementia, and diabetic retinopathy (Ambrose, 2016; Garner, 1993; Love & Miners, 2016). In AMD, the occurrence of vascular loss has been noted for decades, but prior to the advent of optical coherence tomography (OCT) angiography, assessing the relative timing of RPE and choriocapillaris degenerative changes has relied on histological samples. Hayreh (e.g. (Hayreh, 1990)) suggested that the developmental anatomy of the choroid lent itself to areas of nonperfusion that, under a generalized loss of flow, would result in particular hypoxia of the central macula. Pauleikhoff and colleagues (Pauleikhoff et al., 1990; Pauleikhoff et al., 1999) found that eyes with drusen showed a decreased choroidal perfusion with fluorescein and indocyanine green. More recent studies using the novel technique of OCT angiography have found similar loss of flow (described as flow voids) in eyes with geographic atrophy, including outside of the area of RPE loss (Marsh-Armstrong et al., 2019; Moult et al., 2016). In a longitudinal study in which patients were imaged before and after the onset of nascent geographic atrophy, flow deficits in the areas that subsequently developed lesions was significantly increased at the prior visit compared to adjacent areas that did not develop atrophy. Choriocapillaris flow changes are also associated with physiological changes that predict AMD; measurements of rod-mediated dark adaptation measurement showed a correlation between increased flow deficit of the choriocapillaris and longer rod intercept time, suggesting to the authors that flow density might be used as a prognostic marker for AMD progression (Kar et al., 2024; Kar et al., 2023).

Because commercial OCT angiography is sensitive only over a window of velocities, loss of signal could represent loss of flow due to loss of vasculature or it could be demonstrating alterations in velocity of erythrocytes within intact vasculature. Histological images, while lacking dynamism of clinical studies and being limited to a single time point, are useful for understanding these relationships. Biesemeire and colleagues used light and electron microscopy of donor eyes and identified choriocapillaris loss in all stages of AMD prior to loss of RPE and neural retina, which could be observed by examining border areas between different stages of AMD within the same eye (Biesemeier et al., 2014). Work from the laboratory of Lutty and colleagues has greatly advanced this understanding (McLeod et al., 2009). For example, Seddon et al. studied donor eyes with clinically documented early AMD, intermediate AMD, GA, and macular neovascularization compared to aged control eyes; the choriocapillaris vascular density was decreased in all AMD stages, with the most loss observed in GA and neovascularization. The choriocapillaris in eyes with GA also had significantly lower diameters than control eyes (Seddon et al., 2016). Overall, eyes with neovascularization showed the greatest loss of choriocapillaris vasculature (Bhutto & Lutty, 2012), however vascular losses were noted in even early AMD on UEA-I-labeled wholemounts (Lutty et al., 2020).

Both the shape of ghost vessels and their frequent occupancy by leukocytes can lead to their misidentification as healthy capillaries, requiring high resolution imaging (e.g., transmission EM or at least 1μm sections) or labeling for viable endothelial cells (e.g., UEA-I, which is a more consistent marker than anti-CD34 in aging eyes (Sohn et al., 2014). Attempts to characterize choroidal vascularity without such a marker, for example studying H&E stained sections, can result in inconsistent results (Ramrattan et al., 1994; Spraul et al., 1996).

In the current study we quantified ghost vessels in a cohort of 99 donor maculae and found progressive degenerative changes in association with early/intermediate AMD and GA, specifically in areas with intact RPE cells. This finding, along with reduced normal vascular density in this cohort (Sohn et al., 2019), supports the notion that the loss of signal in OCT angiography is due to death of capillaries rather than altered flow parameters. Eyes with early/intermediate AMD had approximately 1.8x more ghost vessels per length than age matched controls, and eyes with geographic atrophy had approximately 5x more ghost vessels than controls. Although the RPE monolayer was intact in the quantified samples, it is possible that RPE molecular changes help drive this pathology through, for example decreased trophic support due to either impaired diffusion of growth factors across the lipid laden Bruch’s membrane or inherent changes in RPE physiology (Butler et al., 2021; La Cunza et al., 2021). It is apparent, however that choriocapillaris degeneration often precedes that seen in the RPE.

There are a number of polymorphisms associated with AMD, the strongest of which are in the CFH and ARMS2 genes. The CFH Y402H variant does not appear to affect the expression of the Factor H protein but it does alter its binding to the extracellular matrix in the choroid (Langford-Smith et al., 2014; Toomey et al., 2018). Eyes homozygous for the Y402H risk variant have increased choroidal deposition of C-reactive protein (Chirco et al., 2016; Johnson et al., 2006) and the membrane attack complex of complement (Mullins, Dewald, et al., 2011)). We have suggested that the membrane attack complex, which accumulates in intercapillary pillars and to a lesser degree on the choriocapillaris endothelium, is a strong candidate for vascular injury in AMD (reviewed in (Whitmore et al., 2015)). In this study we obtained genotypes for most of the measured samples and found that, when the entire cohort was assessed, homozygosity for the high risk variant (His/His) was associated with increased ghost vessel densities. In order to separate the effects of AMD from genotype, the analysis was performed on controls alone. His/His eyes showed a trend toward possessing more numerous ghost vessels, but this trend did not reach statistical significance.

Consequences of vascular loss to the RPE and outer retina may be profound. It is notable that oxygen delivered to the highly metabolically active retina is completely consumed by photoreceptor cells, especially in dark conditions in which rods are most active (Linsenmeier, 1986; Wangsa-Wirawan & Linsenmeier, 2003). Thus even a small decrease in perfusion could be very harmful to photoreceptor physiology. In this context it is especially interesting that delayed dark adaptation, a striking risk factor for the development of AMD (recently reviewed in (Curcio et al., 2024)), is altered in association with reduced capillary flow (Kar et al., 2024). In addition to oxygen, the choriocapillaris delivers glucose and other nutrients to the RPE and photoreceptor cells, which rely entirely on a healthy choriocapillaris for their metabolic needs.

Other consequences of vascular dropout include altered trafficking out of the choroid. Serial labeling of eyes with UEA-I to identify ghosts and Oil red O, Sudan black B, or filipin did not show consistent increases in lipid staining near areas of vascular loss, although the sensitivity of this detection would require a large and obvious change (data not shown).

The finding by us and others that vascular degeneration occurs in eyes with AMD in advance of RPE loss has therapeutic ramifications. If indeed choriocapillaris loss is primary in AMD, then addressing this injury while the RPE and photoreceptors are still intact could be a crucial early intervention. It is especially important in this case to understand the causes and mechanisms of choriocapillaris loss. Although RPE transplantation for geographic atrophy has been proposed, the loss of choriocapillaris vessels (even where the RPE is still intact) suggests that replacing the RPE alone may not be beneficial in many cases. The prospect of regenerating the choroid through iPSC derived endothelial cells (Mulfaul et al., 2020) or circulating endothelial progenitor cells (Bhatwadekar et al., 2009) is an exciting prospect for the treatment of AMD. In addition to replacing lost endothelial cells in cases of advanced disease, it is hoped that fortifying remaining endothelial cells against ongoing complement injury may arrest further retinal injury, and a more complete understanding of why capillaries in some individuals degenerate while others survive is an important area of study. Experiments to identify compounds that impact cell survival (Zeng et al., 2018) and to uncover the molecular basis of endothelial cell susceptibility (such as cellular senescence and membrane stiffening (Cabrera et al., 2016; Cabrera et al., 2022) can provide important translational insights into endothelial cell loss in AMD.

There are a number of limitations to this study. While the collection is well-powered, we had relatively few eyes with geographic atrophy. This is in part because eyes without intact RPE in the section were excluded to avoid measuring the secondary loss of choriocapillaris after loss of the overlying RPE (as shown by Korte and others (Korte et al., 1984)). Our AMD phenotyping was predominantly from clinical chart review. While we believe this is a robust “rule in” for the presence of AMD, donors were seen at multiple clinics by different doctors, and clinicians use varied terminology for AMD. Because of this, early and intermediate AMD (corresponding to AREDS grades 2 and 3 respectively) were by necessity grouped into a single category, while geographic atrophy eyes were their own category. Finally, eyes with macular neovascularization, which have been shown to exhibit considerable choriocapillaris degeneration (Bhutto & Lutty, 2012), were not included in this study.

5. Conclusion

In summary, we quantified ghost vessels in a cohort of human eyes and found that between normal aging and early/intermediate AMD, the choriocapillaris undergoes significant degeneration beneath an intact layer of RPE, which is much more pronounced in eyes with geographic atrophy. Complement complexes persist around ghost vessels, however they are much more robust around intact vasculature, which suggests an active process of complement activation requiring an intact vasculature. This may also suggest that the complement activating surfaces of the choriocapillaris require an intact cell. A genetic risk factor (Y402H of the CFH gene) was associated with more ghost vessels than in homozygous low risk individuals, but this association was significant only when eyes with AMD were included. Identifying the causes and consequences of vascular loss in AMD, and identifying methods for supporting the remaining endothelial cells and replacing those that have degenerated, is an important area of investigation for this disease.

HIGHLIGHTS.

Ghost vessels in the choriocapillaris are nonperfused lumens that persist after endothelial death

Density of ghost vessels was measured in a set of human donor maculae with early-intermediate age-related macular degeneration (AMD), advanced atrophic AMD (geographic atrophy), and age-matched controls using a marker of viable endothelial cells

Genotyping was also performed at the CFH Y402H locus

Advancing AMD and to a lesser extent genotype was associated with density of ghost vessels

Relevance to therapy is discussed

Acknowledgments

The authors gratefully acknowledge the eye donors and their families for their selfless contribution to science. We also thank the Iowa Lions Eye Bank for their ongoing support of our research program. This manuscript was written entirely by human beings and without the use of generative AI. Finally, we thank Dr. Jerry Lutty for his monumental role in advancing the study of the choroid. He was an inspiration, advocate, role model, supporter, and friend. While he is deeply missed, we strive to honor and carry on his passion for research and his kindness toward others.

Funding

Supported in part by NIH grants EY-024605, EY-025580 and the Sramek Charitable Trust.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Ambrose CT (2016). The Role of Capillaries in the Lesser Ailments of Old Age and in Alzheimer’s Disease and Vascular Dementia: The Potential of Pro-Therapeutic Angiogenesis. J Alzheimers Dis, 54(1), 31–43. 10.3233/JAD-160303 [DOI] [PubMed] [Google Scholar]
  2. Barthel LK, & Raymond PA (1990). Improved method for obtaining 3-microns cryosections for immunocytochemistry. J Histochem Cytochem, 38(9), 1383–1388. 10.1177/38.9.2201738 [DOI] [PubMed] [Google Scholar]
  3. Benedicto I, Lehmann GL, Ginsberg M, Nolan DJ, Bareja R, Elemento O, Salfati Z, Alam NM, Prusky GT, Llanos P, Rabbany SY, Maminishkis A, Miller SS, Rafii S, & Rodriguez-Boulan E (2017). Concerted regulation of retinal pigment epithelium basement membrane and barrier function by angiocrine factors. Nat Commun, 8, 15374. 10.1038/ncomms15374 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Berenberg TL, Metelitsina TI, Madow B, Dai Y, Ying GS, Dupont JC, Grunwald L, Brucker AJ, & Grunwald JE (2012). The association between drusen extent and foveolar choroidal blood flow in age-related macular degeneration. Retina, 32(1), 25–31. 10.1097/IAE.0b013e3182150483 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bhatwadekar AD, Glenn JV, Curtis TM, Grant MB, Stitt AW, & Gardiner TA (2009). Retinal endothelial cell apoptosis stimulates recruitment of endothelial progenitor cells. Invest Ophthalmol Vis Sci, 50(10), 4967–4973. 10.1167/iovs.09-3616 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bhutto I, & Lutty G (2012). Understanding age-related macular degeneration (AMD): relationships between the photoreceptor/retinal pigment epithelium/Bruch’s membrane/choriocapillaris complex. Mol Aspects Med, 33(4), 295–317. 10.1016/j.mam.2012.04.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bhutto IA, Baba T, Merges C, Juriasinghani V, McLeod DS, & Lutty GA (2011). C-reactive protein and complement factor H in aged human eyes and eyes with age-related macular degeneration. Br J Ophthalmol, 95(9), 1323–1330. 10.1136/bjo.2010.199216 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Biesemeier A, Taubitz T, Julien S, Yoeruek E, & Schraermeyer U (2014). Choriocapillaris breakdown precedes retinal degeneration in age-related macular degeneration. Neurobiol Aging, 35(11), 2562–2573. 10.1016/j.neurobiolaging.2014.05.003 [DOI] [PubMed] [Google Scholar]
  9. Butler JM, Supharattanasitthi W, Yang YC, & Paraoan L (2021). RNA-seq analysis of ageing human retinal pigment epithelium: Unexpected up-regulation of visual cycle gene transcription. J Cell Mol Med, 25(12), 5572–5585. 10.1111/jcmm.16569 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cabrera AP, Bhaskaran A, Xu J, Yang X, Scott HA, Mohideen U, & Ghosh K (2016). Senescence Increases Choroidal Endothelial Stiffness and Susceptibility to Complement Injury: Implications for Choriocapillaris Loss in AMD. Invest Ophthalmol Vis Sci, 57(14), 5910–5918. 10.1167/iovs.16-19727 [DOI] [PubMed] [Google Scholar]
  11. Cabrera AP, Stoddard J, Santiago Tierno I, Matisioudis N, Agarwal M, Renner L, Palegar N, Neuringer M, McGill T, & Ghosh K (2022). Increased cell stiffness contributes to complement-mediated injury of choroidal endothelial cells in a monkey model of early age-related macular degeneration. J Pathol, 257(3), 314–326. 10.1002/path.5892 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Chirco KR, Whitmore SS, Wang K, Potempa LA, Halder JA, Stone EM, Tucker BA, & Mullins RF (2016). Monomeric C-reactive protein and inflammation in age-related macular degeneration. J Pathol, 240(2), 173–183. 10.1002/path.4766 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Curcio CA, Kar D, Owsley C, Sloan KR, & Ach T (2024). Age-Related Macular Degeneration, a Mathematically Tractable Disease. Invest Ophthalmol Vis Sci, 65(3), 4. 10.1167/iovs.65.3.4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Curcio CA, Presley JB, Malek G, Medeiros NE, Avery DV, & Kruth HS (2005). Esterified and unesterified cholesterol in drusen and basal deposits of eyes with age-related maculopathy. Exp Eye Res, 81(6), 731–741. 10.1016/j.exer.2005.04.012 [DOI] [PubMed] [Google Scholar]
  15. Garner A (1993). Histopathology of diabetic retinopathy in man. Eye (Lond), 7 (Pt 2), 250–253. 10.1038/eye.1993.58 [DOI] [PubMed] [Google Scholar]
  16. Glendenning LM, Reynero KM, Kukan EN, Long MD, & Cobb BA (2023). IgG sialylation occurs via the FcRn-mediated recycling pathway in endothelial cells. bioRxiv. 10.1101/2023.06.30.547255 [DOI] [Google Scholar]
  17. Goverdhan SV, Howell MW, Mullins RF, Osmond C, Hodgkins PR, Self J, Avery K, & Lotery AJ (2005). Association of HLA class I and class II polymorphisms with age-related macular degeneration. Invest Ophthalmol Vis Sci, 46(5), 1726–1734. 10.1167/iovs.04-0928 [DOI] [PubMed] [Google Scholar]
  18. Grebe R, Mughal I, Bryden W, McLeod S, Edwards M, Hageman GS, & Lutty G (2019). Ultrastructural analysis of submacular choriocapillaris and its transport systems in AMD and aged control eyes. Exp Eye Res, 181, 252–262. 10.1016/j.exer.2019.02.018 [DOI] [PubMed] [Google Scholar]
  19. Grunwald JE, Hariprasad SM, DuPont J, Maguire MG, Fine SL, Brucker AJ, Maguire AM, & Ho AC (1998). Foveolar choroidal blood flow in age-related macular degeneration. Invest Ophthalmol Vis Sci, 39(2), 385–390. https://www.ncbi.nlm.nih.gov/pubmed/9477998 [PubMed] [Google Scholar]
  20. Handa JT, Verzijl N, Matsunaga H, Aotaki-Keen A, Lutty GA, te Koppele JM, Miyata T, & Hjelmeland LM (1999). Increase in the advanced glycation end product pentosidine in Bruch’s membrane with age. Invest Ophthalmol Vis Sci, 40(3), 775–779. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=10067983 [PubMed] [Google Scholar]
  21. Hayreh SS (1990). In vivo choroidal circulation and its watershed zones. Eye (Lond), 4 (Pt 2), 273–289. 10.1038/eye.1990.39 [DOI] [PubMed] [Google Scholar]
  22. Hayreh SS (2010). Submacular choroidal vascular bed watershed zones and their clinical importance. Am J Ophthalmol, 150(6), 940–941; author reply 941–942. 10.1016/j.ajo.2010.08.011 [DOI] [PubMed] [Google Scholar]
  23. Johnson PT, Betts KE, Radeke MJ, Hageman GS, Anderson DH, & Johnson LV (2006). Individuals homozygous for the age-related macular degeneration risk-conferring variant of complement factor H have elevated levels of CRP in the choroid. Proc Natl Acad Sci U S A, 103(46), 17456–17461. 10.1073/pnas.0606234103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kar D, Amjad M, Corradetti G, Swain TA, Clark ME, McGwin G Jr., Sloan KR, Owsley C, Sadda SR, & Curcio CA (2024). Choriocapillaris Impairment, Visual Function, and Distance to Fovea in Aging and Age-Related Macular Degeneration: ALSTAR2 Baseline. Invest Ophthalmol Vis Sci, 65(8), 40. 10.1167/iovs.65.8.40 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kar D, Corradetti G, Swain TA, Clark ME, McGwin G Jr., Owsley C, Sadda SR, & Curcio CA (2023). Choriocapillaris Impairment Is Associated With Delayed Rod-Mediated Dark Adaptation in Age-Related Macular Degeneration. Invest Ophthalmol Vis Sci, 64(12), 41. 10.1167/iovs.64.12.41 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Korte GE, Reppucci V, & Henkind P (1984). RPE destruction causes choriocapillary atrophy. Invest Ophthalmol Vis Sci, 25(10), 1135–1145. https://www.ncbi.nlm.nih.gov/pubmed/6480292 [PubMed] [Google Scholar]
  27. La Cunza N, Tan LX, Thamban T, Germer CJ, Rathnasamy G, Toops KA, & Lakkaraju A (2021). Mitochondria-dependent phase separation of disease-relevant proteins drives pathological features of age-related macular degeneration. JCI Insight, 6(9). 10.1172/jci.insight.142254 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Langford-Smith A, Keenan TD, Clark SJ, Bishop PN, & Day AJ (2014). The role of complement in age-related macular degeneration: heparan sulphate, a ZIP code for complement factor H? J Innate Immun, 6(4), 407–416. 10.1159/000356513 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Li M, Huisingh C, Messinger J, Dolz-Marco R, Ferrara D, Freund KB, & Curcio CA (2018). HISTOLOGY OF GEOGRAPHIC ATROPHY SECONDARY TO AGE-RELATED MACULAR DEGENERATION: A Multilayer Approach. Retina, 38(10), 1937–1953. 10.1097/IAE.0000000000002182 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Linsenmeier RA (1986). Effects of light and darkness on oxygen distribution and consumption in the cat retina. J Gen Physiol, 88(4), 521–542. 10.1085/jgp.88.4.521 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Linsenmeier RA, & Padnick-Silver L (2000). Metabolic dependence of photoreceptors on the choroid in the normal and detached retina. Invest Ophthalmol Vis Sci, 41(10), 3117–3123. https://www.ncbi.nlm.nih.gov/pubmed/10967072 [PubMed] [Google Scholar]
  32. Love S, & Miners JS (2016). Cerebrovascular disease in ageing and Alzheimer’s disease. Acta Neuropathol, 131(5), 645–658. 10.1007/s00401-015-1522-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Lutty GA, McLeod DS, Bhutto IA, Edwards MM, & Seddon JM (2020). Choriocapillaris dropout in early age-related macular degeneration. Exp Eye Res, 192, 107939. 10.1016/j.exer.2020.107939 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Marsh-Armstrong B, Migacz J, Jonnal R, & Werner JS (2019). Automated quantification of choriocapillaris anatomical features in ultrahigh-speed optical coherence tomography angiograms. Biomed Opt Express, 10(10), 5337–5350. 10.1364/BOE.10.005337 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. McLeod DS, Grebe R, Bhutto I, Merges C, Baba T, & Lutty GA (2009). Relationship between RPE and choriocapillaris in age-related macular degeneration. Invest Ophthalmol Vis Sci, 50(10), 4982–4991. 10.1167/iovs.09-3639 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. McLeod DS, Lefer DJ, Merges C, & Lutty GA (1995). Enhanced expression of intracellular adhesion molecule-1 and P-selectin in the diabetic human retina and choroid. Am J Pathol, 147(3), 642–653. https://www.ncbi.nlm.nih.gov/pubmed/7545873 [PMC free article] [PubMed] [Google Scholar]
  37. Molins B, Fuentes-Prior P, Adan A, Anton R, Arostegui JI, Yague J, & Dick AD (2016). Complement factor H binding of monomeric C-reactive protein downregulates proinflammatory activity and is impaired with at risk polymorphic CFH variants. Sci Rep, 6, 22889. 10.1038/srep22889 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Moult EM, Waheed NK, Novais EA, Choi W, Lee B, Ploner SB, Cole ED, Louzada RN, Lu CD, Rosenfeld PJ, Duker JS, & Fujimoto JG (2016). Swept-Source Optical Coherence Tomography Angiography Reveals Choriocapillaris Alterations in Eyes with Nascent Geographic Atrophy and Drusen-Associated Geographic Atrophy. Retina, 36 Suppl 1(Suppl 1), S2–S11. 10.1097/IAE.0000000000001287 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Mulfaul K, Giacalone JC, Voigt AP, Riker MJ, Ochoa D, Han IC, Stone EM, Mullins RF, & Tucker BA (2020). Stepwise differentiation and functional characterization of human induced pluripotent stem cell-derived choroidal endothelial cells. Stem Cell Res Ther, 11(1), 409. 10.1186/s13287-020-01903-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Mullins RF, Dewald AD, Streb LM, Wang K, Kuehn MH, & Stone EM (2011). Elevated membrane attack complex in human choroid with high risk complement factor H genotypes. Exp Eye Res, 93(4), 565–567. 10.1016/j.exer.2011.06.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Mullins RF, Johnson MN, Faidley EA, Skeie JM, & Huang J (2011). Choriocapillaris vascular dropout related to density of drusen in human eyes with early age-related macular degeneration. Invest Ophthalmol Vis Sci, 52(3), 1606–1612. 10.1167/iovs.10-6476 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Mullins RF, Kuehn MH, Faidley EA, Syed NA, & Stone EM (2007). Differential macular and peripheral expression of bestrophin in human eyes and its implication for best disease. Invest Ophthalmol Vis Sci, 48(7), 3372–3380. 10.1167/iovs.06-0868 [DOI] [PubMed] [Google Scholar]
  43. Mullins RF, Schoo DP, Sohn EH, Flamme-Wiese MJ, Workamelahu G, Johnston RM, Wang K, Tucker BA, & Stone EM (2014). The membrane attack complex in aging human choriocapillaris: relationship to macular degeneration and choroidal thinning. Am J Pathol, 184(11), 3142–3153. 10.1016/j.ajpath.2014.07.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Nakanishi M, Grebe R, Bhutto IA, Edwards M, McLeod DS, & Lutty GA (2016). Albumen Transport to Bruch’s Membrane and RPE by Choriocapillaris Caveolae. Invest Ophthalmol Vis Sci, 57(4), 2213–2224. 10.1167/iovs.15-17934 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Pauleikhoff D, Chen JC, Chisholm IH, & Bird AC (1990). Choroidal perfusion abnormality with age-related Bruch’s membrane change. Am J Ophthalmol, 109(2), 211–217. 10.1016/s0002-9394(14)75989-6 [DOI] [PubMed] [Google Scholar]
  46. Pauleikhoff D, Spital G, Radermacher M, Brumm GA, Lommatzsch A, & Bird AC (1999). A fluorescein and indocyanine green angiographic study of choriocapillaris in age-related macular disease. Arch Ophthalmol, 117(10), 1353–1358. 10.1001/archopht.117.10.1353 [DOI] [PubMed] [Google Scholar]
  47. Ramrattan RS, van der Schaft TL, Mooy CM, de Bruijn WC, Mulder PG, & de Jong PT (1994). Morphometric analysis of Bruch’s membrane, the choriocapillaris, and the choroid in aging. Invest Ophthalmol Vis Sci, 35(6), 2857–2864. https://www.ncbi.nlm.nih.gov/pubmed/8188481 [PubMed] [Google Scholar]
  48. Seddon JM, McLeod DS, Bhutto IA, Villalonga MB, Silver RE, Wenick AS, Edwards MM, & Lutty GA (2016). Histopathological Insights Into Choroidal Vascular Loss in Clinically Documented Cases of Age-Related Macular Degeneration. JAMA Ophthalmol, 134(11), 1272–1280. 10.1001/jamaophthalmol.2016.3519 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Sohn EH, Flamme-Wiese MJ, Whitmore SS, Wang K, Tucker BA, & Mullins RF (2014). Loss of CD34 expression in aging human choriocapillaris endothelial cells. PLoS One, 9(1), e86538. 10.1371/journal.pone.0086538 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Sohn EH, Flamme-Wiese MJ, Whitmore SS, Workalemahu G, Marneros AG, Boese EA, Kwon YH, Wang K, Abramoff MD, Tucker BA, Stone EM, & Mullins RF (2019). Choriocapillaris Degeneration in Geographic Atrophy. Am J Pathol, 189(7), 1473–1480. 10.1016/j.ajpath.2019.04.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Spraul CW, Lang GE, & Grossniklaus HE (1996). Morphometric analysis of the choroid, Bruch’s membrane, and retinal pigment epithelium in eyes with age-related macular degeneration. Invest Ophthalmol Vis Sci, 37(13), 2724–2735. https://www.ncbi.nlm.nih.gov/pubmed/8977488 [PubMed] [Google Scholar]
  52. Tamari T, Kawar-Jaraisy R, Doppelt O, Giladi B, Sabbah N, & Zigdon-Giladi H (2020). The Paracrine Role of Endothelial Cells in Bone Formation via CXCR4/SDF-1 Pathway. Cells, 9(6). 10.3390/cells9061325 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Toomey CB, Johnson LV, & Bowes Rickman C (2018). Complement factor H in AMD: Bridging genetic associations and pathobiology. Prog Retin Eye Res, 62, 38–57. 10.1016/j.preteyeres.2017.09.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Tuckermann J, & Adams RH (2021). The endothelium-bone axis in development, homeostasis and bone and joint disease. Nat Rev Rheumatol, 17(10), 608–620. 10.1038/s41584-021-00682-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Uno K, Bhutto IA, McLeod DS, Merges C, & Lutty GA (2006). Impaired expression of thrombospondin-1 in eyes with age related macular degeneration. Br J Ophthalmol, 90(1), 48–54. 10.1136/bjo.2005.074005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Voigt AP, Whitmore SS, Mulfaul K, Chirco KR, Giacalone JC, Flamme-Wiese MJ, Stockman A, Stone EM, Tucker BA, Scheetz TE, & Mullins RF (2020). Bulk and single-cell gene expression analyses reveal aging human choriocapillaris has proinflammatory phenotype. Microvasc Res, 131, 104031. 10.1016/j.mvr.2020.104031 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Wangsa-Wirawan ND, & Linsenmeier RA (2003). Retinal oxygen: fundamental and clinical aspects. Arch Ophthalmol, 121(4), 547–557. 10.1001/archopht.121.4.547 [DOI] [PubMed] [Google Scholar]
  58. Whitmore SS, Braun TA, Skeie JM, Haas CM, Sohn EH, Stone EM, Scheetz TE, & Mullins RF (2013). Altered gene expression in dry age-related macular degeneration suggests early loss of choroidal endothelial cells. Mol Vis, 19, 2274–2297. https://www.ncbi.nlm.nih.gov/pubmed/24265543 [PMC free article] [PubMed] [Google Scholar]
  59. Whitmore SS, Sohn EH, Chirco KR, Drack AV, Stone EM, Tucker BA, & Mullins RF (2015). Complement activation and choriocapillaris loss in early AMD: implications for pathophysiology and therapy. Prog Retin Eye Res, 45, 1–29. 10.1016/j.preteyeres.2014.11.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Yuan X, Gu X, Crabb JS, Yue X, Shadrach K, Hollyfield JG, & Crabb JW (2010). Quantitative proteomics: comparison of the macular Bruch membrane/choroid complex from age-related macular degeneration and normal eyes. Mol Cell Proteomics, 9(6), 1031–1046. 10.1074/mcp.M900523-MCP200 [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Zeng S, Wen KK, Workalemahu G, Sohn EH, Wu M, Chirco KR, Flamme-Wiese MJ, Liu X, Stone EM, Tucker BA, & Mullins RF (2018). Imidazole Compounds for Protecting Choroidal Endothelial Cells from Complement Injury. Sci Rep, 8(1), 13387. 10.1038/s41598-018-31846-z [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES