Retinal Pigment Epithelium: The Eye’s First Line of Defense Against Macular Degeneration
BrightFocus-funded Work at ARVO 2016
This research was supported by BrightFocus
Research suggests that the retinal pigment epithelium (RPE) is where macular degeneration begins. This pigmented layer of cells next to the retina serves as a pass-through between the light-sensitive photoreceptors of the retina and a layer of blood vessels, called the choroid, lying below.
The RPE’s role is to nourish the fragile nerve tissue of the retina and maintain its health by transporting molecules in and out, getting rid of dead cells, secreting hormones, modulating immune factors, and more. All this has magnified the RPE’s role in macular degeneration.
Things can go wrong when the RPE succumbs to chronic “sterile” inflammation of uncertain origin. That confuses RPE cells and sometimes impairs their ability to clear away debris that includes toxic inflammatory enzymes and the cell’s own damaged parts. This breakdown can lead to further inflammation and eventual cell death.
The triggers of this “inflammatory cascade” are still being explored at the cellular level. They include factors such as genetic susceptibility; age (the biggest risk factor for age-related macular degeneration, or AMD); and direct stressors on RPE cells, including prolonged exposure to environmental toxins like sun, chemicals, tobacco smoke, and other pollution.
The response of the RPE to all these factors seems to be a tipping point that can either defend against or lead the eye toward macular degeneration. Several BrightFocus projects are investigating this critical part of the eye.
What Causes the RPE to Stop Taking Out the Trash?
Aparna Lakkaraju, PhD, of the University of Wisconsin, is trying to understand how defective recycling of proteins and lipids leads to the accumulation of debris or “garbage” in the RPE — a characteristic sign of early AMD. With age, this causes drusen to form above and beneath the RPE and conspires with environmental and genetic factors to damage the retina.
With her 2015-17 BrightFocus grant, Lakkaraju and her and her co-principal investigator (PI), Kimberly A. Toops, PhD, have been tracing the complex cell biological processes that give rise to drusen and how they impact normal RPE functions. At ARVO 2016, she presented evidence that vitamin A dimers, which are by-products of the visual cycle, form aggregates in the RPE that also trap cholesterol. Vitamin A, while a vital eye nutrient, sometimes undergoes a chemical transformation in the eye causes it to form clumps. This leads to insoluble deposits that may accumulate in the eye over decades.
Lakkaraju and team showed that in inherited and age-related macular degenerations, these vitamin A dimers and the excess cholesterol that accumulates, disrupt the function of compartments in the cell responsible for recycling and garbage removal and initiate pro-inflammatory cross-talk in the retina (Abstract No. 1405). In another presentation, she identified FDA-approved drugs that can help strengthen innate mechanisms that are essential for preserving RPE health and function (Abstract No. 5783).
Are Other Cells Recruited to Participate in the RPE’s Downfall?
Two grantees at Johns Hopkins’ Wilmer Eye Institute are trying to trace the path to RPE dysfunction.
Debasish Sinha, PhD (2014-16), seeks to further our understanding of how impaired clearance in the RPE can be an early trigger of AMD that activates the immune system. “Clearance” refers to autophagy and phagocytosis, the primary means cells have to discard and remove unwanted substances.
At ARVO, Sinha and colleagues presented compelling evidence showing that inflammation is a necessary preface to macular degeneration in the retina (Abstract No. 483). In mice models of AMD, there is increased RNA expression of numerous inflammatory factors and a host of other changes associated with a pro-inflammatory environment. These included an increase in toll-like receptor 4 pathway proteins and increased phosphorylation of serine-threonine protein kinase AKT1 in the retinas of AMD mice.
Phosphorylation refers to the addition of a phosphoryl group to a molecule and is one way protein enzymes are turned “on” and “off.” The authors suggest that another protein, βA3/A1-crystallin, may regulate AKT2 phosphorylation, thereby contributing to the onset of a pro-inflammatory condition.
Crystallins are abundant proteins in the ocular lens that determine the transparency and refractivity required for lens function. While initially thought to have evolved solely for that purpose, they are also found outside of the lens, where they may play other roles, including to acidify the lysosomes involved in the breakdown of cellular “garbage.”
In another ARVO poster (Abstract No. 5006), Sinha provided novel evidence that βA3/A1-crystallin is essential for regulating clearance mechanisms (eg, phagocytosis and autophagy) that help maintain a healthy RPE. This multi-purpose role of βA3/A1-crystallin also was laid out in a paper of Sinha’s last year, one that draws from his BrightFocus-funded work (Zigler JS and Sinha D, Prog Retin Eye Res, January 2015).
Dr. Sinha is the 2014 recipient of the Carolyn K. McGillvray Award for Macular Degeneration Research.
Also at Wilmer, Imran Bhutto, MD, PhD, (2014-16) and co-PI Gerard Lutty, PhD, are studying a different type of inflammatory cells, called mast cells, that resides below the RPE and may contribute to its decline. Mast cells are found in the choroid, a layer of blood vessels and connective tissue that lies next to the RPE and is the only source of oxygen and nutrition for both it and the photoreceptor cells it helps maintain.
This project is unique in that the association of choroidal mast cells with pathological changes in AMD has not previously been studied. At ARVO, Bhutto reported that mast cells, when activated, release granules rich in proinflammatory factors and proteases into the surrounding area (Abstract No. 1407). Their investigation showed that the number of mast cells was significantly increased in all choroidal areas in early AMD, and that this discharge of proinflammatory granules by mast cells may contribute to AMD pathogenesis. Drugs already exist to control mast cell activation, so this study could build interest in this new treatment approach.
In addition to being presented at ARVO, these findings which relate to the authors’ BrightFocus-funded work were published in the British Journal of Ophthalmology in March (Bhutto et al, 2016).
Nature’s Way of Muting the Immune Response
The resident immune cells of the retina, called microglia, act to clear the tissue of cellular debris by a process called phagocytosis (cell devouring).Their job is to routinely monitor tissue health and integrity, and if they find damage, they actively migrate to the outer retina, where they can cause a pronounced inflammatory response.
Notably, both the RPE and retinal microglia also express surface proteins, called Mertk, which in other immune cells have a modulating effect of either increasing or decreasing inflammation. Debra Thompson, PhD, of the University of Michigan (2014-16), with co-PI, Steven Abcouwer, PhD, are studying whether Mertk also acts to blunt inflammatory responses in the retina, thus helping to prevent damage from long-term, low-grade inflammation such as that which damages the retina in AMD. Thompson is a member of the BrightFocus MDR Scientific Review Committee.
For their study, they engineered a new microglial cell line that exhibits properties of native microglia, including the ability to phagocytose outer segments shed from photoreceptor cells. This shedding is normal; photoreceptors have an inner segment that contains the cell’s metabolic machinery, and an outer segment consisting of membrane-like discs that are densely packed with the light-sensitive pigment, rhodopsin. Rhodopsin permits us to see in low-light conditions but quickly photobleaches when it becomes exposed to light. These rhodopsin-filled discs of the OS are completely replaced once every ten days, and this continuous renewal of the light-sensitive machinery of the eye continues throughout our lifetime.
Thompson et al hypothesized that when OS phagocytosis was mediated by Merck, the inflammatory response of microglia would be self-limiting. And indeed, their experiments showed that Mertk effectively blocked an inflammatory response (Abstract No. 2234) . These results lend hope that it might be possible to use Mertk, or a biologic mediator like it, to calm or repair a pathologic inflammatory response in AMD.
The Gatekeeper’s Role in Sterile Inflammation
Typically, inflammatory responses of the immune system, such as that seen in AMD, are caused by infection. However, in chronic conditions, a form of “sterile” inflammation can exist in areas of the body, even in the absence of infection, due to other types of damage. This type of “sterile” inflammation is central to the progression of AMD.
Sarah Doyle, PhD, of Trinity College, Dublin, and her co-PIs there, Matthew Campbell, PhD, and Luke O’Neill, PhD, have hypothesized that this reaction is due to uncontrolled activation of immune system sensors known as toll-like receptors (TLR). Their investigation of this pathway could lead to the targeted development of new AMD treatments. Doyle has been honored with a fellowship from the ARVO Foundation to continue work started under her 2014-16 BrightFocus grant.
Toll-like receptors (TLR), a relatively new discovery in mammals, are a class of proteins that play a key role in the innate immune system. They have a special talent for recognizing patterns in foreign microbes, and thus have earned the distinction of being “sentries” of the immune system, capable of activating a host defense. So far, at least 13 different TLR types have been identified in humans. TLRs may also play a role in Alzheimer’s disease, and BrightFocus is funding research in that area.
Typically, TLRs are found on white blood cells (leukocytes), which include monocytes, macrophages, and other killer cells that serve as the standing army of the immune system. When activated, TLRs recruit other proteins within these immune cells to amplify molecular signaling. This hyped-up signaling leads to the upregulation or suppression of genes that orchestrate inflammatory responses. The response can range from adaptive immunity, to clearance of damaged cells (eg, phagocytosis), or even programmed cell death.
Doyle and her colleagues believe that in early AMD, TLRs are activated in response not to infection, but to sterile danger associated molecular patterns (DAMPs) found in the degenerating retina. DAMPs, they believe, activate TLR signaling pathways to induce an inflammatory trigger known as complement factor 3 (C3), which is well-known in AMD. Despite the fact that both the complement cascade and TLRs are critical components of the innate immune response, little is known about the interaction between these two pathways.
At ARVO, Doyle and Kelly Mulfaul, a PhD student in her lab, presented their newest findings (Abstract No. 5010). They’ve learned that C3 is significantly upregulated in the RPE in response to several types of TLRs. They hypothesize that the degenerating retina provokes a strong C3 in response to TLR activation, and that the large volume of C3 secreted in response may potentially compromise the integrity of the RPE outer blood-retina barrier, contributing to a “wet AMD” response.
If the blood-retina barrier is breached in AMD in response to TLR activation of the complement cascade, Doyle benefits from having Matt Campbell, her spouse, on the grant as her co-PI. He specializes in studying barriers between blood and nervous system tissue, and currently has a BrightFocus grant to study the blood-brain barrier (BBB) in Alzheimer’s disease. His goal in that research is to see whether the BBB can be manipulated to improve clearance of toxic amyloid beta. Read more about the BrightFocus-funded research of this husband-wife team.
Pursuing an Enzyme that Protects Against Oxidation
Hongli Wu, PhD, a 2015-17 BrightFocus grantee at the University of North Texas Health Science Center in Fort Worth, is investigating an enzyme that protects RPE cells from oxidation, a form of wear and tear that contributes to macular degeneration. Oxygen has a tendency to combine with other molecules; however, as part of that process, other molecules typically give up an electron when combining with oxygen, and become destabilized. In the material world, burning is an example of rapid oxidation, whereas corrosion and rust seen on metals and other substances are examples of slow oxidation.
In the human body, oxidation creates unstable molecules known as “free radicals,” which try to recover their missing electrons by binding to other molecules. This all occurs naturally with aging, but can be accelerated by stress, cigarette smoking, alcohol, sunlight, pollution and other factors. And while the body is able to neutralize many free radicals, exposure to high amounts can damage cells over time.
Dr. Wu describes her research as an attempt to understand “how oxidative stress defense agents and enzymes work” in order to be able to harness their protective powers in potential therapies for AMD. She is particularly interested in glutaredoxin 2 (Grx2), an enzyme discovered in mitochondrial cells that protects cells from oxidative damage. Her BrightFocus-funded project is to learn whether and how Grx2 might also protect RPE cells facing similar threats.
At ARVO she reported on early experiments using cells drawn from mice engineered not to express Grx2 (the protective enzyme). Her results showed that Grx2, when administered to those cells lacking it, rescued them from lethal oxidative damage, possibly through more than one pathway. First, her results showed that RPE cells lacking Grx2 were more sensitive to oxidative damage, whereas Grx2 treated cells were to a higher degree spared from oxidation-induced damage and death. Also, in cells reinforced with Grx2 , there were inhibited levels of a key protein linked to autophagy, a self-induced form of cell death that can be triggered by oxidation and other causes. (Abstracts No. 248 and 5800)
Is AMD a Lipid Disorder?
The fact that drusen, the tell-tale early signs of AMD found in the RPE , are composed largely of cholesterol has led some to speculate that AMD’s proinflammatory environment may in fact be linked to a breakdown in the RPE’s ability to metabolize and clear lipids, resulting in a sort of atherosclerosis of the eye. Jing Chen, PhD, of Boston Children’s Hospital, Harvard, used her 2013-15 BrightFocus grant to study the links between lipid metabolism, inflammation, and immune functions.
In preliminary studies, Chen and collaborators were able to link genetic changes to a nuclear receptor called RORα (retinoic acid receptor-related orphan receptor alpha) with the risk of developing wet AMD. RORα influences circadian rhythms in the liver, kidney, retina, and lung, and also works to regulate lipid metabolism and immunity in humans.
In her BrightFocus work, Chen and her team set out to determine whether RORα, as a sensor of lipid homeostasis, influences the development of wet AMD by causing changes in tissue inflammation. For her experiments, she used a mouse model of oxygen-induced retinopathy (OIR) with pathologic neovascularization, ie, the type of leaky blood vessels that mimics the abnormal vessels as seen in wet AMD.At ARVO, she reported that RORα was significantly increased in inflamed mouse retinas, and that conversely, when RORα expression was suppressed using either gene therapy or drugs, inflammatory factors decreased and pathologic neovascularization was suppressed. RORα may represent a novel druggable target for treating retinopathy, Chen and coauthors conclude (Abstract No. 3630).
[On a related note, a similar inquiry was undertaken in a 2010-12 BrightFocus grant to Rajendra Apte, MD, PhD, of Washington University of St. Louis. His project explored the possibility that cholesterol in drusen causes dysfunctional activation of macrophages and influences the progression to wet AMD. Results were reported in Cell Metabolism (Sene et al, 2013).]
A few years ago, in her BrightFocus grant profile, Chen stated that “a more integrated treatment approach that corrects more than one of the mechanisms that can lead to AMD, such as problems with inflammation and lipid metabolism, would be of great benefit.” Studying that might “provide clues to a new therapeutic approach to treat this blinding eye disease,” she added.
Recently she brought that idea a step closer to reality. The results of her BrightFocus-funded project, and other research, were used to compose, with dozens of expert collaborators, a well-documented, far-reaching argument in support of the still-novel idea that AMD is, in part, a lipid disorder. It advances the hypothesis that dysregulated lipid and glucose photoreceptor energy metabolism may be a driving force in AMD and other forms of retinal disease. Their lengthy “letter” was published in Nature Medicine earlier this spring (Joyal et al, 2016).
Hydroxyapatite: The Hard Core at the Center of Drusen
If lipids represent the “soft” side of drusen, there’s a harder side. Literally.
The research team of Richard Thompson, PhD, University of Maryland, and Imre Lengyel, PhD, University College London, discovered a microscopic mineral scaffolding around which drusen may begin to form in the RPE.
Before now, no one’s fully understood how drusen formed and grew. Thompson’s 2014-16 MDR grant (with Lengyel as co-PI) has enabled them to explore how mineralized calcium phosphate, called hydroxyapatite (HAP), may lead this process.
HAP is common in the body, and forms the hard part of bones and teeth, but it had never been identified in the eye in this form before. Thompson, Lengyel, et al, believe these mineralized spheres attract proteins and fats to their surface. It starts when an insoluble shell made of HAP forms around naturally occurring lipid droplets; then protein and fat molecules form around that. Over years, with an age- or disease related slowdown in the RPE’s ability to get rid cellular debris, the globules build into drusen.
Their discovery was published last year in the Proceedings of the National Academy of Sciences (PNAS).
At ARVO, Thompson reported their newest results comparing HAP deposition in human and macaque retinas. Molecular events involved in deposit formation in these nonhuman primates and human eyes are similar, but compared with human tissues, the macaque tissues generally had smaller and less numerous HAP depositions, with fewer hollow spherules in sub-RPE deposits. However the pattern of distribution did not differ from those in human tissues. (Abstract No. 5799)
Up ahead, the work could have a major impact. To quote Lengyel from a UCL news release: “The fluorescent labelling technique that we used can identify the early signs of drusen build-up long before they become visible with current methods. If we could develop a safe way of getting these dyes into the eye, we could advance AMD diagnoses by a decade or more, and could follow early progression more precisely.”
It’s even possible that further strategies to prevent drusen build-up could potentially stop AMD from developing altogether.
“Our discovery opens up an exciting new avenue of scientific research into potential new diagnostics and treatments, but this is only the beginning of a long road,” Thompson said.
Read more about this exciting work.