Lipofuscin and the complement system in retinal pigment epithelial cell biology

Age-related macular degeneration (AMD) is the leading cause of blindness in the elderly in developed countries. The disease is characterised by progressive degeneration of retinal pigment epithelial (RPE) cells and photoreceptors, leading to vision-threatening complications. Genetic studies a decade ago revealed that variants in a number of genes involved in the complement system are associated with increased susceptibility to AMD. The complement system is part of our immune system, and while it normally plays a vital role in defence against infection, it is now clear that defects in this system can lead to the development of AMD. In normal healthy individuals the complement system works by assembling a number of complement proteins into a complex known as C5b-9, on the surface of the target cells such as bacteria. The C5b-9 complex forms a pore that helps to destroy the target cell. However, over-activation of the system can result in C5b-9 forming on our own cells, leading to changes in cell function and ultimately cell death. RPE cells have evolved mechanisms to protect themselves from complement attack, such as the internalization and destruction of C5b-9, but with advancing age cells appear to be more susceptible to complement attack. The hypothesis to be tested in this proposal is that this increased susceptibility is due, at least in part, to lipofuscin accumulation.

The slow accumulation of lipofuscin within RPE cells is a feature of normal physiological ageing but high levels are strongly correlated with AMD. Lipofuscin is a complex mixture of oxidised lipids, proteins and autofluorescent material mainly derived from breakdown products of the visual cycle and photoreceptor outer segments (POS). The tips of the POS detach from photoreceptors every day, to be engulfed and digested by the adjacent RPE cells. Waste products are disposed of, but useful POS constituents are recycled back to the photoreceptors. POS are rich in polyunsaturated fatty acids, which are highly susceptible to oxidation, for example by exposure to ultraviolet (UV) light, and changes in these lipids may lead to their incomplete digestion by the RPE cells. This would favour the accumulation of insoluble deposits such as lipofuscin within the cell. To test our hypothesis we will establish an AMD model that combines lipofuscin accumulation and complement attack, in order to gain insight into the cellular mechanisms of RPE dysfunction in AMD. In preliminary work we have shown that POS may be exposed to an ultraviolet (UV) light source to alter their structure. Here, cultured RPE cells will be fed with UV-irradiated POS and then treated with complement proteins to induce C5b-9 formation. Biochemical tests on these cells will reveal whether this combination of insults leads to changes similar to those associated with AMD. We will also establish an in vivo model that mimics the features of AMD by crossing a mouse that accumulates lipofuscin with another that has increased complement activation in the retina. We will analyse whether these mice exhibit the signs of inflammation and/or abnormal blood vessel growth that are seen in AMD. In addition we will use ocular imaging techniques to study subretinal deposits, and light and electron microscopy to investigate the integrity of subcellular structures. These investigations will provide insight into the underlying mechanism of how lipofuscin-loaded RPE cells react to increased complement activation, and whether lipofuscin-free cells are better able to defend themselves.

The results of our studies will help to explain some of the underlying mechanisms involved in the pathogenesis of AMD, a key step toward the development of more effective and better targeted therapeutics.

Primary porcine RPE cells will be grown as monolayers in transwells in order to mimic their topography in the retina and permit manipulations at the basal and apical surfaces. Monolayers will be induced to accumulate lipofuscin by being cultured with UV-irradiated photoreceptor outer segments (POS) at the apical side, while being treated with purified complement proteins in the basal chamber. Lipofuscin accumulation within the cells will be quantified by autofluorescence measured with a plate reader or flow cytometry. C5b-9 assembly on the basal cell surface will be measured by immunofluorescence and confocal microscopy. To investigate whether C5b-9 formation on lipofuscin-loaded RPE cells stimulates a pro-inflammatory and/or pro-angiogenic environment as seen in AMD pathogenesis, nuclear translocation of NF-kappaB will be examined using immunofluorescence, inflammasome activation will be analysed by quantifying caspase-1 and IL-1beta and IL-18, zymography will be used to analyse the expression of MMP-2 and -9, and the VEGF-PEDF ratio will be determined via ELISA.

We will also investigate RPE properties and functions in a mutant mouse containing a targeted deletion of complement factor H in the RPE, crossed with the ABCA4/RDH8 double KO mouse. The retinal phenotype will be examined by fundus photography, optical coherence tomography, electroretinography and fluorescein angiography (all in the living mouse eye), and post mortem immunostaining of retinal sections and flatmounts to investigate RPE ultrastructure, phagosome maturation, complement activation and lipofuscin accumulation. To characterise tissue histology such as thickness of the outer nuclear layer, hematoxylin and eosin staining will be performed on paraffin-embedded sections. Periodic acid Schiff staining will serve for visualisation of mucopolysaccharides and glycogen, Sudan black will be used to detect lipids, and immunostaining will be used to identify Amyloid-beta. Sections will also be stained for vitronectin, since vitronectin accumulates in subretinal desposits in AMD and may be involved in complement-mediated immune responses.

Taken together, this programme of in vitro and in vivo studies will provide new insight into the interplay between the complement system and lipofuscin accumulation in the pathogenesis of AMD.

1. Who will benefit from this research?
The outcomes of this project will benefit i) other workers investigating the molecular mechanisms of AMD pathogenesis, ii) investigators studying innate immunity and lipofuscin, and iii) pharmaceutical and biotech companies seeking new therapeutic targets for the treatment of AMD.

2. How will they benefit from this research?
We focus here on group iii) since these are the beneficiaries whose activities are most likely to have a direct impact on the nation's health and wealth. In this regard, AMD carries a significant cost in terms of disability, management and therapy and in western societies this socio-economic burden is rising with increased longevity. Among diseases of the eye, AMD is the most common in the over 60s, and its prevalence is increasing with some 30 - 50 million affected worldwide. Therapeutic options for AMD are limited and there remains a significant unmet clinical need. For these reasons there is strong interest in Pharma in the development of therapies aimed at arresting, slowing and curing AMD. In the eye the best available treatment for wet (neovascular) AMD is VEGF blockade via biologics such as Lucentis and Eylea, but these drugs are not effective in all of the target patient group and there are growing concerns over adverse side effects. It is already known that lipofuscin accumulation and complement activation are important contributory factors in AMD, so we anticipate that the outcomes of this project will be of interest to Pharmaceutical companies seeking new therapeutic options in the treatment of retinal degeneration.

The wealth implications are substantial with regard to the commercial value of therapeutics in these areas. Thus, the humanized monoclonal antibody Lucentis (Roche-Genentech), which is used to treat wet AMD, grossed almost $2B worldwide in 2013. As wet AMD accounts for only around 15% of all AMD, and Lucentis is effective in only about half of those patients treated, the unmet clinical need creates enormous scope for future revenue-generating drugs. Therapeutics that target the complement system or that reduce lipofuscin accumulation would be useful in both wet and dry AMD, and would have the added attraction of targeting the actual cause of AMD rather than one of the symptoms. If the work proposed here reveals new potential targets for the treatment of AMD then the impact, in terms of health wealth implications for the UK, would be considerable.