Annexin 8 and differentiation of the retinal pigment epithelium

There are two broad aims to this research. One is to learn more about the function of a protein named annexin A8, and the other is to find out how this protein influences vision. Annexin A8 is only expressed in a few cell types, and is most abundant in the cells that line the airways of the lungs. These are termed epithelial cells, and annexin A8 is also expressed in epithelial cells in other parts of the body including the breast and eye. A consistent observation is that annexin A8 is present at high levels in epithelial cells that are mature and functioning normally. The maturation process for epithelial cells requires their development from dividing immature cells into a sheet of tightly connected non-dividing cells. As this maturation process occurs, annexin A8 levels steadily increase. Interestingly, annexin A8 levels are much lower in cancerous epithelial cells, most notably in breast cancer, and the disappearance of annexin A8 from these cells seems to be associated with a reversal of the maturation process. These observations led us to the hypothesis that annexin A8 is somehow required for epithelial cell maturation and for maintenance of the normal characteristics of the mature epithelial sheet. Support for our hypothesis came unexpectedly from a study in which we were investigating the growth and maturation of epithelial cells from the eye. In the retina, which is the part of the eye that captures light and enables us to see, there is a layer of vitally important epithelial cells that is essential for retinal health. Without these cells the neighbouring light-sensitive photoreceptors rapidly die (which is what happens in age-related macular disease). We examined the expression of all known genes when the cells underwent a drug-induced reverse-maturation process, and we found that annexin A8 levels were massively reduced when this occurred. We also found that by forcing the cells to express annexin A8, we could prevent the reverse-maturation. Here, we plan to use these epithelial cells from the eye to investigate the function of annexin A8, to find out how it maintains the mature epithelial state, and why it needs to be switched off or suppressed for maturation to reverse. Some of this work we can do using cultured cells, but we now also have the necessary strains of mice to investigate this process in vivo. We have developed a way of deleting a gene in the retinal epithelial cells, and will use this approach to delete annexin A8 in the living eye. We can then investigate the consequences of this on vision, and also examine whether or not the annexin A8-depleted epithelial cells undergo maturation reversal in the living eye. If they do, and thus behave in a manner similar to their behaviour in culture, then they may convert from epithelial cells to neurones. This would constitute a remarkable developmental feat, and in the future may pave the way for new treatments for forms of blindness caused by the loss of photoreceptors.

In this proposal we will use both in vitro and in vivo approaches to investigate annexin A8 function, in which the former will employ the widely used ARPE19 cell line, and the latter transgenic mice. For the in vitro studies we will control annexin A8 expression using either siRNA, or via ectopic expression of annexin A8-GFP, and then examine the cellular responses to fenretinide with regard to proliferation, senescence, and development of a neuronal phenotype. In addition, we will examine the effects of manipulating annexin A8 expression on retinal phagocytosis and fenretinide signalling via c-raf and ERK1/2. These studies will use techniques that include western blotting, quantitative fluorescence, live cell imaging, and quantitative morphometric analysis. To determine whether annexin A8 responds directly to fenretinide we will conduct standard promoter-reporter assays in control RPE and RPE cells induced to transdifferentiate using fenretinide. We will also use a mouse BAC to generate a construct containing annexin A8-GFP-6xHis, to affinity-isolate candidate binding partners of annexin A8 that will be identified using mass spectrometry. For the in vivo studies we will use a transgenic mouse that we recently generated that contains an inducible form of the Cre recombinase under the control of the RPE-specific Monocarboxylate Transporter 3 gene promoter. These mice will be crossed with floxed annexin A8 mice generated by Professor Volker Gerke (Muenster), and the annexin A8 gene will then be deleted in the RPE by administration of tamoxifen. Mouse retinas will be examined in section and flatmount using immunohistochemical and histological techniques, with particular attention to the organisation and anatomy of the RPE and photoreceptors. Depending on the speed with which a phenotype develops mice will be examined using tests of visual function including electroretinography, acuity and contrast sensitivity.

1. Who will benefit from this research? The outcomes of this project will benefit i) other workers investigating the cell biology of the retinal pigment epithelium, ii) investigators studying epithelial cell differentiation and development, iii) pharmaceutical and biotech companies seeking new therapeutic targets for the treatment of retinal diseases, iv) the clinicians and their patients who might use such therapeutics. 2. How will they benefit from this research? I focus here on groups iii and iv, since these are the beneficiaries whose activities are most likely to directly impact on the nation's health and wealth. With regard to health, RPE cell dysfunction is heavily implicated in several forms of blinding eye disease, in particular age-related macular disease (AMD). Blindness and visual impairment carry a significant cost in terms of disability, management and therapy, and in western societies this socio-economic burden is rising with increased longevity. There is currently strong interest in pharma in the development of therapies aimed at arresting or even reversing retinal degeneration, and in this context RPE proteins with roles in maintaining the differentiated cell phenotype are attractive therapeutic targets. This is because de-differentiation of RPE cells may permit either division or transdifferentiation. The former would be useful in repopulating a retina in which RPE cells have died, and the latter in the generation of non-RPE cell types that may have died or become dysfunctional, such as photoreceptors. The wealth implications are significant. To give one example, the humanized monoclonal antibody Lucentis (Roche-Genentech), which is used to treat neovascular AMD, is expected to gross ~$2bn worldwide next year. Importantly, neovascular AMD accounts for only 10% of all AMD, and Lucentis is effective in only half of those patients. Thus, even therapies that target only a small fraction of people with visually impairment can yield substantial revenues. The timescales for realizing these benefits cannot be accurately calculated. A best guess would be 3-5 years to establish proof of concept in animal models of retinal disease, onto which one can add another 3-5 years for clinical trials and ultimately commercialization. This is therefore no quick fix, but note that I have recently been awarded a MRC Developmental Pathways Funding Scheme grant to develop a therapeutic monoclonal antibody, and that this followed some 5 years of basic research leading to validation of a novel target. The timescales estimated here may be guesses, but they are at least based on personal experience.