Jennifer A. E. Williams

Retinal changes precede visual dysfunction in the complement factor H knockout mouse.

We previously reported that aged mice lacking complement factor H (CFH) exhibit visual defects and structural changes in the retina. However, it is not known whether this phenotype is age-related or is the consequence of disturbed development. To address this question we investigated the effect of Cfh gene deletion on the retinal phenotype of young and mid-age mice. Cfh −/− mouse eyes exhibited thickening of the retina and reduced nuclear density, but relatively normal scotopic and photopic electroretinograms. At 12 months there was evidence of subtle astroglial activation in the Cfh −/− eyes, and significant elevation of the complement regulator, decay-accelerating factor (DAF) in Müller cells. In the retinal pigment epithelium (RPE) of young control and Cfh −/− animals mitochondria and melanosomes were oriented basally and apically respectively, whereas the apical positioning of melanosomes was significantly perturbed in the mid-age Cfh −/− RPE. We conclude that deletion of Cfh in the mouse leads to defects in the retina that precede any marked loss of visual function, but which become progressively more marked as the animals age. These observations are consistent with a lifelong role for CFH in retinal homeostasis.

Exposure to the complement C5b-9 complex sensitizes 661W photoreceptor cells to both apoptosis and necroptosis

The loss of photoreceptors is the defining characteristic of many retinal degenerative diseases, but the mechanisms that regulate photoreceptor cell death are not fully understood. Here we have used the 661W cone photoreceptor cell line to ask whether exposure to the terminal complement complex C5b-9 induces cell death and/or modulates the sensitivity of these cells to other cellular stressors. 661W cone photoreceptors were exposed to complete normal human serum following antibody blockade of CD59. Apoptosis induction was assessed morphologically, by flow cytometry, and on western blotting by probing for cleaved PARP and activated caspase-3. Necroptosis was assessed by flow cytometry and Sirtuin 2 inhibition using 2-cyano-3-[5-(2,5-dichlorophenyl)-2-furyl]-N-5-quinolinylacrylamide (AGK2). The sensitivity of 661W cells to ionomycin, staurosporine, peroxide and chelerythrine was also investigated, with or without prior formation of C5b-9. 661W cells underwent apoptotic cell death following exposure to C5b-9, as judged by poly(ADP-ribose) polymerase 1 cleavage and activation of caspase-3. We also observed apoptotic cell death in response to staurosporine, but 661W cells were resistant to both ionomycin and peroxide. Interestingly, C5b-9 significantly increased 661W sensitivity to staurosporine-induced apoptosis and necroptosis. These studies show that low levels of C5b-9 on 661W cells can induce apoptosis, and that C5b-9 specifically sensitizes 661W cells to certain apoptotic and necroptotic pathways. Our observations provide new insight into the potential role of the complement system in photoreceptor loss, with implications for the molecular aetiology of retinal disease.

In situ regeneration of retinal pigment epithelium by gene transfer of E2F2 : a potential strategy for treatment of macular degenerations

The retinal pigment epithelium (RPE) interacts closely with photoreceptors to maintain visual function. In degenerative diseases such as Stargardt disease and age-related macular degeneration, the leading cause of blindness in the developed world, RPE cell loss is followed by photoreceptor cell death. RPE cells can proliferate under certain conditions, suggesting an intrinsic regenerative potential, but so far this has not been utilised therapeutically. Here, we used E2F2 to induce RPE cell replication and thereby regeneration. In both young and old (2 and 18 month) wildtype mice, subretinal injection of non-integrating lentiviral vector expressing E2F2 resulted in 47% of examined RPE cells becoming BrdU positive. E2F2 induced an increase in RPE cell density of 17% compared with control vector-treated and 14% compared with untreated eyes. We also tested this approach in an inducible transgenic mouse model of RPE loss, generated through activation of diphtheria toxin-A gene. E2F2 expression resulted in a 10-fold increase in BrdU uptake and a 34% increase in central RPE cell density. Although in mice this localised rescue is insufficiently large to be demonstrable by electroretinography, a measure of massed retinal function, these results provide proof-of-concept for a strategy to induce in situ regeneration of RPE for the treatment of RPE degeneration.

Regulation of C3 Activation by the Alternative Complement Pathway in the Mouse Retina

The purpose of this study was to examine the retinas of mice carrying hemizygous and null double deletions of Cfb-/- and Cfh-/-, and to compare these with the single knockouts of Cfb, Cfh and Cfd. Retinas were isolated from wild type (WT), Cfb-/-/Cfh-/-, Cfb-/-/Cfh+/-, Cfh-/-/Cfb+/-, Cfb-/-, Cfh-/- Cfd-/-, and Cfd+/- mice. Complement proteins were evaluated by western blotting, ELISA and immunocytochemistry, and retinal morphology was assessed using toluidine blue stained semi-thin sections. WT mice showed staining for C3 and its breakdown products in the retinal vasculature and the basal surface of the retinal pigment epithelium (RPE). Cfb-/- mice exhibited a similar C3 staining pattern to WT in the retinal vessels but a decrease in C3 and its breakdown products at the basal surface of the RPE. Deletion of both Cfb and Cfh restored C3 to levels similar to those observed in WT mice, however this reversal of phenotype was not observed in Cfh-/-/Cfb+/- or Cfb-/-/Cfh+/- mice. Loss of CFD caused an increase in C3 and a decrease in C3 breakdown products along the basal surface of the RPE. Overall the retinal morphology and retinal vasculature did not appear different across the various genotypes. We observed that C3 accumulates at the basal RPE in Cfb-/-, Cfb-/-/Cfh-/-, Cfb-/-/Cfh+/-, Cfd-/- and WT mice, but is absent in Cfh-/- and Cfh-/-/Cfb+/- mice, consistent with its consumption in the serum of mice lacking CFH when CFB is present. C3 breakdown products along the surface of the RPE were either decreased or absent when CFB, CFH or CFD was deleted or partially deleted.