Developing a patient-derived induced pluripotent stem cell model to understand the clinical and pathological changes in macular degeneration

Abstract

Late-onset retinal macular degeneration (L-ORMD) is a fully penetrant autosomal dominant macular degeneration resulting from a Ser163Arg substitution in the gene encoding the protein C1QTNF5. Clinically L-ORMD results in dark adaptation delay in the fifth decade, central visual loss in the sixth decade and further progressive visual field loss in successive decades of life. Pathologically the disease results in thick sub-retinal deposits, which have a similar composition to drusen seen in AMD, retinal pigment epithelial (RPE) loss, and neuro-retinal atrophy. The function of C1QTNF5 is incompletely understood however within the eye it is expressed most strongly by the RPE cells.

An in vitro model for L-ORMD was developed using human induced pluripotent stem cells (hiPSCs) derived from patients and with stem cells from patient’s unaffected siblings used as controls. The hiPSCs were differentiated to RPE (hiPSC-RPE). L-ORMD hiPSC-RPE shared baseline characteristics with sibling control hiPSC-RPE. In order to model in vivo conditions hiPSC-RPE were grown on permeable supports in human serum enriched media. Case hiPSC-RPE cell lines were found to activate the complement pathway resulting in increased deposition of the terminal complement complex (TCC) C5b-9 when compared to control hiPSC-RPE. Using depleted serum, deposition was not affected by depletion of classical and lectin pathway components but was reduced by depletion of alternative complement pathway components. Depletion of complement components C3 and C5 abolished TCC deposition. The addition of a monoclonal antibody against C5 also reduced TCC deposition.

The role of complement dysregulation in L-ORMD pathogenesis was confirmed by immunostaining of L-ORMD and age-matched control human donor retinal sections. L-ORMD retinal sections displayed increased C3d and C5b-9 deposition. Using mutant and wild type-protein generated from a bacterial expression system it was found that the mutant protein was less stable than the wild-type. In addition the wild type protein formed multimers whilst the mutant was mainly monomeric. A surface plasmon resonance (SPR) study showed an increased affinity of wild-type C1QTNF5, especially in multimeric form for complement factor H (CFH), a key regulator of the alternative complement pathway when compared to mutant protein. Taken together these studies implicate dysfunction of the alternative complement pathway in L-ORMD disease mechanism and have suggested a role for C1Q TNF5 in the extracellular matrix. The studies also show that L-ORMD and AMD share a pathogenic and clinical similarities.