Investigating molecular mechanisms underlying Marfan syndrome, based on fibrillin-1 secretion and assembly

Fibrillin-1 is a major glycoprotein component of 10-12 nm microfibrils, consisting predominantly of calcium binding epidermal factor-like (cbEGF) and TGFß binding protein like (TB) domains. Mutations within the gene encoding fibrillin-1 cause Marfan Syndrome (MFS), an autosomal dominant disease of the connective tissue, characterized by a wide range of severity, with neonatal MFS (nMFS) lying at the most severe end of the spectrum. nMFS substitutions are clustered into domains TB3-cbEGF18, the ‘neonatal’ region of fibrillin-1. Thus far, the molecular pathology of MFS does not explain the severity associated with a dysfunctional neonatal region, especially since ‘classical’ MFS (cMFS) substitutions are also found in these domains. This study aims to investigate the molecular and cellular factors that might contribute to disease severity in MFS. A set of cMFS and nMFS substitutions were investigated for their effect on the secretion of full-length GFP-Fbn expressed transiently in HEK293T cells, and of recombinant fragment NPro-TB3-cbEGF19 expressed stably in MSU1.1 fibroblasts and transiently in HEK293T cells. A microfibril assembly assay then tested the ability of the full-length GFP-Fbn variants to incorporate into FS2 assembled microfibrils. Secretion results from the different systems were consistent, showing no effect of most of the cMFS and nMFS substitutions on the trafficking of fibrillin-1, with the exception of two cysteine variants that showed partial intracellular retention. Opposing secretion results of the same substitutions in different recombinant constructs provided insight into the role of domain TB3 in the folding of adjacent cbEGF domains. Microfibril assembly results distinguished, for the first time, the behavior of cMFS and nMFS variants. cMFS variants clearly assembled into FS2 microfibrils, suggesting that a dominant-negative effect involving mutant incorporation could contribute to the molecular pathology of MFS. nMFS variants, in contrast, did not incorporate into microfibrils, despite having the N- and C-termini implicated in early fibrillin multimerization, indicating that they may form non-productive complexes that are precluded from further assembly steps. These results indicate that phenotypic severity is correlated with the extent of microfibril assembly perturbation and suggest an important role for domains TB3-cbEGF19 in the lateral association of fibrillin-1 monomers during microfibril assembly.