Evidence for an ancient origin of the FGA p.Glu545Val (E526V) amyloidosis-causing mutation endemic in Northern Portugal

• Main Author: Costa, P.P.
• Publication Date: 2020
• Other Authors: Lacerda, P.C., Oliveira, M.E., Maia, N., Lobato, Luísa, Santos, Rosário, Tavares, Isabel
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/10400.18/7674
• topic: Amyloidosis; Fibrinogen Alpha Chain Amyloidosis; FGA p.Glu545Val (E526V); Genetic Diseases; Mutation Age; Doenças Genéticas; Determinantes da Saude e da Doença

Fibrinogen alpha chain amyloidosis is widely spread throughout the world, and is most frequently associated with the FGA p.Glu545Val (E526V) mutation, particularly in European countries, with endemic foci of the disease identified in the UK and in northern Portugal. All identified Portuguese patients are from the same region, and a preliminary attempt to characterize the disease-associated haplotype hinted at a common ancestor, but whether this is true and how far back in time the founding event would have taken place is still much an open question. In order to address these questions we studied all available Portuguese patients and relatives, 56 individuals in total, 33 of which were mutation carriers, belonging to 12 extended families. Thirteen polymorphic short tandem-repeats spanning 5.8Mbps over the FGA gene in chromosome 4 were genotyped. A control population of 67 unrelated individuals was also genotyped for the same polymorphisms. Haplotype phasing was carried out using an empirical linkage disequilibrium-based method implemented in the Beagle 4.1 computer program, with some manual adjustments to take into account pedigree constraints and to preserve parsimony. In total 7 different but closely related disease-associated haplotypes were identified, the most frequent of which (I), represented in 5 families, was presumed to be the ancestral haplotype. The age of the E526V mutation in this population was estimated by fitting a multipoint LD model, as implemented in the DMLE+ program. While this model is somewhat sensitive to estimates of population growth and other parameters, it consistently predicted a mutation age above 100 generations (2500 years). These results point to a relatively ancient mutation, which could explain, at least in part, its wide dissemination throughout the world. It would be interesting to extend this study to other populations, to see if there is evidence for a common ancestor, and to try to establish a pattern of mutation dissemination.