Dissecting the genetic heterogeneity of familial dementias

Dementia is a clinical syndrome associated with progressive deterioration of intellectual functions including loss of memory, difficulties with language, simple calculations, planning and judgment, and motor skills with eventually loss of autonomy. Currently there are an estimated 36 million people worldwide with dementia and this figure is set to increase to more than 115 million people by 2050. Given the prevalence of dementia and the associated significant financial and human costs, in recent years there has been a huge burst of studies aimed to identify the causes of this disorder and its underlain pathological mechanisms, in order to define therapeutic treatments to replace the nowadays available palliative cares. Among different subtypes of dementia, Alzheimer's Disease (AD) is the most frequent form, followed by Vascular Dementia and Frontotemporal Lobar Degeneration (FTLD) which represents the second most common form in people younger than 65 years. All three of these diseases may have a genetic component and, despite considerable progress and efforts made in recent years to clarify their molecular basis, little is known about the pathological mechanisms determining these diseases. Three forms of dementia, which may have a genetic component, were included in this study: Alzheimer's Disease, Frontotemporal Lobar Degeneration and Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leuokoencephalopathy (CADASIL) which is a subtype of vascular dementia. The main objectives of this study included mutational analysis of the genes associated with these diseases with the dual purpose of defining their mutational frequencies in an Italian series and to evaluate the presence of possible genotype-phenotype correlations. The secondary purpose is to be able to draw diagnostic algorithms useful to clinicians in selecting which patients submit to genetic testing. Two clinical cohorts were set up; one consisting in patients with diagnosis of AD, FTLD or related disorders, and the other made up of patients with CADASIL. The mutational analysis was performed by two main techniques: Denaturing High Performance Liquid Chromatography and direct sequencing. The analyzed genes encompassed: Amyloid Precursor Protein (APP), Presenilin1 (PSEN1) and Presenilin2 (PSEN2) for Alzheimer's Disease; Progranulin (PGRN), Microtubule Associated Protein Tau (MAPT), Valosin Containing Protein (VCP), Charged Multivesicular Body Protein 2B (CHMP2B), TAR DNA Binding Protein (TARDBP) for Frontotemporal Lobar Degeneration and NOTCH3 for CADASIL. The mutational analysis of AD- and FTLD-associated genes led to the identification of 8 mutations: three novel variations, PSEN1 p.Ile437Asn, PSEN2 p.Thr18Met and PGRN p.His400ThrfsX12 and five already described substitutions, PSEN2 p.Arg71Trp and p.Met174Val, PGRN p.Phe86SerfsX170 and p.Thr272SerfsX10 and MAPT c.IVS10+16C>T. The molecular data and in silico analyses performed in this study argue in favour of pathogenetic nature for Presenilins variations even though the role of some of them is debated in literature. The substitution identified in PGRN and MAPT are mutations whose pathogenic mechanism has already been described. The clinical phenotype associated to identified mutations was not unique and, especially mutations in the PGRN gene showed a marked variability in clinical presentations, even within the same family. Furthermore, the identification of a FTD-MND family linked to a locus on chromosome 9 further emphasizes the genetic heterogeneity of FTLD. The mutational screening of NOTCH3 gene led to the identification of 21 different mutations, including 7 novel variations, distributed unevenly along the gene. A geographical clustering was observed with mutations identified only in patients living in North-East Italy, a few in North-West and other in Central Italy. Haplotype analysis was performed to assess a possible founder effected underlying this regionalization but, although consistent, it was not confirmed as the majority of mutations was associated with the most common haplotype. The results of this study together were useful to define diagnostic algorithms that could help clinicians to identify patients suggestive of a molecular basis of disease to address to genetic testing.