Genetics in dementia : impact of sequence variations for families and populations

Even though the human genome sequences are remarkably similar, there is room for genetic variability that makes every human unique. The most common form of sequence variation in the genome is an exchange of a single nucleotide. The effect of sequence variations on the phenotype can be considered to be a continuum, from common polymorphisms with none or relatively mild effects, to severe and dramatic effects for mutations.

Dementia is a devastating disorder, which severely affects cognitive functions and eventually leads to death. Dementia is, not always but most of the time, caused by neurodegeneration, as in the case of Alzheimer disease (AD) and Frontotemporal dementia (FTD). The genetics in dementia is complex. The majority of cases are caused by a combination of variations in many genes, polymorphisms, in addition to an increased vulnerability due to exposure to harmful environmental factors during the life course. The effects of these variations are studied in populations, where both genetic and environmental factors can be assessed. In a few percent of dementia cases, the disease is caused by mutations in single genes. By identifying mutations in the affected families, important features about the disease etiology can be revealed.

In this thesis, variations in four genes were studied. Study I focuses on the impact of the widely confirmed genetic risk factor for dementia, APOE, on mortality in a prospective community-based study, The Kungsholmen Project. The risk for mortality was increased for epsilon4 carriers, and decreased for epsilon2 carriers. The increased mortality was mainly explained by dementia. In addition, a gender specific effect was observed. In Study II, a G35fsX19 mutation was identified in the GRN gene, causing FTD in a Swedish family. Members in this family developed dementia with behavioral disturbances and progressive aphasia with an age at onset around 55 years. At autopsy, neurodegeneration and immunoreactivity for TDP-43 was observed. The G35fsX19 mutation resulted in a frameshift and was predicted to create a premature stop codon. Functional analyses of mRNA showed about 50% less expression of GRN. The mutated mRNA was not detected by cDNA sequencing, suggesting it was degraded by nonsense mediated mRNA decay. In Study III, an I143T mutation was identified in PSEN1 in a Swedish family with early onset AD. The onset age was around 36 years. The mutation carriers were severely affected by cognitive deficits, in addition to neurological symptoms such as myoclonia. Neuropathologically, they were severely affected by Alzheimer pathology. Since one of the pathological hallmarks of AD, the amyloid plaques, consists of the Abeta peptide, the distribution of Abeta species in the postmortem brain of mutation carriers was investigated. Abeta42 was abundantly present both in plaques and vessels while Abeta40 was mainly present in vessels. Interestingly, we found Abeta43, which has rarely been studied in AD, to be present in all investigated brain regions, emphasizing a role for Abeta43 in the disease etiology. In Study IV, the FTO gene, which has been known for its involvement in body weight, was shown to influence the risk for dementia and AD for the first time, in persons 75+ years derived from the Kungsholmen project. The AA-genotype of the FTO rs9939609 polymorphism increased the risk about 50% compared to TT-carriers. This effect could not be explained by vascular risk factors measured at baseline, such as diabetes, high BMI, CVD or physical activity. Interaction between FTO and APOE was found, and together the two risk alleles increased the dementia risk almost three times. This finding supports a role for metabolic dysregulation in the dementia etiology.

To conclude, four genomic sequence variations were investigated for their impact onneurodegenerative diseases and mortality. Two mutations were identified, in GRN and PSEN1 in two families suffering from FTD and AD, respectively. The APOE gene was found to increase mortality whereas FTO, the obesity-associated gene, was for the first time shown to increase the risk of dementia and AD in the old Kungsholmen population.

List of scientific papers

I. Rosvall L, Rizzuto D, Wang HX, Winblad B, Graff C, Fratiglioni L (2009). "APOE-related mortality: effect of dementia, cardiovascular disease and gender." Neurobiol Aging 30(10): 1545-51. Epub 2008 Jan 30
https://pubmed.ncbi.nlm.nih.gov/18237822

II. Chiang HH, Rosvall L, Brohede J, Axelman K, Björk BF, Nennesmo I, Robins T, Graff C (2008). "Progranulin mutation causes frontotemporal dementia in the Swedish Karolinska family." Alzheimers Dement 4(6): 414-20
https://pubmed.ncbi.nlm.nih.gov/19012866

III. Keller L, Welander H, Chiang HH, Tjernberg LO, Nennesmo I, Wallin ÅK, Graff C (2010). "The PSEN1 I143T mutation in a Swedish family with Alzheimer disease: Clinical report and quantification of Abeta in different brain regions." (Manuscript)

IV. Keller L, Xu W, Wang HH, Winblad B, Fratiglioni L, Graff C (2010). "The obesity related gene, FTO, increases the risk for Alzheimer disease in a prospective cohort study." (Manuscript)