A short story of the long telomere : evidence in aging epidemiology

Abstract

Telomeres are the sequences of nucleotides at the end of chromosomes. Each time a cell divides, telomeres become shorter. The length of telomeres can be replenished by an enzyme, telomerase. Telomere shortening is hypothesized as the biological origin of aging. Indeed, telomeres are shorter in people with various diseases than others. Whether these observed associations are causal or due to other factors that could be the common causes of both telomere shortening and diseases are largely unknown. In this thesis, we aimed to disentangle the relationship between telomere length and several aging-associated diseases and traits to enhance our understanding of the biology underpinning disease pathogenesis.

In Study I, we performed a Mendelian randomization (MR) study examining the association between telomere length and Alzheimer’s disease, using genome-wide association study (GWAS) summary statistics data released by the International Genomics of Alzheimer’s Project Consortium. We used seven single-nucleotide polymorphisms (SNPs) identified to be of genome-wide significance for telomere length as instrumental variables. The MR analysis showed that shorter telomere length was associated with higher odds of Alzheimer’s disease.

In Study II, we explored the potential pathways from telomere length to coronary heart disease using network MR method. The same seven SNPs were used as instrumental variables as in Study I. Various GWAS summary statistics data of metabolic biomarkers and coronary heart disease were used. The MR analyses found that shorter telomeres were associated with higher levels of insulin, which also had an effect on coronary heart disease. Overall, this study indicates that insulin lies in the pathway from shorter telomeres to coronary heart disease.

In Study III, we investigated the association of telomere length with trajectories of general cognitive abilities in two Swedish cohorts (SATSA and GENDER) and two US cohorts (MCSA and HRS). Telomere length was measured once, while general cognitive abilities were assessed repeatedly at up to seven occasions. Latent growth curve models were applied to examine the associations. We found that shorter telomere length was associated with lower mean levels of general cognitive ability in the age-adjusted models, but not in models when other covariates were further considered. We did not find evidence to support that telomere length was associated with the decline of general cognitive ability.

In Study IV, we revisited the association of telomere length with all-cause mortality, allowing for time-varying effects in a Swedish twin sample where shared familial confounding could be controlled for. Telomere length was measured using Southern blot method, and data of all-cause mortality was obtained from the Swedish Population Registry. We applied between-within analyses in a shared-frailty generalized survival model framework. We found that shorter telomere length was associated with higher mortality rate when controlling for shared familial confounding. Further, we found significant time-varying effects of telomere length on mortality.

In summary, we presented novel evidence about the role of shorter telomere length in aging related diseases and traits in this thesis. We took advantage of publicly available GWAS summary statistics data as well as individual-level cohort data and used various innovative designs and statistical methods to achieve this. This compiled thesis could contribute substantially to the literature of the short and long story of telomeres.