The individual blood cell telomere attrition rate is telomere length dependent

Abstract

Age-associated telomere shortening is a well documented feature of peripheral blood cells in human population studies, but it is not known to what extent these data can be transferred to the individual level. Telomere length (TL) in two blood samples taken at approximately 10 years interval from 959 individuals was investigated using real-time PCR. TL was also measured in 13 families from a multigenerational cohort. As expected, we found an age-related decline in TL over time (r = -0.164, P<0.001, n = 959). However, approximately one-third of the individuals exhibited a stable or increased TL over a decade. The individual telomere attrition rate was inversely correlated with initial TL at a highly significant level (r = -0.752, P<0.001), indicating that the attrition rate was most pronounced in individuals with long telomeres at baseline. In accordance, the age-associated telomere attrition rate was more prominent in families with members displaying longer telomeres at a young age (r = -0.691, P<0.001). Abnormal blood TL has been reported at diagnosis of various malignancies, but in the present study there was no association between individual telomere attrition rate or prediagnostic TL and later tumor development. The collected data strongly suggest a TL maintenance mechanism acting in vivo, providing protection of short telomeres as previously demonstrated in vitro. Our findings might challenge the hypothesis that individual TL can predict possible life span or later tumor development.

Figure 1. RTL and yearly telomere loss over a 9–11 year period.

(A) There was a highly significant and inverse correlation between RTL in sample 1 and telomere attrition rate. The inverse correlation between RTL in sample 1 and attrition rate was similar for (B) future tumor patients and (C) controls.

Figure 2. Correlation between RTL and attrition rate in 13 separate families.

Each family contained 10 or more (maximum 28) related individuals (i.e., no in-laws). (A) Illustration of telomere shortening with age in 13 individual families. The separate regression lines were obtained by plotting RTL vs. age for each individual in a family. (B) Telomere attrition rate (i.e. slope) in the 13 families in relation to the estimated RTL at age 14 in each family, showing a strong negative correlation.

Figure 3. Blood RTL could not predict future tumor development.

There was no difference between future tumor patients and controls regarding RTL measured ≥9 years before diagnosis (RTL sample 1) or measured 0–11 years before diagnosis (RTL sample 2).

Figure 4. Individual telomere attrition rate was not a marker for future tumor development.

Yearly telomere loss at the individual level did not differ between future tumor patients and controls.

Figure 5. Blood RTL and prognosis in prediagnostic samples from prostate cancer patients.

Shorter than median RTL ≥9 years before diagnosis was associated with decreased survival.

Figure 6. Schematic drawing of blood draws for sample 1 (baseline) and sample 2 (follow up).