Sci. Aging Knowl. Environ., 5 February 2003
Vol. 2003, Issue 5, p. nw24
[DOI: 10.1126/sageke.2003.5.nw24]


When Tips Disappear, the End Is Near

Short telomeres presage mortality, but do they cause it?

R. John Davenport;2003/5/nw24

Key Words: quantitative PCR • retrospective study • proportional-hazards regression model

Like a biblical sign of Armageddon, diminished chromosome ends portend death, according to a new study. The work is the first to link telomere length to human mortality, although whether stubby telomeres cause death or result from another lethal process remains unclear.

Telomeres, which protect the ends of chromosomes, shrink each time a cell divides. When they erode too far, cultured cells enter a shutdown state known as senescence. Telomere-induced senescence might age an animal by paralyzing crucial cells that divide frequently (see "More Than a Sum of Our Cells"). However, no one had looked for an association between human mortality and telomere length until now.

To probe that issue, molecular geneticist Richard Cawthon of the University of Utah, Salt Lake City, and colleagues exploited blood samples previously collected from 143 60- to 97-year-olds. The team measured the length of telomeres in DNA purified from each blood sample, using a new method based on the polymerase chain reaction that takes less time and material than standard techniques do. Because telomeres in blood cells tend to shorten as people get older, the researchers created two groups with roughly the same distribution of ages but with a different distribution of telomere lengths. First, they grouped the subjects into 5-year age brackets. Then, they further divided each group, splitting those with the longest telomeres from those with the shortest. The researchers pooled the "long-telomere" subjects from every age group together and the "short-telomere" subjects from every age group together.

The investigators then used death and census records to calculate how likely individuals in each population were to die as a function of time after sample collection; enough individuals perished to allow a statistical analysis of mortality. The short-telomere bunch died nearly twice as fast as the long-telomere folks did. Causes of death varied between the two groups. People with short telomeres died from infectious diseases nearly nine times as often as did those with heartier telomeres, and they perished from heart trouble more than three times as frequently. Stroke and cancer killed members of both groups at approximately the same rate. Pruned telomeres might hasten death, or another factor that increases mortality might whittle away chromosome ends, says Cawthon, but the study doesn't distinguish between those possibilities. For example, immune cells with truncated telomeres might repel invaders poorly, proposes Cawthon, but oxidative stress could spur heart disease and shrink telomeres.

"Even though you can't tell whether it's cause or effect, any relationship between mortality and telomere length is immensely interesting," says cell biologist Peter Hornsby of the University of Texas Health Science Center in San Antonio. "It suggests that telomere shortening over life isn't something to be ignored." Statistician Hans-Georg Mueller of the University of California, Davis, cautions that the analysis doesn't fully adjust for age differences. However, demographer S. Jay Olshansky of the University of Illinois, Chicago, says the study's design is the best way to approach the problem without investing decades of labor into collecting samples. Everyone agrees that further work is necessary to determine if short telomeres mark people for death or just harbinger the coming of the end.

--R. John Davenport

R. M. Cawthon, K. R. Smith, E. O'Brien, A. Sivatchenko, R. A. Kerber, Association between telomere length in blood and mortality in people aged 60 years or older. Lancet 361, 393-395 (2003). [Abstract] [Full Text]

February 5, 2003

Suggested ReadingBack to Top

Citation: R. J. Davenport, When Tips Disappear, the End Is Near. Science's SAGE KE (5 February 2003),;2003/5/nw24

Science of Aging Knowledge Environment. ISSN 1539-6150