Published online 22 November 2010 | Nature | doi:10.1038/news.2010.625


The birthday candles in your veins

DNA artefacts from white blood cells offer forensic clues to a person's age.

Blood test – forensic sceneBlood found at crime scenes could help to narrow down the age of a suspect.Howard Barlow / Alamy

A drop of blood can provide a rough estimate of a person's age, helping forensic investigators to draw physical profiles of suspects and victims who leave few other traces behind.

Conventional forensic DNA analysis matches samples gathered from crime scenes and compares them with those of people identified in an investigation or in a database. Increasingly, however, investigators are building physical profiles of individuals on the basis of their DNA alone. For instance, six genetic markers can indicate whether a person has blue or brown eyes1.

In a paper published online today in Current Biology2, researchers based in the Netherlands report a genetic signature for a person's age — to within a decade or so — in a type of white blood cell known as a T cell. Other means of accurately determining a person's age rely on skeletal remains. But, in "most cases you don't have bones or teeth you simply have stains", says Manfred Kayser, a geneticist at Erasmus University Medical Center in Rotterdam, the Netherlands, and a co-author on the paper.

T-cell trace

While scouring the scientific literature for a molecular signature of age present in blood, Kayser and his team realized that the organ that pumps out T cells, the thymus, is gradually replaced with fat tissue as people age.

Previous research has shown that this process leaves behind genetic artefacts. Every time a T cell matures in the thymus it rearranges its DNA to create a molecular receptor that can recognize pathogens and other foreign molecules, leaving loops of excised DNA behind. These loops are present only in newly made T cells, says Kayser, so they provide a reliable stopwatch to time the decline of the thymus and ageing in general.

Kayser and his team quantified the levels of one particular T-cell loop sequence in 195 Dutch volunteers, and plotted them against their biological ages, which ranged from a few weeks to 80 years old. The correlation wasn't perfect, but the researchers found that they could estimate a person's age to within 9 years fairly accurately. When the researchers divided the participants into age groups spanning 20 years, the T-cell loop sequence proved an even better proxy for generational age.

"The correlation is pretty impressive," says Mark Jobling, a geneticist at the University of Leicester, UK. "How useful it will be in practice as a forensic tool remains to be seen, although there will certainly be forensic cases where it will help as an investigative tool."

Disease doubts

Kayser doesn't expect that simply identifying a suspect's age, give or take a decade, will break open many cold cases. But the technique could be combined with other sources of evidence to rule suspects in or out. The approach could also help investigators to identify victims from disasters, he says. Field ecologists could adopt the technique to identify the ages of animals based on blood that they leave behind.


Christopher Phillips, a forensic genetics researcher at the University of Santiago de Compostela in Chile, calls the approach "a breakthrough". But its wide margin of error means that it will be most useful for distinguishing young from old people, he says.

Before that happens, researchers will need to determine whether T-cell loops correlate with age in other ethnic groups, says Kayser. A potential limitation of the technique is that it could be a poor judge of age in people with HIV, diabetes and other conditions that perturb T cells. "There are a huge number of disease states that affect thymic function," says Alice Lorenzi, a rheumatologist at Newcastle University, UK.

Kayser also stresses that his team's technique would be used to identify suspects during investigations, not to convict them. "This will never be a tool that ends up in front of a court," he says. 

  • References

    1. Liu, F. et al. Curr. Biol. 19, R192-R193 (2009).
    2. Zubakov, D. et al. Curr. Biol. 20, R970-R971 (2010).
Commenting is now closed.