Smokers leave a chemical “footprint” of their addiction history on the surface of their DNA, and this may serve as a measure of their risk for developing cancer, say researchers from the UK and Italy who presented their findings at a conference in Liverpool, UK, this week.

While the underlying genetic code of DNA remains unaltered, a history of smoking leaves a footprint on the surface of DNA, something that is referred to as an “epigenetic modification”. Epigenetic modifications don’t change the underlying genomic code as such, but they influence how it is interpreted.

James Flanagan, Breast Cancer Campaign scientific fellow at Imperial College London, and colleagues, identified a number of locations in the DNA of blood that carry chemical tags put there as a result of smoking.

The tags gradually vanish once the smoker gives up the habit, but the DNA never quite matches the unmarked DNA of the never-smoker.

Flanagan says in a statement:

“This research may help to build a test that will be able to look at a person’s epigenetic information at the molecular level and measure in great detail the added risk of cancer from exposures such as smoking.”

The study was prompted by the recent discovery that a single site on the F2RL3 gene was found to be chemically tagged (hypomethylated) in blood genomic DNA from smokers compared to former and never-smokers.

For their investigation, Flanagan and colleagues carried out two epigenome-wide association studies (EWAS) on blood DNA to look for chemical tags in a healthy population comprising matched pairs of 374 healthy individuals, of whom half went on to develop breast or colon cancer. The technology they used is called “Illumina 450K Methylation Beadchip”.

They examined links between presence of tags and smoking status, as well as cancer risk.

As well as the same previously discovered tag in F2RL3, they discovered several others that were present in the smokers but not as strong in the former and never-smokers.

Using a separate test (bisulphite pyrosequencing) on a group of 180 healthy individuals, they confirmed six of the tags.

They also found one of the six tags (in the region of the aryl hydrocarbon receptor repressor gene AHRR) in the lung tissue of current smokers but not never smokers. They confirmed this in tests on mice bred for smoke exposure experiments.

There was one significant association with breast cancer risk (after adjusting for smoking status) for one of the tags (“an intergenic CpG island on 2q37.1”), but not for the others.

Previous studies have found a link between smoking and bowel cancer risk, but evidence linking smoking and breast cancer is weak. However, Flanagan and colleagues suggest this is likely to be because those studies didn’t use the same genetic or epigenetic measures of smoke exposure.

The team is making the findings available to other scientists who wish to do further research into links between these tags and any as yet unidentified risks.

Flanagan explains how their discovery opens a new avenue for using a person’s smoking history as a way to assess cancer risk:

“Previous research into smoking has often asked people to fill out questionnaires, which have their obvious drawbacks and inaccuracies. Using this approach, we will be able to read the fingerprint on a person’s DNA to tell us a story of how their habit may have changed over the course of their life.”

The researchers say smoking is just the start: there is no reason why further studies could not open up this approach to look at how a person’s history of alcohol consumption changes their epigenome, and to what extent those modifications might affect cancer risk.

Written by Catharine Paddock PhD