The systems the body uses to detect infections or foreign invaders such as bacteria and viruses are highly complex, and not fully understood.

This means that doctors do not know what is going wrong in 'autoimmune' diseases, such as multiple sclerosis (MS), and cannot develop drugs to target the problem.

'Friendly fire'

Massive projects in the UK and US are looking for differences in the genes of people with these illnesses compared with healthy people.

The latest research, carried out at the Whitehead Institute and the Dana-Farber Cancer Institute, took a different approach, looking closely at a cell which plays an important role in switching immune defences on and off.

However, another type of T-cell has the job of making sure that this attack does not spill over into healthy tissue, causing unwanted damage.

If these 'regulatory T-cells' aren't doing their job properly, then the other T-cells can destroy cells we need to be healthy, such as nerve sheath cells in MS.

Scientists already know that a gene called Foxp3 is important in controlling regulatory T-cells.

By looking at the entire T-cell genome, the latest research identified a group of 30 genes in the cell that Foxp3 appears to influence.

One of them, Ptpn22, has already been linked by other studies to type 1 diabetes, lupus, Grave's disease and rheumatoid arthritis, so the discovery strengthens the evidence that this is the mechanism that is going wrong in these diseases.

Treatment hope

Professor Richard Young, who led the research, said: "This may shorten the path to new therapies for autoimmune disease.

"With this new list of genes, we can now look for possible therapies with far greater precision."

Professor John Todd is part of a wider project looking for genetic differences between people with diabetes and healthy people, working from the Juvenile Diabetes Research Foundation and Wellcome Trust Diabetes and Inflammation Laboratory at the University of Cambridge.

He described the project as 'very interesting'.

He said: "In the near future, the combination of the approach taken in this research, and searches for 'susceptibility genes' across the entire human genome will be highly productive in helping us investigate these complex diseases."