(Nov. 22, 2006) —  Autoimmunity can be triggered in the thymus, where the immune system's Thymus or T cells develop, if T cells fail to recognize just one of the body's thousands of proteins as "self." Then that unrecognized part will be attacked .The research confirms an emerging view that autoimmunity can start in this cradle of the immune system, and not only at the sites where autoimmune diseases target, such as the pancreas in the case of type 1 diabetes, or the joints in rheumatoid arthritis.

The discovery, from a mouse model of a human autoimmune condition, suggests that effective strategies to treat autoimmune disease should target not only the "peripheral" sites where autoimmune disease is active, but also the thymus -- the organ where T cells and self-proteins, or self-antigens, first interact.

The research was led by investigators at the University of California, San Francisco (UCSF). It was published online November 20 by the Journal of Experimental Medicine and will appear in the journal's print edition November 27.

T cell soldiers encounter the body's full array of proteins in the thymus, and those T cells with receptors that recognize "self" proteins, or antigens, normally are purged to avoid autoimmune attacks in the body later on. The new research showed that if just one of the body's antigens is not recognized as "self," this single failure can lead to a severe autoimmune disease in the retina.

"The thymus is like a filter," said Mark Anderson, MD, PhD, assistant professor of medicine at the UCSF Diabetes Center, and senior author of a scientific paper describing the discovery. "It is removing or pulling out autoreactive T cells. What this new study shows is if just one self-antigen is missing as the T cells go through the filter, it looks like this alone can lead to an autoimmune disease."

"The finding supports the promise of treatments targeting individual body proteins or antigens since we have shown that a single self-antigen can trigger disease," he added.

A similar mechanism may be at play involving other autoimmune diseases such as type 1 diabetes, Anderson said. Immunologists have demonstrated that insulin is expressed in the thymus -- not just in the pancreas. Studies have shown that people who are protected from diabetes express high levels of insulin in the thymus, while those who are predisposed express lower levels of insulin in this organ.

"What we think is that 'more is better' in the thymus," Anderson says. "If you have more insulin in the thymus, then there is a better chance that potentially destructive insulin-specific T cells will encounter insulin as self and be filtered out."

In the thymus, immature T cells display on their surface many thousands of unique receptors, generated by random gene rearrangements. This strategy allows the receptors to recognize the tremendous diversity of invading pathogens. In the process, however, they also develop receptors that bind to the body's own proteins. These T cells are normally eliminated, avoiding the plague of autoimmunity.

A clue to how the elimination process is controlled came from previous work involving a protein in the cell nucleus called Aire (for autoimmune regulator), which regulates the expression of some 300 to 1,000 antigens in the thymus. Humans and mice lacking the normal Aire gene suffer from multiple autoimmune diseases including diseases that target the thyroid, adrenal, ovary, and eye.

In 2002, Anderson, then at Harvard Medical School, and colleagues there demonstrated that knocking out the Aire gene in the mouse thymus led to failures of expression of a number of genes in peripheral tissues, resulting in autoimmune diseases in those tissues -- the first direct evidence linking gene knockouts in the thymus to autoimmune defects in body tissues. The study, however, did not link a specific organ autoimmune attack with a specific protein missing in the thymus.

In the new study, the researchers carried out a detailed analysis of the autoimmune attack that is directed against the eye in Aire-deficient mice. What the team found was that the immune system was mainly targeting a single eye protein called IRBP despite the fact that several eye-specific proteins were missing in the thymus of Aire knockout mice. The team then went on to show that IRBP was expressed in the thymus under the control of Aire and that knockout mice lacking the IRBP protein were protected from the disease because they don't express the protein that the immune system is targeting.

In a key, final part of the new study, Anderson and his colleagues showed that if mice without a thymus gland -- so-called "nude" mice -- received a normal thymus lacking only IRBP, they developed the autoimmune eye disease. The autoimmune attack occurred even though the mice had normally functioning IRBP in their retinas. The final finding demonstrated that failure of T cells in the thymus to recognize IRBP as a self-protein was sufficient to cause the autoimmune disorder in the retina.

The scientists hope that better understanding of interactions in the thymus can lead to earlier, more effective treatment of autoimmune diseases.

"When we see autoimmune disease in the clinic, we are usually looking at it in a relatively late stage. Tissue is already damaged, antigen expression is ramped up and the immune response is spreading to other self-antigens," Anderson said. "If we can also train our focus on the thymus, where we know at least some of the autoimmune disease is triggered, we may be able to determine just what self-antigens are important and shut down the autoimmune process targeting those self -antigens."

The team is collaborating with Jeffrey Bluestone, PhD, director of the UCSF Diabetes Center, in preclinical studies to see if T cell autoimmune attacks on IRBP can be modulated to prevent the autoimmune eye disease.

Lead author is Jason DeVoss, PhD, a postdoctoral scientist in Anderson's laboratory.

Co-authors include Lawrence Fong, MD, PhD, UCSF assistant professor of hematology and oncology; Yafei Hou, PhD, a postdoc in Fong's lab; Wen Lu, BS, and Kellsey Johannes, BA, both research assistants in Anderson's lab.

The research is supported in part by the National Institutes of Health. Reported by ScienceDaily