Much Remains to Be Learned

For the most part, we still do not know what causes an autoimmune disease. That is, we do not know the specific factors that initiate the process (or processes) ultimately leading to an autoimmune disease.

Since the official acceptance in the mid 1950s that the immune system can attack the very own body of an individual (rather than just invading microbes), several different theories have been proposed to explain the induction (initiation) of an autoimmune disease.

Each theory is supported by experimental data and some clinical evidence, but none has the ability to explain the entire clinical picture in a satisfactory manner. Indeed, minimal change (if any) in the diagnosis and management of patients with autoimmune diseases has resulted from these theories of causation.

Quite interestingly, it also remains unknown why in autoimmune diseases the autoreactive lymphocytes become specific for a particular antigenic target, as opposed to other possible targets (the so-called conundrum of specificity). For example, in patients with type 1 diabetes mellitus, why is the insulin-producing beta cells of the endocrine pancreas targeted by the patient's immune system rather than the nearby glucagon-producing alpha cells?

Theories of Causality: the Role of Infections

Many of the theories about the causes of autoimmune diseases invoke a preceding infection (such as a viral infection) as the initiating factor.

An excellent review of these theories can be read in this paper.

An Example of The Causality Theories: Molecular Mimicry

A popular theory about the causes of autoimmunity is the molecular mimicry theory.

It was first proposed by Raymond T. Damian in the early 1960s and then adapted to explain autoimmunity by Robert Fujinami and Michael Oldstone in 1983. The theory states that pathogens share epitopes (i.e., amino sequences or structures) with the host; therefore, when the immune system of the host reacts against those epitopes of the pathogens, it will also accidentally cross-react with the same epitopes of the host. More simply put, molecular mimicry is the sharing of epitopes between self (the patient) and foreign (the microbe) antigens. Perhaps the best example of molecular mimicry is that between the M proteins of the group A Streptococcus pyogenes and the myosin form expressed in heart valves. Recurrent throat infections with this bacterium cause the development of antibodies and CD4 T cells directed against the streptococcal M proteins that accidentally cross-react with myosin expressed in the cardiac valves. This self-reactivity is believed to ultimately lead to detrimental immune pathology (in this case, the chronic valvular disease known as rheumatic heart disease).

Molecular Mimicry Theory Illustration

Problems with the molecular mimicry theory

The molecular mimicry theory, however, faces several problems. Infection with group A Streptococcus is very common, yet only a minority of people go on to develop rheumatic heart disease, even if one restricts the denominator to individuals who carry genes that make them susceptible to rheumatic heart disease.

In addition, molecular mimicry can be found not only between cardiac valve myosin and group A Streptococcus, but also between many cardiac antigens and a variety of other microbes that do not lead to rheumatic heart disease.

Finally, convincing evidence of molecular mimicry in the majority of classical autoimmune diseases, such as type 1 diabetes mellitus, is lacking despite decades of studies.