By David N. Leff

Have you ever noticed that in research reports and news stories about autoimmune diseases - notably, lupus, rheumatoid arthritis, insulin-dependent diabetes, multiple sclerosis - there's usually a disclaimer? After discussing the latest findings on how the immune system's autoantibodies attack the body's own autoantigens, comes the caveat: Despite scientific progress, the root cause and molecular mechanisms of autoimmune diseases remain unknown. That's not for lack of trying.

"People have thought many things about where autoantibodies come from," observed rheumatologist and immunologist Paul Plotz, chief of the Arthritis and Rheumatism Branch of the National Institute of Arthritis and Muscoloskeletal and Skin Diseases in Bethesda, Md. "One set of ideas," he added, "holds that the creation of autoimmunity really reflects a generalized breakdown of the mechanism by which the body maintains immune tolerance of its 'self'.

"So a lot is known about why one doesn't make an autoimmune response to everything all the time. And a big theory about autoimmune diseases - particularly lupus - is that the tolerance system has broken down. The trouble with that idea," Plotz continued, "is that, contrary to popular belief, you don't make responses to everything. You make very limited specific reactions that are repetitive from individual to individual; they're not random. So most patients with lupus will have a particular set of autoantibodies, and only that set, not everything.

"In the mouse disease of lupus, of which there are several models," Plotz continued, "the same observation has been made, that there's a limited set of autoantibodies, but nobody really understands what the genetic molecular basis of that murine disease is. One lupus mouse, for example, cannot cause the apoptosis [programmed cell death] of its own lymphocytes. So those antibody-producing lymphocytes grow beyond control, but they are not what causes the autoimmunity. There are many other genetic lesions in that mouse, which are responsible, and none them has been identified. There's a substantial effort in the immunological community to find the genes - of which there are probably dozens - whose alteration contributes to autoimmunity.

"There are many reports of simulating autoimmune diseases," he pointed out, "but the exact mechaisms are not always clear. In my laboratory, we think we know exactly what we're doing: We're turning on our major histocompatibility [MHC] class I molecules. That is, in a single manipulation, we've created in mice an autoimmune inflammatory muscle disorder - polymyositis - that is utterly characteristic of human myositis, but whose features are not found in other conditions."

How To Make Murine Muscles Mimic Human Myositis

Plotz is senior author of a paper in today's Proceedings of the National Academy of Sciences (PNAS), dated Aug. 1, 2000, which reports this feat. Its title: "Conditional upregulation of MHC class I in skeletal muscle leads to self-sustaining autoimmune myositis and myositis-specific autoantibodies."

Myositis holds credentials as an autoimmune ailment, but it's not in the A league. "Myositis is a fairly uncommon disease," Plotz told BioWorld Today. "There are approximately 10 new cases per million individuals in the U.S. each year. That means probably 3,000 - and multiply that prevalence by the world population." Its principal feature is generalized muscle weakness.

To grasp how Plotz and his co-authors created a transgenic mouse that faithfully mimics human myositis, he proffered this run-down of MHC class I molecules 101:

"It's a two-chain molecule," he recounted, "which assembles on the surface of the cell - in our case, a muscle cell - to form a molecule that has a domain that can bind peptides. The normal state of MHC class I is to be made inside cells, to assemble both chains, and pick up either peptides made from broken-down proteins from within the cell, or peptides made when the cell is virally infected. When the molecule goes on to the surface of the cell, it has two chains and a groove, which is on the outside of the cell, and in that groove is a peptide that was made inside that cell.

"People have imagined," Plotz went on, "that MHC class I is the molecule that signals to the outside world: 'Hey! I've been invaded by a virus, and right here in this groove is a peptide of that virus to prove it. Now come along and kill me - because I represent a danger!' So that cell is duly targeted and destroyed by T lymphocytes."

Turning to the PNAS paper's myositis mouse, he narrated: "We used the mouse's own MHC class I molecule and put it in a transgene controlled by the tetracycline [tet] transactivator. When the tet is present, these mice are exactly like any other mouse. When you take the tet away, the transactivator protein made in the muscle is now able to switch on the transgene, and MHC class I appears on the surface of muscle cells - where it doesn't belong. But in patients with myositis it's there.

"So we asked the question: In what way does that appearance contribute to the muscle disorder? Could it be the primary starting point of human myositis, with a heretical view that it's likely the starting point? Indeed, we got a disease with much like the critical features of myositis, including the self-sustainability."

Muscle Is Only The Beginning

"The truly startling finding," Plotz recounted, "is that we find the induction of an autoantibody, which is not in itself against an antigen that is peculiar to muscle. It's an autoantibody that can react with a protein that's in every tissue, every cell - histidyl-tRNA synthetase (HRS). There are many such antibodies around, and nobody knows why they're there, or how they get there, or what their connection to disease is. They are found in humans only, many in lupus for example.

"So we're suggesting," he said, "that the autoimmunity process - in this case from a non-specific stimulus - can become self-sustaining. And that's really what reflects the human disease. Once begun, whether it's rheumatoid arthritis or lupus or myositis, it is a life-long process. And we have found in this limited model, that we can turn it on, we can get a characteristic autoantibody - and we can't turn it off."

Plotz and his laboratory "are now going to see whether we can simulate similar disease-specific autoantibodies in other tissues. If this were just myositis, it would have only small interest. But the general implication that a nonspecific perturbation of a tissue leads to the unrolling of a program that can eventuate in autoimmunty - that's really what we're after - the idea that this has a generality. And I'm a clinician," he concluded, "so I think about therapy. Yet I don't know its implications for clinical medicine - yet."