By David N. Leff

Suffering from brain and blood vessel inflammation of uncertain origin, the 33-year-old Wisconsin man had been referred by his physician in 1997 to the National Institutes of Health (NIH) in Bethesda, Md. One likely risk factor for the patient's puzzling symptoms was his history of avid recreational hunting in the heavily wooded areas of his home state.

Wisconsin is one of the hot-spot states that account for 90 percent of Lyme disease cases in the U.S. The others are Minnesota, California and Oregon, plus the Eastern seaboard from Massachusetts to Maryland. The remaining 10 percent are scattered around the rest of the country. Every year, 16,000 new cases of the tick-borne infection are reported to the Centers for Disease Control.

Four weeks on intravenous antibiotic therapy cleared up the young man's meningoencephalitis and vascular cellulitis, but attacks of these Lyme disease aftermaths have flared up repeatedly, ever since they began in 1993. "To this day," observed neurologist Roland Martin, at the National Institute of Neurological Diseases and Stroke (NINDS), "he remains pretty much disabled."

Infectious disease specialist Adriana Marques pointed out that, "If people infected by the Lyme disease bacterium, Borrelia burgdorferi, get antibiotic treatment early on, then the vast majority of them will not have problems. The minority who get missed may develop the late, chronic form of the disease." Marques heads the Clinical Studies Unit at NIAID, the National Institute of Allergy and Infectious Diseases. She is co-author of a paper in the December 1999 issue of Nature Medicine titled: "Identification of candidate T-cell epitopes and molecular mimics in chronic Lyme disease." Martin, who is acting chief of the Cellular Immunology Section at NINDS, is the article's senior author.

NIAID director Anthony Fauci hailed the Nature Medicine finding as "a major advance for Lyme disease researchers and their patients. We now have a powerful new tool to investigate what role autoimmune mechanisms play in the development of chronic symptoms associated with Lyme disease." He added, "We can also use this strategy to study other infectious and immunological diseases."

Process Kicks Off With Spinal Tap

Marques described this innovative strategy to BioWorld Today step by step:

"We began by withdrawing cerebrospinal fluid from our Wisconsin patient, and looking in it for what T lymphocytes, with importance for the disease, were increased in his fluid.

"We grew these T cells with Borrelia lysate, a mixture of the bacterial proteins. So those T cells that recognized the proteins on that admixture would grow in 96-well plates. We diluted it till we could find only one type of T cell in each well.

"Using that single T-cell clone, we went to a combinatorial peptide library. There we started with 200 different mixtures of peptides. There are 20 essential amino acids," Marques explained, "and the peptides that we use are 10 amino acids in length. So there were 10 peptide sequences and 20 positions, and we fixed one amino acid in each position. Let's say, for example, that you fix an alanine amino acid in position one, and all the other ones are random. You do it again in positions 2, 3, 4 and so on. Then you repeat that for the 19 other amino acids in all positions.

"Next we looked to see which amino acids stimulated the T-cell clone - which ones had the best stimulation in which position. Then we set up a mathematical model of a matrix. Let's say you have three or four amino acids in position five. What are the possible peptides the clone will react against? A search on the database come up with various peptides that our patient's T cell could react against. It identified the ones we thought might be implicated in the disease.

"What we had done," Marques recounted, "was to make a peptide, and prove that the T cell recognized that peptide. But this T-cell clone can actually recognize multiple peptides, including some that are potentially autoantigens - that is, parts of human proteins." She cautioned, however, "This work was all done in vitro; we have no proof that it's what happens in vivo." Martin of NINDS believes it does happen in vivo.

From Lyme Disease To Cancer To Multiple Sclerosis

"The potential clinical applications," he told BioWorld Today, "are really vast. You can imagine using the approach described in this paper in tumor immunology. One would take cells from a tumor and find more or less the entire spectrum of antigens - which are usually self- or autoantigens - that tumor-specific cells may recognize. That is one area in which it will have applications for the future, and is already being explored.

"Another area," he went on, "is Lyme disease vaccine design. It may be suitable in the future if one uses peptides as a vaccine. Multiple peptides from one organism would be much more efficient than using individual single ones.

"A third area," Martin continued, "is autoimmune research. For many autoimmune diseases the triggering antigens - self-antigens or even foreign agents - that start something like diabetes or rheumatoid arthritis are not known. Again, one might use that approach to get to that question."

Martin said, "Together with Dr. Marques and people at the NIAID, we are primarily establishing our technique for use in multiple sclerosis [MS] and Lyme disease. We have used it for taking cells specific for the myelin sheath of the brain, which are considered targets in MS, and with this peptide-library approach found cross-reactive foreign agents that may have triggered these cells in vivo. We hope that we can develop treatments for MS as good as or better than what's currently available."

Martin and his colleagues, he said, "have a research and development agreement with Mixture Sciences, a company in San Diego that has patented this approach, using the combinatorial peptide libraries for deciphering receptor-ligand interactions. And together with them, we have developed this strategy further, to use for understanding T-cell recognition.

"That in turn," he concluded, "should help us understand what is probably going on in this patient - why he has a disease that is damaging his nervous system, rather than being self-terminated after the organism is eliminated, which is the usual course. But it's a long way from being applicable in humans."