One often-used method of fingering a criminal suspect is the police lineup. The likely perpetrator is presented to a witness in a row of randomly selected individuals, and the witness to the crime points out the culprit.

A similar strategy serves to identify molecular miscreants — infectious bacteria, parasites, viruses. Cells of the body's immune system zero in on antigenic proteins in or on the pathogenic microbes, and chop them up into small but still immunogenic peptides.

In this immunological lineup, specialized molecules called Major Histocompatibility Complexes (MHC), class I or II, present these antigens for fingering by the body's two immune executioners — CD-8-reactive T cells for class I MHCs, and CD-4 inflammatory T cells for class II.

That's the way things stood until the mid-1990s, when a pair of immunologists at Harvard Medical School, Michael Brenner and Steven Porcelli, discovered the function of the CD-1 molecule — a third immune-system antigen-presentation pathway. These CD-1s for the first time added immunogenic lipids to the protein and peptide targets thus far known.

Monday afternoon in New York Brenner told the audience at an all-day seminar on vaccines of this third pathway's first potential clinical application — a lipid-antigen-based antituberculosis vaccine. The seminar was part of the 12th International Biotechnology Industry Organization conference.

From his latest research at Harvard-affiliated Brigham and Women's Hospital, in Boston, Brenner showed that CD-1-presented antigens from Mycobacterium tuberculosis are recognized in guinea pig models of TB infection. The study compared animals immunized with his team's prototype vaccine with unvaccinated controls.

"For the first time now," Brenner told his BIO seminar audience, "we show that we can immunize animals with these antigens and elicit CD-1-reactive T cell responses."

These specific killer T lymphocytes are programmed to break and enter the lung-infecting macrophages in which M. tuberculosis holes up, and kill the pathogen dead.

Brenner explained to BioWorld Today why his third-pathway CD-1 vaccine is adept at tracking down and terminating pathogens that do their dirty work inside cells.

"Microbes protect themselves from immune destruction by preventing their peptides from getting into antigen-presenting pathways," he pointed out. Current antibacterial vaccines are antibody-based. Take, for example, pneumococcal pneumonia vaccination. The reason it works is that pneumococci bacteria don't grow in cells, but in the bloodstream, where antibodies can reach them.

"But there are many bacterial and parasitic infections, like TB or the AIDS virus, that grow and multiply inside cells. Antibodies are useless," Brenner continued. "They can't get to them because they are not floating around in the serum and bodily fluid, but hidden inside a cell. They actually move from cell to cell without ever going outside. So for such intracellular infections, you need a T cell that will kill the cell harboring these pathogens. That CD-8 subset has most of the T cells' cytotoxic activity."

Aquila Pharmaceuticals Holds Exclusive CD-1 License

Brenner and Porcelli are co-inventors of U.S. Patent No. 5,679,347, issued October 21, 1997, and assigned by them to Brigham and Women's Hospital. It protects "Methods of isolating CD-1-presented antigens, vaccines comprising CD-1-presented antigens, and cell lines for use in said method."

The hospital has licensed this invention exclusively to Aquila Biopharmaceuticals Inc., of Worcester, Mass. (See BioWorld Today, April 15, 1998, p. 1.) Brenner and Porcelli are members of Aquila's scientific advisory board. The firm is developing the TB vaccine based on the researchers' ongoing in vivo preclinical work.

"We don't have proof in our guinea pigs yet," Brenner said, "that we have made a TB vaccine that works, but we continue to gather pieces of the story, all of which point in that direction."

That story, as he described it, begins with the two traditional antigen-presenting pathways, class I and II MHC molecules. "Peptide fragments processed through the class I pathway," Brenner explained, "typically result in a cytotoxic CD-8 T cell response. Those processed by the class II pathway typically result in a CD-4 inflammatory T cell response."

He continued: "T cells recognize the MHC antigen-presenting molecule, which has bound a small peptide. That was the paradigm for T cell recognition until our work.

"What we discovered for the first time is that T cells could recognize not only proteins and peptides in their presenting molecules, they could recognize lipids and glycolipids in CD1-presenting molecules. So now it made sense for us to ask — knowing that T cells mediate the protection to certain kinds of infections, and that they recognize lipids and glycolipids — couldn't we use lipids and glycolipids to make T cell-based protective vaccines?

"The answer was that perhaps we could, and that's what we're doing. That's our technology, using this class of antigens to elicit T cell responses as a means to prophylactic immunity against pulmonary tuberculosis."

Brenner told the BIO seminar how major histocompatibility complexes work at the atomic level.

"The class I MHC molecule," he showed his audience, "has at its top a groove into which sit the peptide fragments that are being presented to the T cells. The molecular nature of that groove is that it has multiple small pockets for the side chains of the amino acids of the peptides.

"This really provides the molecular explanation," he went on, "as to which peptides are immunogenic. That is, some peptides have the right amino acids in them to fit in these side pockets, and others don't. The ones that fit are bound to an MHC molecule, and can be presented to a T cell. The ones that don't, can't.

"For CD-1's 3-D structure," he added, "instead of this MHC groove with small side pockets, there are two large sacs, or pockets — deep pockets. And they are very hydrophobic. That's the nature of a pocket for binding a lipid, for example, the lipid tail of a fatty acid's long carbon chain.

"So the key to this whole story," Brenner observed, "is that the CD-1 molecules have two lipid-binding pockets, rather than one peptide-binding groove. That explains our functional data over the last couple of years, which revealed that CD-1 molecules present lipid and glycolipid antigens, not proteins or peptides."

He concluded: "Presentation of lipids and glycolipids by CD-1 activates all kinds of T cells that are likely to have far-reaching effects in host defense. This means vaccinations against microbial infection and against cancer, because T cells in part mediate tumor surveillance; and against autoimmunity, because T cells are responsible for many of the autoimmune diseases." *