By David N. Leff

Chlamydia bacteria get you coming and going.

In the earliest days after birth, a newborn infant stands a 2 percent to 4 percent chance of incurring neonatal trachoma, an eye inflammation caused by Chlamydia trachomatis. The bacteria are bestowed down the birth canal by an infected pregnant mother.

Later in life, C. trachomatis is known as the leading sexually transmitted disease in the U.S. It inflames the reproductive organs of both sexes, and is the principal perpetrator of sterility in women.

Then there's C. pneumoniae, which inflicts (usually nonfatal) pneumonia and respiratory disease on young and old, plus C. psittacosis, the "parrot disease" that humans can also catch.

Lately, epidemiologists have blamed chlamydia for heart disease, but no one knew its mechanism. Now we do.

Here's how today's issue of Science, dated Feb. 26, 1999, breaks that story: "Chlamydia infections and heart disease linked through antigenic mimicry." The paper's lead author is immunologist Kurt Bachmaier, of the Amgen Institute in Toronto.

"We found the mechanism," Bachmaier told BioWorld Today, "by which infection with chlamydia bacteria actually leads to cardiac disease. The scientific link we saw in the mouse model is what we call 'antigenic mimicry.'"

He explained how this connection works: "In order to find the chlamydial infection, the mouse's immune system activates its B and T lymphocytes. Once those cells get turned on to fight the bacterial infection, they find - in the mouse's own heart - something that looks similar to the chlamydial antigen they're fighting.

"So in this Science article," Bachmaier continued, "we established that this was antigenic mimicry. We showed that T cells and B cells, which react against chlamydia, can also react against heart-muscle protein. This showed that T cells from these mice can react against their own endogenous proteins as well as the foreign chlamydial proteins."

He pointed out that this phenomenon of antigenic mimicry is not unique to mice or to chlamydia. "It's especially important," he said, "in finding out how autoimmune diseases like diabetes mellitus or encephalomyelitis [inflammation of brain and spinal cord] might occur. Some people have proposed that such diseases are caused by bacterial or viral infection, and that the mechanism by which such infections contribute might be due to antigenic mimicry.

"So I think the significance of this paper," he observed, "is that we defined a pathogenic epitope present in both the bacterium and the heart. And we showed that there can be cross-reactivity, or mimicry, between those two at the B-cell [antibody-producing] level."

Bacterial Surface Antigen Apes Heart Muscle

Bachmaier recounted how he and his co-authors uncovered this unsuspected nexus:

"We knew that certain mouse strains are susceptible to Chlamydia trachomatis infection. The strain we used is prone to heart disease as well. So, we first mapped the endogenous peptides that can induce severe heart disease in such mice. Once we had those antigenic sequences defined, we went into the computer databases and compared them with bacterial and viral sequences.

"Amongst many other things, chlamydia came up. So we synthesized those bacterial sequences and injected them into mice. To our surprise, the animals developed heart disease.

"But when we infected the mice," Bachmaier continued, "we realized there was the possibility that the effect we saw on the heart was not caused by antigenic mimicry - meaning it could as well be due to the chlamydial bacteria directly infecting the heart. We distinguished between those two possibilities by using just the peptides and not the infection.

"Then we went to one of our co-authors, Luis de la Maza, at the University of California, Irvine. He infected mice with Chlamydia trachomatis via the respiratory and genital tracts, sacrificed these animals, and sent us their sera.

"When I looked for autoantibody production against endogenous - own-heart-specific - epitopes, I found that 100 percent of the infected mice carried those anti-self antibodies. Which indicated that T and B lymphocytes reacting against the animals' own heart muscle protein, myosin, had been activated as a consequence of the bacterial infection."

Those mice developed cardiac symptoms, Bachmaier observed, but didn't drop dead of them. "It's a progressive, ongoing disease," he explained. "Three weeks after the first injection with the antigen, we could see damage to the heart muscle, especially around cardiac blood vessels, of which some were partially occluded."

He cited some up-to-date "preliminary data where surprisingly, four months after injection, some of the mice still have the ongoing cardiac inflammatory process."

At the molecular level, Bachmaier pointed out, "The primary 15-amino-acid sequences of the peptide are not remarkably similar. We knew that if one injected the mice with purified skeletal-muscle myosin, they remain completely healthy. But there are certain defined differences between heart-muscle myosin and skeletal-muscle myosin. Because of those variations, we focused on those amino acid sequences that were different.

"We discovered that the crucial amino acids, which are found only in the heart-muscle-specific form of myosin, are also present in a chlamydia bacterium's outer membrane protein.

This brought his account to the critical immunological level:

"T lymphocytes, the main immune effector cells in this disease model," Bachmaier pointed out, "recognize the molecular shapes of antigens, but not their primary sequences. This recognition comes when molecules of MHC - the major histocompatibility complex - present the antigens to the immune cells. If a mouse strain has a different form of that MHC, it is protected from disease."

"Similarly, this might happen in human populations, where we have a huge variety of inherited MHC-complex molecules. Clearly there seems to be some link between certain MHC variant alleles and development of heart disease in humans."

How to translate this finding from mouse to man?

Bachmaier and his co-authors are now "designing a next-step study, taking sera from patients with a known history of heart disease, and from patients with a long history of chlamydia infections, but no history of heart disease. We will compare those groups with completely healthy controls, to see whether there are antibodies present, and if so do they have any significance with regard to development of heart disease."