Return to the Table of Contents.
F E B R U A R Y 1 9 9 9
WALD closed the medical textbook on his knee. "This was published twenty years ago," he said. "If we looked up 'atherosclerosis' in a textbook from ten years ago, we'd find the same kind of things -- stress, lifestyle, lots about diet, nothing about infection."
Heart disease is now being linked to Chlamydia pneumoniae, a newly discovered bacterium that causes pneumonia and bronchitis. The germ is a relative of Chlamydia trachomatis, which causes trachoma, a leading cause of blindness in parts of the Third World. C. trachomatis is perhaps more familiar to us as a sexually transmitted disease that, left untreated in women, can lead to scarring of the fallopian tubes, pelvic inflammatory disease, ectopic pregnancy, and tubal infertility.
Pekka Saikku and Maija Leinonen, a Finnish husband-and-wife team who have evoked comparisons to the Curies, discovered the new type of chlamydial infection in 1985, though its existence was not officially recognized until 1989. Saikku and Leinonen found that 68 percent of Finnish patients who had suffered heart attacks had high levels of antibodies to C. pneumoniae, as did 50 percent of patients with coronary heart disease, in contrast to 17 percent of the healthy controls. "We were mostly ignored or laughed at," Saikku recalls.
While examining coronary-artery tissues at autopsy in 1991, Allan Shor, a pathologist in Johannesburg, saw "pear-shaped bodies" that looked like nothing he'd ever seen before. He mentioned his observations to a microbiologist colleague, who had read about a new species of chlamydia with a peculiar pear shape. The colleague referred Shor to an expert on the subject, Cho-Chou Kuo, of the University of Washington School of Public Health, in Seattle. After Shor shipped Kuo the curious coronary tissue, Kuo found that the clogged coronary arteries were full of C. pneumoniae. Before long, others were reporting the presence of live C. pneumoniae in arterial plaque fresh from operating tables. Everywhere the bacterium lodges, it appears to precipitate the same grim sequence of events: a chronic inflammation, followed by a buildup of plaque that occludes the opening of the artery (or, in the case of venereal chlamydia, a buildup of scar tissue in the fallopian tube). Recently a team of pathologists at MCP-Hahnemann School of Medicine, in Philadelphia, found the same bacterium in the diseased sections of the autopsied brains of patients who had had late-onset Alzheimer's disease: it was present in seventeen of nineteen Alzheimer's patients and in only one of nineteen controls.
By the mid-1990s a radical new view was emerging of atherosclerosis as a chronic, lifelong arterial infection. "I am confident that this will reach the level of certainty of ulcer and H. pylori," says Saikku, who estimates that at least 80 percent of all coronary heart disease is caused by the bacterium. Big questions remain, of course. Studies show that about 50 percent of U.S. adults carry antibodies to C. pneumoniae -- but how many will develop heart disease? Even if heart patients can be shown to have antibodies to C. pneumoniae, and even if colonies of the bacteria are found living and breeding in diseased coronary arteries, is it certain that the germ caused the damage? Perhaps it's an innocent bystander, as some critics have proposed; or a secondary, opportunistic infection.
But suppose that a Chlamydia pneumoniae infection during childhood can initiate a silent, chronic infection of the coronary arteries, resulting in a "cardiovascular event" fifty years later. Could antibiotics help to address the problem?
A few early studies suggest they might. Researchers in Salt Lake City infected white rabbits with C. pneumoniae, fed them a modestly cholesterol-enhanced diet, killed them, and found thickening of the thoracic aortas, in contrast to the condition of uninfected controls fed the same diet. Additionally, treatment of infected rabbits with antibiotics in the weeks following infection prevented the thickening. Saikku and colleagues reported a similar finding, also in rabbits. Coronary patients in Europe who were treated with azithromycin not only showed a decline in antibodies and other markers of infection but in some studies had fewer subsequent cardiovascular events than patients who were given placebos. (These findings are preliminary; in a few years we may know more. The first major clinical trial is under way in the United States, sponsored by the National Institutes of Health and the Pfizer Corporation: 4,000 heart patients at twenty-seven clinical centers will be given either the antibiotic azithromycin or a placebo and followed for four years to gauge whether the antibiotic affects the incidence of further coronary events.)
Smoking, stress, cholesterol, and heredity all play a role in heart disease. But imagine if our No. 1 killer -- with its vast culture of stress-reduction theories, low-fat diets, high-fiber cereals, cholesterol-lowering drugs, and high-tech bypass surgery -- could in many instances be vanquished with an antibiotic. Numerous precedents exist for long-smoldering bacterial infections with consequences that appear months or years later. Lyme disease, leprosy, tuberculosis, and ulcers have a similar course. Ewald is confident that the association of C. pneumoniae and heart disease is real. He doesn't believe that the germ is an innocent bystander. "It reminds you a lot of gonorrhea in the 1890s," he says. "When they saw the organism there, people said, 'Well, we don't know if it's really causing the disease, or is just living there.' Every month the data are getting stronger. This is a smoking gun, just like Helicobacter."
HAVE a motto," Gregory Cochran told me recently. "'Big old diseases are infectious.' If it's common, higher than one in a thousand, I get suspicious. And if it's old, if it has been around for a while, I get suspicious."
The fact that Ewald has dared to conceive of a big theory for the medical sciences owes much to Cochran's contributions. A forty-five-year-old Ph.D. physicist who lives in Albuquerque with his wife and three small children, Cochran makes a living doing contract work on advanced optical systems for weaponry and other devices. Whereas Ewald is an academic insider, with department meetings to attend and honors theses to monitor, Cochran is a solo player, with an encyclopedic mind (he is a former College Bowl contestant) and a manner that verges on edginess. These days he spends a lot of time at his computer, as rapt as a conspiracy theorist, cruising Medline for new data on infectious diseases and, one imagines, almost cackling to himself when he finds something really good. Cochran's background in a field dominated by grand theories and universal laws may serve as a valuable counterpoint to the empirical and theory-hostile universe of the health sciences.
To illustrate his thinking about infectiousness and disease, Cochran not long ago gave me a tour of his conceptual bins, into which he sorts afflictions according to their fitness impact. Remember that fitness can be defined as the evolutionary success of one organism relative to competing organisms. Only one thing counts: getting one's genes into the future. Any disease that kills host organisms before they can reproduce reduces fitness to zero. Obviously, fitness takes a major hit whenever the reproductive system itself is involved, as in the case of venereal chlamydia.
Consider a disease with a fitness cost of one percent -- that is, a disease that takes a toll on survival or reproduction such that people who have it end up with one percent fewer offspring, on average, than the general population. That small amount adds up. If you have an inherited disease with a one percent fitness cost, in the next generation there will be 99 percent of the original number in the gene pool. Eventually the number of people with the disease will dwindle to close to zero -- or, more precisely, to the rate produced by random genetic mutations: about one in 50,000 to one in 100,000.
We were considering the bin containing diseases that are profoundly antagonistic to fitness, with a fitness cost of somewhere between one and 10 percent by Cochran's calculations. My eye took in a catalogue of human ills -- some familiar, some exotic, some historically fearsome but close to extinct, some lethal in the tropics but of little concern to inhabitants of the temperate zones. This list also showed prevailing medical opinion about cause. Each name of a disease was trailed by a lower-case letter: i (for infectious), g(genetic), g+(genetic defense against an infectious disease), e (caused by an environmental agent), or u (unknown). I read, "Atherosclerosis (u), ... chlamydia (venereal) (i), cholera (i), diphtheria (i), endometriosis (u), filariasis (i), G6PD deficiency (g+), ... hemoglobin E disease (g+), hepatitis B (i), hepatitis C (i), hookworm disease (i), kuru (i), ... malaria (vivax) (i), ... pertussis (i), pneumococcal pneumonia (i), polycystic ovary disease (u), scarlet fever (i), ... tuberculosis (i), typhoid (i), yellow fever (i)."
Of the top forty fitness-antagonistic diseases on the list, thirty-three are known to be directly infectious and three are indirectly caused by infection; Cochran believes that the others will turn out to be infectious too. The most fitness-antagonistic diseases must be infectious, not genetic, Ewald and Cochran reason, because otherwise their frequency would have sunk to the level of random mutations. The exceptions would be either diseases that could be the effect of some new environmental factor (radiation or smoking, for example), or genetic diseases that balance their fitness cost with a benefit. Sickle-cell anemia is one example of the latter.
Though sickle-cell anemia is strictly heritable according to Mendelian laws, it is widely believed to have persisted in the population in response to infectious selective pressures. It heads the list of genetic diseases that Ewald dubs "self-destructive defenses," in which a disease fatal in its homozygous form (two copies of the gene) carries an evolutionary advantage to heterozygous carriers (with one copy), protecting against a terrible infection: in this case falciparum malaria, common in Africa. Similarly, cystic fibrosis, some argue, evolved in northern Europe as a defense against Salmonella typhi, the cause of typhoid fever. Infection thus explains why these deadly genetic diseases have remained in the human gene pool when they should have died out.
But what about something like atherosclerosis? I asked. Leaving aside the evidence concerning C. pneumoniae, it is not apparent why a genetic cause for atherosclerosis should be dismissed out of hand on evolutionary grounds. If it hits people in midlife or later, after they have launched their genes, how could it possibly affect fitness?
Cochran's response illustrates some of the intricacies of evolutionary thinking. "Well, obviously, it's not as bad as a disease that kills you before puberty, but I think it does have a fitness cost. First of all, it's really common. Second, people think that all you have to do to pass your genes along is have children, but that's not true. You still need to raise the offspring to adulthood. In a hunter-gatherer or subsistence-farming culture, the fitness impact of dying in midlife might be considerable, especially during bad times, like famines. You've got to feed your family. Also, cardiovascular disease is a leading cause of impotence, and any disease that makes males impotent at age forty-five has got to affect reproduction somewhat."
But fifty-year-olds? Sixty-year-olds?
Grandmothers do a large proportion of the food-gathering in some tribal cultures, according to recent anthropological reports. "They aren't hampered by babies anymore, and they don't have to go around chucking spears like the men," says George C. Williams, a professor emeritus of ecology and evolution at the State University of New York at Stonybrook, and one of the pillars of modern evolutionary biology. "They contribute substantially to the family diet." If long-lived elders historically have made a difference by fostering the survival of their descendants, and therefore their genes, Cochran figures, then a disease that kills sixty-year-olds could have a fitness impact of around one percent.
NOW what that is?" Ewald asked. We were standing in the main corridor of the Life Sciences Building, gazing up at a decorative metalwork frieze that runs along the walls just above door height. A pair of hummingbirds chase each other in a circle. A human eye and an octopus eye face off. A human hand is juxtaposed with a chimpanzee hand. Ewald pointed to something that looked like a daddy longlegs with a video camera for a head. "Some kind of insect?" I ventured. "It's a virus," he said. "See, it's like a spaceship. That" -- he pointed at the head -- "is its DNA. It injects it inside the cell."
There is something unsettling and fascinating about a virus, an organism that is neither strictly alive nor strictly inanimate, and that replicates by sneaking inside a host cell and commandeering its machinery. "Viruses are essentially bits of nucleic acid -- either DNA or RNA -- wrapped in a protein capsule," Ewald explained. "A retrovirus, like HIV, is an RNA virus with a protein called reverse transcriptase built into it, and once it gets into a cell, it uses the reverse transcriptase to make a DNA copy of its RNA. This viral DNA copy can insert itself into our DNA, where it can be read by our protein-making machinery the same way our own instructions are read."
The modus operandi of the world's most feared virus, HIV, is clever, killing its hosts very, very slowly. A sexually transmitted pathogen, without the luxury of being spread through sneezes or coughs, must await its few opportunities patiently; if those infected have no symptoms and don't know they are sick, so much the better. A mild, chronic form of AIDS had in all likelihood been around for centuries in Africa, according to Ewald. Suddenly in the 1970s -- owing to changing patterns of sexual activity and to population movements -- deadly strains spread in the population of Central and East Africa.
HIV has an extremely high mutation rate, which means that it is continually evolving, even within a single patient, producing competing strains that fight for survival against the weapons produced by the immune system. If selective pressures -- in this case a high potential sexual transmission -- have forced the virus to evolve toward virulence, the opposite selective pressures could do the reverse. Conceivably, we could "tame" HIV, encouraging it to evolve toward comparative harmlessness. It was already known that preventive measures such as safe sex, fewer partners, clean needles, and so forth could curb the spread of the disease. But Ewald pointed out early on that social modification was a far more potent weapon than anyone realized. Once HIV was cut off from easy access to new hosts, milder strains would flourish -- ones that the host could tolerate for longer and longer periods. Indeed, Ewald argues, given limited public-health budgets, it might make sense to put more money into transmission-prevention programs and less into the search for vaccines. (He also has strong opinions about how drugs should be used to treat AIDS. He asserts that every time we use an antiviral drug like AZT, we produce an array of AZT-resistant HIVs in the population; if viral evolution is taken into account, antiviral drugs can be used more judiciously.)
Ewald's theories tilt him decidedly toward the optimistic camp. Even in the absence of a vaccine the AIDS epidemic will not inevitably worsen; it can be curbed without reducing transmission to zero. A natural experiment now occurring in Japan, he says, could be a test case for his theories. In the early 1990s highly virulent strains of HIV from Thailand took root in Japan, but Ewald predicts that low rates of sexual transmission in that country -- due to widespread condom use and other factors -- will act as a selective pressure on these strains so that they evolve toward mildness. If this is true, the trend should become evident over the next ten years.
Like HIV, many other viruses have an indolent course, with a long latency between infection and the development of symptoms. Herpes zoster, the agent of chicken pox, lingers in the body forever, capable of erupting as painful shingles decades later. There are also so-called hit-and-run infections, in which a pathogen or its products disrupt the body's immunological surveillance system; once the microbes are gone (or when they are present in such low frequency as to be undetectable), the immune response stays stuck in the "on" mode, causing a lingering inflammation. By the time symptoms occur, the microorganism itself has disappeared, and its genome will not be detectable in any tissue.
"The health sciences are still grappling with the masking effects of long delays between the onset of infection and the onset of disease," Ewald says. "Any time you have hit-and-run infections, slow viruses, lingering or relapsing infections, or a time lag between infection and symptoms, the cause and effect is going to be very cryptic. You won't find these newly recognized infections by the methods we used to find old infectious diseases. We have to be ready to think of all sorts of new, clever ways to identify pathogens. We will have to abandon Koch's postulates in some cases."
The online version of this article appears in three parts. Click here to go to part one. Click here to go to part three.
Judith Hooper, a former newspaper reporter and magazine editor, is the author of Would the Buddha Wear a Walkman? (1989) and The 3-Pound Universe (1986).
Illustrations by Dave Jonason
Copyright © 1999 by The Atlantic Monthly Company. All rights reserved.
The Atlantic Monthly; February 1999; A New Germ Theory; 283, No. 2; pages 41 - 53.