In the United States, 70 percent of all deaths are due to chronic afflictions. Unlike the classic scourges of humankind, these illnesses usually strike insidiously. Until the past few years, the mechanisms of their mayhem were all but invisible. But today, a growing number of researchers claim that these disabling conditions may be caused by infection. If true, the repercussions are almost beyond imagining.

To get a sense of this rapidly shifting ground, consider the chronic illnesses not merely suspected but proven to result from infection. All are open-and-shut cases, a single agent provoking a single disease. The human papillomavirus, for instance, causes more than 90 percent of cervical cancer. The hepatitis B virus accounts for more than 60 percent of liver cancer. The hepatitis C virus claims 8,000 to 10,000 lives here each year from complications of chronic liver disease, a number

that could triple by 2010. The first identified retrovirus, HTLV-1, sets the stage for adult T-cell lymphoma decades after the original infection. (Retroviruses, including HIV, are RNA viruses able to make DNA that incorporates itself into the chromosome of an infected cell.) The Epstein-Barr virus, a herpesvirus that triggers mononucleosis in teenagers and young adults, produces in people simultaneously infected with malaria a cancer known as Burkitt’s lymphoma, the leading cause of childhood cancer deaths globally. Human herpesvirus 8 causes Kaposi’s sarcoma, a malignant complication of AIDS. Helicobacter pylori, a spiral-shaped bacterium, induces peptic ulcers. And a host of foodborne pathogens trigger chronic and autoimmune diseases, such as Guillain-Barré syndrome and reactive arthritis.

This may be merely the starting point. Just as the germ theory cleared the way for a deluge of discoveries about the sources of acute infections, today’s changing ideas about the nature of infectious disease, coupled with new techniques in the lab, have led to a rush of claims about the origins of chronic diseases. In a kind of medical Klondike, some of these theories are based on wild hope and dubious information, and will yield nothing. Other discoveries may change the practice of medicine. Researchers have found the Epstein-Barr virus, for example, in Hodgkin’s disease patients and in aggressive breast cancers. Multiple sclerosis (MS) acts suspiciously like an infection, with its odd clustering across populations, its high antibody levels, and its surges and retreats. Juvenile onset diabetes, also known as Type I diabetes, may arise when a Coxsackie B enterovirus elicits an immune response that damages the pancreas. Crohn’s disease, a chronic and painful inflammation of the bowel, looks so much like an infection, the physician for whom it was named remarked in the early 1900s that it resembled an intestinal form of tuberculosis; today, one suspected pathogen is Mycobacterium paratuberculosis, a cousin of the TB bacterium. Up to half of the asthma in this country, as well as a large portion of juvenile rheumatoid arthritis, may be caused by Mycoplasma pneumoniae, a free-living organism smaller than most bacteria. The sexually transmitted organism Ureaplasma urealyticum appears to cause meningitis, pneumonia, and even death in newborns—and it too may

be linked to childhood asthma. Patients with gallstones have far higher numbers of clostridia and eubacteria in their intestines than do healthy people. Kidney stones may be caused by nanobacteria, microbes so tiny they are dwarfed by viruses. A woman’s chances of delivering a premature baby rise dramatically if she has bacterial gum disease; according to one estimate, curing periodontal infection in pregnant women would prevent 45,000 premature low weight births in the United States each year. Schizophrenia, which strikes about one million Americans and impels 10 percent of victims to take their own lives, may be remotely caused by viral infection just before or after birth, which may derail key neural connections in the developing brain; another theory ties the disease to a parasite carried by cats.

Fifty years ago, this list might have sounded preposterous. “People forget how new this history is. We’ve only known about infections for a little over a hundred years,” says Julie Parsonnet, a physician and epidemiologist at Stanford University. “For the first ninety years, all we thought about was: There is this acute infectious agent and it causes this disease. So you get infected with Vibrio cholerae and you get cholera, you get infected with Salmonella typhi and you get typhoid. These are short-term diseases, they come and go, and the organism is gone. This has been most infectious disease research.” Now the questions are different. Can sly viruses and bacteria set off more subtle damage? Do the organisms that live in us all the time, traditionally dismissed as “benign commensals,” actually betray us? Is our old view of infection—the view stamped by the hallowed germ theory—just half of the picture?

Even if scientists link only some chronic conditions to viruses or bacteria, doctors would draw on vaccines and antibiotics to prevent and cure, rather than heavy-duty last-hope drugs or lifestyle therapies that may or may not slow symptoms. “One huge advantage of looking for infections is that you can do something about them,” says Robert Yolken, director of the Stanley Neurovirology Laboratory. “In general, we can deal with infectious diseases much better than we can deal with almost any type of human disease. And it’s easier to treat an infection than to get somebody to change their diet or their lifestyle.”

So dramatic is this turning point in medicine, enthusiasts with an eye to history have already christened it the Second Golden Age of Bacteriology, Koch’s Postulates Part II, and the New Germ Theory. In 1999, David Morens, a medical epidemiologist at the National Institute of Allergy and Infectious Diseases, convened a group of leaders in this emerging field. Before the meeting, he jotted down a long list of recently discovered infectious causes of disease and highly suggestive associations. “When a bureaucrat like me makes a list like that, it can really open your eyes. A large number of diseases that I was taught were chronic when I was in medical school have proven to be infectious.” Morens realized that the quest for hidden agents behind chronic disease is no longer confined to some dreamy future. “It’s happening,” Morens says, “but it’s not happening fast enough.”

The patron saint of the reformed church of infectious causation is Barry Marshall, the Australian gastroenterologist who in the early 1980s discovered the bacterial origins of peptic ulcers, one of the most societally inflected diseases around. The traditional litany of ulcer causes had included stress, smoking, alcohol, excess gastric acid, and genetic predisposition.

Like many in the pathogen discovery game, Marshall and his collaborator, pathologist J. Robin Warren, were unwittingly following in the footsteps of their predecessors. As early as 1892, reports in the medical literature had described a spiral-shaped bacterium that inhabited the stomach. But conventional wisdom then and for the next century stated that bacteria could never survive the sterile, acidic environs of the stomach. In the early 1950s, when a leading American gastroenterologist failed to find signs of infection in more than 1,000 gastric biopsies, the infection hypothesis was put to rest for 30 years. When analyzing biopsied tissue, pathologists set their microscopes at 250 power, since there was nothing to look for but cancer. Beginning in 1979, however, Warren cranked up his microscope to 500 power and noticed in gastritis specimens what looked like S-shaped bacteria. In

1981, he shared his discovery with Marshall, who was immediately fascinated. They wondered if they had stumbled across, in Marshall’s words, “some interesting Australian phenomenon.” By 1982, after a year of analyzing tissue from ulcer patients, Marshall, only 30 years old and still in training at Australia’s Royal Perth Hospital, and Warren, the more seasoned physician to whom he was assigned, were convinced that the bacteria were living brazenly in a zone the medical texts had declared off-limits. “We didn’t set out to find the cause of ulcers,” Marshall says. “It was just an academic exercise.”

Their attempts to grow the bacteria in various culture media failed again and again. That April, an outbreak of drug-resistant Staphylococcus aureus hit the hospital, and the microbiology lab was barraged with specimens. What transpired evokes shades of Fleming’s serendipitous discovery of penicillin. Preoccupied with the staph outbreak, the lab’s technicians left Marshall’s and Warren’s petri dishes to languish in the dark, humid incubator over the long Easter holiday—for five days instead of the usual two. It was enough time for a crop of translucent bumps, known as “water spray colonies,” to grow. After the microbes were smeared on a slide, Marshall peered at them through a microscope and saw dozens of organisms—what he would call, in the title of his now classic 1983 Lancet paper, “Unidentified curved bacilli on gastric epithelium.”

But did they cause ulcers? Marshall took biopsies from healthy people and from ulcer-afflicted patients and, without saying which were which, asked Robin Warren to look for bacteria. As it turned out, everyone with bacteria had an ulcer. And so strong was the correlation between bacteria and stomach inflammation—a prelude to ulcer— Warren “had to buy a new kind of calculator to work it out,” says Marshall. “An eight noughts one.” Put another way, the statistical chance that the association was random was one in one billion. People require bacteria to get inflammation, and inflammation to get ulcers. Marshall reasoned that any drug that removed inflammation would cure ulcers. In 1983, he began successfully treating patients with antibiotics and bismuth. That same year, at an infectious disease conference in Belgium, someone asked Marshall if he thought bacteria caused

at least some ulcer disease. Marshall shot back that he believed bacteria caused all stomach ulcers. “It was annoying to have me standing up pontificating,” he recalls. “With so many ‘experts,’ it was impossible to displace the dogma. Their agenda was to shut me up and get me out of gastroenterology and into general practice in the outback.”

Unfortunately, Marshall couldn’t produce the crowning proof. Try as he did, he couldn’t induce ulcers by feeding the bacterium to piglets. So in June 1984, he threw out the animal model. As he later reported in the Medical Journal of Australia, “a 32-year-old man, a light smoker, and social drinker who had no known gastrointestinal disease or family history of peptic ulceration”—a superb test subject, in other words—“swallowed the growth from a flourishing three day culture of the isolate.” The volunteer was Marshall himself. He had poured five milliliters of beef broth into a petri dish, swirled it around until all the colonies of bacteria dissolved into what looked like murky chicken soup, closed his eyes, threw back his head, and swallowed, gagging a few times. Five days later, Marshall began vomiting every morning— classic symptoms of gastritis. A photo of him taken that week shows a glum-looking young man with dark circles under his eyes. For seven mornings in a row he was sick, suffering the rest of the days and nights with a gnawing sensation somewhere between hunger and nausea.

Helicobacter pylori has since been blamed, not only for the seething inflammation of ulcers (all but the relatively small number of cases produced by high doses of nonsteroidal pain relievers), but also for virtually all stomach cancer cases worldwide. Marshall’s antibiotic cure is widely accepted, and today ulcer treatment is as straightforward as that for strep throat or sinus infection. Researchers in chronic disease regard Marshall’s work as totemic. Marshall himself believes the

prospects for more breakthroughs are endless. “I think a lot of these chronic diseases are in the cards for being infectious,” he says. “If you said to me, ‘What do you think the chances are that antibiotics could protect you against cardiovascular disease?,’ I’d say, ‘Quite a good chance.’”

***

If Barry Marshall is the patron saint, Brent Muhlestein is a recent convert. Muhlestein is one of the leading researchers on the mechanisms of heart disease, the accumulation of fatty plaque in coronary arteries. An interventional cardiologist at Latter Day Saints Hospital in Salt Lake City, he used to perform angioplasties and then helplessly monitor his patients as their scooped-out vessel walls would block up again. In the hopes of predicting whose vessels would and would not close up, he saved the plaque he had cut out and analyzed the specimens’ composition. Then, in 1993, Muhlestein picked up a copy of a publication he usually didn’t peruse, the Journal of Infectious Diseases. Inside was an article that, as Muhlestein remembers, “sounded to me heretical.” In heart vessel plaque taken from patients who had died mostly in automobile accidents, the authors had found Chlamydia-like organisms. “My mind said, ‘What? Atherosclerotic plaque is sterile.’ I couldn’t believe it. But then my thought was, ‘Well, I can prove them wrong by testing my ninety specimens.’” And so, he took his frozen plugs of plaque and looked for Chlamydia pneumoniae, which he knew simply as a respiratory pathogen. “Lo and behold,” he reports, “we found that they were anything but wrong.” Embedded in 79 percent of his plaque samples was the unsuspected bacterium.

Like Barry Marshall, Muhlestein had happened upon a well-trod historical path. In 1859, the pathologist Rudolf Virchow noted in coronary arteries “a stage of irritation preceding the fatty metamorphosis, comparable to the stage of swelling, cloudiness, and enlargement which we see in other inflamed parts.” Virchow wrote that he was siding with what was, even then, considered “the old view in this matter” by holding “an inflammation of the inner arterial coat to be the starting point

of the so-called atheromatous degeneration.” By the end of the nineteenth century, researchers could induce heart disease in animals by injuring the coronary vessels and experimentally infecting the animals. In his 1908 book Modern Medicine, William Osler listed “acute infections” among causes of arteriosclerosis. But during the first few decades of the twentieth century, the focus shifted away from germs and toward the multifactorial or “black box” paradigm, which holds that heart disease results from many different causes working in synergy, from high cholesterol and hypertension to smoking and stress and (what was later disproven) just being male.

In the 1970s, however, the infection theory again began to percolate as scientists pursued new viral and bacterial leads. When chickens were experimentally infected with a herpesvirus, the birds developed arterial lesions resembling those in human atherosclerosis. Then a 1985 Finnish study revealed that heart attack patients had higher levels of antibodies to Chlamydia pneumoniae than did healthy subjects. All this made researchers look twice at the mysterious decline in heart disease that had commenced in the United States in the late 1960s—an unprecedented drop that continues to this day. Doctors knew the drop wasn’t due to a sudden society-wide passion for distance running and macrobiotic cuisine. Could it have come from the widespread use of broad-spectrum antibiotics, which inadvertently killed bacteria in the heart?

Muhlestein decided to explore the bacterial hypothesis in his own lab. He infected rabbits through the nose with C. pneumoniae and found he could induce atherosclerosis. When he raised cholesterol levels in the animals’ diets, the blockage worsened. Other studies have found rising levels of antibodies to the bacterium during the months before a heart attack. People infected with C. pneumoniae are also more than four times as likely to suffer a first ischemic stroke—the kind caused by blockage of a blood vessel—than are uninfected people. As these discoveries piled up, the medical world had finally sanctioned Barry Marshall’s once-dissident belief that bacteria cause ulcers. “The idea that another chronic inflammatory process like peptic ulcer disease was caused by a single infectious agent made me start to think,

‘Well, maybe it’s the same story,’” Muhlestein says. “Maybe there’s just one infectious organism that’s causing atherosclerosis that kills half the world. And if we kill that, maybe we can eliminate this terrible scourge.” As Muhlestein and others would later come to realize, heart disease doesn’t hew to such simple mechanisms.

Yet because it afflicts more than 50 million Americans and each year kills nearly one million, and because by 2020 it will be the leading cause of disability in the world, heart disease remains the prized trophy for microbe hunters. Are Chlamydia pneunomiae the causes of inflammation or merely innocent bystanders conveyed to the site by the immune cells in which they camp out? And what about the rest of the catalog of human misery—the chronic diseases and autoimmune disorders and psychiatric illnesses that continue to defy explanation? Will infections prove to be the long-sought answer or just a mirage?

Until quite recently, the standard in infectious diseases for separating hope from hype, truth from speculation, have been guidelines known as Koch’s postulates. The German country doctor Robert Koch, an enormously gifted lab man with an obsessive drive to answer the big questions in medicine, founded the field of bacteriology. To understand the importance of his postulates, you need to know a little about previous theories of infectious disease.

Ever since Aristotle’s time, observers had believed that minute organisms arose spontaneously in decaying organic matter. One of Louis Pasteur’s great achievements was his insistence that living organisms could not emerge without living precursors—that, as he put it, “the hypothesis of spontaneous generation is a chimera.” This belief stemmed from his work in the 1850s and 1860s on yeasts and fermentation. In 1859, he declared: “Everything indicates that contagious diseases owe their existence to similar causes.”

By demonstrating the life cycle of the anthrax bacterium, Robert Koch showed for the first time that a specific microorganism caused a specific infectious disease. In 1882, before the Physiological Society in

Berlin, he delivered an elegant paper on the etiology of tuberculosis. Paul Ehrlich, a future Nobel laureate then just in his twenties, proclaimed the occasion his “greatest scientific event.” Koch’s work ushered in the fruitful days of the germ theory, in which microbes suddenly seemed to unlock every puzzle. Researchers believed that yeasts could transmute into bacteria and that the hordes of bacteria normally found in tissue all contributed to disease. The enthusiasm engendered by an unsuspected world of disease agents was not unlike today’s. Koch’s sober postulates—a series of steps or procedures that had to be followed in order to prove that a microorganism causes a particular infectious disease (and actually the ideas of Koch’s teacher, the German anatomist Jakob Henle)—was just the antidote to such irrational exuberance. Medicine needed techniques that would permit rigorously controlled experiments in which microscopic organisms could be isolated and handled.

Ironically, Koch himself never formally stated his namesake postulates, though the ideas shine through all his work. The German bacteriologist Friedrich Loeffler published them in a paper on diphtheria. Briefly, the postulates state that:

1. The organism must be present in diseased tissue.

2. The organism must be isolated and grown in pure culture outside the animal host.

3. The organism must be shown to induce the same disease when injected into a healthy animal.

Some reviewers have tacked on a fourth postulate: that the microbe must be isolated from the experimentally inoculated animal in pure culture and shown to be the same as the original agent.

Each infectious disease could be attributed only to a single species of microbe, not to every member of the microbial retinue found at the site. Microbe hunters had to isolate the bacterial culprit and prove its malignity. Those scientific hurdles were really a blessing. Between 1877 and 1897, investigators harvested a microbial bounty, including the agents of typhoid fever, leprosy, glanders, cholera, diph-

theria, tetanus, gas gangrene, plague, dysentery, gonorrhea, cholera, common bacterial pneumonia, bacterial meningitis, and a main cause of food poisoning.

When he volunteered himself as a human guinea pig, Barry Marshall neatly fulfilled all of the postulates. Indeed, Koch’s rules work beautifully whenever disease agents are easily cultured and transmitted, and when symptoms explode soon after infection.

But those same ground rules don’t apply to more subtle medical mysteries. In chronic ailments, the links between infection and symptoms are obscure and covert. This is true for several reasons. First, suspect bacteria and viruses are more difficult to grow in the lab— sometimes impossible, as in the case of the hepatitis C virus. Even at the beginning of the twentieth century, scientists had complained of “less docile” organisms that didn’t thrive in the usual culture media. Second, many human diseases suspected to have an infectious cause don’t have animal equivalents, another roadblock in proving causation. Human studies are out, of course—it’s unethical to subject people to experimental doses of agents that could cause chronic disease years later. Third, there can be yawning delays between infection and disease, so that by the time disease shows up, the agents that planted the original infection may have fled the scene. The seeds of some psychiatric diseases with onset in adulthood, for example, may be viral or bacterial infections that occurred just before or after birth, distorting events during a vulnerable point in the development of the brain.

Most discoveries about chronic infectious diseases over the past two decades—AIDS, adult T-cell leukemia, cervical and liver cancers— have for one reason or another failed to meet Koch’s postulates. Those standards may even have deflected investigators from the new set of clues before them. Writes biologist Paul Ewald, “We need not, however, let the inability to meet the gold standard . . . lead us to be overly skeptical of assigning infectious disease causation based on silver or bronze standards.” Or as microbiologist Hal Nash commented, “The nature of acute disease allowed for the development of ‘rules’ and ‘postulates.’ The nature of chronic disease is likely that there are no rules.”

Rules or no rules, the evidence for infection in chronic diseases is

too obvious to ignore. When an illness arises mostly in people whose immune systems are down, it’s a tipoff to infection (take Kaposi’s sarcoma, a complication of AIDS and transplant patients). When a disease improves with antibiotics (as does strep-induced rheumatic fever), it’s a good bet it’s infectious. The same is true when a disease occurs more often among people living in poverty (as does H. pylori– induced stomach cancer). Or when disease prevalence varies according to time of year (some data suggest that schizophrenia develops slightly more often in people born in the winter months—particularly February and March, the heart of the respiratory disease season). Or when the disease clusters geographically (multiple sclerosis rises in a steady gradient the farther one travels away from the equator, consistent with winter respiratory infections; an upsurge of MS from the early 1940s to the early 1970s on the Faroe Islands, in the North Atlantic, may have been sown by British troops during World War II).

And there’s another clue—a kind of negative clue, if you will— that infections may play a central role in chronic diseases: scientists are beginning to realize how much they don’t know about what’s living in and on the human body. It is estimated that less than 1 percent of all bacterial species have been identified; only a tiny fraction of bacteria and viruses can be cultured with standard methods. Part of the problem is that we humans came late to the party. Prior to the last billion years, virtually all life on Earth was bacterial. From their very genesis, animals have probably carried indigenous bacteria. Even earthworms, which have been around for perhaps 800 million years, harbor such organisms. Having dominated the planet for three-quarters of its existence and having made themselves comfortable in just about every ecological niche—including Homo sapiens—microorganisms not only bombard our bodies but actually make up the fabric of our physical selves. Humans, you might say, evolved in order to haul around bacteria. A piece of doggerel from 1912 titled “On the Antiquity of Microbes” summed up the situation: “Adam had ’em.”

How exactly do we become colonized by these tiny creatures? The fetus normally remains sterile as long as it resides in the uterus, protected by the placenta. Its first exposure to microorganisms usually takes place during passage through the birth canal. Immediately after delivery more organisms swarm in, some becoming permanently established in or on the baby’s body. By the end of the first week or two, the infant has acquired most of the organisms found in the adult. And in most cases, that’s lucky for us. The bowel alone contains hundreds of microbial species, some of which break down food into absorbable nutrients and may disarm potential carcinogens in the diet. In addition, intestinal microorganisms synthesize vitamins. Indigenous flora also stave off colonization by more virulent bacteria and stimulate the immune system. About 1 percent of the human genome consists of endogenous retroviruses—viruses that millions of years ago insinuated themselves into the human genome when they infected the progenitors of humans.

So abundant is our microbial haul that no one has even tried to draw up a complete bill of lading. Our bodies contain at least ten times more bacterial cells than human ones, making us walking petri dishes, and blurring the line between where microbes end and humans begin. Joshua Lederberg has described each human host and its complement of parasites as “a superorganism,” with myriad genomes “yoked into a chimera of sorts.” As Martin Blaser, chairman of the Department of Medicine at New York University, sees it, “The great dilemma of vertebrate biology is: How can we live with all these bacteria we’re carrying? The short answer is that over millions of years nature has provided a way of regulating our immune response so that we’re not so unresponsive that these guys kill us, and we’re not so overresponsive that we destroy normal organs trying to fight these guys.”

In other words, the one germ/one disease idea is too neat to encompass the teeming hordes that make up the human body. A supercomputer could barely keep up with the pathogens that are hitting us from without and from within, agents that potentially work in

concert. “What may be causing chronic diseases,” says Stanford University’s Julie Parsonnet, “is the interplay between our responses to all those organisms simultaneously.”

David Relman, a physician and microbiologist at Stanford, began to grasp the enormity of science’s cluelessness when he set out to catalog bacteria in one of the most prolific bacterial niches: the subgingival crevice, which is the pocket between the gum and tooth. The abundance of action at this site had long been known. In a 1683 letter to the Royal Society, Anton van Leeuwenhoek, the Dutch lens grinder and jack-of-all trades, included drawings of bodies observed in scrapings from his own teeth. Peering through the microscope, he reported “many very little living animalcules, very prettily a-moving”—the first bacteria ever seen and described. Three centuries later, Relman took a sample from a healthy mouth and divided it in two. One half he gave to technicians in an academic hospital clinical microbiology lab, asking that they culture everything they could, separating all the organisms that appeared distinct and different. The other half Relman analyzed using molecular biology tools. While there was a fair amount of overlap in the two sets of results, Relman’s approach uncovered almost ten times as many novel organisms as did traditional culturing. In one tiny sample, he found between 30 and 40 never-before-seen species. The statistical margin of error suggests that at least a dozen more were left undiscovered.

Relman concluded that the old idea of microorganisms either parasitically sapping the human body or benignly renting space doesn’t capture reality. Some of the organisms that compose “endogenous flora”—our tiny corporeal cohabitants—could actually make us sick. Maybe it’s not just barbarous pathogens from outside, ill-adapted to the human body, that give us grief, but also the quiet habitués that seem to biologically blend in and speak the same language. “There may be a whole lot more about what we carry around within our bodies, and what our bodies are in fact made up of, that could be factors in chronic disease,” Relman says, “even in acute disease, for that matter.”

How might these infinitesimal homebodies cook up disease while eluding detection? One strategy is known as persistence. Scientists

used to think that viruses gave themselves away by telltale signs: they killed cells, they turned off protein synthesis, they replicated like mad. Under the microscope, virus-infected cells looked bloated and crummy until they finally fell apart. But viruses that cause persistent infection do not destroy cells. Rather, they can replicate continuously in so-called differentiated cells—those with special functions, such as the neurons that make neurotransmitters and synapses, or the endocrine cells that make hormones such as insulin. Such viral strategies “have been clearly established in animal models and are very likely to occur in humans,” says Michael Oldstone, a virologist at the Scripps Research Institute in La Jolla. “What you wind up with are tissues or cells that look normal under a microscope but in fact contain infectious agents which are turning off the function of the cell. That would reflect, we think, the majority of human conditions that don’t have any kind of etiology associated with them.” Moreover, such diseases as MS, juvenile diabetes, and various psychiatric illnesses may be initiated by certain viral infections and promoted by the immune system’s response to subsequent infections from unrelated agents, a kind of “multiple hit” operation that scientists haven’t been able to pin down.

Microorganisms can also mimic components of the human body and thus spur autoimmune disease. A common denominator in these afflictions—from MS and rheumatoid arthritis to Crohn’s disease and lupus—is chronic inflammation, a sign of possible infection. “One of the ways that infectious agents survive in humans is by making themselves look human, so that our body doesn’t combat them,” explains Julie Parsonnet. “But this system may go slightly awry, where the organisms look just enough off so that the body will respond to both the human cell and the infectious agent, thinking that they’re both bad. The infectious agent may eventually be gone, but the human body will still think a foreigner is present because the human cells will be there.” Autoimmune disease, adds David Relman, is the “consequence of this imperfect balance between the immune system’s need for vigilance and its need for tolerance of structures that should not be attacked.”

Yet exactly which infectious agents provoke autoimmunity, and how, is a mystery. Human immune defenses “are acting as if they would

be reacting to a bug,” says Thomas Quinn, a senior investigator at the National Institute of Allergy and Infectious Diseases. “But we’ve not been able to find the bug.”

One scheme behind autoimmune disease is what scientists call “molecular mimicry.” Organisms have evolved surface substances, called antigens, that closely resemble substances found in human tissue. Antigens are what raise specific antibody and T-cell (or T-lymphocyte) responses from the immune system. Most autoimmune diseases target one organ. The T cells that recognize an antigen on a Coxsackie B virus, for instance, sometimes also react with part of an enzyme in the cells of the pancreas that make insulin. When the T cells attack and destroy the pancreatic cells, diabetes results. Regardless of what specific viruses set off an autoimmune cascade, says Michael Oldstone, subsequent infection from an unrelated virus may exacerbate the damage.

One of the more surprising revelations of immunity gone awry is a childhood form of obsessive–compulsive disorder (OCD). Susan Swedo, a neuropsychiatrist at the National Institute of Mental Health, has found that infections with Streptococcus pyogenes, or Group A strep, in some children lead to a form of OCD sometimes accompanied by tics. The disorder may arise when proteins on the outside of strep bacteria stimulate the immune system to attack human brain tissue containing nearly identical proteins—again, the old trick of molecular mimicry. Swedo named the disorder PANDAS, for pediatric autoimmune neuropsychiatric disorders with strep. But clues to this unusual complication were known as early as the turn of the twentieth century. William Osler, the Canadian-born dean of American medicine, noticed that patients suffering from Sydenham’s chorea—involuntary movements from neurologic complications of rheumatic fever, caused by streptococcus A—exhibited “a certain perseverativeness of behavior.”

“In general,” says Swedo, “other types of obsessive–compulsive disorders start slowly, gradually increase in severity, and if you ask somebody to tell you when it started, they’re not able to do that. It’s

normally, ‘Oh, I had it all my life,’ or ‘I think it was sometime in the first or second grade.’ In these kids, the parents could say, ‘It was the Tuesday before Thanksgiving.’ It came on very quickly, very abruptly, and very severely.” Fortunately, many children return to normal—but if they contract back to back strep infections, their symptoms can last a lifetime. Swedo estimates that about 10 percent of OCD in the United States stems from childhood strep infections—meaning that if 1 in 100 Americans have OCD, 1 in 1,000 have PANDAS. Swedo is now running an antibiotic trial to see if the drugs prevent second strep infections, and thus a lifelong disorder.

The immune system’s attack against retroviruses embedded in the human genome may induce other psychiatric conditions. Pediatrician and virologist Robert Yolken has found that 30 percent of patients suffering first episodes of schizophrenia, and 7 percent with chronic schizophrenia, show signs that an ancient retrovirus in their brain is turned on; the molecular footprint didn’t appear in the brains and cerebrospinal fluid of people who did not have schizophrenia. The virus—which can be activated by herpes simplex virus infections, as well as by hormones and immune cells—may or may not actually cause schizophrenia. To find out, Yolken plans studies in which he administers to these patients antiviral drugs in addition to antipsychotics, in the hope that their symptoms diminish. Yolken and psychiatrist E. Fuller Torrey have also proposed that some cases of schizophrenia and manic depression may result when the fetal brain is infected through the mother with the parasite Toxoplasma gondii, an organism transmitted by cat feces.

If true, such theories may cast a new light on older medical literature. Many nineteenth-century doctors surmised that schizophrenia could be an infectious disorder. “It is certain that there are years when . . . insanity seems suddenly to extend to a great number of individuals,” wrote the French neurologist Jean E. Esquirol. In 1845, when Esquirol observed that “mental alienation is epidemic,” he meant it literally.

Amherst College’s Paul Ewald offers an even more provocative twist on evolution. He contends that many viruses and bacteria have evolved in such a way that it’s in their interest to cause latent, languishing disease—to be “cryptic,” as he puts it, rather than conspicuous. Ewald is not a virologist or molecular biologist or physician; his ideas are rooted in the more speculative realm of evolutionary biology. Nevertheless, a number of researchers and theorists at the front lines of pathogen discovery have been struck by the resonance between their work and Ewald’s. As one virologist put it, Ewald’s theories could raise “red flags” that point scientists to certain infectious agents behind chronic disease.

Like a good evolutionary biologist, Ewald assumes at the outset that the raison d’être of microorganisms and humans alike is to pass on their genes. He goes on to divide human disease into three categories: genetic, environmental, and infectious. Deleterious genes—those that harm survival or reproduction by causing disease—are weeded out by natural selection. Environmentally caused illnesses—such as lung cancer from smoking cigarettes or certain cancers from Agent Orange— are limited both in time and space. And so, Ewald reasons, if a serious and widespread disease is not genetic—that is, if it is too common to have sprung up by random mutation and too ruinous to have survived the culling process of natural selection—and if it is not environmental, it must therefore be infectious. He applies this to today’s most damaging diseases: atherosclerosis, stroke, many if not most cancers, brain disorders, and autoimmune conditions.

Even some diseases we put under the “genetic” heading—such as sickle-cell anemia or cystic fibrosis—are indirectly forged by infectious threats, Ewald theorizes. People with two copies of the sickle cell anemia gene, for instance, suffer the painful blood disorder—but those with one copy of the sickle-cell anemia gene and one normal gene are not only spared sickle-cell anemia but are also less susceptible to the most serious type of malaria. Likewise, healthy people who have one cystic fibrosis gene and one normal gene appear more resistant to ty-

phoid fever. Seen through the lens of evolutionary biology, genetic mutations that occasionally gum up a human machine fine-tuned over millennia by natural selection may actually benefit us by thwarting the equally fine-tuned mechanisms of the microorganisms besieging us.

According to Ewald, if one considered only their effects on human health, scourges such as smallpox, TB, malaria, typhoid, bubonic plague, yellow fever, and cholera could only have been infectious. By the same token, it should have been no surprise that the rising rate of female infertility in the 1970s turned out to be primarily caused by a sexually transmitted agent, Chlamydia trachomatis; more liberal attitudes about sex, coupled with the Pill and other nonbarrier methods of birth control, gave the bacterium a free ride. Even the common diseases of older individuals that don’t impair reproduction—diseases such as atherosclerosis and various forms of dementia—must by a process of elimination be infectious, Ewald contends; since identical twins don’t automatically suffer the same condition, genes can’t explain it all, and since risk factors like diet and smoking fail to predict many cases, neither is environment the last word.

Bucking the conventional wisdom that disease-causing organisms evolve toward benign coexistence with humans, Ewald early in his career proposed that infectious agents can become either more or less virulent over time, depending on how durable they are and how they are transmitted. Drawing on this earlier work, Ewald makes a new prediction: the pathogens behind many chronic diseases are probably transmitted either through sex or through some other intimate contact, such as kissing or hugging. After all, these agents—unlike, say, flu or Ebola viruses—depend on mobile humans to spread. And because they rely on people to get around, it’s not in their interest to quickly dispatch their hosts. According to Ewald, sexually transmitted pathogens “have to have tricks up their sleeves for avoiding the immune system.” They must be infectious for months and transmissible after a long period of infection. No wonder, he says, that today’s leading cast of characters in chronic diseases—the Epstein-Barr virus, HHV8, the human papillomavirus, the hepatitis B virus, as well as HIV—are all spread through intimate contact. Those cunning pathogens are most

likely to escape the immune system’s security forces. “A tremendously disproportionate number of agents responsible for causing chronic diseases,” says Ewald, “will be STPs—sexually transmitted pathogens.”

Delving into medical history, Ewald has concluded that the more cryptic a disease’s chain of transmission, the more reluctant medicine has been to acknowledge an infectious cause (though there are exceptions—dengue and cholera, for instance, were widely speculated to be infectious, though their chains of transmission were elusive). “We’re doing the same thing with cancer, heart disease, diabetes,” he says, “that people were doing in the 1840s with yellow fever and gonorrhea and malaria and diarrheal diseases that were transmitted by water.” Moreover, the triumph of the germ theory led to an arbitrary dichotomy between acute and chronic diseases. Back then, newly found infectious diseases had a defined incubation period and fever as a prominent symptom. Chronic afflictions such as heart disease and cancer were often set aside as degenerative, age-related, and by definition noninfectious. Doctors pictured their complex origins as a web of causation, a crowd of risk factors—in drier language, as “multifactorial.”

But to Ewald, “multifactorial” is a weasel word. He contends that the beauty of scientific medicine is in declaring causal hypotheses, then proving or disproving those claims. “Risk factors, in and of themselves, don’t tell you anything about causation,” he says. “The real answers in nature do involve causes.” Ewald’s stark and simple definition of an infectious disease is as follows: a disease that would be eliminated if its infectious agent were eliminated. And he applies it to acute and chronic diseases alike.

The only way to get to the real answers—and here, Ewald and others in the field agree—is to perfect new methods of ferreting out disease-causing organisms, to catch them in their tracks. “New tools will allow us to make cryptic infectious agents less cryptic,” says Ewald—words that could easily have been the rallying cry of van Leeuwenhoek or Koch.

The father of bacteriology was also the father of invention. To abet his rural medical practice, Robert Koch built a state-of-the-art microbiology lab. An ingenious innovator, he adapted the light microscope to better magnify bacteria, contrived fixing and staining methods to better see them, developed solid nutrient media techniques for growing them, and was the first to publish their photomicrographic portraits. As much as anything, it was Koch’s technical achievements that established the germ theory. Today, the petri dish and tissue staining have given way to nucleic acid amplification and gene chips. These molecular biology techniques go beyond appearances to plumb pathogens’ genetic makeup, an infinitely more precise way to distinguish one agent from another. “It’s like the difference between looking at a computer screen that only has three hundred lines versus twelve hundred,” says Stanford University microbiologist Stanley Falkow. “You can see things more sharply and clearly.” Though the leads for today’s explorations have been known for decades, adds Stanford’s David

Relman, “They’ve been buried within the pathology literature and some of the clinical literature. They could only remain theories. There weren’t the means of pursuing these ideas experimentally—until now.” Relman helped carve out the nascent field of pathogen discovery. His interest was first piqued in 1989 at a weekly clinical conference. Doctors had noticed hordes of curved rod bacilli in a disease called bacillary angiomatosis, a complication that causes skin lesions and systemic problems in AIDS patients and other immunocompromised people—but the bacterium refused to replicate in culture. Relman tackled the problem by throwing out Koch’s postulates. Instead, he employed a technique called consensus polymerase chain reaction (PCR). Using snatches of genetic sequences common to all bacteria as a kind of bait, he snagged and then copied a small piece of the unknown bacterium’s genome. Comparing that piece with the genes of known agents, Relman stumbled upon a surprise: the bacillary angiomatosis bacterium was closely related to Bartonella quintana, the

agent of World War I trench fever. When Relman further refined the identification, he came up with another surprise: bacillary angiomatosis came from the same bacterium—Bartonella henselae—that causes cat scratch disease, a mild, flu-like illness in humans. In the ever-growing annals of emerging infections, here was another case of an old bug showing up in a new host. Relman followed up his tour de force by using the same genetic angling techniques to find the bacterium that causes Whipple’s disease, a puzzling wasting syndrome long suspected to be infectious. But other scientists were reluctant to accept his findings. “We simply had a DNA sequence out of these tissues—we didn’t have a living organism,” Relman recalls. “For a lot of people, that didn’t feel quite right.” Today, however, Relman’s experiments are considered critical turning points.

Another technique—representational difference analysis, or RDA—selectively amplifies rare genetic fragments that show up in diseased tissue but not in otherwise matched normal tissue. This was how Yuan Chang, a pathologist at Columbia University, discovered the infection that causes Kaposi’s sarcoma, or KS.

Before the era of AIDS, Kaposi’s sarcoma, a rare malignancy that causes purple blotches on the skin, was known mainly as a benign cancer in elderly men of the Mediterranean and the Middle East, and in men of Jewish descent (the so-called classic form of the disease); as endemic disease in equatorial Africa; and as an occasional side effect in immunosuppressed organ transplant patients. But in 1981, the cancer entered the spotlight when it alerted doctors to the AIDS epidemic. Early in the epidemic, the risk of developing KS was 10,000 times higher among HIV-positive patients than in the general population. (In fact, one of the original names for AIDS at the CDC was Kaposi Sarcoma–Opportunistic Infection.)

For years afterward, doctors assumed these KS lesions resulted from HIV’s own effects on endothelial cells. But by the early 1990s, Chang and her husband, epidemiologist Patrick Moore, had other ideas. Certain peculiarities hinted that KS sprang from an infection. Epidemiologically, it seemed to spread like one. While HIV-positive gay men often developed KS, hemophiliacs with HIV had very low

rates of the cancer, suggesting that it might be sexually transmitted. The classic and endemic forms of the disease followed a patchy geographic pattern, another clue to infection.

Moore, who had researched arboviruses for the CDC, worked for the New York City health department. Chang, a pathologist specializing in brain cancer, had just set up, on a shoestring budget, a molecular diagnostic lab at Columbia. As they were pondering the problem in 1993, scientists at Cold Spring Harbor Laboratory published the first paper on representational difference analysis. “It was a really beautiful technique,” says Chang. “We just had to try it.”

Theirs was a magnificent obsession. To carry out their experiment, they each worked their regular jobs and then holed up until late at night in Chang’s lab. With no padding in her budget, Chang borrowed reagents from other laboratories. “We were working on average fourteen to sixteen hours a day,” Patrick Moore says, “getting three or four hours’ sleep in between the chores. That went on seven days a week for weeks and weeks.” Adds Chang, “We knew all the gypsy cab drivers around Columbia.” Most of their time was spent trying to prove themselves wrong—“because,” explains Moore, “you can fall in love with a hypothesis and it can really trick you.”

Using RDA on KS lesions, Chang found two small viral fragments. She and her husband later showed that the virus was present not only in AIDS patients with Kaposi’s sarcoma but in all patients who had KS. The scientists eventually devised a blood test and a genetic test that picks up the virus and predicts disease before symptoms ever arise. Searches through a gene library revealed that the virus resembled the Epstein-Barr herpesvirus, one of just a handful of viruses known to cause tumors. Eventually, the 34-year-old Chang proved the cancer was caused by a never-before-seen herpesvirus, which now variously goes by the names human herpesvirus 8 (HHV8) and Kaposi’s sarcoma herpesvirus (KSHV). The Chang and Moore theory may have been bolstered by a 2001 report from University of Michigan scientists, who found that another herpesvirus, the Epstein-Barr virus, disables a brake on cell migration and thus permits cells to spread and metastasize.

Key discoveries so far about the biology of cancer have all come from learning how tumor viruses convert normal cells into cancer cells. In 1909, the Rockefeller Institute’s Peyton Rous discovered the first known tumor virus in chickens—though it was not until 1966, when Rous was in his mid-eighties, that he finally was awarded the Nobel Prize in medicine and physiology. Indeed, five recent Nobel prizes awarded for cancer research have gone to scientists studying tumor viruses (in 1926, the Danish pathologist Johannes Fibiger won for his discovery of a parasitic cause of stomach cancer in rats, which was used as a method of inducing cancer). What makes these viruses so illuminating is that they turn on and off many of the same genes that, when spontaneously mutated, cause cells to turn malignant. In a somewhat controversial theory, Chang and Moore now propose that the same mechanisms designed by the immune system to control viral infections are intertwined with those designed to hem in tumors. “When the virus is preventing an anti-viral response from occurring—to save its own life, to maintain itself, and to be transmitted—it accidentally can trigger a cell into a transformed or cancerous phenotype,” says Moore, who now teaches at Columbia. This implies that if researchers want to find out how all cancer starts, they should study the cellular checkpoints erected to halt viral infections.

Chang and Moore’s work, now embraced by their colleagues, was the first high-profile discovery of a new agent using methods that left cell and tissue culturing in the dust. “It was very high risk. It was a new technique,” says Chang. “A group of people had been looking for a long time and were naysaying the idea of an infectious agent.” When her first paper on the virus was published in Science in 1994, an accompanying editorial cast some doubts on the finding. “It’s really good work, and it could be a substantial advance,” commented the National Cancer Institute’s Robert Gallo, one of the co-discoverers of the AIDS virus. “But I have major questions.” Chang vividly remembers reading those words. “It was like the 600 pound gorilla stomping on a bug.” Nevertheless, she has embarked on more searches. “I feel very strongly that other chronic diseases and cancers are caused by infectious agents

5 that have not been identified yet, or where the association has not yet been made.”

An even newer generation of technologies—known as DNA chips or DNA microarrays—can register tens of thousands of genetic sequences in viruses or bacteria at once. These chips are rectangles of glass or silicon roughly the size of a microscope slide that contain thousands of known gene sequences, to which complementary sequences from a patient’s tissue will bind. David Relman and others suggest that such chips could be used to regularly survey normal bacteria in the mouth or bowel or on the skin—creating a detailed portrait of healthy tissue. Doctors could then compare at a molecular level this inventory of “normal” body flora against disease aberrations as they unfold, when they are presumably more treatable. Other chips can simultaneously register the turning on and off of thousands of human genes in response to infectious agents. Indeed, scientists envision a not-too-distant future when they will actually be able to diagnose disease by analyzing the unique pattern of expression as genes react to infections. As David Relman and colleague Craig Cummings have written, “We are on the verge of being able to listen to the two-way conversation between pathogen and host through devices of immense power.”

As these new techniques suggest, the easy part will be finding new bacteria and viruses, or cornering them in surprising hideouts. The hard part will be proving in a scientific court of law that they really do cause disease and are not merely hanging around the scene of the crime. “That’s the downside to all this,” says the National Institute of Allergy and Infectious Diseases’ Thomas Quinn. “If you rush to publish without really good evidence and support, it discredits the field.” So many agents have been floated as the “germs” behind so many conditions—even obesity—headline writers have had a field day. But unlike H. pylori and ulcers, further research often paints a more ambiguous picture.

The search for the genesis of heart disease illustrates these problems. Researchers know that patients with atherosclerosis have higher levels of antibodies against not only C. pneumoniae but also H. pylori, cytomegalovirus, herpes simplex viruses, Mycoplasma pneumoniae, and the hepatitis A virus. Brent Muhlestein, an early believer in the Chlamydia hypothesis, now concedes, along with many cardiologists, that the issues are messy. “Maybe coronary heart disease is not going to be due to a simple single infectious agent,” he says. “It would more likely be associated with a variety of infectious agents and probably with other risk factors. Atherosclerosis is a tremendously multifactorial disease. We may never find one single unifying cause. Although one can still dream of such a thing.” Supporting this idea, a 2001 study by Austrian and Italian researchers found that chronic infections of all kinds— from gum disease and bronchitis to sinus and urinary tract infections— triple the risk of atherosclerosis.

Other heart specialists go so far as to cast doubt on the whole infection theory. While atherosclerosis is clearly an inflammatory disorder, they say, the inflammation could spring from lots of things besides infection—the oxidation of “bad” LDL cholesterol, toxins from cigarettes, a misdirected autoimmune reaction. That people who have had heart attacks in the past exhibit higher levels of antibodies to certain respiratory organisms could just mean that they were sicker to begin with—that their lungs were weaker and they were more prone to infection. When Paul Ridker, a heart researcher at Boston’s Brigham and Women’s Hospital, published a study in 1997 linking heart disease to inflammation, Newsweek blared on a lurid red-and-yellow cover a story about “The Hidden Causes of Heart Attack,” including bacteria—a theory that Ridker himself all but dismisses. “My experience with the Newsweek cover was: It’s so easy for the lay public to equate inflammation, which they don’t understand, to infection, which is so intuitive,” he says. Many of the tantalizing reports linking microorganisms and heart disease that do hit the daily news cycle “would not pass scientific muster,” Ridker adds, because they narrowly focus on one experiment and do not take into account the massive backlog of contradictory studies.

Even if the infection hypothesis does turn out to be true, treatment would be tricky. Vaccines targeted against the organisms that cause heart disease could actually aggravate any autoimmune process triggered by the pathogen, since the vaccine would provoke the immune system in the same way as the infectious agent. Antibiotics present problems of their own. For one thing, since these drugs don’t kill specific bacteria but rather a broad range of microbes, doctors may not know what particular pathogen may be causing heart disease. And, of course, as we’ve seen in Chapter 4, long-term use of broad-spectrum antibiotics forces the evolution of bacterial strains that can withstand the drug. “If I give 10,000 people an antibiotic for a year or two years or five years or ten years, the one thing I can guarantee you is they will start developing resistant organisms,” says Ridker. “Until you have definitive evidence that you are doing some benefit, you damn well better do no harm.” To help settle these questions, scientists have enrolled more than 10,000 people in prospective clinical trials that will try to determine, as a starting point, whether antibiotics do reduce heart disease. Similarly, in 2001, Australian researchers began a clinical trial in which Crohn’s disease patients were treated with antibiotics.

The quest to eradicate chronic diseases may also bring unanticipated problems. Take the case of Helicobacter pylori. Barry Marshall’s curative treatment for ulcers, together with the wider use of broad-spectrum antibiotics, has killed off loads of H. pylori among residents of developed nations. That’s good news if you’re prone to ulcers. But some scientists believe that the bacterium, while causing ulcers, also protects against gastric reflux by producing an enzyme that reduces acidity (and makes it possible for H. pylori to survive in the stomach). So as ulcers decrease, reflux and its complications may rise. In the United States, adenocarcinoma of the esophagus is the fastest increasing cancer, and reflux is a key risk factor. “For the first time in human history, you have large numbers of people who are reaching their fourth, fifth, sixth decade of life without Helicobacter who still have full acid-producing mechanisms, and that’s leading to reflux and its sequelae,” says New York University’s Martin Blaser. “One is a known risk that is decreasing, and the other is an unknown risk that’s

increasing and we don’t know where it’s going to end.” Blaser is convinced that H. pylori evolved in our species not so much to protect against reflux as against infant diarrhea. “We have evolved with a certain microbial flora over the last millions of years,” he says. “Helicobacter was normal flora, and the twentieth century is the aberration. In brief, don’t mess with Mother Nature.”

Perhaps the biggest paradox about the emerging field of pathogen discovery is that, despite its promise, funders are wary. “When you do fishing expedition research, the hits are few and the misses are many. Science wants to be safe,” explains the National Institute of Allergy and Infectious Diseases’ David Morens. Mainstream institutions favor “hypothesis-driven” rather than “hypothesis-generating” research. In the former, a scientist starts with a supposition and conducts the experiment to prove or disprove the idea; whatever the results, at least in the end there’s a paper to write up, something to show for the work. But in hypothesis-generating research—the engine behind today’s search for new pathogens—the scientist inches forward by hunch and intuition, gathering clues, speculating on what they mean. “Nobody’s funding hypothesis-generating research,” says Morens. “Within science, that’s considered the lowest level of research—out of the primeval ooze.”

But anyone who searches for new pathogens must wander over that muddy, uncharted terrain. Yuan Chang took the risk—and it paid off. But while “KS was great at perking people’s ears up and alerting them to the fact that it can be done,” she says, “most associations that will be found in the future will be a little less clear than the epidemiology for KS. The only way to look for new infectious agents or a new association between an infectious agent and a disease is to have a lab devoted full-time to that.”

Scientists who go out on a limb often complain that the researchers who volunteer to sit on the National Institutes of Health study sections that approve or reject research proposals are not sophisticated or imaginative enough to assess more daring scientific visions. That bureaucratic process, they say, leaches innovation from the field. In pathogen discovery and in other realms, the answer may be to finan-

cially support the best minds for long stints, letting them go wherever their ideas take them.

The science establishment’s reluctance to pursue unproven ideas is a hurdle medical oncologist Beatriz G.-T. Pogo knows well. At the Mt. Sinai School of Medicine in New York, Pogo has tested more than 1,000 breast tumor samples from the operating room and has found that more than a third have genetic swatches nearly identical to mouse mammary tumor virus, a slow-growing retrovirus that is transmitted through mother’s milk in mice and causes cancer in female progeny. In these human tissues, the virus appears to be exogenous—i.e., it is not part of the women’s own genome but rather comes from outside—and it virtually is never found in normal breast tissue. Pogo doesn’t know how the virus is transmitted, nor exactly how it infects cells. “I am the first one to acknowledge there are many holes in this story,” says Pogo. But the very suggestion that some breast cancer could be caused by infection raises the hackles of the medical community. Pogo’s hypothesis must compete with theories about genetics and hormone exposure—ideas that have a lot of good data and adherents behind them. Before she embarked on this research, in the early 1990s, Pogo specialized in studying pox viruses, another group known to cause cancer, and she was well sponsored. Not so once she turned to potential infectious causes of breast cancer. “All my life, prior to this subject, I have been funded by NIH,” she lamented. “But this subject has been considered high risk.”

Barry Marshall, who broke the ground that others are now sowing, understands these feelings. Had he not discovered in the 1980s that bacteria cause ulcers, he says, today he might not get funding to look. Proceeding on intuition, Marshall says, “is a luxury that not many researchers have. It was always said to me that if you want to get research funding you better make sure that you’ve got the experiment half done. You have to prove it works before they’ll fund you to test it out.” And it helps to be something of an outsider. “The people who have got a stake in the old technology are never the ones to embrace the new technology. It’s always someone a bit on the periphery—who

hasn’t got anything to gain by the status quo—who is interested in changing it.”

Even if only some of these hunches prove right, the payoff could be enormous. Doctors will be able to define chronic diseases by the pathogens that cause them and single out who is at risk. Indeed, they may someday be able to predict who is likely to develop a chronic disease even before it starts. If antivirals or antibiotics stop the problem, the next step would be immunization to prevent it. This is what Robert Yolken calls Koch’s Fifth Postulate: “The true test will be removing these organisms and then seeing what happens to the disease. In Koch’s time, they didn’t have this luxury.”

“If we were to eliminate all these infectious influences, maybe people would be quite healthy,” says Paul Ewald. “They wouldn’t have arthritis, they wouldn’t have atherosclerosis, they wouldn’t have Alzheimer’s, they wouldn’t have much of any cancer. If you take out all of those things that look like they might be caused by infection, maybe you would end up with people who are quite vigorous. Maybe at the age of ninety, one hundred, one hundred and ten, everything would just fall apart. We just don’t know what real encoded senescence is.”

But others are more reserved in their enthusiasm. “We’ve gone through these cycles before, where everyone thinks science is going to cure and solve everything. And it just doesn’t,” says Stanford’s Julie Parsonnet. Two decades after scientists discovered the AIDS virus, for instance, there is still no reliable drug cure or vaccine. Unlike the years just after the germ theory was accepted, when acute infections fell like dominoes, chronic diseases may stand their ground for a long while.