Missing Microbes: How the Overuse of Antibiotics Is Fueling Our Modern Plagues (16 page)

In 1910 a German physiologist, Dragutin Schwarz, recognized that for an ulcer to occur, the stomach must contain acid. Elderly people whose stomach acid had naturally dissipated never got ulcers. Schwarz’s dictum was no acid, no ulcer. So doctors figured that the way to treat ulcers was to reduce stomach acidity. Generations of patients were advised to drink milk, take anti-acids, or undergo surgery that eliminated or reduced the stomach’s ability to produce acid. Moreover, stress seemed to make ulcers worse, which would explain why they waxed and waned. People were urged to control their stress along with their stomach acid. In fact, as a medical student I learned that men with ulcers had trouble getting along with their mothers and that ulcers were one of the best examples of psychosomatic illness. This particular lecture was given by a prominent psychiatrist whose ulcer treatment involved psychotherapy. Not surprisingly, each of the many popular remedies had important limitations, and peptic ulcer disease, as it came to be called, remained a major problem.

Then in 1979 Dr. Robin Warren, a pathologist in Perth, Australia, again noticed bacteria present in the mucous lining of the stomach. Using routine and later specialized stains, he could clearly see the comma and S-shaped bacteria. He further noted that stomach walls of people with bacteria showed signs of inflammation under the microscope or what pathologists like Warren typically call gastritis. Nearly a century after the initial discovery of bacteria in the stomach, Warren realized that the stomach was not sterile after all. It contained bacteria, and he correctly deduced that they must be involved somehow in the inflammation. But what kind of bacteria were they? Why didn’t stomach acid kill them off?

Within a few years, Warren shared his observations with Dr. Barry Marshall, a young trainee who had an “aha moment” of his own. He learned from reading the medical literature that almost everyone who has peptic ulcer disease also has gastritis. If the bacteria were related to gastritis, he reasoned, they also might be related to ulcers. They might even cause these peptic ulcers.

The two researchers studied biopsies from patients with and without ulcers. Nearly everyone with an ulcer had both the S-shaped bacteria and gastritis. But many people without an ulcer also had gastritis and bacteria. They concluded that the mysterious bacteria might be necessary but not sufficient to cause ulcers, just as in the case of gastric acidity.

Doctors (me included) were taught that gastritis is a pathological inflammation of the stomach. But hindsight allows me to question whether it really is pathological or instead a normal condition of the stomach in reaction to coexistence with bacteria. We’ll come back soon to this distinction, which is not just academic but is in fact central to understanding our relationship with
H. pylori
.

In April 1982, using methods developed over the previous few years to isolate
Campylobacter
organisms from fecal specimens, Warren and Marshall cultured the S-shaped gastric bacteria for the first time. They accomplished the feat that had eluded German, Dutch, and Japanese scientists nearly a century earlier. As noted in chapter 1, they first called these bacteria “gastric campylobacter-like organisms” (GCLO), then
Campylobacter pyloridis
, then
Campylobacter pylori
. Several years later, after more extensive study, it became clear that these organisms were not campylobacters at all but previously unknown cousins. That’s when they received their new name:
Helicobacter pylori.
Within months of Warren and Marshall’s first publication in 1983 in the
Lancet
, other investigators began finding these “new” organisms in the stomach and reporting their association with gastritis.

But Marshall wanted proof that these organisms could play a causal role in ulcers and were not just passengers. So in 1984 he used himself as a guinea pig. After testing showed that his stomach was
H. pylori
–free, he swallowed a culture of the organisms. At first nothing happened. But after a few days he developed indigestion. A new biopsy of his stomach revealed the presence of
H. pylori
, but even more important he had gastritis; his stomach hurt, and he had bad breath.

A few days later, a second biopsy showed that the gastritis was largely gone. But because Marshall worried that the organism might persist, he took a single antimicrobial agent, tinidazole, and, as far as what has been published, he was never bothered again by
H. pylori
.

Marshall’s self-experiment showed that
H. pylori
caused the gastritis rather than merely thriving in an environment created by it. But his acute gastritis lasted only a few days before getting better on its own. His condition was different from the usual chronic gastritis, which is present for decades in people with
H. pylori
in their stomachs. Moreover, Marshall took an antibiotic that we now know is ineffective in clearing
H. pylori
when taken alone. So with the benefit of hindsight we know that the infection and inflammation had spontaneously cleared. Most important, Marshall never developed an ulcer.

Nevertheless this dramatic experiment convinced most skeptics to accept the idea that this common organism was indeed a pathogen. Since
H. pylori
caused inflammation, it was obviously a bad microbe. Most people remember the experiment as one in which a crazy but brave Australian drank bacteria and caused an ulcer, thus proving his theory. Of course that is incorrect, but it caught the world’s attention.

Next, to see whether
H. pylori
might have a direct role in causing ulcers or might be just bystanders, Marshall and Warren treated ulcer patients with regimens containing bismuth, an antibacterial agent, or with regimens without bismuth. The results were clear: the rate of ulcer recurrence was much lower in the patients who received bismuth. And other investigators found the same relationships in their own studies.

Doctors could now treat ulcer patients with antibacterial agents, including antibiotics. This was revolutionary. Ulcers could be cured. Good-bye to the idea that stress caused ulcers; hello to microbes.

For their isolation of
H. pylori
in pure culture
,
for establishing its association with gastritis and with peptic ulcer disease, and for changing the treatment of ulcer disease, Marshall and Warren were awarded the Nobel Prize in Physiology or Medicine in 2005. This recognition solidified the notion that
H. pylori
was a major human pathogen and that anyone who had it their stomach would be better off without it.

But many mysteries about ulcer disease remained. Why does it affect men so much more than women, although they carry
H. pylori
at about the same frequency? Even though
H. pylori
is carried from early childhood to old age, why does ulcer disease start to appear in the third decade of life, peak over the next twenty years or so, and then decline? Why does an ulcer form and then heal after a few days or weeks and then recur weeks, months, or years later? Finding the link with
H. pylori
enabled us to better treat ulcers and to prevent their recurrence, yet we still understood little about the biology of the disease.

*   *   *

Having heard the first presentation of Warren and Marshall’s work at the International Campylobacter Workshop in Brussels in 1983, I was initially skeptical, in particular about Marshall’s assertions. Clearly they had discovered a new microbe, but Marshall’s claims about ulcers were not convincingly supported by the evidence he presented. Yet as Marshall and others in the field kept showing relationships between the organism and gastritis and ulcers, I decided that my lab should get involved. In 1985, we began to study the organisms themselves (still called campylobacters) and found that they were diverse, but that people who had them in their stomachs formed antibodies to them in their blood.

In 1987 my longtime collaborator Guillermo Pérez-Pérez and I developed the first blood test to accurately identify carriers of
H. pylori
based on their having antibodies to the organism. Like most scientists, we wanted to know our own status. One of the first things we discovered is that I was positive. I must admit, I was surprised. Like most people in the world who have
H. pylori
in their stomachs, I had no symptoms. My belly felt fine, although when I learned the test results I began to feel a little queasy. But the test opened many windows for us. We could obtain blood specimens from people of all ages all over the world, with different kinds of diseases or none, and, with our test, determine who had the organisms hidden in their stomachs, so we could look for relationships with various illnesses.

I wanted to know why only some of the people with the microbe developed ulcers. We had shown that
H. pylori
strains varied considerably, but we did not know whether these differences would determine whether or not a particular strain would cause disease. For example, nearly all of us carry
E. coli
, which is mostly harmless. Only a few types are very dangerous because they carry genes that code for particular proteins, called virulence factors, that make us sick. We wondered whether any
H. pylori
strains had virulence factors. Could such differences explain who got sick and who did not? Was the observed diversity clinically relevant?

After two years of study, we identified a protein in
H. pylori
that fit the bill. It was essentially always present in the strains found in people with ulcers. People without ulcers had it about 60 percent of the time. So while it seemed necessary for ulcers, it was not sufficient. Still it was a very good lead. Could we find the gene that encoded this protein? In 1989 we created a “library” of
H. pylori
genes within
E. coli
cells. This simply means that we could use the
E. coli
cells as microscopic factories to produce
H. pylori
proteins. Each cell churned out only one or two of the estimated 1,600
H. pylori
proteins. Then we took the blood serum of a person who tested positive for the microbes (again it was me) and screened the library to see if any of the
E. coli
cells produced any proteins recognized by my antibodies. In other words, we went fishing, and we landed a big fish. The very first clone that my serum recognized coded for the same protein that we had associated with ulcers. We named it
cagA
, for cytotoxin-associated gene.

Later we learned how clever these microbes are. These virulent strains contain a cluster of genes that not only make highly interactive proteins, such as CagA, but also form a system for injecting these materials from the bacterial cells into host cells. This meant that my
H. pylori
cells were churning out the CagA protein and constantly injecting it into the cells of my stomach wall. This revved up inflammation, which as far as I was concerned at that time was not a good thing.

A second finding we made at that time was that all
H. pylori
strains possess a protein that in sufficient quantity pokes holes in the epithelial cells that line the stomach wall. Some strains make bigger holes than others by secreting a protein that we discovered and named VacA.

*   *   *

After studying Marshall and Warren’s work showing that
H. pylori
played a role in ulcer disease and gastritis, we had another relationship on our minds: whether the microbe was associated with stomach cancer. Cancer is the main scourge of the human stomach. It is an awful disease. After diagnosis, a person has less than a 10 percent chance of being alive five years later. In 1900 stomach cancer was the leading cause of cancer death in the United States. It still is the number-two cause of cancer death in the world, after lung cancer.

In 1987 we tried to convince the National Cancer Institute to work with us on a possible relationship between
H. pylori
and stomach cancer; however, we were turned down. But two years later I was contacted by Dr. Abraham Nomura, principal investigator of the Japan-Hawaii Cancer Study based in Honolulu. He and his colleagues had done pioneering work on the disease risks of Japanese-Americans living in Hawaii, and he wanted to use our blood test to study the risk for stomach cancer related to
H. pylori
. I jumped at the chance.

Between 1965 and 1968, more than 7,400 Japanese-American men born between 1900 and 1919 enrolled in the Honolulu Heart Study. These men, veterans of the 442nd Regiment who fought in the U.S. Army with great distinction during World War II, were heroes of mine ever since I had learned about them from reading James Michener’s book
Hawaii
as a boy. When Japanese-Americans were being rounded up and incarcerated on the West Coast of the United States, these men risked (and some lost) their lives and limbs to defend their country. The late Senator Daniel Inouye was one of them.

By 1989, blood samples from nearly 6,000 of these veterans had been obtained and frozen. During the interim, more than 137 men developed stomach cancer, and of these 109 could be studied. We matched them with 109 men who did not develop stomach cancer and examined their blood for antibodies to
H. pylori
. One strength of the study is that the blood specimens were obtained an average of more than twelve years before the cancer had been diagnosed. This window of time could help establish a causal relationship.

We asked two simple questions: who had
H. pylori
in their stomach in the 1960s, and did having the organism relate to getting cancer later on?

Our findings were dramatic. We discovered that those who carried
H. pylori
back then were six times more likely to develop stomach cancer over the next twenty-one years than those who were negative. I presented this finding as a “late breaker” at the same conference where Marshall had presented his findings about ulcers eight years earlier. Other parallel studies conducted in California and in England yielded similar results. Later we found that those who had the
cagA
-positive type of strains had double the risk.