Filariasis

In the late 1960s, when I was a student at the All India Institute of Medical Sciences in New Delhi, my classmates and I had a microbiology professor who enjoyed taunting us as we struggled to identify badly preserved, poorly stained slides of parasite larvae and eggs. "You don't know what this is, do you?" he would say, cackling gleefully. "The eye does not see what the mind does not know." In truth, we scientists often don't understand what is staring us in the face. Like everyone else, we see what we see through the lens of a conceptual framework. The history of the treatment of filariasis, and of the research that has been done on the disease, is a perfect example of how a framework can guide, but also limit, our thinking.

The disabling and often disfiguring tropical disease known as lymphatic' filariasis is one of the multitude of diseases for which mosquitoes are the vector. Elephantiasis--the grotesque enlargement of a limb, breast, or scrotum, caused by blockage of the lymph vessels--is one of its most conspicuous manifestations. According to the World Health Organization, filariasis afflicts some 120 million people worldwide, and more than a billion may be at risk of contracting it. Surpassed only by malaria as a cause of human suffering from disease, filariasis imposes an enormous burden of illness, lost productivity, and economic hardship on already-poor countries of the global South.

The nematodes that cause this non-lethal but devastating illness are threadlike parasitic worms, primarily of the species Wuchereria bancroft and Brugia malayi. As with nearly every infection caused by a parasite, the precise mechanism that gives rise to the clinical disease is unknown. One can say with some confidence that none of the most obvious mechanisms are to blame: not the increasing population of larvae inside the human host; not the substances produced by the larvae, either living or dead; not the constant motion of the adult nematodes.

Transmission begins when a female mosquito siphons off a few microliters of blood from an infected individual. Two weeks later, when the ingested nematode larvae have developed into a stage that is infectious to humans, the larvae enter the insect's head. When she bites again, she transfers the nematode larvae to a second person. But the illness may remain asymptomatic for months or even years, leaving many of its carriers hard to identify.

On the basis of their own experiences in treating lymphatic filariasis, many of my medical mentors in India asserted that certain antibiotics were effective against the acute symptoms of the disease. Yet a quarter century ago (and, to a large extent, today as well) Western physicians pooh-poohed the Indian approach and held firmly to the admittedly logical, though in the end incomplete, position that infections caused by nematodes could not be treated with antibiotics. And here the story begins to take some twists.

Antibiotics are small molecules made primarily by soil-dwelling microorganisms of the genus Actinomyces, which compete with bacteria in the same ecosphere. These molecules can kill the bacteria that Actinomyces encounter, but they cannot kill eukaryotic cells--that is, any cell with a true nucleus enclosed by a membrane. Hence most living things made up of eukaryotic cells--and that includes nematodes, people, trees, and virtually anything else nonmicroscopic--are unharmed by antibiotics. So if antibiotics cannot destroy nematodes, how could the Indian physicians have treated a nematode-caused illness by administering them?

Filarial infections, it should be said, have some unusual features. Most people picture a patient with an infectious disease looking feverish, exhausted, and generally sick. Those and other "constitutional symptoms" of infectious illnesses are manifestations of the body's reaction to the invading microorganism; they are not caused by the infectious agent itself. When they detect the presence of alien organisms, the body's white blood cells synthesize proteins that cause a rise in temperature. The response is protective, enhancing the efficacy of the body's defense mechanisms. But one of the cardinal features of many parasitic diseases, particularly infections caused by nematodes, is the near-absence of constitutional symptoms. Nematodes can live in the body without eliciting such responses; even in the face of an active infection, many people do not experience acute symptoms.

Investigators have suggested that the longer two species live together symbiotically, the less chance that either one will disrupt the other's physiology. After all, the parasite needs a living home, not a dead one. Because many nematode infections seem to have coevolved with people over the aeons, most nematodes cause few if any disruptions of human physiology, hence few symptoms of infection. Yet many patients who contract filariasis suffer episodes of high fever, chills, trembling, and rigor. Acute filarial fever, in fact, can often look like an attack of an other disease that is rampant in many of the same countries where filariasis is common: malaria.

Here is another oddity: While the nematodes are living out their four- to six-year life spans in their human hosts, they produce vast numbers of larvae that circulate in the blood. When a mosquito transmits some of those infective larvae to a new human host, the larvae migrate almost immediately to the person's lymph vessels. Because the lymphatic system is a critical component of the mammalian immune system, the nematode's choice of home base might seem peculiar: an invader doesn't usually position itself in the midst of a defending army. Yet the nematodes have clearly adapted to that hostile locale all too well.

The central peculiarity of lymphatic filariasis--the apparent usefulness of antibiotics in treating it--should have been resolved a quarter century ago. At that time, several groups of parasitologists interested in the microanatomy of filarial parasites examined the organisms with electron microscopes. One of the investigators, Wieslaw J. Kozek of the University of Puerto Rico School of Medicine in San Juan, noticed something he had not seen in other nematodes, whether parasitic or free-living. Within the vacuoles, or membrane-bound cavities of the nematode's cells, were even smaller organisms, resembling several genera of intracellular bacteria collectively known as rickettsia.

Bacteria in this group lack cell walls and cannot survive outside the cells of the organisms they parasitize. Kozek not only concluded that what he had detected were bacterial symbionts; he also noted that the bacteria were more numerous in female nematodes, particularly in the uteri of the worms and in their developing embryos. Here, then, was a possible explanation for the effectiveness of antibiotics against filarial nematodes.

Kozek's findings, however, had the misfortune of being unfashionably morphological. Ever since the emergence of what is widely referred to as quantitative biology, any observation that cannot be expressed as statistical analyses or as DNA sequences has generally been greeted with skepticism, if not outright indifference.

As it happened, a more "quantitative" study was done by another parasitologist at about the same time, and it yielded a complementary, though inadvertent, result. Thomas R. Klei, a parasitologist at Louisiana State University in Baton Rouge, had initiated an experiment with the jird Meriones unguiculatus--a docile, gerbil-like desert animal that is susceptible to many of the same parasitic infections that afflict people. After being infected with filariasis, the jirds in Klei's experiment developed an unrelated skin infection that he treated with tetracycline, a broad-spectrum antibiotic. When he and his students examined the jirds at the end of the experiment, they found that the animals treated with tetracycline were free of nematode parasites.

The reigning biological dogma of the time made the finding thoroughly puzzling. After repeating the experiment, with the same result, Klei contacted John W. McCall, a colleague at the University of Georgia in Athens, who had been supplying investigators with the infective larvae of a variety of nematode parasites for several years. It turned out that McCall, too, had noted that filarial parasites did not grow in animals treated with broad-spectrum antibiotics. But McCall told Klei that because he could neither explain nor understand his result, he hadn't published it.

Yet despite this cluster of independent, mutually consistent, and biologically exciting laboratory observations made by Western investigators--and despite the clinical successes achieved by practicing physicians in South Asia--no one picked up the thread until two decades later.

The story resumes in the mid-1990s, when Claudio Bandi of the University of Milan, an expert on bacteria living in insects, sought to determine how commonly other life-forms harbor bacteria within their cells. He was aware of studies done by Kozek and others, noting the presence of bacteria in filarial nematodes. Were these bacteria, Bandi wondered, related to the ones that live in insects? He and his colleagues chose a standard technique for answering such questions: they looked at DNA sequences that code for ribosomal RNA (rDNA). These sequences are present in the cells of all living organisms. Some of the rDNA sequences from heartworms were highly homologous to the rDNA of the arthropod-dwelling bacteria. Undoubtedly, those rDNA sequences had come from the genome of the bacteria that live in the worm, not from the genome of the worm itself.

A second reason for the renewed interest in filarial bacteria was the sequencing of entire genomes of biologically important organisms, such as the laboratory mouse, the worm Caenorhabditis elegans, and, of course, Homo sapiens. The filarial parasite B. malayi was part of the next wave of organisms whose entire genomes were to be sequenced. Even in the early stages of the work on the nematode's genome, Steven A. Williams of Smith College in Northampton, Massachusetts, and Mark Blaxter of the University of Edinburgh in Scotland noted that some of the sequences resembled the ribosomal genes of bacterial cells rather than those of eukaryotic organisms such as B. malayi. But because bacteria such as Escherichia coli are ubiquitous in molecular biology laboratories, the investigators initially thought the sequences were just contaminants.

It soon became clear, however, that the resemblances were not caused by contamination; Williams and Blaxter continued to extract bacterial rDNA from the nematodes even when the nematode samples were extremely clean. Even more telling, the sequences did not resemble the rDNA of E. coli. Instead, they were most homologous to DNA sequences from rickettsia, particularly from members of the genus Wolbachia. Combing through the literature to see if they could learn why the sequences were present, Williams, Blaxter, and others encountered the papers of Kozek, Klei, and McCall. Molecular biologists had rediscovered something that had been known to clinicians and morphologists for a quarter century.

Wolbachia bacteria infect at least 20 percent of all known insect species, disrupting their reproductive lives [see "Invasion of the Gender Benders," by John H. Werren, page 58]. For instance, the sperm of a male insect infected with Wolbachia do not function properly when they fertilize the ova of an uninfected female. But if a female insect is infected with Wolbachia, her ova are compatible with the sperm of both infected and uninfected males. Thus females infected with Wolbachia have a reproductive edge: they produce more progeny. Furthermore, Wolbachia is transmitted from the mother to her progeny, which suggests there will be more infected than uninfected progeny in the next generation. The process has no reproductive benefits for the insects, but it does ensure the rapid spread of the bacteria. Another of Wolbachia's tricks is to turn some insects that were genetic males into sexually functioning females, and that leads to the same end result: an increase in the pool of infected females within the insect population.

All the symbiotic microorganisms being studied in current research on filariasis are Wolbachia. Most insects, however, seem to get along just fine without the bacterium. One insect species may harbor Wolbachia, whereas a second species, belonging to the same genus, may remain entirely uninfected. And when the bacteria within an individual insect are killed by antibiotics, the insect shows no obvious deleterious effects. By contrast, every individual filarial worm belonging to a species known to harbor Wolbachia has been found to be infected with the bacterium. And, as I suggested earlier, neither the worm nor the bacterium can live without each other. Killing the bacteria (by administering antibiotics) leaves the worms unable to develop, to mate, or to generate progeny.

A complementary finding reinforces the same conclusion. The genomes of the Wolbachia species that live within filaria are much smaller than the genomes of the bacteria that inhabit insects. That pattern is common when the relationship between two interacting species becomes fixed and mutually dependent. The smaller organism often jettisons substantial parts of its genome, having come to depend on the larger organism for most of its metabolic requirements.

The rediscovery of the fact that bacteria live within nematodes poses exciting medical possibilities, not only for elephantiasis but also for onchocerciasis, or river blindness, a disease that afflicts millions of people in sub-Saharan Africa. Much of the interest centers on two prescient suggestions made by Kozek and Horacio Figueroa Marroquin in their 1977 paper on Onchocerca volvulus, the filarial worm that causes onchocerciasis. First, they suggested that if the worm depends on the bacteria living inside it for some critical metabolic function, one could treat the disease by killing the bacteria. That suggestion has proved to be entirely warranted. Achim Hoerauf and his colleagues at the Bernhard Nocht Institute for Tropical Medicine in Hamburg, Germany, have shown that giving tetracycline to victims of river blindness destroys the Wolbachia inside O. volvulus. Tests on animals have led to the same result. In addition, a former student of mine, Heidi Smith, who is now a resident in internal medicine at the Dartmouth-Hitchcock Medical Center in Lebanon, New Hampshire, and I have shown that tetracycline prevents the filarial larvae from molting. Hence the antibiotic may be useful as a preventive as well as a treatment.

Kozek and Figueroa Marroquin's second suggestion was that some of the acute inflammation that accompanies filarial infections might be caused by the bacteria living inside the nematodes. Mark J. Taylor, a parasitologist at the Liverpool School of Tropical Medicine in England, and his associates have supported this hypothesis by demonstrating that the inflammation can be attributed to molecules called lipopolysaccharides, which are released by Wolbachia bacteria. More recently Eric Pearlman, a microbiologist at Case Western Reserve University in Cleveland, Ohio, and an international group of collaborators demonstrated that the severe eye pathology that occurs in patients with onchocerciasis might be caused by the same molecules.

The presence of Wolbachia in insects as well as in filarial nematodes raises an intriguing evolutionary possibility. Perhaps, at some stage long ago, the bacteria were transferred from insects to nematodes, since filarial nematodes reside in insects during some stages of their life cycles. But for reasons that are not yet clear, some filarial nematodes do not contain Wolbachia.

Physicians often disregard or even reject certain treatments because they don't "make sense" in the context of mainstream thinking. "The tomato effect: Rejection of highly efficacious therapies," a paper published in the Journal of the American Medical Association in 1984, addresses this troubling, though perhaps understandable, phenomenon within the medical community. In the paper, James S. Goodwin of the University of Texas Medical Branch in Galveston and Jean M. Goodwin note that British colonists refused throughout much of the eighteenth and early nineteenth centuries to cultivate tomatoes in North America, on the grounds that the fruit (a member of the nightshade family) was allegedly poisonous. Yet in Italy people had eaten tomatoes for hundreds of years with no ill effects. The Goodwins compare the episode to the rise, fall, and resurrection of such treatments as giving the plant extract colchicine for the pain of gout.

As the title of the Goodwins' paper implies, the treatments they studied were not among the questionable or even useless remedies that untrained or irresponsible "healers" may offer to desperate people. On the contrary, the treatments were provably effective. Clinicians simply avoided them because accepted theories of disease mechanism and drug action offered no explanation for their efficaciousness.

Today, of course, tomatoes are eaten across the globe. Could a similar reversal be in store for antibiotics in treating lymphatic filariasis? It is hard to understand the persistent lack of interest in exploring such a use of antibiotics unless what we have here is an instance of the tomato effect.

I cannot help but conclude that the scientific community is a microcosm of humanity--unable to appreciate the importance of anything until we are ready to do so, not seeing with the eye what we have not accepted with the mind. I wish it were different. I wish we were more truly scholarly, more humble about our limited understanding of the universe, more ready to accept that a number of things work despite our inability to explain why.

T. V. Rajan is a professor and interim chair of the department of pathology at the University of Connecticut Health Center in Farmington.

COPYRIGHT 2003 American Museum of Natural History
COPYRIGHT 2003 Gale Group