The Devil's Disease: Dengue Fever

From witnesses hundreds, even thousands, of years ago to recent research observations, a theory begins to take shape about the origins of ka-dinga pepo.

Some 2,000 years ago in the Nile region of Egypt, a deadly pathogen confined within a specific species of mosquito found a way to thrive in a new host: human beings. Curiously, a legend among peoples in those ancient times bears striking parallels. In it, Allah punishes a sinful leader called Nimrod by inserting a mosquito into his brain. Driven mad by the insect’s buzzing, Nimrod begs a servant to crack open his skull, allowing the mosquito to fly free.

Ka-dinga pepo, marked by its bright red rash, first appeared in isolated epidemics in tropical and subtropical regions. But as the centuries passed, the mosquito that transmitted the virus stowed away with slave traders and rum-runners, slowly taking hold in new surroundings across the world. By the 17th century, it reached the docks of Boston and Philadelphia.

Everywhere, it infected humans—most frequently children—with spiked fevers, terrible pain in joints, muscles, bones and behind the eyes. It could even cause blood to ooze through the pores. It acquired a variety of graphic names, the best known of them attributed to Dr. Benjamin Rush, a signer of the Declaration of Independence. Rush treated an outbreak in 1780 Philadelphia and, observing the misery afflicting its victims, called it “Break Bone Fever.” In Swahili, however, it was still known as ka-dinga pepo, a disease of the devil. It was but a short linguistic jump for it to become known worldwide as dengue fever.

Whatever its origins, scientists, medical professionals and disease control experts are concerned now with the most recent history of dengue (pronounced DEN-ghee). Before 1950, a typical world map depicting affected regions contained few flecks of color. A dab in Africa, a small glob in Southeast Asia, a sliver of color in South America. Today, it’s as if a can of paint spilled across the bottom half of the map.

As it has spread, the most dangerous form of the disease—dengue hemorrhagic fever (DHF)—has appeared with alarming regularity. According to Anna Durbin, MD, an associate professor in International Health, the often-fatal DHF causes vascular leak syndrome where fluid in the blood vessels leaks through the skin and also into spaces around the lungs and belly. Blood pressure falls, shock sets in and death often follows.

DHF ran rampant in Southeast Asia in the 1960s and 1970s, and has increased its presence ever since. The WHO estimates that 2.5 billion people are at risk annually for dengue infections. Between 50 million and 100 million contract the disease each year. More than half a million of those are diagnosed with DHF.

Compared to malaria, dengue has a relatively low mortality rate. Malaria, caused by a parasite transmitted by a different mosquito species, infects up to 400 million people each year and causes 1 million deaths. While dengue’s annual mortality rate is about 26,000, it’s the dengue morbidity rate that causes widespread concern. WHO reports a 30-fold increase in cases since 1960. It has become the world’s most common viral disease spread by blood-sucking insects. Editors of the recent book Frontiers in Dengue Virus Research call it “an old disease spreading with a new vengeance.”

Many trace the spread of dengue to the upheavals of World War II. Fighting in the Pacific meant more mosquitoes were transported via wartime shipping. It was also during the war that scientists used a potent chemical against the Anopheles gambiae mosquito that transmits malaria. In the mid-’50s this chemical—DDT—became the main tool of a worldwide malaria eradication program promoted by WHO. It killed not only the malaria-carrying Anopheles gambiae but the Aedes aegypti mosquito that carried the dengue pathogen, effectively eliminating malaria and dengue fever through Central and South America.

By the late ’60s, however, DDT became anathema for health and environmental reasons. This fact coupled with the decline of malaria eradication programs allowed the mosquitoes and dengue to return.

The only way dengue can spread is when the Aedes aegypti mosquito bites someone who is infected and then bites a second human who is not. The more people who live in close surroundings where mosquito control is weak or nonexistent, the more likely a mosquito will transmit the virus to others.

Dengue has reached endemic status, says the CDC, in many tropical and subtropical regions. It reappears annually during the rainy season when mosquitoes breed.

According to the WHO, dengue is the leading cause of hospitalization among children in eight tropical Asian countries. It often overwhelms the national health budgets and fragile health care systems of developing countries. Many scientists refer to it as a disease of poverty.

That doesn’t mean that a wealthy nation like the U.S. is immune. In fact, a small outbreak of dengue fever was reported in the Texas border city of Brownsville in 2005. But generally, dengue is rare in the U.S. even though Aedes aegypti is common in the South. Experts attribute that to infrequent contact between people and the mosquito, due to better sanitation conditions in American cities, as well as window screens and central air conditioning.

“We don’t think actual transmission will happen in the next year or two in the U.S., but we recognize that it could happen,” she says.

Durbin is conducting clinical trials in Baltimore—and later this summer in Brazil—with an experimental dengue vaccine created by NIH. She is also part of the Pediatric Dengue Vaccine Initiative (PDVI), an international consortium of scientists, drug manufacturers and public health institutions. It was founded in 2003 by Scott Halstead, MD, a world authority on dengue research and an adjunct senior scientist in the W. Harry Feinstone Department of Molecular Microbiology and Immunology (MMI).

Dengue’s spread is only part of the story. For dengue fever is not just a virus, it is four viruses. And in a given year, a region may experience an outbreak of one or two of those virus serotypes, and sometimes all four at the same time.

The challenge of a virus, according to George Dimopoulos, PhD, an MMI associate professor, “is its ability to rapidly mutate, making it difficult for the human immune system to resist it.”

With most viruses, a survivor is usually immune to subsequent infections because of antibodies that alert the body to resist new viral invasions. Vaccines—usually weakened forms of a virus—build similar antibodies for resistance. With the eradication of smallpox and near-eradication of polio, vaccine-fostered immunity is a sacred given in public health.

But dengue fever commits sacrilege and it even has its own name: antibody-dependent enhancement. “There’s one thing we are fairly certain of,” says Durbin. “If you have an antibody to a serotype, say Dengue 1, and you are infected with Dengue 2 virus, your Dengue 1 antibody won’t protect you.”

That’s because, says Durbin, the antibody binds itself to the new dengue strain, in effect joining forces with the new virus and helping it gain entry into target cells where it can then replicate. As a result, the bloodstream now contains higher levels of the virus than during the first infection. Often this triggers dengue hemorrhagic fever. Any of the four dengue serotypes can combine with the antibody from any other serotype to cause DHF.

It is, according to Halstead, “a most amazing perversion of the immune response.”

“The cells that are supposed to scout out and kill viruses and the antibodies that are supposed to destroy viruses form an unholy complex to defeat our immune system and promote the life of the dengue virus,” he told a New York Academy of Sciences gathering last year.

The ramifications for dengue fever vaccine development are daunting: A dengue vaccine must be able to combat all four serotypes at the same time. But the hazards only begin there.

“In the vaccine world, two big questions overhang our studies,” says Durbin. “One is close to being answered: If we introduce a live vaccine into people who have a pre-existing antibody, can that vaccine cause severe disease?”

Because a vaccine will essentially be a weakened form of the disease, she says, most researchers believe it’s not likely to occur.

“The greater concern,” she adds, “is what happens if the vaccine we introduce doesn’t produce a balanced immune response in individuals and, over time, their antibodies to various serotypes decline at different levels? And supposing an individual never did develop a good response to one of the serotypes. Are we going to put people at risk for more severe disease months or years down the road? Nobody knows the answer. The only thing that will answer it is long-term surveillance studies of vaccinated populations.”

There are other troubling questions. “We think a vaccine will reduce transmission of dengue or reduce the circulation of dengue in endemic areas,” she says. “But what happens if it reduces the transmissibility below the level that will sustain dengue in a given community? If people aren’t being exposed, will their immunity drop and make them more susceptible should somebody come into their region and reintroduce dengue? We just don’t know. We won’t until we follow them over years and see if they are able to maintain antibody levels or not.”

Most dengue research comes on what scientists refer to as “the human side” of the disease. Far fewer researchers attempt to solve the dengue riddle by understanding the biology of Aedes aegypti, the territory held by those on “the mosquito side.”

George Dimopoulos has built an international reputation as an authority on mosquitoes that carry pathogens for malaria and dengue. In 2001, he started the Dimopoulos Group, a research lab at the Imperial College of London. In 2003, his group became part of the Johns Hopkins Malaria Research Institute in the School’s MMI Department.

“I truly believe,” he says, “that we will never be able to design an efficient strategy to eliminate dengue and other mosquito-borne diseases if we don’t understand the biology of the pathogen and how it interacts with the human and the mosquito.”

Dimopoulos carries out his research in CDC-approved, Biosafety level 2 and 3 facilities housing tens of thousands of mosquitoes. Collected during field trips to tropical regions, some of the mosquitoes are Anopheles gambiae that can carry Plasmodium, the malaria-causing parasite; others are Aedes aegypti.Most are harmless, not yet infected with a pathogen.

Those that have been infected—in one of four brightly lit, white-on-white contaminant labs—are kept in carefully screened cages in small closets off the labs. Temperature and humidity level have been adjusted to tropical conditions. Here the mosquitoes lay eggs in special trays and are fed on human or mouse blood.

To an untrained eye, the Anopheles and Aedes mosquitoes look the same. But, it’s what’s inside of them that intrigues Dimopoulos. He notes that the Plasmodium parasite that causes malaria cannot infect the Aedesmosquito nor can the dengue virus infect the Anopheles.

“It’s not known why that is,” says Dimopoulos. “A pathogen may, for example, require certain receptors in the mosquito’s gut—factors that are present in one species but not in the other—to establish infection.”

A virus cannot propagate independently of its host. The malaria parasite, with its 6,000 genes, replicates its DNA on its own while dengue, with only 10 genes, has to use the machinery of the host cell to propagate.

Unlike the malaria parasite, once the dengue virus has taken hold in the mosquito’s cells, says Dimopoulos, the way it interacts parallels the way it behaves in a human cell. “It’s probably utilizing very similar cellular machinery in [both] its hosts,” he says. “That is why research we do on dengue in the mosquito may also be used to understand how dengue infects humans. The actions are quite similar.”

In a manner very close to the way the human immune system tries to fight off an invasion by a dengue virus, the Aedes aegypti flexes its own defenses against viral pathogens. Researchers know of three immune-related cellular pathways in the Aedes mosquito. In previous work, Dimopoulos’s group has shown that two of them—the Toll and JAK-STAT pathways—defend against the dengue virus. “These defense systems do not seem to be sufficient in the Aedes mosquito,” says Dimopoulos. If they were, the mosquito could fight off the infection. “But these pathways are doing something.”

It’s even probable, that this anti-dengue response may be sufficient against dengue in different mosquito strains and species, explaining why they can’t be infected by the virus. “But we have just started to understand these defenses and their pathways and we don’t really understand the resistance mechanisms yet.”

Historically, notes Dimopoulos, wherever dengue and malaria have been controlled it has been through the mosquito—either through avoidance strategies or with insecticides. But he believes that if there were a way to manipulate some of the factors in the cells of the mosquito’s gut—those that allow these pathogens to propagate, or those that act against the pathogens—the virus and the parasites could not survive. 

“So we are pursuing a different strategy,” he says. “Our idea is to cure the mosquito—by making it resistant to the pathogens.”

Even so, Dimopoulos believes it will take a combination of strategies to solve the dengue problem. One of those strategies comes from neither the human nor the mosquito side. Derek Cummings, PhD, an assistant professor of Epidemiology and International Health, comes down on the side of algorithms.

Using self-designed computer models, Cummings crunches and analyzes dengue numbers based on data kept by the government in Thailand going back 40 years.

“The task is to try to understand what’s driving the dynamics of a dengue pathogen spreading through the population,” says Cummings. “Ultimately, what you want to do is understand how you can control it. If I have a model that captures the features of the epidemiology I can start to test ideas.”

For instance, he says, if he knew of a vaccine that performed in a certain way, he could build a mathematical model that captures the basic dynamics of how the incidence of dengue cycles through a population. He could then hypothesize as to how much of an impact that vaccine might have, or what is the best age to use it—at what time and in what place—to minimize transmission.

Cummings splits his time between computer modeling in Baltimore and field studies in Thailand. There he works with the Ministry of Public Health and with the U.S. Army’s Armed Forces Research Institute of Medical Sciences, which partners with the Royal Thai Army.

Among the records he reviews are longitudinal studies of dengue involving schoolchildren. Blood tests show which children have serological evidence of dengue. Analyzing data going back three decades, Cummings discovered that in Bangkok the incidence of dengue showed a much larger peak than normal every two to four years.

“It’s thought to be driven by the cycling of immunity in the population,” he says, using the predator-prey model to illustrate. “You have foxes and rabbits. The fox eats the rabbits, the rabbit population goes down. So then the fox population crashes. Because foxes have disappeared, the rabbits begin to grow and then the foxes return.”

A large dengue outbreak immunizes people to a specific serotype. When that serotype shows up again the next year, not as many people are infected. Cummings also found a shift in the age group most affected by dengue. Historically in Thailand, 7- to 9-year-olds have shown the highest incidence. Surprisingly, surveillance data shows that the mean age of dengue cases is now 18.  Cummings used a dengue transmission model to determine the cause of this shift. The model considered factors such as climate, socioeconomic indices and demography. Cummings says the most plausible explanation is that changes in population structure have altered dengue transmission to reduce the rate at which people become infected. The results were published last fall in the journal PLoS Medicine.

The results are classic for a developing country, says Cummings. “From the 1960s to today, Thailand went through some pretty dramatic changes in its population structure. What that means for dengue transmission is that the number of kids being born who are completely naïve to dengue and can be infected by any serotype is proportionally less than used to be because birth rates are down. Older people have immunity because they had it when they were kids,” he says. “It makes it less likely that a mosquito that’s bitten someone infectious will go and bite a naïve kid. As a population, Thailand has hung onto its immunity longer because of lower death and birth rates.”

Because the only treatment for dengue is supportive care, efficient case management of the disease can mean the difference between a patient surviving the severest form of dengue or dying. “Thailand really does case management well,” says Cummings, “but the doctors who learn to manage dengue cases are pediatricians.” Because of his study, Thai health officials recognize the need to train regular internists in dengue care.

Cummings has also identified patterns in which dengue epidemics spread through Thailand. “It appears to peak first in the center of the country in Bangkok,” he says. “Then there is a lag as the pathogen moves north and south. It sort of appears as a traveling wave moving out into the country.

“I’m interested in what drives those waves,” says Cummings. “If you can understand that, you might understand what’s really driving transmission to a particular location. You then might understand how best to allocate resources to reduce incidence of dengue in the country.”

While computer models can help analyze data and lead to an understanding of transmission dynamics, he says, sometimes it takes old-fashioned, face-to-face detective work to get a fuller picture.

“There’s a lot of interest in investigating people’s behaviors during outbreaks,” says Cummings. “You try to understand who interacts with who. You ask people over the course of 24 hours who they talked to in the morning, how old were they, how often do you see them. If you find one person who’s infected, how much more likely are you to find another infected person right around them, defining the spatial scale of transmission?”

In the Nile valley two millennia ago, people knew nothing of transmission waves or spatial scales or disease dynamics. Slowly, across most of those 2,000 years, scientific knowledge based on the accumulation of incremental advances in research has struggled to outpace ancient legends of evil spirits. But when circumstance triggered a new life for dengue, it also sparked a new urgency to come to grips at last with this complex virus. With the fight fully engaged comes the question: how close are we to winning?

“That’s a tough question,” says Cummings. “I think we know a lot about dengue. But there are still basic things we don’t know. It’s incremental. The thing I’m proud of at the end of a year is if I can explain something new that could potentially be important in designing control measures to reduce incidence.”

Ultimately, just as a victim of dengue has but one course to follow, so goes the course for science. Says Cummings: “You have to be patient.”