A Double Dose of Hope

On an autumn morning in his lab at the Johns Hopkins Malaria Research Institute, Jürgen Bosch gets down to work.

He removes a pencil-sized metal wand from its case and places it on his lab bench to the left of a light microscope. He hoists a heavy container of liquid nitrogen and pours some into a foam dish. Fog billows from the frigid liquid. Then he sets a plastic laboratory dish underneath the microscope’s lens. The dish is divided into 96 wells.

Bosch takes a seat, interlaces his fingers and stretches out his arms so that his knuckles crack. “Okay,” he says. “Now comes the fun part.” He wheels his chair in close to the lab bench, picks up the wand and peers into the microscope.

The object of his attention?

A protein crystal, about one ten-thousandth of a meter long, delicate and brittle. The protein comes from the malaria parasite Plasmodium. Bosch, PhD, an assistant professor of Biochemistry and Molecular Biology, is studying the structure of this protein and others like it to design new malaria drugs. This work involves growing crystals in laboratory dishes, and then plucking individual crystals from the dishes and flash freezing them in liquid nitrogen.

Through the lens, two crystals come into view, floating in their salt/sugar bath. They resemble rectangular blocks but with more sides. Bright streaks of green and red light reflect off their mirrored surfaces.

Slowly, Bosch angles the end of the wand toward the crystal. At the end of the wand is a fine nylon loop visible only under magnification. Bosch gingerly moves the wand to bring the loop directly over one crystal. He allows it to hover there a moment, until the surface tension of the fluid in the loop sucks up the crystal.

“Got one.”

An Elusive Foe

Antimalarial drug research has had lifesaving success stories. Scientists have generated several highly effective agents for treating or preventing the disease—including chloroquine and pyrimethamine-sulphadoxine, and, most recently, chemical derivatives of the plant-based drug artemisinin.

However, resistance to popular antimalarials such as chloroquine has rapidly spread worldwide, and there have been reports in some regions that the artemisinin derivatives are taking longer to clear the parasite from the body (not true resistance but enough to vex health officials). And no one has produced and marketed a new class of antimalarial in 15 years.

There are 225 million clinical cases of malaria each year. The most deadly parasitic disease, it causes nearly 800,000 deaths per year—and the majority of those deaths occur in young African children. Malaria accounts for approximately one out of five child deaths there.

With the looming specter of drug resistance reducing treatment options even further, scientists are seeking new drugs to add to the malaria medicine chest.

At the Bloomberg School, two  scientists are employing radically different strategies. Jürgen Bosch is using a relatively new strategy called structure-based drug design. He makes crystals of different protein components of Plasmodium and uses a technique called X-ray crystallography to determine their structure. Using the structure as a sort of cast for a mold, he then designs small molecules that will fit in certain gaps, grooves or crevices of the proteins, bind there and inhibit the parasite.

His colleague David Sullivan, MD, associate professor of Molecular Microbiology and Immunology(MMI), is using more of a brute-force approach: Pull a drug off the shelf, aim it at the parasite and see if it kills.

While the two schemes differ significantly, they also have something in common: Both look great on paper, but neither comes with a guarantee that it will work against an opponent that has eluded medicine time and time again

Finding the Perfect Fit

Bosch’s first encounter with malaria came when he was a child.

He lived first in South and Central America, and then spent 11 years in Nigeria, where his father worked for a multinational corporation. Malaria was pervasive there, and young Jürgen was struck with the disease several times. Fortunately, he says, “it was always a mild case, like a flu, the same type of symptoms.” He also had access to good health care and medications, he notes, which cleared up his infection in two or three days. Many in Nigeria weren’t so lucky—including the thousands of Nigerian children who succumbed to the disease each year.

A friend of Bosch’s became one such victim. The boy, another member of the ex-pat community, had come down with malaria while on a trip to Germany. “He didn’t come back,” says Bosch. “That’s how we learned.” The boy’s doctors did not recognize the disease until the infection had become severe. By then, it was too late.

“If you want to develop a new drug, you’re going to fail more than 90 percent of the time. That’s the reality of the process.” —David Sullivan

Bosch’s firsthand view of malaria stimulated his interest in tropical diseases; he went on to earn a PhD at the Max Planck Institute in Germany and to specialize in using crystallography to study tropical diseases.

Crystallography is like an art, says Bosch. It also is a skill that comes with its own challenges, number one being constraints imposed by the simple laws of physics: Delicate crystals break easily. Not only that, says Bosch. “Even if you only bend a crystal, you can disrupt its lattice structure.” Growing crystals is also tricky; some require exact temperatures or acidity. And the pace of formation is crucial.

“If a crystal forms too fast, mistakes will happen and the crystal will be more fragile,” says Bosch. All of this, he says, “requires tremendous patience and endurance.”

One recent day in his office, Bosch calls up on his computer screen a brilliantly colored image of a protein called aldolase—a complex protein that Bosch calls part of the Plasmodium’s “invasion machinery,” the molecular apparatus that the parasite uses to enter and invade the liver and red blood cells of its human hosts.

On the screen, aldolase looks something like a lumpy catcher’s mitt. Bosch points to a crevice. “We want a drug that will fit into this crevice and touch this particular area,” he says, pointing. An effective drug would slip into this space the way a baseball flies into the palm of a catcher’s glove. That interaction, Bosch posits, will knock out the parasite’s invasion machinery—and, therefore, prevent Plasmodium from entering a host cell.

Bosch clicks some keys and a green sausage-shaped molecule appears on the screen, seated within the crevice. He has already designed 60 small molecules that he hopes will bind to parts of the invasion machinery, including this one. To examine the extent of binding, he mixes a candidate drug with aldolase or another part of the invasion machinery. He then crystallizes that complex, freezes the crystals and sends them to a synchrotron facility in California. At the facility, X-rays are fired at the crystals, and the diffraction of those beams is used to generate images of the complex’s structure.

“We can see the space in the cavity where the compound binds,” says Bosch. “Then we can ask, do we want to make [the drug] larger, or a different shape? You know which chemical group will fit where and you can rationally make decisions” to build a more effective drug.

It’s an elegant approach. Bosch is not aware of anyone who has ever designed a malaria drug this way. Still, he says, “I’m positive we’ll find something.” In the best-case scenario, that “something” will be a cure. But if not a cure, the experiments, he says, will at least provide new avenues to pursue in the search for a cure.

"Old" Drugs, New Tricks

Three flights down from Bosch's lab, David Sullivan takes a very different tack. Rather than design a malaria drug from scratch, like Bosch, he is looking for one that already exists but has been used for a different purpose.

The current total pharmacopoeia of approved or experimental drugs contains about 10,000 items, says Sullivan. And within that enormous trove, he believes, there surely reside drugs that have as yet unrecognized powers to kill Plasmodium. To identify such hidden pearls, he advocates screening massive collections of drugs. Since the compounds in such drug libraries have already ascended the arduous FDA approval process or been approved for clinical trials, the approach can save enormous amounts of time and money. (Scientists will not waste time on a drug that is toxic or has serious adverse effects.)

"If you want to develop a new drug, you're going to fail more than 90 percent of the time," says Sullivan. "That's the reality of the process."

It can take 10 to 15 years and cost hundreds of millions of dollars to move a drug from bench to bedside, he explains. Most of the candidate malaria drugs do not make it that far.

"Look at the odds," he says. A scientist might mix a certain drug with an enzyme used by Plasmodium and show that the drug inhibits the enzyme. However, the drug might not be able to penetrate the parasite. Or, if it can get into the parasite, it may not work in an animal model. Then, he says, "the biggest hurdle is taking [drugs] that work in mice and pushing those forward to be drugs that work in humans. We have many agents that work beautifully in mice," he notes, that don't cure malaria in people.

So he'd rather seek from among those drugs that are already approved for human use. Recently, for example, he has been conducting studies on FBS0701. The drug is an iron chelator; it binds and removes excess iron from the body, and it is currently in clinical trials as a possible treatment for iron overload (a condition that can occur in certain diseases that require repeated blood transfusions). But Sullivan believes it also has potential as an antimalarial. Plasmodium requires iron to reproduce. So by binding up free iron, FBS0701 would cut off the supply of an element essential to the parasite's survival.

While there are other iron chelators already in clinical use, they must be given intravenously. FBS0701 can be taken by mouth. Moreover, Sullivan hopes, the drug would target the earliest stage of Plasmodiuminfection, when the parasites are invading the liver. At this stage, fewer parasites are present and a lower dose would be required.

Sullivan has tested the drug in Plasmodium-infected mice and shown that a single dose cures lethal infections. "We can even give the drug a couple of days before we infect the mice, and it works," he says.

Those results are promising, says Sullivan, and he is currently drafting a paper about these findings. Since FBS0701 is already in Phase 2 clinical trials as a treatment for iron overload, Sullivan has a head start on his antimalarial studies.

Despite the apparent promise of FBS0701, Sullivan emphasizes that the drug is still a long way from being ready for human use. "I temper my excitement every day," he says. "We like to say 'cautious optimism.' I've seen a lot of promising drugs die on the vine."

About five years ago, with resistance to chloroquine rising, he and a Hopkins MD/PhD student named Curtis Chong decided to search for alternatives to that antimalarial. Their strategy was to do a systematic search—take a whole bunch of already approved drugs and see if any would rid infected cells of Plasmodium. Over two years, the scientists procured 2,687 different drugs (a collection that eventually became the Johns Hopkins Clinical Compound Library) and tested the ability of each to inhibit the parasite.

Sullivan and Chong had some promising leads, in particular an antihistamine called astemizole. In test tube studies it inhibited the growth of chloroquine-resistant Plasmodium and, in infected mice, it significantly reduced the level of infection. However, there were problems. A dose of astemizole killed only about 100 malaria parasites in a 48-hour period, whereas an equivalent dose of artemisinin could kill 10,000. That's important, says Sullivan, considering that a malaria patient showing symptoms can harbor about a trillion parasites. So in the end, says Sullivan, "we did not find the pearl."

Still, Sullivan continues to have faith that "repurposing" old drugs can yield new cures for a raft of diseases, not just malaria. A good portion of the Johns Hopkins Clinical Compound Library, he notes, is available to any scientist who wants to screen it in search of a drug for a particular disease of interest. A sample of each of the available 1,500 drugs can be dispensed into a 96-well laboratory dish and shipped to the requesting scientist. Using the library, colleagues have found leads to diseases ranging from cancer to HIV.

Meanwhile, Sullivan recently took part in an even more ambitious screening project. This time Sullivan and a multidisciplinary team began with almost 310,000 compounds—a wide net that included known drugs but also thousands of trial compounds that companies had produced but that had not yet demonstrated any therapeutic use. That study, which the team reported in the May 20, 2010, issue of Nature, identified many promising leads for fighting malaria, says Sullivan, not necessarily drugs but chemical "scaffolds" that may help steer researchers toward the drug structures most likely to defeat the parasite.

Best of Both?

The research strategies taken by Sullivan and Bosch could not be more different. But it's possible that their scientific paths may converge at some point.

While Bosch continues to spend most of his time deducing crystal structures, he has also recently begun to take an interest in drug and chemical collections of the sort that Sullivan uses. This interest began when Bosch read a Nature article by scientists who had screened a GlaxoSmithKline chemical compound library that contains 2 million chemicals. They found several thousand that showed some ability to inhibit Plasmodium growth.

When Bosch looked closely at the results, he became excited. Some of the compounds appeared to strike Plasmodium's invasion machinery—direct hits to the proteins whose structures he had studied so closely. Bosch immediately emailed the researchers and asked them to send him samples of their compounds.

None of the chemicals is ready to serve as a malaria drug, says Bosch. They inhibit Plasmodium, but only weakly. But the structure of those chemicals could help him improve the potency of his own drug candidates.

Bosch is also planning to use the Johns Hopkins Clinical Compound Library to search for agents that might bind to aldolase or other parts of the Plasmodium invasion machinery.

"We'll run through them and see if any might potentially hit," he says.