Curiosity killed the plasmodium

This article was originally published in April 2014.

Esther Schmitt, a researcher at the Natural Products Unit of the Novartis Institutes for BioMedical Research in Basel, received a sample of roughly 300 hits that were the result of a medium through-put screening at the Genomics Institute of the Novartis Research Foundation (GNF) that in 2006 tested some 12 000 natural compounds regarding their activity against Plasmodium falciparum, the parasite that causes malaria in humans.

Her task was to check those candidates that looked most promising to be developed into a compound. Had she worked according to protocol and looked only at those candidates that met or exceeded a certain activity level, she might have discarded one particular substance that failed to reach the predetermined threshold.

Curiosity takes the lead

Because it looked “very peculiar” to her, she started to delve more deeply into the structure: “I was simply curious to find out more about this substance because it somehow fell out of the usual framework,” Esther Schmitt remembers the start of the journey, which would be full of astonishing twists and turns and would finally lead to the development of antimalarial compound KAE609, which is currently tested in Phase II.

The substance, which was part of the natural compounds library, had been bought a few years earlier. But until the medium throughput screening in the US in 2006 it had not been fully analyzed. Surprise was therefore big after her colleague Philipp Krastel in Basel started to scrutinize the structure and found that, after all, the substance was not pure but a mixture of several molecules. “It was simply out of curiosity that we started the analysis even as the compound did not meet the activity matrix. The guiding principle was to seize an opportunity,” Krastel said.

After an initial mass spectrometry Krastel realized that the compound was different than expected and not pure and therefore needed chromatographic purification.

The purified compound was then sent to the Swiss Tropical and Public Health Institute in Basel for another test. The screening showed increased potency against Plasmodium falciparum and the substance was analyzed further.

Then – another surprise – researchers also found that the supposed substance was a chiral mixture, meaning that it was composed by two compounds looking like mirror images to each other. The team contacted the Novartis researcher Eric Francotte, who has pioneered the complex field of chiral separation. With his team he was able to isolate the two molecules. These were sent to the Novartis Institute for Tropical Diseases (NITD) in Singapore where Bryan Yeung showed that only one of the mirror image structures was responsible for the plasmodicidal activity. Knowing now how the substance exactly looked like the chemist began to improve the structure of the compound.

Fighting malaria resistance

That screening project at GNF was not without a well-founded rationale. After all, the most successful antimalarial medicines had been developed from natural products.

An extract from the bark of the cinchona tree had been the standard malaria therapy for almost 300 years until quinine was isolated in 1820 by French researchers Pierre Joseph Pelletier and Joseph Bienaimé Caventou. Later quinine was replaced by chloroquine, which was discovered in 1934 by Hans Andersag at the Bayer laboratories.

But, like subsequent drugs such as mefloquine and atovaquone, these compounds became almost useless due to parasite resistance. In fact, before the development of Coartem® in the late 1990s, the World Health Organization had feared that some parts of Africa and Asia could become uninhabitable as the parasitic disease killed more than one million people every year because all standard therapies had lost their potency.

Yet, even Coartem, which has helped save more than one million people since Novartis started its Malaria Initiative in 2001, is running into resistance problems. Although the medicine, which includes the active ingredient artemisinin that is extracted from sweet wormwood, has a cure rate of some 96 percent, first cases of resistance have been reported in Asia. It has yet to be seen when parasite resistance will affect other regions.

Hence, it was only consistent that the GNF was screening the Novartis natural compound library for a successor to Coartem, which would be instrumental in eliminating a disease that ev-ery year kills around 600 000 people, most of them children.

That the screening would unearth a synthetic mixture with a chiral structure that belonged to a new class was not to be expected. But this is exactly what science is about: If scientists knew what they were doing, it wouldn’t be called research, to quote Albert Einstein’s famous saying.

Solving the Rubik’s Cube

With so many twists and turns in the development of KAE609, the early compound was greeted with little enthusiasm by many researchers, who believed the substance should not be followed any further. But as the analysis and refinement of the compound proceeded – upon the separation of the active enantiomer the potency was increased 10-fold – hopes were beginning to rise.

Bryan Yeung from the NITD in Singapore also refused to let go. Tasked to optimize the structure of the lead he had to work on the chemical structure to enhance the potency even further. Yet, while this sounds relatively simple, the task is fraught with many difficulties, not unlike solving the riddle of a Rubik’s Cube, at least for a beginner. Similar to aligning the six color sides in a correct fashion, adapting the chemical structure of a compound requires keeping several aspects under control at the same time. While the Rubik’s Cube cannot be solved by only concentrating on one side at a time, enhancement of a compound must be checked against a series of characteristics such as stability, toxicity, metabolic clearance and potency.

Yeung and his team therefore tested many potential structures until they developed a compound that included elements of both the active and the inactive enantiomer. Early lead enhancements led to the creation of compounds such as NITD261, which had excellent in vitro potency but high metabolic clearance. A later compound, NITD246, despite its fifty-fold potency over NITD261, was discarded because Yeung’s team hit upon what later would be called KAE609, which in mouse models showed that it could reduce the parasite almost completely and that the cure rate was a stunning 100 percent.

“Usually about 1000 compounds need to be tested until a satisfactory result can be obtained,” Yeung said. “In this case we were lucky as the structure which we constructed, after some 120 attempts, had a good safety and pharmacokinetic profile and showed that it was fast-acting with potent blood-stage activity against the parasite.”

The Rubik’s Cube was solved! And more than that: In collaboration with the research group of Kiaran Kirk from the Australian National University Yeung’s team established also the mechanism of action, showing that the spiroindolones disrupt the sodium pumps of the parasite, thus killing Plasmodium falciparum.

Making decisions

In the meantime KAE609 has demonstrated an adequate pharmacokinetic and safety profile in humans and as a result was the first molecule with a novel mechanism of action to successfully complete Phase IIa studies for malaria in the last 20 years.

The expectation now is to fully develop a drug that could be administered more easily, probably allowing patients to take fewer pills than in current standard therapies which require several pills a day over an extended period of time.

But if KAE609 makes it to the market one day, it will neither be due to serendipity nor strict scientific protocol. Rather, as Esther Schmitt and Philipp Krastel noted, scientific development is as much driven by curiosity as well as decisions which defy easy classification. “Research cannot be done at the push of a button. Often, it’s about making decisions and the researcher needs to make a choice. But of course, the real momentum is coming from curiosity.”