Nuclear magnetic resonance (NMR) spectroscopy: Understanding antimalarial drugs

Malaria is among the top causes of death in the developing world. The disease is responsible for about one to three million deaths per year. Almost half of the entire population of the world is exposed to various species of Plasmodium, the parasite responsible for this disease. This exposure translates into several hundreds of millions of people carrying the parasite. The disease is therefore a major concern in developing countries. Affordable medicine against the disease is imperative and its non-lucrative nature unfortunately does not draw interest among top pharmaceutical companies. For this reason, efforts in designing drugs against this deadly disease have been confined within academic research institutions in the United States, as an example. The widely used drug against malaria is chloroquine. Unfortunately, there are now strains in all three continents – Africa, Asia and South America – that are resistant to chloroquine. With the emergence of drug-resistant strains, it has become timely to understand how antimalarial drugs in the hope of intelligently designing new drugs against malaria.

The most common drugs used against malaria, including chloroquine, are based on quinoline, a heteroaromatic compound (similar to naphthalene, except with one carbon replaced by nitrogen). Its target is believed to be the heme fragment, a byproduct of the metabolism of hemoglobin. The malaria parasite feeds on this human blood protein for its amino acid source. Unlike humans, the parasite cannot metabolize heme and this turns out to be toxic to the parasite. The heme, however, precipitates out and crystallizes into hemozoin, making it innocuous to the parasite. Crystals of hemozoin, commonly referred to as malaria pigment, are found in red blood cells infected by the malaria parasite. Antimalarial drugs such as quinine, chloroquine and amodiaquine are thought to inhibit heme crystallization. By understanding the underlying mechanisms and chemical properties behind the interaction between the drug and its target heme, it is hoped that new drugs can be designed in an effective and speedy manner. This desired increase in understanding could only come with a detailed knowledge of these molecules. The drugs and their target are very small, about a million times smaller than the parasite, preventing their direct observation by even the most powerful microscope. Spectroscopic techniques are required to probe systems at the atomic resolution level. To understand spectroscopy a little bit, a simple analogy can be made. These techniques are similar to listening carefully to what tunes atoms and molecules are singing. Hearing a love song brings us certain emotions. Similarly, the frequencies atoms and molecules display can reveal information regarding their environment and structure, allowing our naked eyes to probe into the ultramicroscopic world of atoms and molecules. Extraction of chemical information from these techniques, however, is not straightforward. Analysis is required and nowadays, computers are necessary to model the systems and to evaluate the interactions between molecules from first principles.

Our research group at Georgetown University employs nuclear magnetic resonance (NMR) spectroscopy and molecular simulations in the study of antimalarial drugs. NMR spectroscopy is a widely used technique in chemistry. This type of spectroscopy employs a magnet (similar to magnetic resonance imaging, MRI) in which samples containing the molecules to be studied are placed. Inside the magnet, the nuclei (the core of atoms that contain protons and neutrons) begin to spin at their characteristic frequencies, their unique tunes. Sophomores in college chemistry are usually introduced to this method as a tool for characterizing and identifying small organic compounds. In NMR spectroscopy each nucleus in a unique chemical setting will report a particular frequency, a "signature" NMR chemical shift. NMR spectroscopy, similar to other techniques in chemistry, provides us with a glimpse into the world of atoms our naked eyes would normally not see. Since its discovery in the 1950s, the NMR chemical shift, demonstrating a great sensitivity to the electronic environment around a nucleus, has likewise become of value in elucidating the secondary structures of proteins in solution. Proteins can be regarded as large molecules and the arrangement of their atoms, what chemists would refer to as structure, leads to the properties and functions these proteins exhibit in human health and disease. Changes in NMR chemical shifts can be observed by simply altering the solvent, the medium in which we place atoms or molecules. Interactions between molecules manifest in this ubiquitous observable. One example that demonstrates the sensitivity of the NMR chemical shift is the fact that nuclei in the vicinity of molecules that contain aromatic rings exhibit changes in the chemical shift, which can then provide clues regarding their position and distance from these aromatic centers. In chemistry, an aromatic molecule is one in which electrons are free to cycle around circular arrangements of atoms. Aromaticity is relevant to the study of malaria since most antimalarial drugs and their target, heme, are aromatic. The aromatic rings, due to the electrons freely circulating above and below a molecular plane, are able to perturb the NMR spectrum of a nearby molecule. Aromatic compounds make nuclei sing a different tune.

There are other parameters in NMR spectroscopy in addition to the chemical shift. Coupling constants, which arise from either a direct interaction between nuclear spins or from an indirect interaction mediated by electrons, are also excellent reporters of the environment in the vicinity of the nucleus. There are duets, trios, and even choral singing. Equally important are relaxation times, as these provide a wealth of information regarding the structure and dynamics of chemical systems. Relaxation times refer to the rate at which a nucleus returns to its equilibrium spin state after we have forced it to sing its tune. Nuclei are like singers that need some break before they could sing their tune again. Unlike other spectroscopies, relaxation in NMR is not spontaneous, as it requires agents or mechanisms, such as unpaired electrons, other nuclear spins, and molecular motion. And in the presence of unpaired electrons, nuclei relax a lot faster. And the closer they are to the source of unpaired electrons, the less time they take to relax.

At the center of heme is a ferric ion, which is paramagnetic (having unpaired electrons). One can clearly see this by simply seeing that the number of electrons in a ferric ion is not even, thus, no matter how these electrons are distributed, complete pairing is not possible. In molecules with unpaired electrons, how fast a nucleus can return to its equilibrium spin state can serve as a clue of how the various nuclei are situated with respect to the unpaired electron. The most widely studied nucleus in NMR is the proton. Biologically relevant compounds, like other organic compounds, contain plenty of hydrogen atoms. Thus, through an NMR spectrometer, one can see a view of what is going on through the eyes of a hydrogen nucleus. And a compound usually will have several unique hydrogens, with each one offering its own vantage point. In the presence of heme, the NMR spectra of the antimalarial drugs are significantly perturbed. The paramagnetic heme affords an efficient relaxation mechanism for the quinoline nuclear spins, that their lines in the NMR spectrum become much broader when the quinoline molecule becomes closer to a heme fragment. In the presence of a relaxing agent, they can sing very often, but their tunes are not long lasting. The enhancement in the relaxation rates can be carefully measured. Each hydrogen nucleus in the quinoline drug will have a relaxation time, which can be analyzed to extract its distance from the paramagnetic heme. With all the hydrogens in the antimalarial drug reporting a distance from the ferric center of heme, one can perform some sort of a triangulation procedure to draw the position of the drug molecule with respect to its target, the heme. Thus, it has been demonstrated that via NMR spectroscopy, one can determine solution structures, the three-dimensional arrangement of atoms, of heme-antimalarial drug complexes.

The NMR chemical shifts are likewise perturbed when aromatic systems become close to each other. The interaction between heme and an antimalarial drug is suggested to be noncovalent (no chemical bonding is involved). The interaction is believed to arise from the aromatic rings of the porphyrin group in heme and the drug molecule. These aromatic rings are characterized by electrons lying primarily above and below a molecular plane. These electrons extend a bit further compared to other electrons found in systems that have only single bonds. These electrons can influence the NMR chemical shift of nearby nuclei. By monitoring the NMR chemical shift, one can also gain information regarding how far an aromatic ring is from any given nucleus or vantage point since these interactions are very much distance-dependent. Hence, NMR chemical shifts have likewise been utilized in deciphering how antimalarial drugs approach a heme fragment. Furthermore, NMR chemical shifts can be used to study drug-drug interactions as these antimalarial drugs all contain the aromatic quinoline ring. NMR spectroscopy has been utilized, for example, in the study of self-aggregation of various antimalarial drugs. Solution structures for chloroquine dimers have been proposed. Mixed aggregates are also important in cases where one drug interacts strongly with another, suggesting that a mixed cocktail of antimalarial drugs may not be a wise choice at all times.

The ultimate goal in these studies is to understand at the molecular level how antimalarial drugs work. With this understanding, it is hoped that new drugs can be designed in a systematic fashion. Work on antimalarial drugs involves computational work. Since the NMR chemical shift is electronic in origin, one can calculate this NMR parameter from first principles, those we learned from Newton, Schrödinger, Einstein, Dirac, Ramsey and others. The NMR chemical shift reported at each site originates from the electronic or bonding makeup near that atom. Particularly interesting for the antimalarial drugs are the chemical shifts of the aromatic nuclei. Preliminary calculations indicate that there may be a heterogeneity or non-uniformity in the NMR chemical shifts of aromatic nuclei, an indication perhaps of the non-uniform distribution of electrons in the region above and below the aromatic plane. This non-uniform distribution of charges can lead us to the more familiar case of electrostatic interactions, the simple principle that states unlike charges attract. Substitutions on a quinoline ring as exemplified by the well-known drugs chloroquine, quinine and amodiaquine, enhance this non-uniform distribution. These subtle differences may partly explain the interactions found between aromatic rings, the interaction believed to be at the heart of the drug-target complex in malaria. Therefore, NMR spectroscopy may not only provide detailed pictures of drug-target complexes at atomic resolution, but may likewise provide insights regarding the origin of the properties of these drug molecules. With each tune played by a nucleus, we learn more about what atoms see.

* * *

Angel C. de Dios, Ph.D., is currently an associate professor and director of undergraduate studies in the Department of Chemistry at Georgetown University, Washington, DC 20057. He can be reached at [email protected].