COMMENTARY

Quinolines block every step of crystal growth COMMENTARY David J. Sullivan Jr.a,b,1

Malaria is a lethal zoonotic disease that has impacted accumulation in the acidic, lysosome-like specialized human survival and indeed, the history of human civiliza- digestive vacuole designed for rapid he- tions worldwide. The first effective treatment for malaria moglobin digestion and heme crystal formation (10). was reported in 1632 with the use of extracts from Not all drugs that bind heme stop heme crystal growth, the bark of the cinchona tree. Since that time, quinoline nor do all drugs that bind heme and inhibit heme crystal compoundshavebeenusedasbothprophylacticand growth accumulate at the target site to kill Plasmodium. therapeutic drugs in every type of malarial treatment. Sullivan and Goldberg and their colleagues observed Almost 400 y later, the molecular mechanism of quinoline heme–quinoline complexes binding to hemozoin in situ action is brought into focus by Olafson et al. in their study by electron microscopy, subcellular fractionation, and ofthestepinhibitionofhemecrystalgrowth(1). in vitro with hemozoin extension assays, and proposed To put the significance of their findings into per- a quinoline-capping hemozoin growth mechanism (11). spective, identification of hemozoin was integral Unresolved were the diverse equilibriums between to the 1880 malaria diagnosis by Laveran (2). Some- quinoline–heme complexes decreasing heme crystal times referred to as malaria pigment, this crystalline substrate availability, heme substrate incorporation heme is synthesized by parasites intracellularly and into heme crystal, heme and the mu-oxo–dimer, and was identified as a recognizable black round body quinolines binding to heme crystal. with thin filaments on the periphery from exflagella- In a previous paper, Olafson et al. used citrate- tion of the gametocytes in unstained human erythro- buffered saturated octanol to mimic an organic cytes (2). also found malarial pigment environment to produce crystals monitored by atomic outside the Anopheles stomach to implicate the mos- force microscopy (12). The evidence was a strictly classic quito as the vector (3). In 1911, Brown (4) stated that mechanismofhemecrystallization,withnoevidencefor hematin was a component of hemozoin, distinct from a nonclassic incorporation of preformed hematin oligo- . He also postulated malaria pigment was an mers. The authors described four classes of growth sites: impure byproduct of degradation. For (site 1) flat surfaces between steps, (site 2) nuclei of small the next 80 y, malariologists neglected hemozoin numbers of heme dimers protruding from a flat face, (site study until Slater and Cerami proved that hemozoin 3) obtuse and acute kinks located on steps, and (site 4) was pure heme by elemental analysis (5) and, impor- groups of closely spaced steps, shown in Fig. 1. Bohle tantly, that quinolines inhibited hemozoin growth (6). and colleagues (13), by powder diffraction, demon- Decades ago, an initial hypothesis mentioned by strated head-to-tail heme dimers linking the proprionate Olafson et al. (1) concerning the mechanism of quinoline carboxyl group to ferric of the adjacent heme, in inhibition was binding to substrate heme to prevent in- which the pair are ∼1-nm square, as the fundamental corporation into hemozoin. Cohen et al. first observed building brick for heme crystal growth. The heme dimer quinoline–heme complexes that were postulated to be 1-nm-square bricks then stack by hydrogen-bonding in- toxic to the malaria parasites (7). Subsequent work dem- teractions into the crystal, which in Plasmodium falcipa- onstrated quinolone–heme complex interference with rum are ∼100 nm × 100 nm × 500 nm. A neutral lipid Plasmodium and insertion into membranes nanosphere environment within the digestive vacuole (8). However, this substrate hypothesis was not able excludes water, as well other interfering molecules, like to explain why many quinoline drugs that bind heme amino acids or proteins in the diverse intracellular mix- were unable to kill the malaria parasite. For example, ture (14). Kapishnikov and Leiserowitz and their col- the stereoisomer of quinine-9-epiquinine is an ineffective leagues applied crystallographic measures and soft X- malaria drug (9). Egan et al. have proposed a hierarchy ray tomography to postulate a digestive vacuole mem- for structure-activity, which starts with heme binding, fol- brane site of formation when examining late trophozoites lowed by heme crystal inhibition, as well as weak base (15–17). These authors also modeled quinolines binding

aDepartment of Molecular Microbiology and Immunology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD 21205; and bDepartment of Medicine, Division of Infectious Diseases, The Johns Hopkins School of Medicine, Baltimore, MD 21205 Author contributions: D.J.S. wrote the paper. The author declares no conflict of interest. See companion article on page 7531. 1Email: [email protected].

www.pnas.org/cgi/doi/10.1073/pnas.1708153114 PNAS | July 18, 2017 | vol. 114 | no. 29 | 7483–7485 Downloaded by guest on September 28, 2021 Fig. 1. Quinoline blockade of heme crystal growth by step-pinning or kink-blocking. Head-to-tail heme dimers add in a classic crystallization mechanism at growth sites, which are: (1) terraces or open facers, (2) nuclei of a few heme dimers, (3) kink sites, and (4) closely spaced steps. Quinolines block by open flat face attachment in the instance of and quinine, kink-blocking in the case of amodiaquine and mefloquine, and by step-bunching in the case of pyronaridine.

to the surface of crystals (18). Hemozoin in early trophozoites has addition to kink sites. By this process, heme crystal formation is been observed to spin wildly in excess of Brownian motion in live inhibited over broad areas of the crystal surface. Finally, a single parasites, which stop moving in drug-treated dead parasites (19, 20). quinoline-tested pyronaridine inhibits growth by a step-bunching This finding may indicate that a neutral lipid environment within the mechanism, whereby the ability to simultaneously bind at two digestive vacuole not attached to the digestive vacuole membrane step edges leads to a pile-up of the elementary steps into groups exists for an early trophozoite portion of crystal formation. separated by large terraces. In this new work, Olafson et al. (1) use atomic-force microscopy Although not tested here, previous work indicated that a to capture the in situ growth of heme crystals to show how step quinoline–heme complex binds more efficiently to the steps or propagation is inhibited by the quinoline class of antimalarials. kinks of hemozoin than quinolines alone (21, 22). Quinolines can The authors also observed the effects of artemisnin, another im- also form covalent bonds with heme to incorporate into the steps, portant antimalarial that binds heme, but does not inhibit crystal- but the proportion of covalent-bound quinoline–heme versus lization (1). Direct observations of how these compounds affect noncovalent-bound complexes has not been determined (23– growth lead Olafson et al. to detail three classes of quinoline- 25). At present, evidence points to a reversible quinoline–heme inhibition mechanisms. First, the quinolones, amodiaquine and complex binding at the three types of sites rather than irreversible mefloquine, were found to largely only bind to kink growth sites inhibition. This reversible binding is still governed by changes in where molecular units are added to a step edge by an inhibition pH associated with quinoline-resistant parasites or alterations in mechanism that is termed “kink-blocking.” This is the least- PfCRT or Pfmdr1 transporters associated with drug-resistant par- effective mechanism of heme crystal growth inhibition. In a sec- asites. This work opens new opportunities for a lock-and-key ap- ond mechanism, known as “step-pinning,” the quinolines, such as proach to targeted drug development that is rooted in the chloroquine and quinine, bind anywhere on a flat surface face in physical basis for hemozoin crystal growth inhibition.

7484 | www.pnas.org/cgi/doi/10.1073/pnas.1708153114 Sullivan Downloaded by guest on September 28, 2021 1 Olafson KN, Nguyen TQ, Rimer JD, Vekilov PG (2017) Antimalarials inhibit hematin crystallization by unique drug–surface site interactions. Proc Natl Acad Sci USA 114:7531–7536. 2 Laveran A (1880) Un nouveau parasite trouve dans le sang des malade atteints de fievre palustre. Origine parasitaire des accidents de l’impaludisme. Bull Mem Soc Med Hop Paris 7:158–164. 3 Ross R (1897) On some peculiar pigmented cells found in two fed on malarial . BMJ 2:1786–1788. 4 Brown WH (1911) Malarial pigment (so-called melanin): Its nature and mode of production. J Exp Med 13:290–299. 5 Slater AF, et al. (1991) An iron-carboxylate bond links the heme units of malaria pigment. Proc Natl Acad Sci USA 88:325–329. 6 Slater AF, Cerami A (1992) Inhibition by chloroquine of a novel haem polymerase activity in malaria trophozoites. Nature 355:167–169. 7 Cohen SN, Phifer KO, Yielding KL (1964) Complex formation between chloroquine and ferrihaemic acid in vitro, and its effect on the antimalarial actionof chloroquine. Nature 202:805–806. 8 Orjih AU, Banyal HS, Chevli R, Fitch CD (1981) lyses malaria parasites. Science 214:667–669. 9 Slater AF (1993) Chloroquine: Mechanism of drug action and resistance in . Pharmacol Ther 57:203–235. 10 Egan TJ, et al. (2000) Structure-function relationships in aminoquinolines: Effect of amino and chloro groups on quinoline-hematin complex formation, inhibition of beta-hematin formation, and antiplasmodial activity. J Med Chem 43:283–291. 11 Sullivan DJ, Jr, Gluzman IY, Russell DG, Goldberg DE (1996) On the molecular mechanism of chloroquine’s antimalarial action. Proc Natl Acad Sci USA 93:11865–11870. 12 Olafson KN, Ketchum MA, Rimer JD, Vekilov PG (2015) Mechanisms of hematin crystallization and inhibition by the antimalarial drug chloroquine. Proc Natl Acad Sci USA 112:4946–4951. 13 Pagola S, Stephens PW, Bohle DS, Kosar AD, Madsen SK (2000) The structure of malaria pigment beta-haematin. Nature 404:307–310. 14 Pisciotta JM, et al. (2007) The role of neutral lipid nanospheres in Plasmodium falciparum haem crystallization. Biochem J 402:197–204. 15 Kapishnikov S, et al. (2013) Digestive vacuole membrane in Plasmodium falciparum-infected erythrocytes: Relevance to templated nucleation of hemozoin. Langmuir 29:14595–14602. 16 Kapishnikov S, et al. (2012) Oriented nucleation of hemozoin at the digestive vacuole membrane in Plasmodium falciparum. Proc Natl Acad Sci USA 109:11188–11193. 17 Kapishnikov S, et al. (2012) Aligned hemozoin crystals in curved clusters in malarial red blood cells revealed by nanoprobe X-ray Fe fluorescence and diffraction. Proc Natl Acad Sci USA 109:11184–11187. 18 Solomonov I, et al. (2007) Crystal nucleation, growth, and morphology of the synthetic malaria pigment beta-hematin and the effect thereon by quinoline additives: The malaria pigment as a target of various antimalarial drugs. J Am Chem Soc 129:2615–2627. 19 Sachanonta N, et al. (2011) Ultrastructural and real-time microscopic changes in P. falciparum-infected red blood cells following treatment with antimalarial drugs. Ultrastruct Pathol 35:214–225. 20 Sigala PA, Goldberg DE (2014) The peculiarities and paradoxes of Plasmodium heme metabolism. Annu Rev Microbiol 68:259–278. 21 Sullivan DJ, Jr, Matile H, Ridley RG, Goldberg DE (1998) A common mechanism for blockade of heme polymerization by antimalarial quinolines. J Biol Chem 273:31103–31107. 22 Iyer JK, Shi L, Shankar AH, Sullivan DJ, Jr (2003) Zinc protoporphyrin IX binds heme crystals to inhibit the process of crystallization in Plasmodium falciparum. Mol Med 9:175–182. 23 Alumasa JN, et al. (2011) The hydroxyl functionality and a rigid proximal N are required for forming a novel non-covalent quinine-heme complex. J Inorg Biochem 105:467–475. 24 Gorka AP, de Dios A, Roepe PD (2013) Quinoline drug-heme interactions and implications for antimalarial cytostatic versus cytocidal activities. J Med Chem 56:5231–5246. 25 Gorka AP, Sherlach KS, de Dios AC, Roepe PD (2013) Relative to quinine and quinidine, their 9-epimers exhibit decreased cytostatic activity and altered heme binding but similar cytocidal activity versus Plasmodium falciparum. Antimicrob Agents Chemother 57:365–374.

Sullivan PNAS | July 18, 2017 | vol. 114 | no. 29 | 7485 Downloaded by guest on September 28, 2021