Screening of actinobacteria for novel antimalarial compounds

Abstract

The success of our first-line antimalarial treatments is threatened by increased drug resistance in Plasmodium parasites. This makes the development of novel drugs critical to combat malaria. Historically, natural products have been an excellent source of novel antimalarial compounds and thus are an ideal place to search for potential drugs. Filamentous members of the bacterial phylum, Actinobacteria, are well-known antibiotic producers, but their antimalarial potential has not been well investigated. This makes these actinobacteria a potentially valuable source of novel antimalarial compounds. To evaluate the antimalarial potential of the filamentous actinobacteria, uncharacterized environmental actinobacterial strains from the Meyers laboratory culture collection, as well as the type strains of new actinobacterial species identified and characterized in the Meyers laboratory, were screened for antiplasmodial activity against drug-sensitive Plasmodium falciparum, NF54. Liquid cultures were extracted using the mid-polar solvent, ethyl acetate, with the aim of discovering drug-like molecules that can be administered orally. Thirty-one strains of actinobacteria belonging to eight genera (Actinomadura, Amycolatopsis, Gordonia, Kribbella, Micromonospora, Nocardia, Nonomuraea, and Streptomyces) were screened revealing fourteen active strains. Eight strains were identified for further study as the displayed antiplasmodial efficacy matching predefined criteria. Of these eight candidates, Streptomyces strain PR3 was selected, as it showed excellent antiplasmodial efficacy, no cytotoxicity against Chinese Hamster Ovary (CHO) or liver HepG2 cell lines, no haemotoxicity, and was easy to culture. Bioassay-guided fractionation of the crude extracts of strain PR3, supported by high-resolution mass spectrometry (HRMS) and nuclear magnetic resonance (NMR) analysis, was conducted to isolate and identify the compounds responsible for the antiplasmodial activity. During purification by solid phase extraction (SPE), a novel class of compounds was isolated. The structure of these compounds was elucidated by HRMS and NMR analysis and determined to be a series of crown ethers with a methylated backbone. These methylated crown ethers (MCE) were not produced by strain PR3, but by the cyclization of polypropylene glycol (PPG) oligomers from Amberlite® XAD-16N 20–60 mesh resin under aqueous conditions. The MCEs displayed weak antiplasmodial activity against P. falciparum NF54, without cytotoxicity against the Chinese Hamster Ovary, HepG2 cell lines, nor human erythrocytes. To the author's knowledge, the MCEs are novel compounds, and this is the first time the cyclization of PPG oligomers into crown ethers has been reported. As the MCEs were not responsible for strain PR3's potent antiplasmodial activity, further study was conducted. Using the Global Natural Product Social molecular networking (GNPS) workflow, genome mining, and NMR analysis, it was revealed that the cyclodepsipeptides, valinomycin, montanastatin, and nine other novel analogues were responsible for the high antiplasmodial activity detected. A review of the literature revealed that the structure of four of these analogues had been predicted, based on MS/MS and the biosynthesis of valinomycin. Using the same described biosynthetic logic and MS/MS analysis, two new cyclodepsipeptides, compounds 1054 and 1068, were elucidated. Unfortunately, chromatographic systems developed were unable to purify the cyclodepsipeptides, and individual evaluation of their antiplasmodial efficacy and host selectivity was not possible. The fraction containing the cyclodepsipeptides exhibited strong antiplasmodial activity against the drug-sensitive, NF54 and multidrug-resistant K1, strains of P. falciparum. No cytotoxicity was displayed against the CHO cell line and no haemotoxicity was seen against human erythrocytes. Moderate toxicity was exhibited against the liver HepG2 cell line; however, the selectivity index of the cyclodepsipeptides suggested that they are selectively targeting the Plasmodium parasites. Overall, these results are positive, and further study of the individual cyclodepsipeptides is warranted. During the investigation, discrepancies were noticed between different fractions in terms of antiplasmodial activity. These fractions contained both the MCEs and, cyclodepsipeptides along with a range of impurities, yet they displayed potent antiplasmodial activity. Further study suggested that combination of the MCEs and cyclodepsipeptides elicits a synergistic response and improves antiplasmodial efficacy. This was determined independently using two models, the fixed-ratio isobologram method and the CompuSyn programme based on the massaction law principle. The workflow developed during this investigation demonstrates how new technologies can be used to dereplicate and elucidate bioactive natural products. This workflow can be utilized to continue this research and identify new natural products that can combat malaria