Plasmodium falciparum parasites remain the major cause of malaria, a serious and potentially fatal infectious disease worldwide, despite the intermittent introduction of several different successive classes of potent and effective drugs. According to the World Health Organization’s Malaria Report (2011), an estimated 3.3 billion people were still at risk of malaria in 2010.1 Worldwide, in 2010, 99 countries and territories reported ongoing malaria transmission and a further 7 countries were trying to prevent the reintroduction of the disease, comprising a total of 106 countries in which malaria was considered endemic.1

The prevention and successful treatment of malaria is heavily dependent on antimalarial drugs. Chloroquine proved to be a remarkable safer, cheap and extremely effective antimalarial drug, serving as the frontline antimalarial treatment from its introduction in the mid-1940s to the 1990s.2 However, the malaria parasite has continually proved to be particularly efficient at developing resistance to virtually all drugs used to control it. Chloroquine-resistant parasites quickly emerged and spread worldwide, rendering the drug useless in many locations. The same scenario is being witnessed with the latest in the long line of potent antimalarials, artemisinin. Therefore, there is a continuing need for new antimalarial drugs that are effective, safe, affordable and easy to use and which, ideally, have a novel mode of action.

In our institute, we have focused on the screening and synthesis of antimalarial agents from microbial metabolites including antibacterial macrolides.3, 4, 5 In 2007, azithromycin, a 15-membered antibacterial macrolide, was found to possess potent antimalarial activity by Fidock and co-workers.6 Azithromycin is a slow-acting antimalarial that targets the parasite apicoplast and its prokaryotic ribosomes. The effects of its action increase with prolonged incubation time (from one to two generations), giving an IC50 in the nano-molar range. Continuously, new derivatives of azithromycin have been developed with the aim to improve activity and selectivity for the malaria parasite.7 Another macrolide antibiotic, tylosin, was first isolated by McGuire and co-workers8 in 1961, and is a 16-membered macrolide isolated from a culture broth of Streptomyces fradiae. In 1984, McColm and McHardy9 reported that tylosin demonstrated antimalarial activity against P. falciparum strain Liverpool (IC50=0.1 μg ml−1). However, there have been no subsequent reports concerning the antimalarial activity of tylosin or its derivatives. In this short note, we report that novel C-9 oximes of desmycosin express antimalarial activity against chloroquine-resistant P. falciparum K110 malaria parasites.

The removal of mycarose from tylosin can be accomplished under mild acidic hydrolysis to give desmycosin, a known minor component in the fermentation broth of the tylosin-producing strain.11 The antimicrobial activity of desmycosin is potent but slightly lower than that of tylosin, but it has almost no efficacy in vivo when given orally.11 Likewise, C-9 oxime derivatives of tylosin and desmycosin have been evaluated in bioassays, but did not show better antimicrobial activity than tylosin.12 Our work on tylosin analog synthesis produced C-9 oximes of desmycosin (3, 6 and 7), prepared from dimethoxydesmycosin (1)13 by oxime formation with corresponding hydroxylamines HCl14 and acid hydrolysis of dimethoxyacetal moiety in good yields, respectively. These analogs showed antimicrobial activities, similar to or slightly lower than that of desmycosin, (Scheme 1, Table 1). We specifically tested for antimalarial activity in all new analogs derived from tylosin. Moreover, as the discovery of the antimicrobial alkyne-bearing lead compounds (6 and 7) for antimicrobial activities, our efforts have focused on the preparation of new analogs in this series utilizing ‘click chemistry’, which provides an important approach for simple and rapid evaluation of functional activity. The concept of ‘click chemistry’ was originally introduced by Kolb, Finn and Sharpless in 2001.15 It incorporates powerful and selective reactions for efficient synthesis of interesting compounds and combinatorial libraries through heteroatom links. The advantages of ‘click chemistry’ in biological studies of macrolides have recently been clearly demonstrated in our laboratory.16, 17

Table 1 In vitro antibacterial and antimalarial activities of tylosin, desmycosin and its derivatives

The alkyne-bearing oximes (6 and 7) are intermediates allowing generation of targeted triazole compounds. Anti-selective triazole formation was carried out with a catalytic amount of Cu(MeCN)4PF6 and tris-(benzyltriazolylmethyl)amine (TBTA)18 in MeOH at room temperature to afford the corresponding triazoles (8a–e and 9a–c) in good yields, respectively (61–100%, see Supplementary Information; for example, compound 9a was 61% yield) (Representative example (Compound 9a); To a solution of compound 7 (5.2 mg, 6.1 μmol) in MeOH was added 3-azidequinoline (2.2 mg, 12.9 μmol), TBTA (0.04 mg, 0.75 μmol), Cu[Me(CN)4]PF6 (0.01 mg, 0.027 μmol ) and a small piece of Cu(0) turning in MeOH (0.5 ml) at room temperature, and the resulting mixture was stirred for 20 h. After removing of Cu(0) turning, then the reaction mixture was concentrated and purified by preparative TLC (Chloroform: acetone: NH4OH=2:1:0.1) to give 9a (3.8 mg, 61%) as a single isomer (not determined the geometry of the oxime): mp. 116.0117.2 °C, [α]D28–44.4 (c 0.30, CHCl3); IR (KBr) ν cm−1 3437, 2972, 2933, 1718, 1458, 1379, 1217, 1167, 1082, 1063, 960, 754; 1H-NMR (600 MHz, DMSO-d6) δ 9.59 (s, 1H), 9.43 (d, J=2.1 Hz, 1H), 8.87 (d, J=2.1 Hz, 1H), 8.76 (s, 1H), 8.12 (d, J=8.0 Hz, 1H), 8.10 (d, J=8.0 Hz, 1H), 7.84 (apparently t, J=8.0 Hz, 1H), 7.72 (apparently t, J=8.0 Hz, 1H), 7.42 (broad d, J=15.5 Hz, 1H), 5.73 (broad d, J=15.5 Hz, 1H), 5.52 (d, J=10.7 Hz, 1H), 4.84 (m, 1H), 4.79 (d, J=7.2 Hz, 1H), 4.48-4.40 (complex m, 3H), 4.31 (d, J=4.3 Hz, 1H), 4.12-4.05 (complex m, 3H), 3.72 (dd, J=9.8, 3.0 Hz, 1H), 3.70 (m, 1H), 3.55-3.40 (complex m, 4H), 3.37 (s, 3H), 3.27 (s, 3H), 3.13 (m, 1H), 3.06 (m, 1H), 2.99 (m, 1H), 2.94 (m, 1H), 2.90 (dd, J=8.2, 2.8 Hz 1H), 2.86-2.71 (complex m, 4H), 2.77 (t, J=7.3 Hz, 1H), 2.40-2.17 (complex m, 3H), 2.38 (s, 6H), 2.07-1.93 (complex m, 4H), 1.78 (m, 1H), 1.71 (s, 3H), 1.60-1.47 (complex m, 3H), 1.23 (m, 1H), 1.09 (d, J=6.9 Hz, 3H), 1.02 (d, J=6.1 Hz, 6H), 0.90 (d, J=6.5 Hz, 3H), 0.82 (t, J=7.4 Hz, 3H); 13C-NMR (125 MHz, DMSO-d6) δ 204.1, 172.6, 147.9, 146.8, 143.1 (2C), 137.2, 135.2, 130.3 (2C), 128.9, 128.5, 128.1 (2C), 127.1, 125.4, 120.8, 112.8, 103.7, 100.7, 80.6, 80.1, 74.4, 72.7 (2C), 72.6 (2C), 70.6, 70.4, 69.7, 69.2 (2C), 68.8, 61.0, 58.1, 44.0, 43.1, 41.6 (3C), 35.2, 31.9, 28.5, 24.5, 21.7, 21.6, 19.3, 17.7 (2C), 17.7, 12.1, 9.7, 8.6; ESI-MS: calcd. for C53H79N6O14: 1023.5654 [M+H]+, found m/z: 1023.5643.). The efficiency of the present ‘click chemistry’ process allowed the preparation of each triazole candidates with sufficient purity for in vitro antibacterial and antimalarial testing as shown in Table 1.

In in vitro tests using Staphylococcus aureus and Enterococcus faecalis, O-(3-quinolyltriazolylmethyl)oxime (8a) and O-phenylthiomethyltriazolyloxime (8c) were effective against both strains with MICs of 0.5–1 μg ml−1, similar to results obtained with the alkyne-bearing oximes (6 and 7). No superior analogs were found from this triazole series with respect to antibacterial properties. However, we obtained results for antimalarial activity in vitro that were different from what we expected. Although all C-9 oxime analogs (3, 6 and 7) showed no activity against the P. falciparum K1 strain, the newly synthesized triazoles demonstrated effective antimalarial properties. Our analysis indicated that the 3-quinolyltriazole function on the desmycosin framework is a promising structure responsible for expression of antimalarial activity in vitro (IC50=0.7–2.0 μg ml−1). The cytotoxicity of triazole analogs were measured by colorimetric 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay with human MRC-5 cells.10 All new analogs (8a–e, 9a and 9b) showed comparatively low cytotoxicity with IC50 value of 43–>100 μg ml−1, except for thioether (9c), and Selectivity Indexes (cytotoxicity [IC50 for the MRC-5 cell]/ antimalarial activity [IC50 for the K1 strain]) of these analogs were promising. Furthermore, the in vivo antimalarial activity of 9a and 9c per s.c. (conducted by using a rodent malaria-derived P. berghei strain according to the 4-days suppressive test established by Peters and co-workers)10, 19, 20 was evaluated. Test compounds (9a and 9c) were solubilized in 10%DMSO-0.5%Tween 80 aqueous solution, each 30 mg kg−1 of which were administered s.c. to mice 2 h after the infection (Day 0) and once a day for three consecutive days (Days 1–3). One day after the last treatment (Day 4), thin blood films were made from the tail blood of mice, and the parasitaemia was determined. The results showed 67% and 51% inhibition of parasite growth, respectively (data not shown).

In conclusion, we have described a highly efficient approach to the synthesis of modular type analogs at C-9 of desmycosin through the use of ‘click chemistry’. The triazole-possessing analogs of desmycosin turned out to be potent antimalarial agents. Based on these findings, further structural optimization and structure-activity relationship studies of this class of compounds are currently in progress.

scheme 1

Preparation of C-9 oxime derivatives via desmycosin from tyrosin: a) NH2OH • HCl, pyrdine; b) TsOH, PPTS, CH3CN; c) MeOH, pyridine; d) Azide reagents, Cu(MeCN)4PF6, TBTA, Cu(0) turning, MeOH. Ts, p-toluenesulfonyl; PPTS, pyridinium p-toluenesulfonate; TBTA, tris-(benzyltriazolylmethyl)amine.