A novel class of fast‐acting antimalarial agents: Substituted 15‐membered azalides

Background and Purpose Efficacy of current antimalarial treatments is declining as a result of increasing antimalarial drug resistance, so new and potent antimalarial drugs are urgently needed. Azithromycin, an azalide antibiotic, was found useful in malaria therapy, but its efficacy in humans is low. Experimental Approach Four compounds belonging to structurally different azalide classes were tested and their activities compared to azithromycin and chloroquine. in vitro evaluation included testing against sensitive and resistant Plasmodium falciparum , cytotoxicity against HepG2 cells, accumulation and retention in human erythrocytes, antibacterial activity, and mode of action studies (delayed death phenotype and haem polymerization). in vivo assessment enabled determination of pharmacokinetic profiles in mice, rats, dogs, and monkeys and in vivo efficacy in a humanized mouse model. Key Results Novel fast‐acting azalides were highly active in vitro against P. falciparum strains exhibiting various resistance patterns, including chloroquine‐resistant strains. Excellent antimalarial activity was confirmed in a P. falciparum murine model by strong inhibition of haemozoin‐containing trophozoites and quick clearance of parasites from the blood. Pharmacokinetic analysis revealed that compounds are metabolically stable and have moderate oral bioavailability, long half‐lives, low clearance, and substantial exposures, with blood cells as the preferred compartment, especially infected erythrocytes. Fast anti‐plasmodial action is achieved by the high accumulation into infected erythrocytes and interference with parasite haem polymerization, a mode of action different from slow‐acting azithromycin. Conclusion and Implications The hybrid derivatives described here represent excellent antimalarial drug candidates with the potential for clinical use in malaria therapy.

Conclusion and Implications: The hybrid derivatives described here represent excellent antimalarial drug candidates with the potential for clinical use in malaria therapy.

K E Y W O R D S
antimalarial, azalide, in vivo efficacy, macrolide, malaria, mode of action, pharmacokinetics

| INTRODUCTION
Malaria incidence continues to overwhelm the tropics with estimated 228 million new infections and almost 405,000 deaths in 2018, with antimalarial drug resistance additionally hindering control of the disease (WHO, 2019). Of the five Plasmodium species infecting humans, Plasmodium falciparum is the leading cause of mortality, and its decreasing susceptibility to current antimalarial drugs increases the risk of inadequate therapy (Ashley et al., 2014;Fairhurst et al., 2012;Imwong et al., 2017;Menard & Dondorp, 2017). Even though prevention and control efforts produced measurable effects on public health, new management tools and intensified drug discovery efforts are crucial in controlling malaria and expanding available treatment options (Anthony, Burrows, Duparc, Moehrle, & Wells, 2012;Wells, Hooft van, & Van Voorhis, 2015).
Azithromycin (Figure 1), a 15-membered macrolide antibiotic belonging to the semi-synthetic azalide subclass, is characterized with a broader spectrum of antibacterial activity, improved PK/PD properties, and safety profile (approved for use in children ≥6 months of age) than its forerunners (Schönfeld & Kirst, 2002).
Azithromycin mainly inhibits the growth of asexual forms of Plasmodium strains in the blood. The target of its antimalarial mode of action is in the apicoplast, a plant-like organelle, where it binds to the prokaryote-like ribosomes and inhibits protein synthesis (Dahl & Rosenthal, 2007;Sidhu et al., 2007). This antimalarial mode of action exerts a slow killing effect on the parasites known as "a delayed death phenotype" (Fichera & Roos, 1997). Through this mode of action, azithromycin shows a marked increase in in vitro potency upon prolonged drug exposure (from 48 to 96 h). Although clinical evaluation showed the safety and usefulness of azithromycin in malaria, its future as a new antimalarial agent is uncertain as strong confirmation for the equivalence or superiority to other antimalarials is lacking (Rosenthal, 2016;van Eijk & Terlouw, 2011).
In the recent years, progressively challenging requirements are set up for new antimalarial drug candidates. Ideally, new antimalarial compounds should show fast and high reduction in parasite load, therapeutic efficacy after ideally a single dose, oral bioavailability (F), activity against all known clinical strains, block parasite transmission, prevent relapse, be safe in adults, children, and during pregnancy plus providing affordable therapy for the patients in the developing countries (Burrows, van Huijsduijnen, Mohrle, Oeuvray, & Wells, 2013; malERA Consultative Group on Drugs, 2011;Wells, Alonso, & Gutteridge, 2009). Azalides, as a class of compounds, have the optimal potential to generate the lead drug candidates with most of such desirable properties for the treatment and/or prophylaxis of malaria (Paljetak et al., 2017), particularly showing promising long half-life suitable for single-dose treatments. In this work, we present results from the discovery efforts of fast-acting and potent 15-membered antimalarial azalides (Bukvic et al., 2011;Peric et al., 2012;Pesic et al., 2012;Starcevic et al., 2012), focusing on the profiling of the four representative compounds generated within this class. The compounds presented here (Figure 1) are the main leads (compounds 1, 2, What is already known • New antimalarials are urgently needed to eradicate malaria, one of the most devastating infectious diseases.
• Ideal antimalarial characteristics include fast-acting, resistant parasites coverage, orally available, no relapse and single-dose efficacy.

What this study adds
• A novel class of antimalarial hybrid molecules with improved performance over starting azithromycin and chloroquine.
• These derivatives show characteristics consistent with ideal new antimalarial compounds.

What is the clinical significance
• These compounds could be used for treatment and/or prophylaxis of malaria.
• Their pharmacological properties ensure affordability for patients in developing countries. and 3) resulting from our discovery program showing the most balanced potency and drug metabolism and pharmacokinetics (DMPK) properties. Compound 4 is presented and discussed because of its intriguing mode of action.
The synthesis of compound 4 is schematically presented in Scheme 1.

| Cytotoxicity assay
The detailed protocol for cytotoxicity measurement is described elsewhere (Verbanac et al., 2005). In brief, HepG2 cells were maintained in complete RPMI 1640 medium supplemented with 10% FBS at 37 C in a 5% CO 2 atmosphere. Each culture in the 96-well plates contained 50,000 cells which were exposed to serial dilutions (1:2) of compounds initially dissolved in DMSO and subsequently diluted in supplemented RPMI 1640 medium. Compounds were tested in duplicates, and plates were incubated for 24 h at 37 C in 5% CO 2 . The cytotoxicity assay was performed using the MTS CellTiter 96 AQueous One Solution Cell Proliferation Assay (Promega, USA).
After the addition of MTS reagent and 2 h of incubation at 37 C in 5% CO 2 , the absorbance at 490 nm was recorded and Tox 50 determined based on the obtained response curves.

| Accumulation and retention in human cells
The human biological samples were sourced ethically, and their research use was in accord with the terms of the informed consents under an IRB/EC approved protocol. Human erythrocytes were isolated from fresh whole blood. Blood cells were removed from the plasma by centrifugation at 1,650 g and 4 C for 15 min. The pellet was re-suspended in saline and centrifuged at 3,000 g and at room temperature for 10 min. Supernatant and the upper part of pellet containing leukocytes were discarded. The procedure was repeated three to four times until there were no traces of leukocytes. The quality of erythrocyte isolation was determined on Sysmex SF-3000 haematological analyser. To measure compound accumulation in cells, erythrocytes (2 × 10 9 ) were re-suspended in 3-ml RPMI 1640 (Gibco, Invitrogen) medium containing 20 μM of tested compounds and incubated at 37 C for 3 or 24 h. Cells were subsequently washed twice in ice-cold PBS (Sigma) followed by centrifugation at 3,000 g and 4 C for 10 min. The cells were lysed by freezing in 0.5% Triton X-100 (Sigma) in deionized water. To determine retention in cells, erythrocytes were firstly loaded with tested compounds (20 μM, 3 h), washed in ice-cold PBS, and then incubated in pure RPMI medium for the next 0.5 or 3 h. Cells were then washed and lysed as described above.
Concentrations in lysates were determined by LC-MS/MS. Samples were diluted 10-fold in 0.5% Triton X-100. Preparation of standards for calibration curve, sample preparation for LC-MS/MS, and analysis were carried out as described previously (Munic, Kelneric, Mikac, & Erakovic Haber, 2010), except that roxithromycin was used as an internal standard. To determine intracellular concentration, measured concentration in lysates was normalized to the volume of erythrocyte pellet in samples.
Accumulation and retention in human primary cells was determined as described in detail previously (Stepanic et al., 2011). Briefly, were obtained by cultivating monocytes with 5 ngÁml −1 rhGMCSF for 10 days. To determine compound accumulation cells were incubated with 3-10 μM of compounds in their corresponding culture medium S C H E M E 1 The synthesis of compound 4. Reagents and conditions: i, 2-naphthyl isocyanate, DCM, room temperature, 1 h, yield = 89%; ii, I 2 , MeOH, 500-W lamp for 2.5 h then room temperature, 24 h, yield = 83%; iii, acetyl chloride, TEA, DCM, room temperature, 4 h, yield = 63% for 3 h at 37 C in 5% CO 2 , washed, and lysed. To measure cellular retention of compounds, after being washed, drug-loaded cells were incubated in fresh medium for 3 h, washed, and lysed. Compound concentrations were determined by HPLC-MS/MS analysis and expressed as % of azithromycin measurements.

| Haem polymerization
Experiments were performed according to Tripathi, Khan, Walker, and Tekwani (2004). Briefly, a solution of sodium acetate containing mono-oleoyl glycol and the corresponding inhibitor at different concentrations is placed in 1.5-ml reaction tubes. The reaction is triggered by adding a solution of haem (100-μM final concentration), and after mixing, the tubes are incubated at 37 C overnight to allow the reaction to take place. Multiple washing steps with Tris/SDS and bicarbonate, to remove monomeric haem and haem aggregates, allow quantification of de novo formed β-haematin.

| Metabolite identification in vitro
Metabolite identification in vitro was performed using cryopreserved rat and human hepatocytes (XenoTech, Kansas City, USA). Test compounds (10 μM) were incubated for 4 and 24 h at 37 C in supplemented DMEM containing 0.5 × 10 6 cellsÁml −1 . The reaction was stopped by addition of 80:20 acetonitrile:methanol: 1% formic acid. Samples were analysed by LC-MS/MS and examined by ACD (Advanced Chemistry Development, Toronto, Canada) IntelliXtract software for potential metabolites.

| P. falciparum murine model
The in vivo efficacy evaluation of the compounds was performed using a P. falciparum murine model of malaria in non-myelodepleted engrafted with human erythrocytes as described previously

| Pharmacokinetic studies in mice
For pharmacokinetic (PK) studies in mice, animals were dosed either by intravenous (i.v.) or oral (p.o.) route. For i.v. administration, the dosing volume was 10 mlÁkg −1 for a total dose of 12.5 mgÁkg −1 , and for p.o. administration, the dosing volume was 20 mlÁkg −1 for a total dose of 12.5 mgÁkg −1 . Dosing solutions were prepared in 100% saline

| Pharmacokinetic studies in rats
For the in vivo pharmacokinetic study, compound 1 was dosed i.v. at

| Pharmacokinetic studies in dogs and monkeys
Non-naïve male cynomolgus monkeys or male beagle dogs were used for single-dose PK studies and were fasted overnight prior to administration of test compound. Food was returned 4 h post-dose, and water was provided freely throughout the studies. Dosing solutions were prepared in 0.9% saline supplemented to a final concentration of 1% acetic acid. An i.v. dose of 2 mgÁkg −1 (dose volume of 1 mlÁkg −1 ) and a p.o. dose of 10 mgÁkg −1 were administered (dose volume of 5 mlÁkg −1 ) to monkeys and dogs (three animals per dose group). After i.v. dosing, blood samples were collected at 5, 10, 15, and 30 min, 1, 2, 4, 6, 8, 24, 48, 72, 96, 144, and 168 h post-dose. After oral dosing, blood samples were collected 15 and 30 min, 1,2,4,6,8,24,48,72,96,144, and 168 h post-dose. Blood samples were stored at −80 C until analysis and were assayed as described for mouse PK.

| Data analysis
The data and statistical analysis comply with the recommendations of the British Journal of Pharmacology on experimental design and analysis in pharmacology. Data collected from animal studies include analyses on samples collected from three animals, so we consider the analysed data to have exploratory value. Declared group size is the number of independent values, and data analysis was done using these independent values. No outliers were identified in reported experiments, and no data were excluded from analyses. The methodology used for data analysis for each experimental design is described in detail in the specified sections above.

| Materials
Azithromycin and chloroquine were supplied by USP (Rockville, USA) and [ 3 H]hypoxanthine was supplied by PerkinElmer (Boston, USA).

| Nomenclature of targets and ligands
Key protein targets and ligands in this article are hyperlinked to corresponding entries in the IUPHAR/BPS Guide to PHARMACOLOGY (http://www.guidetopharmacology.org).

| RESULTS
The four compounds belonging to structurally different azalide classes were tested and their activities compared to azithromycin and chloroquine, the two main chemical "building blocks" used to construct the novel compounds ( Figure 1). Compounds have a 4-amino chloroquinoline moiety covalently linked through a propyl amine chain to the azalide scaffold: at 9a-N position in compound 1 and at 2 0 -Oposition in compounds 2 and 3. Additionally, the cladinose sugar was removed from the 15-membered azalide scaffold in compound 3.
Compound 4, as one of the most active non-chloroquinoline compounds detected in the screening process, was chosen to test beyond the azithromycinchloroquine paradigm and provide a proof-ofconcept for the choice of the azalide scaffold as the basic building block. The synthesis of compound 4 is presented in Scheme 1.

| In vitro activity
Antimalarial activity of compounds was initially determined in P. falciparum, 3D7A (sensitive), and W2 (chloroquine/pyrimethamine resistant) strains (Table 1). After 72 h of incubation, compounds demonstrated similar antimalarial potency against both strains (ranging from 5 to 40 nM) with potencies increased more than 1,900-fold over azithromycin and 62-fold over chloroquine. Also, the activity/toxicity window (the ratio between P. falciparum IC 50 and HepG2 Tox 50 s) showed that the inhibitory concentrations on P. falciparum growth were 1,000-to 7,500-fold lower than concentrations inhibiting the growth of the human cell line. Substantial amounts of the compounds were subsequently retained in the cells following the 0.5-and 3-h washout period in the drug-free medium. The concentrations measured after the washout were 3-19and 9-50-fold higher than for azithromycin and chloroquine, respectively, and up to 45% of the initially accumulated amount was retained in the erythrocytes compared to only 10% for azithromycin.
Interestingly, in a set of primary human cells known to accumulate macrolide antibiotics, accumulation and retention of compound 1 was similar to that of azithromycin (Table S1), rather than threefold to fourfold higher as observed in erythrocytes.

| Pharmacokinetics and in vivo efficacy in mouse models
Compounds were administered intravenously (IV) and orally (PO) to CD-1 mice, and the pharmacokinetic (PK) data are summarized in Table 2. Following i.v. administration to CD-1 mice, compounds 1-3 were characterized by a very low systemic CL, a moderate (compounds 1 and 2) to large (compound 3) Vss, a long half-life, and low (compounds 1 and 2) to moderate (compound 3) oral F. Although the oral F was lower in comparison to azithromycin, the oral exposure measured in whole blood (DNAUC PO ) was comparable and even higher at equivalent doses, likely due to a lower Vss. The B/P ratio was higher in comparison to azithromycin for compounds 1-3, in line with the results obtained from in vitro uptake/retention experiments in the red blood cells. Compound 4 displayed a notably divergent PK profile, with the highest clearance and a low Vss leading to the shortest half-life. Oral exposure was also very low and related to the lowest oral F observed. Also, compound 4 did not exhibit in vivo, the same erythrocyte uptake observed in vitro, with a low B/P ratio.
Overall, PK profile exhibited was discouraging, and compound 4 was The oral exposure for compounds 1-3 was evaluated in P. falciparum-infected and uninfected humanized mice, along with azithromycin and chloroquine (Table 3). The DNAUCs for compounds 1-3 were 1.5-3.9-fold higher in infected versus uninfected mice and 1.4-3.5-fold higher in comparison to azithromycin. Interestingly, the intrinsic in vivo potency of compound 2 expressed as the daily molar exposure necessary to achieve 90% of parasitaemia reduction (AUC ED90 ) was 4.4 μMÁh −1 Áday −1 which is close to the potency estimated for chloroquine of AUC ED90 = 3.1 μMÁh −1 Áday −1 .
An additional study with compound 1 in fed versus fasted mice resulted in a threefold increase in oral F in fasted animals (Table S2), indicating that food may substantially inhibit the absorption of these compounds.
Compounds were further profiled in SD rats following i.v. and p.o. dosing, and PK parameters are summarized in Table 4. PK behaviour of compounds 1 and 3 in the rat was consistent with the profile observed in mice with low systemic CL, moderate Vss, very long halflife, and low oral F. Compound 2 displayed a slightly higher oral F, low CL, large Vss, and longer half-life.
As compound 2 showed the most promising in vivo efficacy, it was selected for further profiling of its DMPK properties.

| Further parasitological profiling and mode of action studies
The observed high potency as well as the hybrid nature of the novel azalides motivated target-oriented studies to better understand their anti-plasmodial mode of action.

| Activity against a panel of chloroquineresistant strains
Testing of six additional chloroquine-resistant P. falciparum strains (Dd2, K1, V1/S, FCR-3, T9/94, and 7G8) was performed, and results confirmed a high level of activity across different resistance genotypes F I G U R E 3 Efficacy of azithromycin (AZ), chloroquine (CQ) and compounds 1, 2, and 3 in Plasmodium falciparum-infected NODscidβ2m −/− mice engrafted with human erythrocytes. (a) Estimated ED 90 after oral administration for four consecutive days (once daily). Data are the mean of the log 10 [% parasitaemia at day 7] in groups of N = 3 mice. SD is indicated as half upper bar. (b) Analysis of P. falciparum in peripheral blood of mice treated with vehicle or compound 2 by microscopy and flow cytometry. Samples were taken 48 h after start of treatment (1 cycle of exposure to drug) and phenotypes (Table 5). Compound 2 showed the best activity with IC 50 s being consistently low (3-6 nM) against all strains tested.

| Antibacterial activity
As the novel azalides are based structurally on azithromycin, a successful antibiotic, the antibacterial activity of these new azithromycin derivatives was confirmed using a panel of five bacterial strains (Table S3) Compound 3 had no antibacterial activity while compounds 2 and 4 showed only minor activity against M. catarrhalis.

| Delayed death phenotype
Azithromycin manifests a reasonable in vitro efficacy as an antimalarial agent, although its true potency is only shown when parasite growth is assessed after two life cycles. This behaviour, known as the T A B L E 3 PK parameters, mean (SD), N = 3, estimated in blood after oral administration to Plasmodium falciparum infected (I) and uninfected (U) humanized mice Note: Compounds were dosed at 12.5 mgÁkg −1 and chloroquine at 10 mgÁkg −1 .
T A B L E 4 PK parameters, mean (SD), N = 3, estimated in blood after intravenous (2 mgÁkg −1 ) and oral gavage (10 mgÁkg −1 ) administration to Sprague-Dawley rats, beagle dogs, and Cynomolgus monkeys  delayed death phenotype, is typical of antimalarials exerting their mode of action through inhibition of parasite organelles (Fichera & Roos, 1997 (Sidhu et al., 2007) when assayed at 96 h. None of the novel azalides were affected by the described mutation, and potencies remained at a similar level for both 48-and 96-h exposure time points (Table 6).

| Cell-free assay
Cell-free haem polymerization assay was performed to confirm the quinoline-like mode of action of the new compounds. Compounds 1, 2, and 3 inhibited haem polymerization process with even higher efficiency than chloroquine while compound 4 demonstrated $10-fold weaker potency (Table 1) in line with the molecular structure containing 4-amino chloroquinoline (compounds 1, 2, and 3) or naphthyl aromate (compound 4).

| DISCUSSION
Chloroquine, one of the most successful antimalarial drugs of all time in terms of its efficacy, good tolerance and low cost, has lost its previous usefulness due to the spread of resistant strains that cause recurrent treatment failures (WHO, 2015). On the other hand, azithromycin is still being tested in clinical studies as a partner drug for treatment and prevention of malaria showing attractive safety features and additional health benefits for children, pregnant women, and neonates (Gilliams et al., 2014;Porco et al., 2009;See et al., 2015;Taylor et al., 2003;Unger et al., 2015). As the right combination partner for azithromycin as well as their therapeutic niche are still under evaluation, the fate of azithromycin as a future antimalarial is not known at present (Rosenthal, 2016;van Eijk & Terlouw, 2011).
In our novel drug design, we envisaged a class of compounds that would combine the best properties of these two antimalarial drugs into one hybrid molecule, not just in terms of antimalarial activity but also in overcoming resistance and taking advantage of combined physicochemical properties. Hybrid molecules are chemical entities with two or more structural domains having different biological functions, ideally dual activity, and acting as two distinct pharmacophores (Meunier, 2008). Our new hybrid molecules consist of the azalide backbone with the chloroquinoline or naphthyl aromate (a pharmacophore enhancing antimalarial activity of azalides) units bound at different positions (Bukvic et al., 2011;Peric et al., 2012;Pesic et al., 2012).
The four compounds reported here exhibited excellent in vitro activity against P. falciparum (low nM range) improving their potency 2-3 orders of magnitude over azithromycin (Table 1). Moreover, these compounds demonstrated consistently high potency against an array of resistant strains with various resistance patterns, including CQ R and AZ R strains (Tables 5 and 6).
Consequently, in the P. falciparum-infected mice, compounds 1, 2, and 3 demonstrated low effective doses and quick parasiticidal effect (pyknotic or aberrant trophozoites were detected already in the first parasite cycle). Interestingly, azithromycin did not show efficacy after 4-day oral administration (two parasite cycles) in P. falciparum-infected humanized mice. These results contrast with those from previously reported animal malaria models (Andersen et al., 1995;Gingras & Jensen, 1993)   The PK analysis in mice (Tables 2 and 3) revealed that despite the lower oral F, the blood exposures of the compounds 1-3 were quite high, presumably a consequence of their low clearance and long halflife. Compound 2 demonstrated consistently very long half-lives, low clearances, and moderate oral F across species with oral exposures in the range or higher than azithromycin (Girard et al., 1987). Such PK properties might ensure infrequent dosing (once daily or even singledose cure) and better compliance to therapy. Also, the new compounds exhibited superior metabolic stability across species, with no products of metabolism observed for compound 2 in cryopreserved hepatocytes.
The uptake and retention in human erythrocytes revealed that compounds are abundantly accumulated and retained in these cells to a much higher extent than azithromycin and chloroquine ( Figure 2).
This was supported by the in vivo PK data measuring B/P ratios where the accumulation in the whole blood was up to 14-fold higher than in plasma (Table 2) leading to the increase of the compound's concentration directly at the site of parasite infection. Also, in the infected P. falciparum humanized mice, oral exposures in the blood were higher than the exposures seen for azithromycin and chloroquine, demonstrating their additional capacity to accumulate in parasitized blood, for example, infected erythrocytes, and ensuring that sufficient amounts are available for the eradication of parasites after oral treatment. As it was previously shown that the critical molecular properties for cellular accumulation and retention of a basic macrolide are its lipophilicity and charge (Stepanic et al., 2011), we propose that for these new azalides the increased number of positively charged centres and the introduced aromatic rings in the scaffold increase the alkaline (cationic) as well as lipophilic properties, compared with those of azithromycin. Consequently, this enhances their ability to cross cellular membranes and become protonated at physiological pH. The gradient between the cytosolic pH of erythrocytes (7.1-7.3) and the pH of plasma (7.4) could, thus, lead these cationic amphiphilic compounds (CADs) towards the erythrocyte cytosol and, following intracellular protonation, trap them inside the cell (Funder & Wieth, 1966;Kaufmann & Krise, 2007;Stepanic et al., 2011). The same mechanism could also lie behind the additional compound accumulation observed in infected blood through further sequestration into the parasite acidic food vacuole (pH 5.4). Physicochemical properties of our novel azalides are apparently quite unique and specific since the high accumulation observed for the uninfected and infected erythrocytes does not translate into other cell types (Table S1) and remains at the level of azithromycin, thus potentially minimizing the risk of CAD-associated side effects and maximizing parasite targeting.
Furthermore, studies of the molecular mechanisms underlying the activity of the novel azalides provided deeper insights into two assumed mechanisms of action: inhibition of prokaryotic protein synthesis (present in the parasite apicoplast and targeted by azithromycin; Sidhu et al., 2007) and haemozoin crystallization inhibited by 4-aminoquinolines (Roepe, 2009). The assessment of the macrolide mode of action by testing for their antibacterial activity revealed that compound 1 inhibited the growth of whole cell bacteria while the other three compounds had no antibacterial activity, thus indirectly pointing to the lack of ribosome targeting in compounds 2-4. It was previously shown that the antibacterial potency could be eliminated by modifying the 2 0 -OH position of the azalide scaffold with the quinoline substituents, as this position is of high importance for target binding and this modification leads to the disruption of the compound-ribosome steric complementarity Schlunzen et al., 2001).
Interestingly, the lack of delayed death phenotype in P. falciparum and retained potency against azithromycin-resistant strains for all compounds implied an antimalarial mode of action distinct from that of azithromycin. This prompted us to carry out the haem polymerization test to evaluate inhibition of haemozoin formation, and the results demonstrated that the compounds with chloroquinoline substituents (compounds 1, 2, and 3) inhibit haem polymerization with potencies equivalent to chloroquine. As chloroquine efficacy as well as resistance mechanisms involve complex and not fully understood biochemistry and physiology of haem chemistry and membrane transport (Roepe, 2009) to other three compounds, it is conceivable that as yet unknown mode of action might also play a role in the activity of this compound, but also of other novel azalide derivatives (Bukvic et al., 2011;Peric et al., 2012;Pesic et al., 2012;Starcevic et al., 2012).
In conclusion, we propose that the novel substituted 15-membered azalides studied here, achieve their anti-plasmodial activity by combining and complementing their chemical, biological, and pharmacological characteristics: (i) hybrid molecule strategy contributing to the favourable physicochemical features, (ii) the mechanism of action shifting from slow to fast parasite elimination, and (iii) favourable pharmacokinetic properties (longer half-life, lower clearance, and increased accumulation into blood compartment and even more into infected erythrocytes) combined with excellent metabolic stability (no compound degradation nor parent compound action) enabling fast and efficient eradication of P. falciparum parasites after oral dosing. Moreover, chemical development studies of compound 2 revealed the crystalline form as non-hygroscopic, solvent-free, highly soluble in biorelevant media, and physicochemically stable for up to 4 weeks in solid and dissolved state (Filic et al., 2011).
The presented results demonstrate that the described macrolide derivatives have excellent efficacy and encouraging drug-like properties. This will facilitate their progression through the drug development pipeline, as well as the search for appropriate drug partner(s). This new therapy could provide successful new medicines in different malaria-related indications with a high likelihood of wide use. Additional work in preclinical development and safety profiling is needed in order to fully explore the potential of these compounds and progress this class as a novel antimalarial treatment option.