Antiparasitic effect of synthetic aromathecins on Leishmania infantum

Background Canine leishmaniasis is a zoonotic disease caused by Leishmania infantum, being the dogs one of the major reservoirs of human visceral leishmaniasis. DNA topology is a consolidated target for drug discovery. In this regard, topoisomerase IB – one of the enzymes controlling DNA topology – has been poisoned by hundreds of compounds that increase DNA fragility and cell death. Aromathecins are novel molecules with a multiheterocyclic ring scaffold that have higher stability than camptothecins. Results Aromathecins showed strong activity against both forms of L. infantum parasites, free-living promastigotes and intra-macrophagic amastigotes harbored in ex vivo splenic explant cultures obtained from infected BALB/c mice. However, they prevented the relaxation activity of leishmanial topoisomerase IB weakly, which suggests that the inhibition of topoisomerase IB partially explains the antileishmanial effect of these compounds. The effect of aromathecins was also studied against a strain resistant to camptothecin, and results suggested that the trafficking of these compounds is not through the ABCG6 transporter. Conclusions Aromathecins are promising novel compounds against canine leishmaniasis that can circumvent potential resistances based on drug efflux pumps.


Background
Canine leishmaniasis (CanL) is a serious zoonotic disease caused by L. infantum in the Old World and L. infantum chagasi in the New World. Dogs affected by this disease become reservoirs of human visceral leishmaniasis, being extremely relevant the presence of L. infantum as its subspecies in Latin America, mainly in Brazil. Although there are several vaccines in use, and preventive measures including collars impregnated with insecticide or spot-on drops contribute to reduce the endemicity of this disease, the use of drugs is strictly necessary when the signs and symptoms appear in the animal. Nowadays, antimony-based (Sb V ) drugs, alone or in combination with allopurinol, are considered the gold standard treatment against CanL in southern European countries, whereas the oral drug miltefosine is used as second-line drug [1]. However, the mandatory parenteral administration of Sb V and the multiple side effects of Sb III (product of enzymatic activation of the drug by host enzymes), are some of the causes of treatment interruption, which favors the emergence of relapses in the first-year post-treatment. Related to drug misuse is the emergence of resistant strains that may be triggered by host or parasite factors. Among the factors linked to the host, the most prevalent are those related to the alterations of pharmacokinetic parameters or the immunological system. On the other hand, factors related to parasites include structural modifications of target proteins, along with overexpression of ABC and multiple drug resistance (MDR) proteins [2]. Therefore, drug discovery research in this field is absolutely necessary to find new drugs for the management of CanL.
DNA topoisomerases are consolidated targets for drug development in cancer and infectious diseases. DNA topoisomerase IB (TopIB) is involved in relaxing supercoiled DNA by a DNA breaking and rejoining process. In this process TopIB cleaves one DNA strand by nucleophilic attack from the catalytic tyrosine placed in the active site, which becomes linked to 3′ phosphate end of DNA generating a reversible DNA-enzyme cleavage complex. The unbroken strand rotates through the gap and finally the DNA backbone is rejoined [3]. Leishmania TopIB (LTopIB) is interesting from a therapeutic point of view due to its heterodimeric structure, which is dissimilar to the monomeric Top IB found in the rest of animal species [4]. TopIB inhibitors have been classified as TopIB poisons and TopIB inhibitors. TopIB poisons, such as camptothecin (CPT) and other non-CPT compounds like indenoisoquinolines, are anticancer drug leads [5,6], and several reports have shown their efficacy as trypanocide [7] and leishmanicide drugs [8]. These TopIB poisons trap the cleavage complex preventing the final step of rejoining, intercalating into the DNAenzyme complex and generating single strand breaks that evolve to double strand breaks when the replication fork collides with the stabilized cleavage complex [3]. On the other hand, TopIB inhibitors do not stabilize the cleavage complex; they inhibit the enzyme, preventing its binding with DNA by the interaction with the enzymatic catalytic domains or with the DNA substrate [3].
Aromathecins are a new class of TopI poisons, described as stable hybrids of indenoisoquinolines and camptothecins that also show similarity to the natural product luotonin A, which is a weaker TopIB poison [9] ( Fig. 1). Several series of modified and substituted luotonins have been published and some analogues have greater antiproliferative activity than the parent compound. 22-Hydroxyacuminatine [10], a rare natural product, contains the 12H-5,11a-diazadibenzo [b, h] fluoren-11-one system, which is known as "rosettacin" (Fig. 1). Rosettacin derivatives generate aromathecins, which are more stable than camptothecins and maintain lower, but mensurable, topoisomerase poisoning activity [11] (Fig. 1). Further structure-activity studies revealed some trends about these compounds, such as substitutions at C-14 that have been related to an increased anti-cancer activity [11][12][13] or the ethylenedioxy bridge between C-2 and C-3 that has been associated with an enhanced antitrypanosomal activity [14].
The only TopIB inhibitors FDA approved (irinotecan and topotecan) are CPT analogs. These compounds have several limitations as chemical instability of the CPT alpha-hydroxy-lactone ring and loss of efficacy due to the drug efflux-mediated resistance generation [6]. Therefore, it is necessary the development of new non-CPT TopIB inhibitors that overcome these limitations such as aromathecins.
In Leishmania have been described two mechanisms of CPT resistance using strains exposed to increasing concentrations of CPT. The first one involves overexpression of the ABCG6 transporter [15]; the other mechanism presumes the amino acid substitutions Gly185Arg and Asp325Glu in the large subunit of the LTopIB enzyme [16]. In mammalian cells the drug efflux pump involved in CPT resistance is the ABCG2 transporter and it has been described that indenoisoquinolines are poor substrates for this transporter besides the multiple drug resistance (MDR)-1 protein [17].
In this report, we describe the antileishmanial activity of two series of aromathecins, which have been kindly provided by Dr. Mark Cushman (Dpt. of Medicinal Chemistry, Purdue University, Indiana, USA), against both stages of L. infantum; free-living promastigotes and intra-macrophagic amastigotes present in splenic explants obtained from infected BALB/c mice. In addition, it was assessed their activity as LTopIB inhibitors and their ability to overcome the transporters involved in CPT resistance.

Results
Two series of aromathecins (Table 1) have been tested against both stages of L. infantum, free-living promastigotes and intracellular amastigotes harbored in mouse splenic cells. Experiments were carried out using a genetically modified L. infantum strain that constitutively produces the iRFP protein. This makes it possible to evaluate the viability of parasites measuring the fluorescence emitted at 708 nm [18].
In order to enhance the antiproliferative activity and TopIB inhibition, the aromathecins used in this study include different substituents (amines, amino alcohols and nitrogenous heterocycles) at the C-14 position which adding solubility and stability to the DNA-enzyme complex [11] (Table 1). In addition, four of them contain an ethylenedioxy bridge (between C-2 and C-3) in the Aring (aromathecins 7, 8, 9 and 10) which is aimed at improving the TopIB inhibitory activity of these compounds [13] (Table 1). Interestingly, all the compounds tested in this work showed lower EC 50 values in amastigotes than in promastigotes within the micromolar to submicromolar range. Interesting SI values (> 9) were found for compounds 1, 8 and 10. The ethylenedioxy bridge does not seem to exert a significant effect on the antileishmanial potency.
In addition to the antileishmanial effect of these compounds, their activity as LTopIB inhibitors was analyzed in vitro (Fig. 2). Recombinant LTopIB was produced in a TopIB-defective yeast platform, and purified by standard chromatographic protocols [20]. The effect of aromathecins (comp. 1 to 10) on LTopIB was addressed by measuring the relaxation of negatively supercoiled pBluescript-SK DNA plasmid in the presence of different concentrations of these molecules. In order to separate the nicked DNA generated by the cleavage-complex stabilization from the relaxed topoisomers, gels were run in the presence of ethidium bromide. Five compounds (1, 3, 4, 5 and 7) showed a partial inhibition of LTopIB, which started at a very low concentration (0.01 μM) and was not concentration dependent. Only a weak band of nicked DNA could be observed with compound 8, suggesting that these compounds would act as LTopIB inhibitors rather than poisons. The total inhibition of the enzyme was only observed at a high concentration (100 μM) of aromathecins 3, 7 and 9; in addition, compounds 6 and 10 did not have any inhibitory effect, indicating that other targets may be involved.
In order to evaluate if aromathecins would share the transporter involved in CPT resistance, one of the major problems related to the loss of effectiveness of CPT derivatives, a CPT-resistant strain was generated. The CPT-resistant strain used was generated by the exposure of L. infantum promastigotes to increasing concentrations of CPT, from 0.1 to 20 μM (CPT-20, from now on). CPT-20 showed cross-resistance to other CPT derivatives poisons, such as topotecan (EC 50 > 100 μM vs EC 50 = 9.54 ± 0.22 μM in the WT strain) and SN38 (EC 50 = 40.98 ± 0.3 μM vs EC 50 = 4.73 ± 0.14 μM in the WT strain). Amplification and sequencing of the LTopIBencoding genes from CPT-20 and WT strains showed no differences between them. Therefore, it is very likely that the resistance mechanism of CPT-20 is related to the overexpression of the ABCG6 transporter previously described as the drug-efflux pump involved in the CPT resistance mechanism [15]. The CPT-20 strain was susceptible to the aromathecins assayed in this work (2, 5 and 10 were not assayed due to their poor effect against iRFP L. infantum promastigotes), showing very similar EC 50 values to those obtained with iRFP L. infantum.

Discussion
All aromathecins tested have shown leishmanicidal activity with stronger effect over the intracellular amastigote form. Aromathecins have been designed as inhibitors of TopIB enzyme and their mechanism of action would be mediated by the stabilization of the intermediate cleavage-complex [11][12][13]. However, the results obtained in this work indicate that most of these compounds would act as LTopIB inhibitors rather than LTo-pIB poisons, due to only with compound 8 we can observe nick DNA from cleavage-complex stabilization.
Unlike the results obtained with T. brucei [14], the ethylenedioxy bridge does not seem to exert a significant effect on the antileishmanial potency. Besides, this bridge does not seem to improve the LTopIB inhibitory activity, especially when we compare the activity of compounds 1 and 9, with the same substitution at C-14 position. The inhibition of LTopIB with compound 9 that has the ethylene bridge only happed at a high concentration whilst compound 1 inhibit the enzyme at very low concentrations.  The leishmanicidal activity of compounds 6 and 10, which do not have LTopIB inhibitory activity, along with the partial inhibition of the enzyme observed with other compounds indicates that other targets may be involved.
The CPT resistant strain generated in this work by exposition to increased CPT concentrations was susceptible to all the aromathecins tested. These results indicate that aromathecins overcome the ABCG6 transporter, the pump responsible of CPT resistance generation in Leishmania [15]. Similar results have been obtained with other non-camptothecin compounds as indenoisoquinolines, which overcome the ABCG2 transporter involved in CPT resistance in mammalian cells, and the multiple drug resistance (MDR)-1 protein [17].

Conclusions
In conclusion, the aromathecins tested in this work showed a good antileishmanial activity. These results suggest the existence of another mechanism of action complementary to the expected LTopIB poisoning. Structural differences of LTopIB and the mammalian enzyme could explain this behavior. These differences could be used to design new molecules with improved selectivity, whose advantage lies in overcoming the transporter involved in the generation of CPT resistance.

Ex vivo murine splenic explant culture
Primary splenocyte cultures containing intramacrophagic amastigotes were obtained from dissected spleens of female BALB/c mice infected intraperitoneally with 1.5 × 10 9 metacyclic iRFP L. infantum promastigotes, 5 weeks before their sacrifice. Mice were obtained commercially (Janvier-Labs). Animals were housed in specific-pathogen free facilities and they were euthanized by cervical dislocation without anesthesia. These protocols were approved by the Animal Care Committee of University of Leon (project license SAF2017-83575-R), which complies with European Union Legislation (2010/ 63/UE) and Spanish Act (RD 53/2013).

Cytotoxicity and selectivity index determination
The viability of cultured parasites in the presence of different concentrations of aromathecins was determined by measuring the reduction of infrared fluorescence emitted by free-living promastigotes/amastigotes with respect to negative controls of each form of the parasite treated with DMSO (up to 0.1% final concentration) as a vehicle. Thus, in order to calculate EC 50 value, promastigotes or splenic mouse explants, which harbored amastigotes, were incubated with 7 different concentrations of each aromathecin starting from 100 μM and one-third dilutions up to 0.13 μM in duplicate and in three independent experiments. After 72 h incubation at 26°C, the infrared fluorescence emitted at 708 nm by viable promastigotes/amastigotes was measured in an Odyssey (Li-Cor) infrared imaging system. Similarly, the cytotoxic effect (CC 50 ) of aromathecins on uninfected splenic explants from BALB/c mice (cells that naturally harbor the amastigote stage) was measured by the Alamar Blue (Invitrogen) assay. The selectivity index (SI) was determined as the relationship between the CC 50 value and the EC 50 value for amastigotes. CC 50 and EC 50 were calculated by nonlinear analysis using the Sigma-Plot 10.0 statistical package. The aromathecins were dissolved in DMSO and stored at − 20°C before their use as fresh aliquots.

CPT resistant strain generation
The CPT resistant strain used in this work was generated exposing L. infantum promastigotes to increased concentrations of CPT, from 0.1 to the solubility limit of the compound, 20 μM in the case of CPT. Parasites were grown at 26°C in M199 medium supplemented as described previously, each culture being inoculated at an initial density of 10 6 cells/mL. The promastigotes cultured in liquid medium in the presence of 20 μM CPT, were plated on semisolid M199 medium containing 20 μM CTP in order to select a single colony.