Design, Synthesis and In Vitro Investigation of Novel Basic Celastrol Carboxamides as Bio-Inspired Leishmanicidal Agents Endowed with Inhibitory Activity against Leishmania Hsp90

Abstract

The natural triterpene celastrol (CE) is here used as lead compound for the design and synthesis of a panel of eleven CE carboxamides that were tested in vitro for their growth inhibitory activity against Leishmania infantum and L. tropica parasites. Among them, in vitro screening identified four basic CE carboxamides endowed with nanomolar leishmanicidal activity, against both the promastigotes and the intramacrophage Leishmania amastigotes forms. These compounds also showed low toxicity toward two human (HMEC-1 and THP-1) and one murine (BMDM) cell lines. Interestingly, the most selective CE analogue (compound 3) was also endowed with the ability to inhibit the ATPase activity of the Leishmania protein chaperone Hsp90 as demonstrated by the in vitro assay conducted on a purified, full-length recombinant protein. Preliminary investigations by comparing it with the naturally occurring Hsp90 active site inhibitor Geldanamycin (GA) in two different in vitro experiments were performed. These promising results set the basis for a future biochemical investigation of the mode of interaction of celastrol and CE-inspired compounds with Leishmania Hsp90.

1. Introduction

Natural products have been used to date as a convenient source of potential lead compounds in the framework of drug discovery and development and they are now re-emerging as pivotal sources of chemical diversity in the post-genomics era [1]. Among them, plant-derived traditional remedies represent promising drug candidates due their potent pharmacological effects and generally encouraging toxicity profiles [2,3,4]. Celastrol (CE, Figure 1) is a naturally occurring biomolecule extracted from the traditional Chinese medicinal herb Tripterygium wilfordii, commonly referred to as ‘Thunder of God Vine’ [5,6]. Chemically, CE belongs to the class of natural quinone-methide triterpenes and possess a six-membered based pentacyclic structure which further classify it as a pentacyclic triterpenoid. Throughout the years the pharmacological profile of CE has been largely investigated highlighting its ability to modulate a variety of molecular targets both via the formation of covalent Michael adducts and/or by establishing non-covalent interactions (Figure 1) [7].

CE potentialities as drug have been reported for the treatment of different conditions including e.g., obesity, inflammatory, auto-immune and neurodegenerative diseases [2,8]. CE was also identified as potent antiproliferative agent for different human cancer cell lines interfering with different biological targets [2]. Among the others, CE can interact with the human variant of the 90kDa Heat Shock Protein (hHsp90), a ubiquitous protein chaperone in Eukarya known to participate in crucial signaling pathways connected with cell growth and survival in stress-inducing conditions [9,10]. More recently, the growth inhibitory activity of CE-containing plant extracts or of the isolated natural compound against both Plasmodium and Leishmania parasites was reported [11,12].

The therapeutic options against leishmaniasis, a poverty-related protozoal disease, which causes 700,000 new cases worldwide with 26,000–65,000 deaths annually, are still limited and fraught with high costs, severe side effects and the rise and spread of drug resistance [13,14,15]. Interestingly, the transmission and the development of the various forms of this parasitosis are strictly related to the ability of the parasites of the genus Leishmania to differentiate into morphologically distinct life cycle stages via heat-shock induced events. Once transmitted from an infected phlebotomine sand-fly to a mammalian host, Leishmania parasites, in the form of flagellate promastigotes, are phagocytized by macrophages. In response to local factors including the rise of temperature, parasites differentiate at this point to the pathogenic, round and aflagellated amastigote stage. It was previously demonstrated that in these primitive eukaryotes the heat-shock induced differentiation stage is regulated, among others, by specific variants of Hsp90 [16,17].

As a general strategy of target-oriented drug discovery, the pharmacological inhibition of Hsp90 chaperone cycle, which relies on ATP hydrolysis to assist the folding of a plethora of client proteins, has long been investigated by targeting its essential ATPase activity using ATP-competitive active inhibitors [18,19,20,21,22,23]. Accordingly, the two natural compounds radicicol (RAD) and geldanamycin (GA), classified as hHsp90 active site inhibitors (Figure 2), were investigated for their growth inhibitory activity against Leishmania spp. promastigotes. The obtained results led to the possibility of exploiting (semi)synthetic analogues or specific formulations of RAD and GA as leishmanicidal agents targeting the Leishmania Hsp90 variants [24,25,26,27]. Recently, novel small molecules were identified as promising inhibitors of Leishmania spp. Hsp90 highlighting the potentialities of this approach in the research for novel antiprotozoal agents [28,29].

In eukaryotic cells, the activity of Hsp90 is finely regulated by modulating its cytosolic expression and by the action of specific allosteric ligands. These are small proteins or peptides able to positively or negatively modulate the kinetics of Hsp90 ATP-dependent chaperone cycle by perturbing its complex conformational dynamics (Figure 3) [30].

Taking inspiration from this, the use of small molecules-based allosteric modulators of the Hsp90 chaperone cycle [31,32,33,34,35], has gained more and more attention in the field of innovative chemotherapies [36,37,38]. Unlike active site inhibitors, this class of compounds possesses in fact the unique ability to interfere with essential protein–protein interactions within the cells avoiding the insurgence of heat-shock responses, pro-survival mechanisms usually triggered by Hsp90 active site inhibitors [39,40]. Recently, detailed investigations revealed the ability of CE to allosterically block the formation of the protein-complex between human Hsp90 and the signal transduction protein Cdc37 resulting in a quite strong cytotoxicity and promising antiproliferative/anticancer activity [41,42,43,44,45,46]. Remarkably, basic CE derivatives have demonstrated a higher degree of selectivity of action in vitro when tested as potential antiproliferative agents interacting with the human Hsp90-Cdc37 system [47]. Kalayeh et al. also discovered a modest, yet promising, modulatory activity of CE against Leishmania Hsp90 chaperone cycle [17,29].

With the aim of developing novel leishmanicidal agents inspired from natural products, we designed and synthetized a small family of CE carboxamide derivatives (Figure 4). The compounds of Figure 4 are characterized by different polarities and lipophilicities thanks to the groups selected as amide decorations. Seven basic derivatives were prepared to assess the importance of the pH-sensitive nitrogen for antiprotozoal activity (compounds 1–7). Among them, the piperidin-1-yl-ethanamine and the pirrolidin-1-yl-ethanamine CE carboxamides (compounds 1 and 2) were previously reported as allosteric disruptors of the hHsp90-Cdc37 protein complex but never tested against protozoa [47]. Moreover, a group of not basic CE carboxamides including three sulfur containing CE analogues (the two methanethiosulfonates 8 and 9 and the allyldisulfide 10) and the more hydrophilic CE-glucosamide 11 were also synthesized and tested. Methanethiosulfonate and allyldisulfide derivatives have been designed because these moieties could react, as previously reported for other proteins [48,49,50], with the free thiol group of exposed cysteines in synergy with CE ability to form Michael adducts with electrophilic protein residues. Finally, the more lipophilic methyl ester derivative of CE, a natural compound named pristimerin (PR), was also included in our investigation.

All the mentioned compounds were screened in vitro for their growth inhibitory activity against cultures of Leishmania tropica and L. infantum promastigotes. Their cytotoxicity against normal cells was assessed using two human cell lines (the immortalized human microvascular endothelial line—HMEC-1—and the human monocytic line—THP-1) and a murine one (the bone marrow-derived macrophage—BMDM). The most promising compounds were also tested against intramacrophage cultures of Leishmania amastigotes and their selectivity of action was investigated, accordingly.

Intrigued by the improved performances as both Hsp90 inhibitors and antiproliferative agents demonstrated by CE basic carboxamides with the respect to the parental compound [47], the modulatory activity of the compounds of Figure 4 was also studied toward the homologous Hsp90 from Leishmania braziliensis. A preliminary in vitro comparison with GA, the known Hsp90 active site inhibitor endowed with leishmanicidal activity, was also conducted.

2. Materials and Method

2.1. Chemistry

2.1.1. General Information

All commercially available solvents and reagents were used without further purification, unless otherwise stated. TLC was carried out on precoated 60 F254 plates (Merck, Darmstadt, Germany) using UV light. Flash column chromatography (flash column chromatography, CC) was performed using silica gel 60 (32–64 Mesh, Merck). Organic phases were dried over anhydrous sodium sulphate and evaporated under diminished pressure (1–2 kPa) at a bath temperature of 40 °C.

Synthetized compounds were characterized by means of ¹H- and ¹³C-NMR spectra, melting point (Mp) and, with the exception of the previously reported compounds 1 and 3 [47], by high-resolution mass spectrometry. ¹H- (300 MHz) and ¹³C-NMR (75 MHz) spectra were recorded on a Mercury 300VX spectrometer (Palo Alto, CA, USA). Peaks were assigned with 2D COSY experiments and are in agreement with the proposed structures; deuterated chloroform (CDCl₃) was used as solvent. High-resolution mass spectra (HRMS) were conducted on a FT-Orbitrap mass spectrometer (ThermoFisher Scientific, Milan, Italy) in positive electrospray ionization (ESI) mode.

Purities of final compounds were determined by HPLC using CH₃CN/H₂O + CF₃COOH gradient and a Purospher RP 18 5 μm column on an Elite Lachrom instrument (Hitachi, Chiyoda, Tokyo, Japan) equipped with a DAD detector; T_R = retention time. Compound characterization and HPLC data are reported in Appendix A.

2.1.2. Preparation of Compounds 1–11

Carboxamides were prepared as follow: under nitrogen atmosphere, HOBt (1 equiv.) and EDC-HCl (1.5 equiv.) were added to an ice-cooled solution of celastrol (1.0 equiv.) in dry DMF (170 mM) and stirred for 15 min. After that, TEA (1.5 equiv.) and 1.2 equiv. of the selected primary amine were added to reacting mixture which was allowed to warm to room temperature and stirred overnight. Target compounds were isolated after flash column chromatography on silica gel working with a gradient of MeOH in DCM.

2.2. Biological Assays

2.2.1. Recombinant Expression of Full-Length Leishmania braziliensis Hsp90 (LbHsp90)

To evaluate in vitro the modulatory activity of CE, PR and CE caboxamides, a full-length recombinant Hsp90 from L. braziliensis (LbHsp90) was produced in E. coli BL21 DE3 harboring the pET28a_LbHsp90 coding for the C-terminal His-tagged target chaperone (genbank number: XP_001567804.1, full sequence is reported in Appendix B). Pratically, the plasmid pET28a_LbHsp90, kindly donated by professor J.C. Borges [51,52], was transformed in E. coli BL21 DE3 using the TransformAid Bacterial Transformation Kit (Thermo Fisher, Waltham, MA, USA). Subsequently, the obtained transformants were grown overnight in LB medium containing 30 μg mL⁻¹ kanamycin (LB_kan30) and then inoculated in 1 L SB _kan30 medium (Yeast extract 30 g L⁻¹, Tryptone 20 g L⁻¹, NaCl 10 g L⁻¹) at 37 °C and 220 rpm. When the OD₆₀₀ reached 0.5–1, gene expression was induced by the addition of 2 mL of 1 M IPTG solution in water and the culture was shifted at 30 °C for 24 h. After cell harvesting by centrifugation (5000 rpm for 30 min), the cell pellet was resuspended in 20 mL of wash buffer (20 mM HEPES buffer, pH 7.0, 500 mM NaCl, 20 mM imidazole) and recombinant E. coli BL21DE3 cells were disrupted by ultrasonication. LbHsp90 was subsequently purified using a Nickel Sepharose 6 Fast Flow agarose resin (Ni-NTA) (GE Healthcare, Milan, Italy) as follows. Soluble protein fraction was separated from cell debris by centrifugation (10,000 rpm for 30min) after cell ultrasonication, and clear lysates were incubated with the Ni-NTA resin for 90 min at 4 °C under mild shaking and loaded onto a glass column (10 × 110 mm). The resin was then washed with 10mL of wash buffer and His-tagged LbHsp90 was eluted using a 3 step gradient (10 mL washing buffer containing 100, 200, and 300 mM imidazole, respectively) and dialyzed against 20 mM HEPES buffer (20 mM, pH 7.5), at 4 °C overnight. The protein content was measured using the Bio-Rad Protein Assay (Hercules, CA, USA) according to the Bradford method and the protein purity was verified by SDS-PAGE analysis (10% T, 2.6% C). It is worth noting that to avoid interferences with the assays described below, proteins were purified and conserved in HEPES buffer (20 mM, pH 7.5) in place of the standard PB-buffer. Reported data are the mean of two experiments run in duplicate.

2.2.2. Determination of LbHsp90 ATPase Cycle Modulation In Vitro

Modulatory activity toward the ATPase cycle of LbHsp90 was determined using a Malachite Green Phosphate Assay Kit (Catalog Number MAK307, Sigma-Aldrich^®, St. Louis, MO, USA). Analyses were run in a 96-well plate directly following the quantitative formation of a green complex between Malachite Green, molybdate, and free orthophosphate (Pi) produced by ATP hydrolysis catalyzed by the chaperone. Color formation from the reaction was measured on a plate reader at 620 nm. Briefly, the IC₅₀ of CE and its derivatives were extrapolated from a set of dose-response curves obtained by calculating the amount of Pi produced by LbHsp90 in the presence of different compounds concentrations.

Reactions were run in HEPES buffer (20 mM, pH 7.5) containing 3 mM MgCl₂ and 50 mM KCl in the presence of different compounds concentrations (12–500 mM), LbHsp90 (7 μM) and ATP (25 mM) at 27 °C for 60 min. OD₆₂₀ was measured 30 min after Malachite Green reactants addition.

LbHsp90 native activity was measured by conducting the reaction in the presence of just LbHsp90 and ATP and extrapolating the concentration of Pi obtained.

Three sets of control experiments, (1) ATP background hydrolysis, (2) compounds-related interferences and (3) protein-related background coloration, were run as follow:

(1)

ATP (25 mM) incubated at 27 °C for 60 min

(2)

Compounds (different concentrations) incubated with ATP (25 mM) at 27 °C for 60 min

(3)

LbHsp90 (7 μM) at 27 °C for 60 min

2.2.3. Evaluation of LbHsp90 ATPase Kinetics In Vitro

LbHsp90 (3.5 μM) was added to a solution prepared in HEPES buffer (20 mM, pH 7.5) containing KCl (100 mM), MgCl₂ (1 mM), NADH (0.18 mM), L-lactate dehydrogenase (4 U mL⁻¹), phosphoenolpyruvate (1 mM), pyruvate kinase (2.5 U mL⁻¹) and the desired compound (dissolved in DMSO to a final concentration of 50 μM). The reaction was initiated by the addition of ATP (1 mM) and absorbance changes at 346 nm were monitored for 30 min at 30 °C.

Native LbHsp90 kinetics was measured as described but in the absence of DMSO. Two control experiments were run as follows:

(1)

Efficiency of the enzymatic cascade by the used compounds: HEPES buffer (20 mM, pH 7.5) containing KCl (100 mM), MgCl₂ (1 mM), ADP (1 mM), NADH (0.18 mM), L-lactate dehydrogenase (4 U mL⁻¹), phosphoenolpyruvate (1 mM), pyruvate kinase (2.5 U mL⁻¹);

(2)

Interference of the tested compound(s) with the enzymatic cascade: HEPES buffer (20 mM, pH 7.5) containing KCl (100 mM), MgCl₂ (1 mM), ADP (1 mM), NADH (0.18 mM), L-lactate dehydrogenase (4 U mL⁻¹), phosphoenolpyruvate (1 mM), pyruvate kinase (2.5 U mL⁻¹) and 50 μM compound.

Appendix B reports the detailed description of the enzymatic cascade used.

2.2.4. Promastigote Stage of Leishmania spp. Cultures and Antileishmanial Activity

Promastigote stage of L. infantum (MHOM/TN/80/IPT1, kindly provided by Dr. M. Gramiccia and Dr. T. Di Muccio, ISS, Roma) and L. tropica (MHOM/SY/2012/ISS3130) were cultured in RPMI 1640 medium (EuroClone) supplemented with 15% heat-inactivated fetal calf serum (EuroClone, Milan, Italy), 20 mM HEPES, and 2 mM L-glutamine at 23 °C. The MTT (3-[4.5-dimethylthiazol-2-yl]-2.5-diphenyltetrazolium bromide) method was used to estimate the 50% inhibitory concentration (IC₅₀), with some modifications [53,54]. Compounds were dissolved in DMSO and then diluted with medium to achieve the required concentrations. Test compounds were placed in 96 wells round bottom microplates and seven serial dilutions made. Amphotericin B was used as the reference antileishmanial drug. Parasites were diluted in complete medium to 5 × 10⁶ parasites/mL and 100 μL of the suspension was seeded into the plates and incubated at 23 °C for 72 h. Then 20 μL of MTT solution (5 mg/mL) was added into each well, after 3 h the plates were centrifuged, the supernatants discarded and the resulting pellets dissolved in 100 μL of lysing buffer consisting of 20% (w/v) of a solution of SDS (Sigma), 40% of N,N-dimethylformamide (Merck) in H₂O. The absorbance of the obtained solutions was measured at a wavelength of 550 nm with a reference at 650 nm with a microplate spectrophotometer (Synergy 4-Biotek, Milan, Italy) and analyzed with the Gen5^® software for data processing.

The results are expressed as IC₅₀, the dose of compound that is necessary to inhibit cell growth by 50%; each IC₅₀ value is the mean ± standard deviation of at least three separate experiments performed in duplicate.

2.2.5. Intramacrophage Amastigotes Stage of Leishmania spp. Cultures and Antileishmanial Activity

For Leishmania infections, THP-1 cells were plated at 5 × 10⁵ cells/mL in 16-chamber Lab-Tek culture slides (Nunc, Frosinone, Italy) and treated with 0.1 μM phorbol myristate acetate (PMA, Sigma) for 48 h to achieve differentiation into macrophages. Cells were washed and infected with metacyclic L. infantum promastigotes at a macrophage/promastigote ratio of 1/10 for 24 h. Cell monolayers were then washed and incubated in the presence of test compounds for 72 h. Slides were fixed with methanol and stained with Giemsa. The percentage of infected macrophages in treated and non-treated cells was determined by light microscopy [55,56].

2.2.6. Evaluation of Morphological Differentiation of Leishmania Cells

L. infantum promastigotes were diluted in complete medium to 4 × 10⁷ parasites/mL and 250 µL of the suspension were seeded into the 24 wells plates and incubated at 23 °C in the presence of geldanamycin, CE or compound 2 at the final concentration of 50 ng/mL. Cultures without drugs incubated at 23 °C or at 37 °C were used as negative and positive control, respectively. After 24 or 48 h of incubation, parasite morphology was evaluated by optical microscopy at 100× (Eclipse Ti Series, Nikon, Minato, Tokyo, Japan) on Giemsa stained smears (a mixture of eosin, methylene blue and Azure-B diluted 1:10). The percent of round promastigotes in treated and non-treated cells was determined. Images were taken at the same magnification by using a digital camera (Nikon Digital Sight). About 500 promastigotes were counted in 4 random fields.

2.2.7. Cytotoxicity Assay on Different Cell Lines (HMEC-1, BMDM, THP1)

Cytotoxicity was evaluated on human microvascular endothelial cells (HMEC-1, provided by the Centers for Disease Control, Atlanta, GA, USA), on immortalized mouse C57Bl/6 bone marrow derived macrophages (BMDM) from wild type lineage (generated in the laboratories of Drs. Douglas Golenbock and Kate Fitzgerald, University of Massachusetts, Worcester, MA, USA) [57] and on a human acute monocytic leukaemia cell line (THP-1).

HMEC-1 were maintained in MCDB 131 medium (Invitrogen, Milan, Italy) supplemented with 10% fetal calf serum (HyClone), 10 ng/mL of epidermal growth factor, 1 μg/mL of hydrocortisone, 2 mM glutamine and 20 mM HEPES buffer, at 37 °C in 5% CO₂. BMDM were maintained in DMEM (Euroclone) medium supplemented with 10% fetal bovine serum (HyClone, Celbio, Milan, Italy), 2 mM glutamine, and 25 mM of HEPES buffer solution (Euroclone). THP-1 were maintained in RPMI supplemented with 10% heat-inactivated FCS, 50 µM 2-mercaptoethanol, 10 µM sodium pyruvate, 20 mM HEPES, and 2 mM glutamine. In order to achieve differentiation into macrophages, cells were treated with 10 ng/mL phorbol myristate acetate (PMA) for 72 h.

For the cytotoxicity assay, cells were seeded in 96 well flat bottom tissue culture clusters (Costar, New York, NY, USA) at the following concentrations: HMEC 10⁵ cell/mL, BMDM at 10⁶ cell/mL, THP1 at 5 × 10⁵ cell/mL. After 24 h, cells were treated with serial dilutions of test compounds in a final volume of 200 µL/well. Cell proliferation was evaluated after 72 h using the MTT assay [53,54].

The absorbances were measured with a microplate spectrophotometer (Synergy 4-Biotek) at a test wavelength of 550 nm and a reference wavelength of 650 nm. The results are expressed as IC₅₀ as the concentration inhibiting 50% of cell growth. Each IC₅₀ value is the mean and standard deviation of at least three separate experiments performed in duplicate.

3. Results

3.1. Leishmanicidal Activity on Promastigotes Cultures and Cytotoxicity Evalutation

The leishmanicidal activity of CE, PR and the carboxamides 1–11 was first measured in vitro in two different species of Leishmania promastigotes (L. infantum and L. tropica) using amphotericin B (AMP) as reference drug. Since CE is reported to possess a plethora of different bioactivities [5,6,7], compounds’ aspecific cytotoxicity was estimated using different mammalian cell lines, both of human and murine origin. Specifically, the immortalized human microvascular endothelial cell line (HMEC-1) and two different macrophage cell lines (human monocytic THP-1 line and the bone marrow-derived murine BMDM line) were used. All the experiments were run as described in the M&M section [53,54]. The results are summarized in

Table 1. Antileishmanial activity on promastigotes of L. infantum or L. tropica and cytotoxicity on human and murine cell lines.

Leishmania Promastigotes (IC50, μM)			Healthy Cell Lines (IC50, μM)
	L. infantum	L. tropica	HMEC-1	THP-1	BMDM
CE	0.21 ± 0.07	0.49 ± 0.17	0.38 ± 0.08	0.90 ± 0.30	1.76 ± 0.43
PR	0.26 ± 0.02	0.24 ± 0.04	N.T. 1	1.10 ± 0.35	1.27 ± 0.09
1	0.06 ± 0.02	0.09 ± 0.03	0.51 ± 0.08	2.08 ± 0.24	3.61 ± 0.38
2	0.26 ± 0.06	0.21 ± 0.04	N.T. 1	1.78 ± 0.48	N.T. 1
3	0.06 ± 0.01	0.08 ± 0.04	0.53 ± 0.17	2.79 ± 0.44	4.53 ± 0.05
4	0.25 ± 0.08	0.20 ± 0.04	0.35 ± 0.16	1.89 ± 0.43	2.59 ± 0.71
5	0.12 ± 0.02	0.12 ± 0.05	0.12 ± 0.03	1.88 ± 0.34	2.68 ± 0.81
6	0.09 ± 0.03	0.14 ± 0.03	0.40 ± 0.03	0.85 ± 0.24	2.15 ± 0.02
7	0.20 ± 0.01	0.15 ± 0.04	0.13 ± 0.01	0.48 ± 0.17	1.78 ± 0.53
8	4.05 ± 1.96	3.44 ± 1.12	3.03 ± 0.01	2.51 ± 0.87	13.88 ± 2.78
9	2.31 ± 0.28	3.07 ± 1.36	4.07 ± 0.05	3.28 ± 0.90	12.96 ± 2.48
10	0.46 ± 0.12	0.34 ± 0.21	0.17 ± 0.04	0.93 ± 0.27	2.86 ± 0.17
11	14.45 ± 1.96	15.15 ± 2.71	>14.17	>15.90	>15.90
AMP	0.16 ± 0.05	0.17 ± 0.02	>20	>20	>20

Reported data are the mean of three experiments run in duplicate. Amphotericin B was used as reference drug.; ¹ N.T.: not tested.

.

The lead compound CE exhibited a micromolar range activity against both L. infantum and L. tropica promastigotes and an almost equal aspecific cytotoxicity. The methyl ester PR showed similar or slightly improved leishmanicidal potency and cytotoxicity profile compared to CE.

In general, tertiary amine-based carboxamides (compounds 1–5) were found to be as active or even more potent than CE itself, corroborating the consolidated hypothesis that a pH-responsive basic head improves the antiprotozoal activity [55,58]. Among them, the ethylene diamine-based piperidinyl and pyrrolidinyl analogues 1 and 3 were by far the most potent leishmanicidal agents with IC₅₀ in the low-nanomolar range. Moreover, the cytotoxicity of this compounds was reduced against all the tested cell lines, with respect to CE. Interestingly, the propylene diamine-spaced piperidine derivative 2 was a less potent leishmanicidal compound than its inferior homologue 1.

Both the pyrrolizidine derivatives 4 and 5 were more potent against L. tropica promastigotes than CE and generally less cytotoxic toward all the tested cell lines. Furthermore, compound 5, the most potent of the two analogues against LbHsp90, was found to be also the least cytotoxic.

The quinoline derivatives 6 and 7 showed different potency and selectivity profiles. While the propylene diamine-based analogue 7 was generally comparable with CE in terms of potency against L. infantum promastigotes and cytotoxicity, the ethylene diamine-containing compound 6 outclassed the parent compound in terms of leishmanicidal activity.

The methanethiosulfonate derivatives 8 and 9 were modestly active toward both Leishmania spp. promastigotes (IC₅₀ = 2–4 μM) and displayed negligible cytotoxicity mainly on BMDM cell line. The disulfide derivative 10 was found to be more potent than CE but of comparable cytotoxicity.

Finally, the glucosamine derivative 11 was totally inactive against all the tested cell lines (IC₅₀ > 14 μM). This could be probably related to an insufficient cellular uptake due to its high polarity.

3.2. Leishmanicidal Activity on Amastigotes Cultures and Selectivity of Action

The four better performing compounds of Table 1 i.e., the CE derivatives 1, 3, 5 and 6, whose nanomolar antileishmanial activity was combined with a reduced profile of aspecific cytotoxicity, were selected to be tested against intramacrophage amastigotes of L. infantum. Their selectivity of action was calculated with the respect of compounds’ IC₅₀ on the human macrophage cell line THP-1 reported in Table 1. Results are summarized in

Table 2. Antileishmanial activity on intramacrophage amastigotes of L. infantum and evaluation of selectivity toward THP-1 cultures.

	L. infantum Amastigotes IC50 (μM) 1	S.I. 2
1	0.19 ± 0.11	11
3	0.13 ± 0.09	21
5	0.52 ± 0.28	3.6
6	0.66 ± 0.06	1.3
AMP	0.18 ± 0.03	>109

¹ Reported data are the mean of three experiments. Amphotericin B was used as reference drug; ² Selectivity index: ratio between compounds’ IC₅₀s on THP-1 cell line (Table 1) and toward L. infantum amastigotes.

.

All the tested compounds were found to be active against the amastigote stage of L. infantum, but with different degrees of selectivity. In detail, compounds 5 and 6, bearing respectively a pyrrolizidine and an aromatic amine moiety showed a sub-micromolar antileishmanial activity but low selectivity. On the contrary, compounds 1 and 3, demonstrated a leishmanicidal activity comparable with the reference drug AMP with higher selectivity when compared with compound 5 and 6. Specifically, the ethylene diamine-spaced pyrrolidine derivative 3 exhibited the higher SI, equal to 21.

3.3. In Vitro Inhibition of LbHsp90

The in vitro performances of CE, PR and the carboxamides 1–11 (Figure 4) as potential inhibitors of the Leishmania braziliensis Hsp90 were investigated using Sigma-Aldrich^® Malachite Green Phosphate Assay Kit. The optimal concentrations of 7 μM LbHsp90 and ATP was 1 mM were used according to the protocol reported in the M&M section. The results expressed as IC₅₀ (μM) are reported in

Table 3. In vitro inhibition of the ATPase activity of LbHsp90.

Compound	IC50 (μM) 1
CE	201 ± 10
1	83 ± 2
2	>125
3	81 ± 1
4	48 ± 3
5	20 ± 2
6	65 ± 4
7	>125
8	67 ± 3
9	>150
10	>125
11	47 ± 5
PR	>500

¹ The reported data are the mean of two experiments run in triplicates.

.

Confirming previous reports [29], a modest yet detectable activity toward LbHsp90 was shown by CE which acted as a negative modulator of the chaperone ATPase activity.

The ethylenediamine-based basic derivatives (i.e., the piperidine, the pyrrolidine and the pyrrolizidine derivatives 1, 3 and 5) showed improved potency against LbHsp90 compared to the parent CE. Compound 5 was the most potent inhibitor of the series with ten-fold increase of activity compared to CE while the propylene diamine-based analogue (2) was the only basic, aliphatic carboxamide to be found inactive toward LbHsp90 up to 125 μM in the assay conditions. Although in compound 4 the two basic nitrogen atoms are separated by three carbon atoms, this compound is still very potent, probably because one carbon is included in the bicyclic moiety making the alkyl chain less flexible than the propylene one.

The quinoline derivatives 6 and 7, inspired by the aromatic portion of the antiprotozoal drug chloroquine, presented the same pattern described for the basic derivatives 1–5. In fact, the ethylene diamine-based analogue 6 worked as an inhibitor of LbHsp90 with a potency comparable with that of the carboxamides 1, 3–5. On the contrary, in analogy to compound 2, the propylenediamine-containing derivative 7 was inactive up to 125 μM.

The chaperone ATPase modulatory activity of the two methanethiosulfonates analogues 8 and 9 was again found to be sensitive to the length of the alkyl chain spacer linking the common CE core and the methanethiosulfonate moiety, with the shorter compound 8 inhibiting LbHsp90 with IC₅₀ = 67 μM in the assay conditions, while the longer one 9 was inactive up to a concentration of 125 μM.

Interestingly, the allyldisulfide ethanamide derivative 10, devoid of any coordinating polar or pH-responsive groups in the carboxamide side chain, did not show any relevant modulatory activity up to a concentration of 125 μM. The glucosamide derivative 11 inhibited LbHps90 ATPase cycle with an IC₅₀ of 47 μM, similarly to other glycosides previously reported as allosteric modulators of the human Hsp90 [33].

Sustaining the hypothesis that hydrogen bonding groups could promote a tighter interaction with LbHsp90 and CE-based scaffolds, Pristimerin (PR), which lacks any H-bonding promoting residues, resulted inactive on the protein (IC₅₀ > 500 μM).

3.4. ATP-Competitve vs. Non-Competitive Hsp90 Modulation

CE and the CE-carboxamides 1 and 3 were previously reported in the literature as compounds endowed with antiproliferative activity in virtue of their ability to allosterically disrupt the protein complex between the human Hsp90 and the protein Cdc37 [41,42,43,44,45,46,47]. By definition, allosteric modulation is a type of non-competitive inhibition/activation of a target protein which is sometimes translated in a detectable variation of an enzymatic activity, like in the case of the ATPase activity of Hsp90 chaperones. To preliminarily investigate the nature of the interaction between this family of compounds and Leishmania Hsp90, the ATPase inhibitory activity of the known active site inhibitor, ATP-competitive GA toward LbHp90 was compared with that of CE and the basic derivative 3.

Accordingly, an in vitro assay developed for the evaluation of non-competitive allosteric modulators, which exploits ATPase kinetics data, was used [31,33]. Briefly, the decrease in absorbance deriving from the continuous oxidation of NADH to NAD⁺ produced by a multi-enzymatic cascade triggered by LbHp90-catalyzed ATP hydrolysis is followed spectrophotometrically over the course of 15 min (for details see Appendix B). All the experiments were normalized using the native ATPase activity of LbHsp90 measured in the presence of DMSO, the co-solvent selected for compounds solubilization. According to literature procedures, the compounds were incubated at the threshold concentration of 50 μM in the presence of 3.5 μM LbHsp90 and 1 mM ATP [31,33].

The LbHsp90 ATPase kinetics modulated by CE and compound 3 (red and green lines, respectively) are reported and compared with the native chaperone cycle of the protein (black line) in Figure 5. As it was extrapolated from the reported curves (Appendix B), CE reduced the native LbHsp90 ATPase activity of 30% while the carboxamide 3 inhibited 95% of it in the presence of 20-fold excess of ATP. These results are in agreement with the higher potency shown by compound 3 on the protein in comparison with CE (lower IC₅₀ in Table 3).

Since GA acts on Hsp90 chaperone targeting their active site via a ATP-competitive mechanism, its ATPase inhibitory activity is, by definition, strictly dependent on ATP concentration. Accordingly, in the assay conditions (1 mM ATP) and when GA is used at the same concentration of CE and compound 3 (50 μM), its inhibitory potency toward LbHsp90 was negligible (Figure 5, orange line). Complete inhibition was instead achieved only conducting the assay in the presence of 1 mM GA (Figure 5, blue line). GA- inhibition of LbHsp90 ATPase kinetics at crescent concentrations–from 10 μM to 1 mM–are reported in Appendix B.

The treatment of the elongated and flagellate promastigotes of L. donovani with sub-IC₅₀ concentrations of the Hsp90 active site inhibitor GA was reported to induce morphological differentiation into a round, amastigote-like stage. This morphological differentiation was also coupled with the expression of amastigote-specific proteins. Moreover, upon long-term exposures to GA, the treated promastigotes activated a series of cellular pro-survival mechanisms, defined as heat-shock responses, which restored and/or made up for Hsp90 chaperon functions resulting in an embryonal drug resistance to GA [16,27]. At variance to GA and Hsp90 active site inhibitors in general, allosteric modulators could be able to interfere with Hsp90-mediated folding events and protein–protein interactions while avoiding the insurgence of heat-shock responses [39,40].

Aiming at collecting additional, yet qualitative, preliminary data comparing GA to the potential non-ATP-competitive Hsp90 inhibitors, CE and compound 3, their morphological effects of the exposure of Leishmania promastigotes were evaluated in vitro. The compounds were incubated (23 °C, physiological pH) with L. infantum promastigotes at a sub-IC₅₀ concentration of 50 ng/mL. IC₅₀ of GA on L. infantum (0.14 ± 0.04 μM) promastigotes was measured as described in the M&M section. After 24 h and 48 h the presence of round, amastigote-like cells, was monitored (Figure 6). A negative and positive control experiment (NC and PC, Figure 6a,b) were also conducted growing the promastigotes at 23 °C and 37 °C, respectively.

4. Discussion

In the quest for novel antiprotozoal agents, the identification of novel targets and mechanisms of actions is highly cherished to surpass the insurgence and spread of drug resistant parasite strains. Not to be forgotten, the cytotoxicity and the lack of selectivity of action of modern and classical leishmanicidal agents is well-known and represents a harming issue of public health [13,14,15,55,56,58].

With the aim of developing novel antileishmanial agents exploiting the rich source of chemical diversity represented by bioactive natural products, a small family of carboxamide derivatives of the natural and bioactive triterpene CE (Figure 4) was designed, synthesized and screened in vitro.

The tested compounds were synthetized applying HOBt/EDC coupling chemistry to be characterized by different polarities and lipophilicities selecting diverse groups as amide decorations: cyclic tertiary amines, 4-aminoquinolines, methanethiosulfonates, an allyldisulfide and a glucosamine.

In vitro investigations were at first focused on the definition of compounds’ aspecific cytotoxicity and leishmanicidal activity against Leishmania promastigotes. Accordingly, the growth inhibitory activity of CE carboxamides, PR and of the parental compound CE were measured against L. infantum and L. tropica promastigotes. Cytotoxicity was estimated using the human HMEC-1 and THP-1 cell lines and the murine macrophage line BMDM (Table 1).

As described in the Results section, the best performing CE carboxamides (compounds 1, 3, 5 and 6) were all characterized by the presence of a basic head as substituent of the amide side chain and all displayed reduced cytotoxicity and improved leishmanicidal potency with the respect to CE itself.

Among them, compound 3 showed a low-nanomolar activity (IC₅₀ < 0.1 μM) against both L. infantum and L. tropica promastigotes and the lowest cytotoxicity. The methanethiosulfonates 8 and 9 and the glucosamine derivative 11 were found to be poorly active against all the tested cell lines probably due to issues related to cellular uptake.

The most active toward Leishmania promastigotes and least cytotoxic carboxamides (i.e., the mentioned compounds 1, 3, 5 and 6) were also tested in vitro against intramacrophage amastigotes of L. infantum (Table 2). While compounds 5 and 6, bearing respectively a pyrrolizidine and an aromatic amine moiety showed a sub-micromolar antileishmanial activity but low selectivity, compounds 1 and 3 demonstrated higher leishmanicidal activity and selectivity. Specifically, the ethylene diamine-spaced pyrrolidine CE-derivative 3 showed a growth inhibitory activity comparable with the reference drug AMP and a SI equal to 21.

CE and CE-related compounds are generally described in the literature as promising candidates for the development of novel drugs targeting different human pathologies in virtue of the different biological activities that have been correlated to the administration of CE itself or CE-enriched plant extracts [2,5,6,7,8,11,12,29]. From a mechanistic point of view, these bioactivities can be ascribed to an aspecific cytotoxicity deriving from a complex of possible and simultaneously operating mechanisms of action potentially involving different protein targets via reversible or covalent binding [7]. As previously described, CE was reported to be endowed with the ability to allosterically disrupt the protein complex between the human Hsp90 and the protein Cdc37 [41,42,43,44,45,46,47]. Moreover, in a following study, CE basic carboxamides (including compounds 1 and 3) have been described as disruptors of the Hsp90-Cdc37 protein complex characterized by an improved antiproliferative action and selectivity against tumor cells [47].

On this basis, to preliminary investigate if Hsp90 inhibition could be involved in their promising described leishmanicidal activity, the modulatory activity of CE itself and of the basic CE carboxamides of Figure 4 was screened toward the ATPase activity of a homologous Hsp90 from Leishmania braziliensis. As a general consideration, the IC₅₀ values reported in Table 3 can be considered as an estimation of the modulatory effect on LbHsp90 ATPase cycle of the tested compounds to facilitate the comparison of their inhibitory potency in the assay conditions applied. The Hsp90 chaperone on which the tests were run is in fact an isolated and purified recombinant protein which is devoid of its natural cellular environment and must be used in a concentration sufficiently high to fulfill the sensibility of the assay. According to its native ATPase activity, 7 μM LbHsp90 was the optimum concentration selected, based on the limit of quantification of the Sigma-Aldrich^® Malachite Green Phosphate Assay Kit.When incubated in the presence of LbHps90 (Table 3), the CE carboxamides either behaved as inhibitors of the chaperone ATPase activity (compounds 1, 3–6, 8 and 11) or they were found to be basically inactive up concentrations of 125–150 μM (compounds 2, 7, 9 and 10). Given their different structural features and the respective IC₅₀ on the target proteins, some qualitative speculations can be made about the requirements for the interaction of this class of CE-derivatives with a still-to-be characterized allosteric pocket LbHsp90. It can be argued in fact that steric hindrance and the presence of a H-bond acceptor on the amide side chain play an important role in the interaction with the protein. Accordingly, compounds 1, 6, and 8 were all active against LbHsp90 while their homologues 2, 7 and 9 (containing just one methylene more) did not show any modulatory activity on the chaperone regardless the analogue nature of their amide side chains. Moreover, the methyl ester derivative PR (devoid of any N/NH or Lewis basic group) and the lipophilic disulfide 10 were inactive up to 500 and 125 μM, respectively.

As previously stated, the use of ATP-competitive active site inhibitors of Hsp90 as potential antiprotozoal agents is well documented in the literature [11,16,25,26]; among them the natural antibiotics GA, known for its high cytotoxicity [59,60], was largely investigated.

Interestingly, upon long-term exposures to GA, the rise of drug resistance, in the form of heat-shock responses, was reported for treated cultures of Leishmania promastigotes. Usually, these pro-survival mechanisms are evident, besides in the pattern of expressed signaling proteins, also at a morphological level, since promastigotes are prone to differentiate into a round shaped amastigote-like state [16,27]. Generally speaking, these effects are usually voided by the use of non-ATP-competitive modulators of the ATPase cycle of Hsp90 chaperones [39,40]. To preliminary investigate whatever the ATPase inhibitory activity of CE and CE carboxamides could be or not ATP-competitive in nature, a comparison between the inhibitory activity toward LbHsp90 ATPase cycle of CE and the basic carboxamide 3, a previously described allosteric disruptor the complex Hps90-Cdc37 in humans [47], was conducted in vitro operating ad hoc designed kinetics assays.

When GA (50 μM) was tested for its ability to modulate LbHsp90 ATPase kinetics in the presence of an excess of ATP (1 mM), no relevant effects were identified (Figure 5, yellow line). According to its ATP-competitive mechanism of chaperone inhibition, when 1 mM GA was used a LbHsp90 ATPase activity was completely inhibited (Figure 5, blue line). In contrast, CE and carboxamide 3 exhibited a strong negative modulatory effect toward the kinetics of the chaperone (30% and 95% of inhibition, respectively; Figure 5 green and red lines). These preliminary results could highly suggest that for CE and CE-carboxamides, as it was reported for compound 3 [47], a non-ATP-competitive mode of interaction with Leishmania Hsp90 could be operating.

Moreover, the performances of GA, CE and compound 3 were compared in an in vitro experiment designed to assess their effects on the cell morphology of cultures of Leishmania promastigotes treated with sub-IC₅₀ concentrations of the listed compounds (Figure 6). Leishmania promastigotes, elongated and flagellate cells, tend in fact to morphologically differentiate into round amastigote-like cells when treated with ATP-competitive Hsp90 active site inhibitors, whose action on the protein “mimics” the heat-shock events and the responses which trigger lifecycle stage progression in parasites [16,27]. Accordingly, GA triggered a morphological differentiation on Leishmania promastigotes which turned into round amastigote-like cells in ca 40% of the total amount of analyzed cells (Figure 6c). On the contrary, neither CE nor compound 3 produced this effect on the treated parasites (Figure 6c,d) which remained in their promastigote state, qualitatively and preliminary sustaining the hypothesis of a non-ATP-competitive mode of interaction with the chaperone, in line with CE well-characterized mechanism of inhibition of the human Hsp90 homologue [41,42,43,44,45,46,47].

5. Conclusions

In this work, the natural and bioactive triterpene celastrol was used as lead compound for the design and synthesis of bioinspired, novel potential leishmanicidal agents. Four compounds, the basic CE-carboxamides 1, 3, 5 and 6 were found to be endowed with a strong growth inhibitory effect on Leishmania tropica and L. infantum promastigotes joined with a reduced aspecific cytotoxicity with the respect of the parental compound CE. Among them, the basic compound 3 showed also a promising nanomolar activity against intramacrophage Leishmania amastigotes together with a good selectivity against the THP-1 human cell line. These features address compound 3 as good candidate for a detailed SAR investigation to develop novel bio-inspired leishmanicidal agents further optimizing the molecular skeleton of CE.

Moreover, the potential modulatory effects of CE and its derivatives toward the ATPase cycle of the Leishmania chaperone Hsp90 was investigated. Depending on their different structural features, the novel CE-analogues were generally more potent than CE itself in inhibiting the ATPase activity of a recombinantly expressed Hps90 from L. braziliensis. Interestingly, the four basic carboxamides 1, 3, 5 and 6, the most potent, least toxic and promising leishmanicidal agents of this series, were also found to be the most active compounds on the chaperone and potentially acting through a non-ATP-competitive mode of action.

Based on these preliminary, yet promising, in vitro data, the role of the ATPase cycle modulation of Leishmania Hsp90 in the leishmanicidal activity of CE-carboxamides (especially 1 and 3) will be deeply investigated to elucidate whether this chaperone could be a biological target of these compound in vivo. In parallel, the biochemical features of their interactions with the protein at a molecular level will be investigated in ad hoc designed in vitro experiments and the results will be reported in a future paper.