New iminodibenzyl derivatives with anti-leishmanial activity

Journal of Inorganic Biochemistry 172 (2017) 9–15

Contents lists available at ScienceDirect

Journal of Inorganic Biochemistry journal homepage: www.elsevier.com/locate/jinorgbio

New iminodibenzyl derivatives with anti-leishmanial activity a

b

c

Anderson Arndt , Cleber Wanderlei Liria , Jenicer K.U. Yokoyama-Yasunaka , M. Terêsa Machinib, Sílvia Reni Bortolin Ulianac, Breno Pannia Espósitoa,⁎ a b c

MARK

Department of Fundamental Chemistry, University of São Paulo, Av. Prof. Lineu Prestes, 748, 05508-000 São Paulo, SP, Brazil Department of Biochemistry, Institute of Chemistry, University of São Paulo, Av. Prof. Lineu Prestes, 748, 05508-000 São Paulo, SP, Brazil Department of Parasitology, Institute of Biomedical Sciences, University of São Paulo, Av. Prof. Lineu Prestes, 1374, 05508-000 São Paulo, SP, Brazil

A R T I C L E I N F O

A B S T R A C T

Keywords: Copper Iminodibenzyl Leishmania Chelation therapy

Leishmaniasis is an infection caused by protozoa of the genus Leishmania and transmitted by sandflies. Current treatments are expensive and time-consuming, involving Sb(V)-based compounds, lipossomal amphotericin B and miltefosine. Recent studies suggest that inhibition of trypanothione reductase (TR) could be a specific target in the development of new drugs because it is essential and exclusive to trypanosomatids. This work presents the synthesis and characterization of new iminodibenzyl derivatives (dado) with ethylenediamine (ea), ethanolamine (en) and diethylenetriamine (dien) and their copper(II) complexes. Computational methods indicated that the complexes were highly lipophilic. Pro-oxidant activity assays by oxidation of the dihydrorhodamine (DHR) fluorimetric probe showed that [Cu(dado-ea)]2 + has the highest rate of oxidation, independent of H2O2 concentration. The toxicity to L. amazonensis promastigotes and RAW 264,7 macrophages was assessed, showing that dado-en was the most active new compound. Complexation to copper did not have an appreciable effect on the toxicity of the compounds.

1. Introduction Leishmaniasis is a disease transmitted by parasites of the genus Leishmania, endemic to 98 countries from both Old and New World. The disease is classified as cutaneous, mucocutaneous or visceral depending upon the species of parasite and organ afflicted [1]. Cutaneous leishmaniasis (CL) is the least aggressive form of the infection, which may subside without treatment (however with the possibility of permanent scarring). However, it can also progress to complicated tegumentary forms with mucosal damage or dissemination. Treatment options include pentavalent antimonials, liposomal amphotericin B or pentamidine [2]. Visceral leishmaniasis (VL) (kala-azar), the most dangerous form of the disease, compromises internal organs. Its treatment depends upon the geography of the disease; for example, in India, where almost all strains of Leishmania are resistant to pentavalent antimonials, liposomal amphotericin B is the recommended treatment [1]. Other treatments for VL include paromomycin and miltefosine. However, all these approved drugs have important side effects (hepatotoxicity, teratogenicity, nephrotoxicity, potentially fatal cardiotoxicity) as well as cost issues [1]. Currently, only two clinical studies on vaccines for leishmaniasis are under way, in Brazil and Uzbekistan [3]. In view of the relatively limited options, new active compounds are being pursued, utilizing strategies such as high content screening of

⁎

commercial and public libraries either targeting specific enzyme candidates or phenotypic screenings. Some of these tests produced compounds undergoing clinical tests. Compounds under screening include candidate metallodrugs based on bismuth, vanadium, ruthenium and others [4–10]. A modern approach for increasing drug efficiency is the design of multi-target molecules (compounds that can hit the target organism by at least two independent mechanisms). Leishmania relies on trypanothione reductase (TRYR), a trypanosomatid-specific enzyme responsible for keeping the intracellular levels of the reducing trypanothione peptide [11]. TRYR has been recognized as a target for antileishmanial therapy where the rationale would be to increase the levels of intracellular oxidants [12,13]. One antidepressant, clomipramine (Fig. 1), was found to be an interesting inhibitor of TRYR [14]. In previous works, we demonstrated that copper complexes with fluorinated ligands have some activity against L. amazonensis [15], and that activity (and TRYR inhibition) may be increased by means of increased lipophilicity and pro-oxidant activity [16]. Apart from TRYR inhibition, there are metabolic pathways that depend upon the homeostasis of trace elements such as zinc or iron that can be disrupted by exogenous chelators to the disadvantage of the parasite [17–21]. Therefore, assembling TRYR inhibitors with redox-active metal ions such as Cu2 +, and/or chelators that can selectively remove other

Corresponding author. E-mail address: [email protected] (B.P. Espósito).

http://dx.doi.org/10.1016/j.jinorgbio.2017.04.004 Received 28 September 2016; Received in revised form 6 February 2017; Accepted 2 April 2017 Available online 04 April 2017 0162-0134/ © 2017 Elsevier Inc. All rights reserved.

Journal of Inorganic Biochemistry 172 (2017) 9–15

A. Arndt et al.

(a)

(b)

(c)

i) n-BuLi

N H

H2N

N

ii) Cl

R

O

O

{

R

-OH (ea) -NH2 (en) -NH(CH2)2NH2 (dien) HO

N

N R

(d)

N Cl

N Fig. 1. Proposed synthetic route for the ligands: (a) iminodibenzyl, (b) dado, (c) general structure of the ligands, (d) structure of clomipramine.

%CHN, calculated: 81.2/6.8/5.6; %CHN, experimental 80.8/6.8/5.6. FTIR (Shimadzu IRPrestige-21 with germanium crystal ATR): disappearing of iminodibenzyl NeH stretching at 3379 cm− 1 and NeH bending at 1522 cm− 1 and appearing of epoxide ring stretching at 1230 cm− 1 and deformation at 917 and 853 cm− 1. 1H NMR (CDCl3, 300 MHz) δ 2.55 (dd, 1H), 2.68 (t, 1H), 3.07 (m, 1H), 3.17 (s, 4H), 3.90 (dd, 2H), 6.89 (d, 2H), 6.92 (t, 2H), 6.96 (d, 2H), 7.09 (m, 2H). 13C NMR (CDCl3, 75 MHz) δ 32.1, 46.7, 50.8, 53.8, 120.2, 122.8, 126.4, 129.8, 134.3 (Varian INOVA). IR and NMR spectra are displayed in Supplementary material.

essential metals involved in parasite metabolism appears as an interesting and novel approach for the treatment of leishmaniasis. In this work, we present novel bifunctional molecules inspired in a TRYR inhibitor linked to a chelator moiety, which can be used either to deliver redox-active Cu2 + to Leishmania parasites or to disrupt their balance of essential trace elements. 2. Materials and methods 2.1. Chemicals Iminodibenzyl, tetrahydrofuran, n-Butyllithium (n-BuLi), epichlorohydrin, deuterated chloroform, hexane, ethyl acetate, calcein, acetonitrile, dimethyl sulfoxide, ethanolamine (ea), ethylenediamine (en), diethylenetriamine (dien), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium (MTT), sodium dodecyl sulphate (SDS) and trifluoroacetic acid were from Sigma-Aldrich. Hydrogen peroxide was from Synth (Brazil). Copper(II) sulphate pentahydrate was from Cromoline (Brazil). HBS (HEPES buffered saline): 20 mM Hepes and 150 mM NaCl aqueous solution, pH = 7.4, treated with 0.01 g/L Chelex™. For the biological studies, it was used Medium 199 (Sigma-Aldrich) and fetal calf serum (FCS) (Invitrogen). 2.2. Synthesis and characterization of the ligands and their copper(II) complexes

2.2.2. Synthesis and chemical characterization of dado-ea, dado-en and dado-dien (Fig. 1) About 0.2 mmol of dado previously synthesized was added to excess of the amines (ethanolamine, ethylenediamine or diethylenetriamine) at room temperature and the solution was stirred for 20 h. After this time, a clear and yellowish homogeneous solution develops. After establishing appropriate conditions by analytical RP-HPLC, the crude dado derivatives were purified on a Waters 600 preparative HPLC system that had an injector Waters Delta 600 and was coupled to a preparative C18 Vydac cat#218TP1022 column, a detector 2487 and a recorder Kipp & Zonen Servogor 124. Elution was done using solvent A (0.1% TFA), solvent B (80% ACN/0.09% TFA), flow rate of 10 mL/ min, detection and the following linear gradient:

2.2.1. Synthesis and characterization of 5H-dibenz[b,f]azepino,10,11dihydro-5-(2-oxyranylmethyl) (dado; Fig. 1) In a dry round flask under nitrogen atmosphere, ca. 20 mmol of iminodibenzyl was dissolved in dry THF. The temperature was decreased to − 72 °C (ethanol/dry ice bath), followed by the addition of ca. 22 mmol of n-BuLi. The system was kept under vigorous stirring until reaching room temperature, then it was again kept at − 72 °C as ca. 25 mmol of freshly distilled epichlorohydrin was added. As the solution reached room temperature, it had a reddish color. The reaction was ended after 12 h by the addition of a saturated solution of NaCl. The organic compounds were separated by liquid-liquid extraction using dichloromethane, which was removed by rotary evaporation. The product of this reaction was purified via silica column chromatography using hexane/ethyl acetate (10:1) as eluent. Fractions containing the product were evaporated, yielding a viscous oil that turned into a yellowish solid when stocked at ca. 5 °C. Formation of dado was confirmed by several techniques. Elemental analysis (Perkin Elmer):

5% → 5%B → 95%B → 95%B. Yields for dado-ea, dado-en and dado-dien: 47%, 31% and 95% respectively. The purified dado-ea, dado-en and dado-dien were then characterized by LC-MS using a Shimadzu HPLC system (composed of two pumps LC-20AD, a degasser DGU-20A3, a column oven CTO-20A, a pre-column C18 Shim-pack GVP-ODS, a column C18 Shim-pack VP-ODS and a detector SDP-20AV) coupled to a mass spectrometer AmaZon Xda Bruker Daltonics and FTIR. dado-ea: m/z = 313.1 g mol− 1; δ (cm− 1) 798 (NeC stretching of secondary aliphatic amines), bands at 1012 and 1184 (NeC stretching of secondary aliphatic amines and CeO stretching of alcohols) and 2400–3500 (NeH stretching of secondary amines) [22]. dado-en: m/z 312.2 g mol− 1; δ (cm− 1) 796 (NeC stretching of secondary aliphatic amines), bands at 1012 and 1184 (NeC stretching of secondary aliphatic amines and CeO stretching of alcohols), 1651 (NeH scissoring of primary amines) and 2650–3350 (NeH stretching of secondary amines) [22]. dado-dien: m/z 355.3 g mol− 1; δ (cm− 1) 796 (NeC stretching of secondary aliphatic amines), bands at 1008 and 1180 (NeC stretching of secondary aliphatic amines and CeO stretching of

10′

10

60′

10′

Journal of Inorganic Biochemistry 172 (2017) 9–15

A. Arndt et al.

0.40 0.35

Cu + dado-ea

Absorbance

0.30 0.25 0.20

Cu + dado-dien

Cu + dado-en

2.0

2.0

1.5

1.5

1.0

1.0

1.0 0.5

0.15

0.5 0.5

0.10

0.0

0.0 0.0

0.05 0.00 400

600

800

1000

400

600

800

1000

400

600

800

1000

Wavelength (nm) Fig. 2. Electronic spectra (HBS/DMSO 1:1) of Cu(II) 10 mM treated with increasing amounts of ligands. Numbers beside the spectra indicate ligand:metal mol ratios.

the test compounds for 48 h in 96-well plates. The viability of triplicate test samples was assessed by measuring the cleavage of 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) (Sigma-Aldrich), as previously described [26]. Results were obtained in five independent experiments. Results presented herein are expressed as the 50% effective concentration (EC50) calculated from sigmoidal regression of the dose-response curves using ORIGIN 7.5 software. Toxicity to the mammalian host cells was evaluated by cultivating 1 × 106 RAW macrophages in RPMI Medium (Invitrogen) supplemented with 10% FCS at 37 °C in 96-well plates for 48 h in the presence of increasing concentrations of each compound. Cell viability was assessed by the MTT assay. The EC50 values for each complex were determined as described above.

alcohols), 1651 (NeH scissoring of primary amines) and 2250–3250 (NeH stretching of secondary amines) [22]. Copper(II) complexes were prepared by adding an aqueous solution of CuSO4 to the ligands. 2.3. Lipophilicity calculations Lipophilicity of ligands and copper complexes was calculated with Marvin Sketch® v. 16.6.27.0 for Microsoft Windows® 7.6.1, by ChemAxon Ltd (Budapest, Hungary) [23]. This software has been successfully used for high throughput screening of antimalarials [24]. 2.4. Competition studies with calcein The complex between copper and calcein (CuCA) was prepared by adding equimolar quantities of calcein and CuSO4 in deionized water. Working solutions of 200 μL CuCA (2 μM in HBS) were treated with 10 μL aliquots of the dado derivatives at increasing concentrations. The fluorescence was recorded at room temperature on a BMG Fluostar Optima instrument (λexc/λem = 485/520 nm).

3. Results and discussion Two new iminodibenzyl-based chelators have been synthesized incorporating a multi-amine binding moiety suited for Cu2 + (dadoen and dado-dien). As a comparison, the compound dado-ea was also prepared where one of the amino groups of dado-en was replaced by the less coordinating hydroxyl group. Both en (logKML = 10.5; logKML2 = 19.6 [27]) and dien (logKML = 15.9; logKML2 = 20.9 [27]) behave as strong bidentate chelators for Cu2 + as this d9 ion is most stable in a tetragonal symmetry resulting from significant Jahn-Teller distortion (which also puts Cu2 + high in the Irving-Williams series of stability of transition metal complexes). Ethanolamine, on the other hand, behaves as a monodentate (N) ligand, establishing up to four metal-ligand equilibria with lower stability (logKML = 5.7; logKML2 = 9.8; logKML3 = 13.0; logKML4 = 15.2 [27]). Electronic spectra of the complexes of the dado-derivatives (Fig. 2) display a hypsochromic shift from 810 nm (aqueous Cu2 +) to 780 nm, 756 nm or 648 nm for ea, en and dien, respectively, reflecting the increase in ligand field splitting caused by the increasing number of strong field amine donor groups. These data also indicate the 1:2 metal:ligand ratio for dado-en and dado-dien, which are in agreement with a tetragonal structure, although higher substitution could theoretically be possible given the extra donor hydroxyl of the dado derivatives. However, it was not possible to observe substituted complexes for dado-ea higher than 1:1, possibly because of the steric hindrance of the iminodibenzyl moiety coupled to the intrinsically weak coordination ability of ea. Conjugation to dado maintained the ligand ability of the chelators.

2.5. Oxidative activity under peroxide The pro-oxidant activity of free Cu2 + and of the copper complexes of dado-ea, dado-en and dado-dien in the presence of peroxide under physiologically relevant conditions of salinity and temperature was evaluated by a fluorescence method described elsewhere [16,25]. The kinetic test measures the relative rates of metal-catalysed oxidation of the fluorescence probe DHR. In the first experiment, aliquots (10 μL) of the test compounds (10 μM) were transferred to transparent, flat bottom 96-well microplates and treated with 10 μL of hydrogen peroxide (0–80 μM) and 180 μL of DHR 50 μM in HBS. In a second experiment, aliquots (10 μL) of H2O2 (50 μM) were transferred to transparent, flat bottom 96-well microplates and treated with 10 μL of the test compounds (0–40 μM) and 180 μL of DHR 50 μM in HBS. Fluorescence was measured on a BMG Fluostar Optima instrument for 40 min at 37 °C (λexc/λem = 485/520 nm). 2.6. Anti-leishmanial activity Leishmania amazonensis (MHOM/BR/1973/M2269) promastigotes, grown in Medium 199 (Sigma-Aldrich) with 10% fetal calf serum (FCS) (Invitrogen) at 25 °C, were incubated with increasing concentrations of 11

Journal of Inorganic Biochemistry 172 (2017) 9–15

A. Arndt et al.

Table 1 Calculated and reported logP values for the compounds under study. n.a.: not available. logP (This work)

logP (Literature)

Reference

Iminodibenzyl ea en dien dado dado-ea dado-en dado-dien [Cu(dado-ea)]2 + [Cu(dado-en)2]2 + [Cu(dado-dien)2]2 +

3.73 − 1.36 − 1.40 − 1.79 3.70 2.18 2.13 1.75 2.48 5.44 4.68

3.8 −1.3 −2.0 −2.1 3.90 n.a. n.a. n.a. n.a. n.a. n.a.

[49] [49] [49] [49] [50] – – – – – –

12000

Fluorescence (a.u.)

Compound

14000

10000 8000 dado-ea dado-en dado-dien ea en dien

6000 4000 2000 0

Lipophilicity, as expressed by logP values, was estimated by the standard method of Marvin Sketch® software with electrolytes (Na+, K+ e Cl−) set at 0.1 M, based on the model of Viswanadhan et al. [23,28]. Whenever available, experimental values were displayed for comparison (Table 1). Good agreement was observed between theoretical and experimental logP values for dado. Attaching ea, en or dien to dado leads to a decrease of lipophilicity, which is expected as the chelators themselves are very hydrophilic. Metal complexation on the other hand renders the compounds less water soluble, as the polar, solvatable eOH or amino groups are now involved in the coordinate bond. According to Wils et al. [29], transepithelial absorption of drugs is proportional to lipophilicity for logP values up to 3.5 (above which absorption fades). This indicates that all the dado derivatives would have good intestinal absorption, while the copper complexes of dado-en and dado-dien would not. Chelating fluorescent probes allow for the assessment of the relative affinities of a given metal ion for other chelators. Calcein is a fluorescent molecule (emission at 515 nm) able to form very stable coordination compounds with several metal ions such as Fe3 + or Cu2 +. Indeed it has long been used as a probe for labile iron in biological systems [30]. Upon metal coordination, calcein fluorescence is stoichiometrically quenched. Therefore, if such metal(calcein) compound is allowed to react with an unknown chelator under equilibrium conditions, its affinity for the metal will show as an increase in the fluorescence of the system [30,31]. In this work, the complex Cu (calcein) (CuCA; logKML = 12.3 [32]) was prepared and its interaction with both free and dado-conjugated en, dien and ea. was studied as a means to assess the relative affinities of these chelators for copper (Fig. 3). As expected, both dien and dado-dien display significant affinity for Cu2 +, even though the presence of the bulky iminodibenzyl moiety attached to dien decreases this affinity. Dado-en on the other hand was not able to force the dissociation of CuCA. Since en forms less stable compounds than dien, this inability could be explained by the trend towards decreased stability for the dado derivatives when compared with the free ligands. The size of the competitor ligand (dado-en or dado-dien) may pose steric constraints to their interactions with the metal center in the CuCA complex (calcein being a relatively large molecule itself), which would translate into decreased affinity constants. For the interaction of CuCA with dado-dien (L), the following equilibria are possible (Eqs. 1–3):

CuCA⇌Cu + CA 1 KCuCA

(1)

Cu + 2 L ⇌ CuL 2 Kapp

(2)

CuCA+2L ⇌ CuL 2 + CA K = Kapp KCuCA

(3)

0

10

20

30

40

50

[ligand] ( µM) Fig. 3. Fluorescence recovery of calcein as a function of ligand concentration. (a.u. = arbitrary fluorescence units).

(2.0 × 1012 [32]) and Kapp is the conditional binding constant for the formation of Cu(dado-dien)2. The K value is given by Eq. 4:

K=

[CA] [CuL 2] [CuCA][L]2

(4)

According to Fig. 3, when [dado-dien] is 13.4 μM there is a 50% fluorescence recovery, indicating that half of the calcein binds to copper and thus [CA] = [CuCA]. Therefore, half (1 μM) of the original copper is complexed by dado-dien. Substitution of these figures in Eq. 4 results in K = 5.6 × 103, which can be used to estimate logKapp (Eq. 3) as 16. This apparent formation constant is lower than the thermodynamic constant since it accounts for salinity and buffering. These figures indicate that dado-dien would be suited to keep copper tightly bound even at the low micromolar range (~ 3% of free metal when Cu (dado-dien)2 is 1 μM). On the other hand, dado-en would release copper(II) to endogenous substrates with high affinity, which could be interesting to overload target organisms with this potentially toxic ion. Macrophages are widely distributed, specialized white blood cells that attack and digest foreign microorganisms and cell debris. The mechanism of action involves the inclusion of the foreign agent into phagolysosomes followed by acidification, enzymatic lysis and reaction with reactive oxygen species. In this sense, H2O2 is kept at a steady state of ca 1–4 μM [33]. However, some parasites are known to be able to circumvent these safeguards, such as Salmonella [33] and Leishmania amastigotes [34]. Copper salts have been used as broad spectrum biocides for a long time, benefiting from the ability of the metal to induce damage to cell membranes and proteins, to inactivate reduced glutathione and to bind strongly to nucleic acids [35]. More recently, the use of copper complexes has been proposed as a means to control the reactivity of the free metal or its delivery to target organisms/biological compartments. Interaction of H2O2 with Cu2 + complexes results in products known to cause oxidative damage in biomolecules such as DNA. The identity of the oxidant species is extensively debated. It may involve Cu (I)OOH peroxide intermediates [36] whose production could be enhanced when a protein is added, suggesting that copper chelation may positively affect the formation of reactive oxidant species [37]. Cu (I) species may also react with H2O2 in a Fenton-like mechanism generating the highly oxidant hydroxyl radicals, although in some cases (eg for some copper(II) complexes with four-nitrogen coordination) the highly oxidant Cu(III) species [38,39] (ECu3 +/Cu2 +0 = 2.4 V) [40] may be formed.

where KCuCA is the conditional binding constant of CuCA at pH 7 12

Journal of Inorganic Biochemistry 172 (2017) 9–15

A. Arndt et al.

Cu(II) free Cu(nta) Cu(dado-ea) Cu(dado-en)2

Initial oxidation rate (a.u.-1 s-1)

3.5 3.0

k×104

Cu(dado-dien)2

2.5

Cu(II) free

6.33±0.47

2.0

Cu(nta)

0.21±0.06

1.5

Cu(dadoea)

9.03±0.54

1.0

Cu(dadoen)2

3.19±0.20

0.5

Cu(dadodien)2

2.55±0.10

0.0 -0.5 1x10-5

0

2x10-5

3x10-5

4x10-5

[complexes] (M) Fig. 4. Initial oxidation rate of DHR (50 μM) catalysed by the Cu(II) derivatives of the ligands, under H2O2 50 μM. Insert: rate constants (a.u. M− 1 s− 1) ± SD. (a.u. = arbitrary fluorescence units).

dado-ea dado-en dado-dien iminodibenzyl

a 140

Cu2+ Cu(dado-ea) Cu(dado-en)2

a 120

Cu(dado-dien)2

120

100

Viability %

Viability %

100 80 60

80 60

40

40

20

20

0

0 0.0

0.5

1.0

1.5

2.0

2.5

0.0

0.5

log10[L] (µM)

1.0

1.5

2.0

log10[CuL] ( µM) dado-en iminodibenzyl

b 140

Cu2+ Cu(dado-ea) Cu(dado-en)2

b 120

120

Cu(dado-dien)2

100

80

80

Viability %

Viability %

100

60 40

60

40

20

20 0

0 0.0

0.5

1.0

1.5

2.0

2.5

0.0

log10[L] (µM)

0.5

1.0

1.5

2.0

log10[CuL] ( µM)

Fig. 5. Dose-response curves for iminodibenzyl and (a) dado-ea, dado-en, dado-dien against Leishmania amazonensis promastigotes; (b) dado-en against RAW macrophages.

Fig. 6. Dose-response curves for the copper derivatives of dado-ea, dado-en, dado-dien against (a) Leishmania amazonensis promastigotes or (b) RAW macrophages.

The oxidation rate of DHR has been proposed as a high-throughput means of assessing pro-oxidant activity of metal complexes (e.g., Fe(III) or Mn(III)) with endogenous ligands such as ascorbate or peroxide [25,41]. Here, hydrogen peroxide was kept at a steady concentration

(50 μM) and the effect of varying concentrations of Cu(II)L under pseudo-first order conditions was assessed to derive the initial rates of oxidation of the DHR probe (Fig. 4). As discussed, ethanolamine is a

13

Journal of Inorganic Biochemistry 172 (2017) 9–15

A. Arndt et al.

increase in the conversion of MTT to formazan. In our experience, this behavior is observed both with parasites and host cells when confronted with several different killing agents. From the three dado derivatives, dado-en is the most similar to clomipramine. The additional eOH of dado-ea affects negatively the lipophilicity, which is required for docking to TRYR [47]. Dado-dien, although as lipophilic as dado-en, is probably too bulky to be an effective inhibitor. Addition of copper to dado-en does not improve its activity (as there are two ligands per metal ion), however it could be interesting to decrease drug side effects [48].

Table 2 EC50 (μM) of the ligands and copper complexes against Leishmania amazonensis promastigotes and RAW macrophages. n.d.: not determined. Compound

L. amazonensis

RAW

Iminodibenzyl dado-ea dado-en dado-dien Cu2 + [Cu(dado-ea)]2 + [Cu(dado-en)2]2 + [Cu(dado-dien)2]2 +

107.4 ± 4.0 > 100 19.9 ± 2.2 > 100 74.9 ± 13.0 > 100 13.7 ± 0.3 38.6 ± 7.3

> 100 n.d. 58.7 ± 9.2 n.d. > 50 > 100 43.3 ± 9.8 > 100

4. Conclusions Derivatives of iminodibenzyl with ethanolamine, ethylenediamine and diethylenetriamine were synthesized, purified and characterized. Dado-dien has the greater affinity for Cu(II) ions. The most probable metal:ligand ratio in solution was [Cu(dado-ea)]2 +, [Cu(dadoen)2]2 + and [Cu(dado-dien)2]2 +, all of them very lipophilic. [Cu (dado-ea)]2 + had the highest DHR oxidation rate, independent of hydrogen peroxide concentration. Only dado-en showed an antileishmania activity in a viable range (EC50 19.9 ± 2.2 μM) when tested against L. amazonensis promastigotes. The complexation with Cu (II) didn't represent an increase in biological activity of the compounds. Thus, these results indicate the potential use of dado-en to combat leishmaniasis.

rather weak chelator for Cu2 + therefore k for both free Cu and Cu (dado-ea) are comparable (6–9 × 104 a.u. M− 1 s− 1; Fig. 4, insert). The fact that they are higher than that for the other chelates is related to the formal reduction potentials of the metal center. Aqueous Cu2 + has an Ef of + 0.167 V/NHE. However, 4N ligands such as en and dien display stronger ligand field splitting, which translates into more negative Ef values (−0.360 and −0.324 V/NHE, respectively) [42] which slow the formation of Cu(I)OOH intermediates (see above) when Cu(dado-en)2 or Cu(dado-dien)2 react with peroxide (assuming that both dado-en and dado-dien behave as bidentade ligands such as en or dien). The redox-inactive complex Cu(nta) [16] was added to this study for comparison; as expected it did not catalyze the oxidation of DHR at a significant rate. Finally, keeping all the complexes at 10 μM and varying peroxide concentration from 0 to 80 μM gave rise to only modest increase in rate for all the tested compounds (highest k = 0.25 ± 0.07 × 104 a.u. M− 1 s− 1; data not shown), meaning that the oxidation rate depends on the concentration of the metal complexes rather than that of peroxide. This would also suggest that the oxidant species is metal-based and not a ROS derived from peroxide. Taken together, these results indicate that dado-en and dado-dien could be scavengers of redox-active ions, and their copper complexes might work as pro-oxidant metallodrugs in peroxide-rich environments. The anti-leishmanial activity (against promastigotes of L. amazonensis) of the new ligands and their copper(II) complexes with the most probable stoichiometry, along with toxicity data for RAW macrophages, is displayed in Figs. 5 and 6 and summarized in Table 2. Dado-en is the most active ligand, in a therapeutically relevant range of concentrations. From the dose-response curves it was possible to determine its EC50 as 19.9 ± 2.2 μM, against 107.7 ± 4.0 μM for the parent iminodibenzyl. The EC50 for dado-en determined against L. amazonensis promastigotes is in the same order of magnitude of effective concentrations determined for tamoxifen (12.6 μM), miltefosine (19.9 μM) or paromomycin (77.0 μM) and larger than for amphotericin B (63 nM) [43–46]. The relatively low toxicity of iminodibenzyl observed here is on par of recent findings that the chlorine substitution in clomipramine is crucial to proper docking to TRYR [47] and might increase the activity of the dado compounds. For dado-ea and dadodien, EC50 is also > 100 μM. All the free chelators (ea, en and dien) had higher activities (EC50 < 3 μM; data not shown), however they cannot be administered to patients due to their toxicity. Iminodibenzyl and dado-en are less toxic to RAW macrophages (EC50 > 100 μM and > 50 μM, respectively; Fig. 5b), the host cell infected by the parasites. For both iminodibenzyl in promastigotes (Fig. 5a) and dado-en in macrophages (Fig. 5b), prior to lethal effects there seems to be an “increase” in cell viability. Drug activity against parasites or macrophages was evaluated by quantifying cell survival. This was measured by quantifying the cleavage of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) by living cells, which is ultimately a measure of mitochondrial activity. When submitted to sub-lethal concentrations of active compounds, cells can increase their mitochondrial activity in order to surpass the noxious effect, which results in an

Acknowledgments Authors are thankful to CAPES (PhD grant for Mr. Arndt) and FAPESP (funding for the projects 2011/50318-1 and 2016/03709-9). We appreciate the assistance of Prof. Dr. Liliana Marzorati in the discussion of NMR data and of Dr. Erica Levatti in the biological tests. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jinorgbio.2017.04.004. References [1] M.P. Barrett, S.L. Croft, Br. Med. Bull. (2012) 175–196. [2] L.F. Oliveira, A.O. Schubach, M.M. Martins, S.L. Passos, R.V. Oliveira, M.C. Marzochi, C.A. Andrade, Acta Trop. (2011) 87–96. [3] S. Srivastava, P. Shankar, J. Mishra, S. Singh, Parasit. Vectors (2016) article # 277. [4] D. Gambino, Coord. Chem. Rev. (2011) 2193–2203. [5] S.K. Hadjikakou, I.I. Ozturk, C.N. Banti, N. Kourkoumelis, N. Hadjiliadis, J. Inorg. Biochem. (2015) 293–305. [6] J. Hess, J. Keiser, G. Gasser, Future Med. Chem. (2015) 821–830. [7] D.M. Keogan, D.M. Griffith, Molecules (2014) 15258–15297. [8] M. Navarro, C. Gabbiani, L. Messori, D. Gambino, Drug Discov. Today (2010) 1070–1078. [9] A.I. Ramos, T.M. Braga, S.S. Braga, Mini-Rev. Med. Chem. (2012) 227–235. [10] A. Tahghighi, J. Organomet. Chem. (2014) 51–60. [11] A.H. Fairlamb, P. Blackburn, P. Ulrich, B.T. Chait, A. Cerami, Science (1985) 1485–1487. [12] V. Olin-Sandoval, R. Moreno-Sanchez, E. Saavedra, Curr. Drug Targets (2010) 1614–1630. [13] C. Pal, U. Bandyopadhyay, Antioxid. Redox Signal. (2012) 555–582. [14] T.J. Benson, J.H. McKie, J. Garforth, A. Borges, A.H. Fairlamb, K.T. Douglas, Biochem. J. (1992) 9–11. [15] R.S. Maffei, J.K.U. Yokoyama-Yasunaka, D.C. Miguel, S.R. Bortolin Uliana, B.P. Esposito, Biometals (2009) 1095–1101. [16] A.S. Portas, D.C. Miguel, J.K.U. Yokoyama-Yasunaka, S.R. Bortolin Uliana, B.P. Esposito, J. Biol. Inorg. Chem. (2012) 107–112. [17] I. Aphasizheva, R. Aphasizhev, L. Simpson, J. Biol. Chem. (2004) 24123–24130. [18] B.S. McGwire, W.A. O'Connell, K.P. Chang, D.M. Engman, J. Biol. Chem. (2002) 8802–8809. [19] C. Mesquita-Rodrigues, R.F.S. Menna-Barreto, L. Saboia-Vahia, S.A.G. Da-Silva, E.M. de Souza, M.C. Waghabi, P. Cuervo, J.B. De Jesus, Plos Neglect. Trop. D. (2013), http://dx.doi.org/10.1371/journal.pntd.0002481. [20] S. Singh, V. Mandlik, S. Shinde, Mol. BioSyst. (2015) 958–968. [21] K. Soteriadou, P. Papavassiliou, C. Voyiatzaki, J. Boelaert, J. Antimicrob. Chemother. (1995) 23–29.

14

Journal of Inorganic Biochemistry 172 (2017) 9–15

A. Arndt et al.

[37] K. Midorikawa, M. Murata, S. Kawanishi, Biochem. Biophys. Res. Commun. (2005) 1073–1077. [38] R.P. Bonomo, R. Marchelli, G. Tabbi, J. Inorg. Biochem. (1995) 205–218. [39] A.N. Pham, G.W. Xing, C.J. Miller, T.D. Waite, J. Catal. (2013) 54–64. [40] D.R. Lide, CRC Handbook of Chemistry and Physics, Internet Version, CRC Press, Boca Raton, 2005. [41] T.A. Pereira, G.E. Teixeira da Silva, R.B. Hernandez, F.L. Forti, B.P. Esposito, Biometals (2013) 439–446. [42] G. Tabbi, A. Giuffrida, R.P. Bonomo, J. Inorg. Biochem. (2013) 137–145. [43] A.C. Coelho, C.T. Trinconi, C.H.N. Costa, S.R.B. Uliana, Plos Neglect. Trop. D. (2014), http://dx.doi.org/10.1371/journal.pntd.0004720. [44] A.C. Coelho, C.T. Trinconi, L. Senra, J.K.U. Yokoyama-Yasunaka, S.R.B. Uliana, Int. J. Parasitol. (2015) 77–83. [45] E. de Morais-Teixeira, M.K. Gallupo, L.F. Rodrigues, A.J. Romanha, A. Rabello, J. Antimicrob. Chemother. (2014) 150–154. [46] C.T. Trinconi, J.Q. Reimao, J.K.U. Yokoyama-Yasunaka, D.C. Miguel, S.R.B. Uliana, Antimicrob. Agents Chemother. (2014) 2608–2613. [47] M.C. O'Sullivan, T.B. Durham, H.E. Valdes, K.L. Dauer, N.J. Karney, A.C. Forrestel, C.J. Bacchi, J.F. Baker, Bioorg. Med. Chem. (2015) 996–1010. [48] J.E. Weder, C.T. Dillon, T.W. Hambley, B.J. Kennedy, P.A. Lay, J.R. Biffin, H.L. Regtop, N.M. Davies, Coord. Chem. Rev. (2002) 95–126. [49] NCBI, National Center for Biotechnology Information — PubChem Compound Database. https://pubchem.ncbi.nlm.nih.gov Access date July 16, 2016. [50] RSC, ChemSpider. http://www.chemspider.com/Chemical-Structure.13271858. html Access date July 16, 2016.

[22] R.M. Silverstein, F.X. Webster, D.J. Kiemle, Spectrometric Identification of Organic Compounds, John Wiley & Sons, 2005. [23] ChemAxon, Documentation of the logP Plugin. https://docs.chemaxon.com/ display/docs/logP+Plugin Access date August 13, 2016. [24] S. Chtita, M. Ghamali, R. Hmamouchi, B. Elidrissi, M. Bourass, M. Larif, M. Bouachrine, T. Lakhlifi, Adv. Phys. Chem. (2016) article # 5137289. [25] B.P. Esposito, W. Breuer, P. Sirankapracha, P. Pootrakul, C. Hershko, Z.I. Cabantchik, Blood (2003) 2670–2677. [26] R.C. Zauli-Nascimento, D.C. Miguel, J.K.U. Yokoyama-Yasunaka, L.I.A. Pereira, M.A.P. de Oliveira, F. Ribeiro-Dias, M.L. Dorta, S.R.B. Uliana, Tropical Med. Int. Health (2010) 68–76. [27] R.M. Smith, A.E. Martell, Critical Stability Constants, Plenum Press, New York, 1975. [28] V.N. Viswanadhan, A.K. Ghose, G.R. Revankar, R.K. Robins, J. Chem. Inf. Comput. Sci. (1989) 163–172. [29] P. Wils, A. Warnery, V. Phungba, S. Legrain, D. Scherman, J. Pharmacol. Exp. Ther. (1994) 654–658. [30] B.P. Esposito, S. Epsztejn, W. Breuer, Z.I. Cabantchik, Anal. Biochem. (2002) 1–18. [31] M.M. Baccan, O. Chiarelli-Neto, R.M.S. Pereira, B.P. Esposito, J. Inorg. Biochem. (2012) 34–39. [32] L.A. Saari, W.R. Seitz, Anal. Chem. (1984) 810–813. [33] J.M. Slauch, Mol. Microbiol. (2011) 580–583. [34] T. Naderer, M.J. McConville, Cell. Microbiol. (2008) 301–308. [35] G. Borkow, J. Gabbay, Curr. Med. Chem. (2005) 2163–2175. [36] K. Yamamoto, S. Kawanishi, J. Biol. Chem. (1989) 15435–15440.

15