Antileishmanial activity of sulphonamide nanoemulsions targeting the β-carbonic anhydrase from Leishmania species

Abstract The β-carbonic anhydrase (CA, EC 4.2.1.1) from Leishmania spp. (LdcCA) is effectively inhibited by aromatic/heterocyclic sulphonamides, in the low nanomolar range, but no in vitro antileishmanial activity was detected for such compounds. We formulated some of these sulphonamides as nanoemulsions (NEs) in clove oil, and tested them in vitro against Leishmania infantum MHOM/BR/1974/PP75 and Leishmania amazonensis IFLA/BR/1967/PH8 strains. Interesting inhibitory concentrations IC50 were observed for some of the sulphonamides NEs, with IC50 as low as 3.90 µM (NE-3F) and 2.24 µM (NE-5B) for L. amazonensis and 3.47 µM (NE-5B) for L. infantum. Some of the investigated NEs displayed toxicity for macrophages beyond the parasites. For the same nonoemulsions, a selective index (SI) greater than for Amphotericin B. Haemolytic assay using human red blood cells indicate that the NEs were less cytotoxic than amphotericin B, a widely used antifungal agent. NEs demonstrated to be an excellent strategy for increasing the penetration of these hydrophilic drugs through membranes, with a huge increase of efficacy over the sulphonamide CA inhibitor (CAI) alone.


Introduction
Carbonic anhydrases (CA, EC 4.2.1.1) are widespread enzymes in organisms all over the phylogenetic tree [1][2][3][4][5] . CAs are metalloenzymes that catalyses the reversible hydration of CO 2 to bicarbonate with a proton release 6 . They are grouped in seven distinct families, named a-, b-, c-, d-, f-, g-, and h-Cas, classified according the sequence similarity/divergence 3 . Due the importance of CAs in cell physiology, their inhibitors possess a range of pharmacologic applications in various fields, such as for antiglaucoma drugs 7 , diuretics 8 , antiepileptics 9 , antiobesity agents 10 , as well as antitumor agents/diagnostics 11 . Recently, the potential use of CA inhibitors (CAIs) as anti-infective started to be considered for obtaining antibacterials [12][13][14] , antifungals 15,16 , and antiprotozoan agents [17][18][19] , with a novel mechanism of action, in the search of agents devoid the resistance problems common to most classes of clinically used such drugs 20 .
Leishmaniasis is a parasitic infectious disease caused by several species of Leishmania, an obligate intracellular protozoan parasite of humans that resides and multiplies in macrophages 21 . It is associated with significant rates of morbidity and mortality in many countries around the world. Leishmaniasis presents three main different clinical forms, visceral, cutaneous, and mucocutaneous 22 . There is no effective vaccine to prevent human leishmaniasis and the drugs available to chemotherapy have several limitations, as side effects and resistance to classical chemotherapy 21,23 . Thus, the search for new drug targets is required to develop newer therapies, and CAs are a promising target. The species used in this work, Leishmania (L.) amazonensis and Leishmania (L.) infantum causes visceral Leismaniosis.
We have reported that Trypanosoma cruzi, the aetiological agent of Chagas diseases 18 , another parasitic protozoan, encodes for an a-CA, called TcCA 24 , which was inhibited in vitro by many sulphonamide CA inhibitors (CAIs), in the low nanomolar or subnanomolar range 24,25 . However, in vivo, the growth of the parasite was not inhibited by such sulphonamides [24][25][26][27][28] . Other protozoa, such as Plasmodium falciparum, encode for CAs belonging to the g-CA class 18 , whereas Leishmania spp. possess b-CAs 29 . In earlier works from our groups we have shown that sulphonamides and thiols, well-known classes of CAIs, effectively inhibit in vitro this enzyme (called LdcCA as it has been cloned from the genome of Leishmania donovani chagasi). The sulphonamides showed inhibition constants varying between 50.2 nM and 9.25 mM, whereas some heterocyclic thiols inhibited the enzyme with K I s in the range of 13.4-152 nM 29 . Some of these thiols were shown to efficiently inhibit the in vivo growth of Leishmania chagasi and L. amazonensis promastigotes, by impairing the flagellar pocket and movement of the parasites and causing their death, whereas the sulphonamides, some of which showed similar inhibitory power in vitro as the thiols, were devoid of any such in vivo effects 31 . We hypothesised that these differences between the two classes of CAIs are due to the very polar nature of the sulphonamides, which interferes with their penetration through biological membranes of the pathogens in order to inhibit the enzyme, responsible for the pH regulation and probably other physiologic effects 30 . This is the reason why we investigated the possibility to enhance the bioavailability of the sulphonamide CAIs, by formulating them as nanoemulsions (NEs) in clove oil 30,31 .
The majority of NEs are dispersions of oil droplets in water with diametre between 20 and 200 nm. A recent study with sulphonamide NEs and Trypanosomas cruzi demonstrated that this represents indeed a good strategy to enhance the penetration of the drugs in the parasites 32 . Most NEs present small droplet size that allows the Brownian motion of the drops retarding their sedimentation or coalescence. Thus, NEs present kinetic stability 30 , promoting tissue permeation and penetration of drugs. Their nanometric droplets have large relative surface area, facilitating the contact of the nano-carrier with the biological membranes or tissues, and consequently favouring drug permeation and retention. In this paper, we present the antileishmanial activity of sulphonamide CAIs formulated as NEs in clove oil.

Chemistry
Sulphonamides 3F, 3G, 3W, 5B, 5C and 5D ( Figure 1) used in the experiments were reported in an earlier work from our groups 25 , but they were not tested earlier as LdcCA inhibitors.

CA activity measurements and inhibition studies
An Applied Photophysics (Oxford, UK) stopped-flow instrument has been used for assaying the CA catalysed CO 2 hydration activity 33 . Phenol red (at a concentration of 0.2 mM) has been used as indicator, working at the absorbance maximum of 557 nm, with 20 mM Tris (pH 8.4) as buffer, and 20 mM Na 2 SO 4 (for maintaining constant the ionic strength), following the initial rates of the CA-catalysed CO 2 hydration reaction for a period of 10-100 s 33 . The CO 2 concentrations ranged from 1.7 to 17 mM for the determination of the kinetic parameters and inhibition constants. For each inhibitor at least six traces of the initial 5-10% of the reaction have been used for determining the initial velocity. The uncatalysed rates were determined in the same manner and subtracted from the total observed rates. Stock solutions of inhibitors (10 mM) were prepared in distilled-deionised water and dilutions up to 0.01 nM were done thereafter with the assay buffer. Inhibitor and enzyme solutions were preincubated together for 15 min at room temperature prior to assay, in order to allow for the formation of the E-I complex. The inhibition constants were obtained by nonlinear least-squares methods using PRISM 3, as reported earlier 34-36 , and represent the mean from at least three different determinations. All CA isoforms were recombinant ones obtained in-house as reported earlier 34-36 .

Nanoemulsion preparation
The oil-in-water (O/W) NEs were prepared by high-energy method using an ultrasound processor (Hielscher model UP100H, Hielscher GmbH, Berlin, Germany), according to the method descripted by Senna et al. 31 . Oil phase was prepared by sulphonamides dissolution in the clove oil. An amount of 5 mg of drug was weighted in a microtube and 1 ml of clove oil was added. The tube was agitated for 1 min for obtaining of the drug solution (5 mg/ml). Aqueous phase was prepared by adding 1 g of Pluronic F127 V R in 8 g of water. Then 1 ml of oil phase (drug dissolved in clove oil) was added to 9 ml of aqueous phase (Pluronic F127 in water) under constant ultrasound homogenisation (amplitude 80%, continuous cycle no. 1) during 5 min in an ice bath at 5 C to prevent heating of the dispersion. A transparent NE was obtained at a concentration of 500 mg/ml.

Determination of droplet size
Determination of droplet size and polydispersity index (PDI) were measured, using the dynamic light scattering (DLS) method with a Malvern model 90S NanoSizer V R (London, UK). NEs were diluted in distilled water at 1:10 and analysed in a cell with 1 cm optical path at room temperature (25 C). These analyses were conducted in three runs with 15 readings. The values shown are the mean ± standard deviation of three measurements for each formulation. The PDI reflects the sample quality in the parameter homogeneity of the droplet diameter. PDI results lower than 0.3 were considered satisfactory 37 .

Antileishmanial assay
The antileishmanial activity of the sulphonamide NEs was evaluated by the microdilution technique. First, polystyrene 96-well plates were used to serially dilute the samples in a 10% FBS-supplemented PBHIL medium. Amphotericin B and NEs prepared without the sulphonamides were used as positive and negative controls, respectively. L. amazonensis and L. infantum promastigote forms were harvest at late log phase of growth (96 h), washed twice with PBS and resuspended in fresh culture medium to a final concentration of 5 Â 10 6 parasites/ml. Then, 100 ml of each parasite suspensions were added to the plates, and the samples were adjusted to final concentrations ranging from 1 to 128 mM. After 120 h incubation period at 26 C, parasites viability was assessed by adding 50 mL of resazurin solution (0.005%) as previously described by Rolon et al. 39 . The minimal inhibitory concentration (MIC) was determined as the lowest concentration capable of inhibiting in vitro growth of the parasites. The 50 and 90% inhibitory concentrations (IC 50 and IC 90 ) were calculated by regression analysis using Microsoft Excel 2013 software.

Cytotoxic assay
Sulphonamide NEs cytotoxicity was performed using tetrazolium dye MTT colorimetric assay. RAW 264.7 macrophages were harvest after confluent monolayer achievement. The cells were washed twice with PBS and a cellular suspension of 10 6 cells/ml was prepared in fresh DMEM culture medium. Aliquots of 100 ml of the cellular suspension were placed into polystyrene 96-well plates, and then incubated at 37 C in a 5% CO 2 atmosphere for 6 h to allow for adherence of macrophages. After this period, the adherent cells were subjected to treatment with several concentrations of the sulphonamide NEs (1-128 mM), and then incubated for additional 48 h. Finally, 20 ml of a MTT solution (5 mg/ml) were added to each well and the plates incubated for 4 h as previously described 40 . Macrophage viability was determined after formazan crystals solubilisation with DMSO followed by the absorbance measurement at 570 nm using a SpectraMax M5 spectrophotometer (Molecular Devices, Los Angeles, CA). The 50% cytotoxic concentrations were calculated by regression analysis using Microsoft Excel 2013 software.

Selective index determination
The selective index (SI) for promastigote forms of L. amazonensis and L. infantum was calculated by the ratio between the CC 50 for RAW 264.7 macrophages and the IC 50 for the parasites. Samples with SI values >10 were considered as low cytotoxic agents 41 .

Haemolytic assay
Haemolytic activity was evaluated as described previously by Ishnava and Shah with a slight modification 42 . Human erythrocytes from healthy individuals were collected in vacuum tubes containing EDTA as anti-coagulant. The erythrocytes were harvested by centrifugation for 10 m at 2500 rpm at 20 C, and washed three times in PBS. To the pellet, PBS was added to yield a 10% (v/v) erythrocytes/PBS suspension. The 10% suspension was then diluted 1:10 in PBS. Aliquots of 100 ml of erythrocytes suspension was added, in triplicate, to 100 ml of a two-fold dilution series of sulphonamides NEs and amphotericin B (at concentrations of 128,

Statistical analysis
The data of the experiments are being carried out through the programme Prism 5.01 GraphPad (GraphPad Software, La Jolla, CA), being considered values statistically significant those with values of the standard deviation (SD), p < .05.

Results and discussion
Carbonic anhydrase inhibition with sulphonamide 3F, 3G, 3W, 5B, 5C, and 5D and preparation of their clove oil NEs Sulphonamides 3F, 3G, 3W, 5B, 5C, and 5D ( Figure 1) were investigated as in vitro inhibitors of LdcCA, the b-CA cloned and characterised earlier by our groups 29 . As seen from Figure 1, they act as highly efficient LdcCA inhibitors in vitro, with inhibition constants ranging between 3.3 and 8.3 nM. All of them are very effective inhibitors of the protozoan enzyme, making structure-activity relationship difficult to delineate. However, as mentioned above, sulphonamides structurally related to the ones investigated here did not show in vivo anti-leishmanial effects 29 . This is why we formulated here sulphonamides 1-6 as NEs in clove oil ( Figure 2). Sulphonamides 3F, 3G, 3W, 5B, 5C, and 5D were easily dissolved in clove oil in the concentration of 5 mg/ml. The NEs were produced with 10% of oil phase. NEs were prepared also without the drug in order to evaluate the stability, droplet size and PDI. The NEs obtained were yellow and transparent suggesting that the system was homogeneous with small droplet size (Table 1). Phase separation and precipitation of the drug were not observed, and the NE was considered stable in the concentration of up to 500 mg/ml.
NEs without the drug presented an average size of 31.54 nm. The NEs containing the drug presented average sizes between 35 and 100 nm, depending on the drug. The lowest average size was exhibited by NE-5D with a diameter of 35.09 nm. NE-3G and NE-3W exhibited the larger average size values with diameters of 100.63 and 97.34 nm, respectively. These NEs presented PDI below 0.3; indicating that the size distribution is homogeneous and monomodal 37 . Thus, we conclude that the inclusion method of the drugs in NEs was adequate, producing nanostructured samples with drops below 100 nm and size distribution homogeneous and monomodal.

Anti-Leishmania infantum/amazonensis activity in vivo
The effect of NEs containing the sulphonamides 3F, 3G, 3W, 5B, 5C, and 5D on L. amazonensis and L. infantum promastigotes viability was assessed in vitro. The obtained results are summarised in Tables 2 and 3. All NEs displayed in vitro antileishmanial activity with great variations in the IC 50 values, which ranged from 3.47 to 51.7 mM for L. infantum. The IC 50 values for L. amazonensis did not vary that much, ranging between 2.24 and 18.26 mM. The best IC 50 against these parasites were presented by 5B-NE, followed by 3F-NE (IC 50 3.90 mM) for L. amazonensis and NE-3G (IC 50 10.72 mM) for L. infantum.
IC 90 values are summarised in Tables 2 and 3. The inhibitors NE-3F, NE-3W, NE-5B showed IC 90 of 105.58 ± 30.63, 92.74 ± 38.23, 22.46 ± 6.80 mM, respectively, for L. amazonensis, and the compound NE-5B induced an IC 90 of 52.03 ± 8.01 mM for L. infantum. The inhibitors NE-3F, NE-3W, and NE-5B against L. amazonensis and NE-5B for L. infantum promastigotes were able to inhibit the growth of the parasites. After 48 h in culture medium, no cell growth of L. amazonensis was observed at the concentration of 128 mM of the following inhibitors 3F, 3W in NEs. NEs containing 5B at the concentrations 32 and 64 mM inhibited completely L. amazonensis and L. infantum, respectively. These values correspond the MIC for both Leishmanias (Tables 2 and 3).
It was not possible to determine the MIC and IC 90 for promastigotes treated with NEs NE-3G, NE-5C, and NE-5D for L. amazonensis and NE-3F, NE-3G, NE-3W, NE-5C, and NE-5D for L. infantum. Probably the MIC and IC 90 for these NEs are above 128 mM (first concentration studied).
On the other hand, the worst compound was NE-5D for L. amazonensis and L. infantum. Only the NE-5B showed IS 50 above two for both parasites. Except for compounds NE-5B and NE-3F, all the others NEs containing inhibitors displayed more toxicity for macrophage cell than for parasites (IS 50 < 1) ( Table 4).
In parallel, NE cytotoxicity assays were performed on 267.4 macrophage cells. At the concentrations tested, NEs were less cytotoxic than Amphotericin B which is a potent leishmanicide drug with anti-Leishmania effect demonstrated in vitro and in vivo studies against the promastigote and amastigote forms of the Leishmania parasite 43 . Typically, assays in promastigote forms of the parasite are always present in the initial screening of candidate compounds for leishmanicidal drugs, since they are simple and inexpensive tests 44 .
Previous work from our group showed that thiols were more effective than sulphonamides in the inhibition of LdcCA from L. donovani chagasi, another Leishmania associated with the visceral form of the disease 29 . Several studies have been demonstrated that nano-emulsified carrier systems have potential to solve problems with poor water solubility and poor membrane permeability of some drugs 45 . Moreover, the toxicity of some drugs can be reduced when they are nano-emulsified with an appropriate carrier 46-48 . Gupta et al. formulated a nano-emulsified carrier system with copaiba oil to improve the antileishmanial activity and oral bioavailability of amphotericin B 46 . They produced their NE by mixing a surfactant (TPGS, d-a-tocopheryl polyethylene glycol 1000 succinate), a cosurfactant (phosphatidyl choline (PC)), the oil and the drug, to form an O/W emulsion with particle size of 127 ± 21 nm and PDI of 0.11 ± 0.02. The IC 50 value of amphotericin nanoemulsified with copaiba oil and plain amphotericin B against L. donovani was 0.018 ± 0.004 and 0.214 ± 0.06 mg/ml, respectively, demonstrating a nearly 12-fold reduction in the IC 50 value. They also reported the haemotoxicity reduction of amphotericin B from 69.8 to 18.2% when formulated in the NE. These results are corroborated by Santos et al. who also observed increase in vivo anti-leishmanial activity and reduction of in vitro haemotoxicity of amphotericin B 47 . However, they did not use natural oil in NE composition, as their NE was prepared with medium chain triglycerides (MCT), Tween 80, and cholesterol in the oily phase and glycerol and amphotericin B in the aqueous phase, forming particles with 134.8 nm of mean size 47 . In our study, the compounds that did not inhibit significantly Leishmania spp. in plain solution but started to inhibit the tested parasites with IC 50 values ranging from 2.24 to 18.26 mM and from 3.47 to 51.7 mM for L. amazonensis and L. infantum, respectively, when formulated as NEs. The improvement of the anti-leishmanial activity through the production of simple NEs of these sulphonamide CAIs is thus remarkable, whereas its preparation and in the number of components required to formulate the NE are also simplified compared to previously reported procedures 46,47 .

Haemolytic assay
Human red blood cells provide a handy tool for toxicity studies of the compounds, because they are readily available, their membrane properties are well known, and their lysis is easily monitored by measuring the release of haemoglobin. At the concentration of 256 and 128 mM, amphotericin B (conventional anti-leishmanial therapeutic agent) exerted 100% haemolysis. In comparison, the NE without drug caused only 15.75 and 13.37% haemolysis. Taken together, our results indicate that oil glove present in the NEs have significantly less cytotoxic effects than amphotericin B. In all sulphonamides NEs, at concentrations below 16 mM, a haemolysis of less than 10% was found. These results showed that NEs containing the CAIs are promising for therapeutic drug trials.
The incorporation of sulphonamides in clove oil NE allowed the drugs to meet their intracellular targets, and to perform their antileishmanial activity. These inhibitors do not penetrate into the parasite without the NEs (data not shown). These results suggest  that 5B-NE present potential activity against amastigotes, since the NE penetrated into macrophages. Other interesting point that could be studied is the synergistic action between clove oil and 5B on the anti-leishmanial activity. Oil clove (Syzygium aromaticum) has been reported to have anti-leishmanial activity due to high eugenol concentration 48 . The values of IC 50 described by Islamuddin et al. against promastigotes and intracellular amastigotes of L. donovani were 21 and 15.24 mg/ml, respectively 48 . The clove oil caused apoptosis in the L. donovani cell parasite.
Although promoting the parasite growth inhibition, the NEs investigated here also reduced the cell viability of macrophages. The macrophages cytotoxicity might be associated principally to the action of sulphonamides, although many such drugs are used for decades without any evidence of such effects 7 . The oil clove cytotoxicity reported by Islamuddin et al. revealed that there was no toxicity of eugenol-rich oil of S. aromaticum on RAW264.7 cells, even at 200 mg/ml 48 . Many substances with potential therapeutic applications have been discarded in the past because they were not administrable in a bioavailable form 48 . This study opens the possibility to formulated NEs containing the CAIs in order to increase the bioavailability of hydrophilic drugs, such as the sulphonamides.

Conclusions
The best IC 50 against both parasites were obtained by 5B-NE, followed by 3F-NE (IC 50 3.90 mM) for L. amazonensis and 3G-NE (IC 50 10.72 mM) for L. infantum NEs have demonstrated potential as a novel vehicle for delivery hydrophophilic drugs such as sulphonamide CAIs improving the bioavailability of the drug and demonstrating a potential use in the treatment of leishmaniasis.

Disclosure statement
No potential conflict of interest was reported by the authors.