INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Leishmaniasis is a major health problem that affects approximately 12 million people worldwide, with 2 million new cases diagnosed every year (26). The causative agents of this disease are parasites of the genus Leishmania, which infect and replicate in macrophages of the vertebrate host. Leishmaniasis presents a broad clinical spectrum, ranging from asymptomatic and self-healing infections to those causing significant mortality (1). Recently, a dramatic increase in the number of cases of leishmaniasis has been observed in patients with compromised T-cell function, such as those infected with the human immunodeficiency virus (25).

Pentavalent antimonials are still the first choice among drugs used for the treatment of leishmaniasis. In general, these compounds are toxic and expensive, and they require long-term use during treatment. Recently, the emergence of antimony-resistant parasites has been reported (1, 10, 14), which has compelled the search for new antileishmanial agents.

Several new antileishmanial compounds are under development, but a drug with the capacity to completely cure these infections has not been discovered. Although most active drugs against infectious agents are derived from medicinal plants, medicinal scientific evaluation of the medicinal properties of plants remains grossly understudied.

Among all families of the plant kingdom, members of the Apocynaceae family have been used for centuries in folk medicine, and many of their compounds have been isolated and are now in clinical use as separate drugs, such as vinblastine, vincristine, and reserpine (18). Among the members of the Apocynaceae family, the genus Peschiera has been used in Brazil, while the synonymous genus Tabernaemontana has been used in Central and South America, and the genus Ervatamia has been used in Australia and Asia. Ethnobotanical sources mention that the most common medicinal uses of this genus involve its antimicrobial action against infectious diseases such as syphilis, leprosy, and gonorrhea, as well as its antiparasitic action against worms, dysentery, diarrhea, and malaria (22). Anti-inflammatory, antitumor, and analgesic activities have also been reported (19, 20, 22). The effective uses described in folklore are probably based on the presence of indole alkaloids, which are the main secondary metabolites in this genus. One of these alkaloids, olivacine, has shown strong activity against human carcinomas (16) as well as parasites (8, 13, 27). Antileishmanial activity has been reported for bis-indole alkaloids isolated from Peschiera van heurkii (17).

The species Peschiera australis (Müll. Arg.) Miers, which flourishes in Brazil, Argentina, Uruguay, and Paraguay, has been poorly investigated with regard to its chemical composition and specific pharmacological activities. The study reported here was undertaken to examine the potential antileishmanial activity of P. australis. We found that an ethanolic extract from the stems of P. australis inhibited the growth of Leishmania amazonensis promastigotes in axenic cultures and of amastigotes in infected murine macrophages. A compound purified by a bioassay-guided chemical fractionation of this extract was identified as the indole alkaloid coronaridine, which exhibited potent antileishmanial activity (2).

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Plant extraction and fractionation. Stem material at the secondary stage of growth was collected from two specimens of P. australis in the Botanical Gardens of Rio de Janeiro (Rio de Janeiro, Brazil), washed, dried at 45°C for 7 days, and ground in an industrial blender. The crude extract was prepared by soaking the material (350 g) with ethanol (Reagen, Rio de Janeiro, Brazil), in a Soxhlet apparatus at 78°C for 36 h. The ethanolic extract was dried, suspended (80% [wt/vol]) in 5% HCl (Reagen), and partitioned with hexane (3× 250 ml) (Reagen) and chloroform (3× 250 ml) (Reagen). The fractions were concentrated in a rotating evaporator (Fission, São Paulo, Brazil), suspended in bidistilled water, lyophilized, and stored at 20°C until used. Organic fractions were diluted in dimethyl sulfoxide (DMSO) (Sigma Chemical Co., St. Louis, Mo.) for the biological assays. Promastigote growth and amastigote survival inhibition assays were used to test the crude extract fractions (hexane, chloroform, and aqueous fractions) and coronaridine for antileishmanial activity. Coronaridine was purified as previously described by Rates et al. (19).

Structure elucidation. As the biological activity was present in the chloroform fraction (CLF), its composition was further analyzed by high resolution (HR) gas chromatography-mass spectrometry (GC-MS), using an HP 5890-a spectrometer equipped with flame ionization detection (290°C), column 25 m, 0.25-µm film thickness, and 0.25-mm fused silica capillary column DB-5 (J&W); a flow rate of 1.2 ml of He per mm; a temperature program of 150°C, rate 4°C/min, to 290°C for 5 min and with the injection port in split mode 1:20 at 240°C. The integrator was model HP 3395. GC-MS was performed under the same conditions in an HP 5970 spectrometer with ionization energy of 70 eV and an ion source at 280°C.

Parasite culture. L. amazonensis (WHOM/BR/75/Josefa) promastigotes were cultured at 26°C in SIM-F (Schneider insect medium [Sigma] plus 10% fetal calf serum [FCS] [Gibco-BRL, Gaithersburg, Md.] and 15 µg of gentamycin per ml) (Schering-Plough, Rio de Janeiro, Brazil).

Antipromastigote activity. Promastigotes were incubated in SIM-F in the presence of different concentrations of the ethanolic extract, the CLF, or purified coronaridine, which were added only once to the cultures. After 3 days at 26°C, parasite survival was estimated by counting viable or motile forms in a hematocytometer. In all tests, 1% DMSO (a concentration five fold higher than that used to dissolve the higher dose of the compounds) and medium alone were used as controls. To investigate the reversibility of the compounds/effects, promastigotes in SIM-F were treated with CLF at 100 µg/ml for 1 h or 20 µg/ml for 24 h. The parasites were then washed and incubated in fresh medium, and their viability was estimated each 24 h. All cultures were performed in triplicate, and the results were expressed as percent growth in comparison to that of the controls.

Antiamastigote activity. Resident peritoneal cells from normal BALB/c mice were harvested in RPMI 1640 medium (Biochrom KG, Berlin, Germany) plus 15 µg of gentamycin per ml. Cells were plated onto 13-mm² coverslips (Thomas Scientific, Swedesboro, N.J.) inside 24-well plates (Nunc, Roskilde, Denmark) and allowed to adhere for 2 h at 37°C in 5% CO₂. Nonadherent cells were removed, and macrophages were incubated overnight in RPMI supplemented with 10% FCS as described above. Adhered macrophages were infected with L. amazonensis promastigotes (stationary growth phase) at a parasite/macrophage ratio of 6:1 and incubated at 37°C in 5% CO₂. After 1 h of incubation, free promastigotes were removed by extensive washing with phosphate-buffered saline (PBS) (0.01 M), and the cultures were incubated for 4 days as described above. Treatment of infected macrophages with crude extract, fractions, and purified coronaridine was done by following two different protocols: (i) one treatment 1 h after the infection and (ii) addition of the compounds once a day for 3 days postinfection without replacing the culture medium. Glucantime (Rhodia, Sao Paulo, Brazil) was used as a control for parasite growth inhibition. After 4 days, the monolayers were washed with PBS at 37°C, fixed in methanol, and stained with Giemsa. The number of amastigotes was determined by counting at least 400 macrophages in duplicate cultures, and results were expressed as percentage of survival in comparison to that of the controls. The survival indices were obtained by multiplying the percentage of infected macrophages by the number of amastigotes per infected macrophage. Only experiments with a survival index of 220 for the untreated macrophages were considered.

Ultrastructural analysis. L. amazonensis promastigotes (10⁸) treated with 100 µg of the CLF per ml or with medium alone for 1 h at 26°C were washed in PBS and fixed with 2.5% glutaraldehyde in 0.1 M sodium cacodilate buffer (SCB) at 4°C. The parasites were washed three times with 0.1 M SCB and postfixed with 1% osmium tetroxide in 0.1 M SCB plus 0.8% potassium ferricianide and 5 mM calcium chloride for 60 min at room temperature. After dehydration in acetone, the material was incubated in an acetone-epon mixture (1:1) at room temperature for 24 h and then transferred to pure epon at 60°C for 72 h. L. amazonensis-infected resident mouse peritoneal macrophages in 25-cm² cell culture flasks (Nunc) were treated with 20 µg of the CLF or coronaridine per ml for 10 h at 37°C in 5% CO₂. CLF and coronaridine effects on noninfected macrophages were also evaluated, and 1% DMSO was used as a control in both situations. The cells were processed for transmission electron microscopy as described above. Sections obtained in a Reichert Ultracut (Leica, Nussloch, Germany) were stained with uranyl acetate and lead citrate and were examined in a Zeiss 900 transmission electron microscope (Carl Zeiss, Oberkochen, Germany).

Phagocytosis assay. Resident mouse macrophages were treated with 20 µg of either coronaridine or Glucantime per ml for 24 h. Stationary-phase L. amazonensis promastigotes used as corpuscular stimuli were added to monolayers at a parasite/macrophage ratio of 6:1, and the preparations were incubated at 37°C in an atmosphere of 5% CO₂. After 1 h, monolayers were washed, fixed, and Giemsa stained as described above. Examing 400 cells at random in duplicate cultures, we determined the percentage of macrophages with adhered or phagocytosed promastigotes. Results were expressed as percentage phagocytosis in comparison to that of the controls.

Cytotoxicity assays. Resident mouse macrophages adhered to 24-well plates were treated with coronaridine and Glucantime at the indicated concentrations for 24 h at 37°C in 5% CO₂. The macrophages were than washed and incubated with 0.3% trypan blue solution, and the number of viable cells was scored in an inverted microscope. This test was also used to check drug cytotoxicity in monocyte-derived macrophages (MDM) obtained from three healthy donors according to the method of (R. G. Lima, E. M. B. Saraiva, J. Van Weyenbergh, M. Barral-Neto, G. Galvão-Castro, and D. C. Bou-Habib, submitted for publication).

Nitric oxide production. Adhered J774.A1 murine macrophage cells (10⁶ cells/well in a 24-well plate) were not activated or were activated with 10% -interferon (-IFN, L1210 cell line, 4-day culture supernatant) and 100 ng of lipopolysaccharide per ml from Escherichia coli O111:B4 (Difco Laboratories Inc., Detroit, Mich.). After 24 h at 37°C in 5% CO₂, the monolayers were treated with 10 and 20 µg of CLF or coronaridine per ml. The nitrite concentration in the culture medium was assayed by the Griess reaction (9). Plates were read at 490 nm, and the NO₂ concentration was determined with reference to a standard curve, using sodium nitrite. The results were expressed as micromolar concentrations of nitrite.

Statistical analysis. Results were statistically analyzed by the Student's t test. P values of 0.05 were considered significant. The 50% inhibitory concentration (IC₅₀) was determined using MATLAB software (Mathworks, Inc., Natick, Mass.) with a specific toolbox for estimating curves.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Partitioning the crude ethanolic stem extract of P. australis with organic solvents yielded three fractions: hexane, chloroform, and aqueous fractions. All of these fractions were assayed for antileishmanial activity against promastigote and amastigote forms of the parasite. CLF was the only fraction that showed significant antileishmanial effect on both forms of the parasite. The HRGC-MS analysis revealed that CLF was rich in alkaloids, mainly indole alkaloids, and that coronaridine was one of the major components identified (Fig. 1). This alkaloid was first isolated and characterized by Gorman et al. (7) from Tabernaemontana coronaria (syn. Ervatamia coronaria). It was classified as an Iboga-type indole alkaloid because it has an isoquinuclidine ring fused to an indole moiety (18).

View larger version (10K): [in this window] [in a new window]   FIG. 1.   Coronaridine, one of the indole alkaloids found in the CLF from P. australis.

Antipromastigote activity. Daily treatment with 1 mg of the crude extract per ml inhibited the growth of the L. amazonensis promastigotes. An inhibition of 30% in the parasite growth and motility was observed after 24 h, reaching 100% inhibition after 6 days of treatment (data not shown). Treatment of promastigotes with the aqueous fraction (100 µg/ml) inhibited the parasite survival by 80% after 3 days.

View larger version (12K): [in this window] [in a new window]   FIG. 2.   Inhibition of L. amazonensis promastigote growth by P. australis CLF (diamonds) and coronaridine (squares). The parasites were treated only once with the different compounds, and their growth was assessed 3 days later. The results are from 1 of 3 experiments done in triplicate, and they are shown as percentage inhibition of parasite growth, measured by counting motile/viable promastigotes. *, P < 0.05.

Antiamastigote activity. The leishmanicidal activity of CLF and purified coronaridine was evaluated in L. amazonensis-infected macrophage cultures by adding the compounds in the first day of culture or once a day for 3 days. The extract and the compounds were added to the cultures without replacing the medium. The stem crude extract inhibited parasite survival by 70% in both protocols (data not shown). The effect of the drugs on amastigote-infected macrophages treated once for 1 h after infection is shown in Fig. 3. A 40% inhibition of amastigote survival was seen using 1 µg of CLF per ml, and 10 and 20 µg of CLF per ml inhibited parasite survival by 82 and 98%, respectively. When coronaridine was used at 10 and 20 µg/ml, the amastigote survival decreased to 38 and 79%, respectively. IC₅₀s of 2.6 and 12 µg/ml were calculated for CLF and coronaridine, respectively. When the monolayers were treated with 10 or 20 µg of Glucantime per ml, an inhibition of 36 or 70%, respectively, was observed with an IC₅₀ of 15 µg/ml.

View larger version (29K): [in this window] [in a new window]   FIG. 3.   Effect of P. australis CLF (Clf), coronaridine (Cor), and Glucantime (Glu) on amastigote survival. L. amazonensis-infected mouse peritoneal macrophages were treated with different concentrations of the drugs 1 h after the infection, and the amastigote survival was assessed 4 days later. Results from four experiments in duplicate are shown as percentage ± standard deviations of survival inhibition in relation to untreated control. All the results were significant (P < 0.05).

View larger version (33K): [in this window] [in a new window]   FIG. 4.   Effect of P. australis CLF (Clf), coronaridine (Cor), and Glucantime (Glu) on amastigote survival. L. amazonensis-infected mouse peritoneal macrophages were treated with different concentrations of the drugs 1 h after the infection. Drug treatment was repeated daily during the next 3 days, and amastigote survival was assessed 4 days later. Results are from four experiments in duplicate and are shown as percentages ± standard deviations of survival inhibition in relation to untreated control. All results were significant (P < 0.05).

Ultrastructural effects on promastigotes, amastigotes, and host cells. Electron microscopy studies were done to determine the ultrastructural changes in the parasites and host cells treated with the drugs. Promastigotes treated with 1% DMSO showed no morphological differences from untreated controls (data not shown). The initial morphological alteration detectable in promastigotes exposed to CLF (100 µg/ml for 1 h) was in the mitochondria, which showed alterations similar to those of the treated amastigotes (Fig. 5B, D, and E).

View larger version (181K): [in this window] [in a new window]   FIG. 5.   Ultrastructural effects of coronaridine on intracellular amastigotes. (A) General view of untreated infected macrophages shows a normal aspect. N, nucleus; P, parasites; arrow, mitochondria. Magnification, ×10,000. (B) General view of infected macrophages treated with 20 µg of coronaridine per ml for 10 h, showing parasites in different stages of damage. Note that the macrophage mitochondria (arrows) show a normal aspect better seen in panel C. P, parasite; arrows, macrophage's mitochondria. Magnifications: B, ×8,000; C, ×30,000 (D and E) Amastigote mitochondria showing a swollen organelle containing fragments of membrane and cristae and remains of the mitochondrial matrix. M, amastigote mitochondria; K, kinetoplast. Magnifications, D, ×30,000; E, ×30,000.

Effects of coronaridine on macrophages. In order to test the safety of coronaridine on mammalian cells, murine and human macrophages were treated with the drug, and their viability was checked by trypan blue dye exclusion. When murine macrophages were treated with 20 and 10 µg of coronaridine per ml for 24 h, 83 and 93% of cells, respectively, remained viable. When treated with Glucantime at 20 µg/ml 89% of cells remained viable. Human macrophages treated with 20 µg of coronaridine per ml for 7 days resulted in less than 5% cell death. Further experiments were carried out to determine the potential toxicity of the drug on human macrophages by the XTT method. We observed that 20 and 40 µg of coronaridine per ml resulted in toxicities of 7 and 17%, respectively, to human MDM after 7 days of treatment.

Nitric oxide production. To determine whether the inhibition of intracellular parasite growth was due to a general activation of macrophage microbicidal mechanisms, we measured nitric oxide production. Nonactivated J774.A1 macrophages treated for 24 h with 20 µg of CLF or coronaridine per ml were able to produce 24 and 10 µM nitrite, respectively. (Fig. 6). Controls treated with 0.2% DMSO induced 10 µM NO₂ production. Activation of J774.A1 macrophages with lipopolysaccharide and -IFN induced a sixfold increase in their NO₂ production in comparison with that of nonactivated macrophages. However, addition of 10 µg of coronaridine or CLF per ml to these already activated macrophages did not alter their NO₂ production (Fig. 6).

View larger version (22K): [in this window] [in a new window]   FIG. 6.   Effect of CLF and coronaridine on the NO production by nonactivated (gray bars) and -IFN-lipopolysaccharide-activated (open bars) J774.A1 macrophages. Supernatants were harvested 24 h after treatment, and NO production was determined by Griess reaction. Data represent the means ± standard deviations of three experiments done in triplicate. NA, nonactivated cells; ACT, activated cells; COR, coronaridine; CLF, chloroform fraction of P. australis.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Here we describe novel pharmacological activity obtained from extracts of P. australis and the pharmacological activity of an indole alkaloid, coronaridine, against promastigote and amastigote forms of the L. amazonensis parasite. Our initial observation that the crude extract of this plant inhibited in vitro growth of both developmental forms of L. amazonensis (data not shown) prompted us to perform a bioassay-guided fractionation of the antileishmanial activity. This activity was identified in CLF that at 25 µg/ml totally inhibited the promastigote growth (Fig. 2) and at 20 µg/ml inhibited 98% of amastigote survival (Fig. 3). Promastigote damage by CLF was irreversible, suggesting a metabolic injury that could not be reversed by culturing treated promastigotes in fresh medium.

Chemical analysis of CLF identified coronaridine, an iboga-type indole alkaloid, as one of its major constituents. Coronaridine, naturally found in P. australis (23), P. laeta (28), P. van heurkii (17), and E. coronaria (11), has also been synthesized and is currently tested as an antiaddictive therapy (6). In our assay conditions, coronaridine (12 µg/ml) showed potent antipromastigote activity (97% killing) and was more active than CLF which at the same concentration killed 65% of the promastigotes (Fig. 2). The contrary was observed with infected macrophages, in which CLF was more active in killing amastigotes than was coronaridine (Fig. 3 and 4). When the IC₅₀s of both compounds were compared, CLF was five- and fourfold more efficient than coronaridine on single- and 3-dose treatments, respectively. Differences could be due to other compounds present in CLF, and the identification and isolation of these compounds are currently being investigated. It is interesting that CLF was six- and fivefold more effective than Glucantime, again based on a comparison of the IC₅₀s of single- and three-dose treatments, respectively.

In order to evaluate the morphological changes induced by CLF and coronaridine, promastigotes and infected macrophages treated with these compounds were analyzed by transmission electron microscopy. The striking ultrastructural change on CLF-exposed promastigotes was the remarkable swelling and disorganization of the mitochondrion. No other changes in parasite organelles were observed in this short treatment. Similar mitochondrial changes were also observed in the intracellular amastigote forms treated with coronaridine (Fig. 5B). It is important that mitochondria, as well as other organelles of macrophages treated with the same concentration of coronaridine, were not affected, which could suggest the specificity of coronaridine for parasite mitochondria (Fig. 5C, D, and E).

The evidence for selectivity of coronaridine against parasites was reinforced by the preserved mitochondrial activity in drug-treated human macrophages measured by the XTT method. In this assay, mitochondrial dehydrogenases metabolized the XTT reagent to a water-soluble formazan dye (3). Cells with damaged mitochondria were unable to metabolize XTT. Moreover, no other significant damage was observed on coronaridine-treated murine or human macrophages as measured by the trypan blue exclusion test. The phagocytic activity of macrophages was also preserved at coronaridine concentrations toxic to the parasites. In general, a drug may act directly against the parasite or indirectly by activating macrophage mitochondrial mechanisms such as NO production, which has been shown to be the most effective antileishmanial mechanism. Purified coronaridine was unable to up-regulate NO production in either activated or nonactivated macrophages (Fig. 6). The slight stimulation of NO production by CLF could be due to different alkaloids present in this fraction. The pronounced changes in the mitochondria of both forms of the parasite suggest that their energy metabolism was affected. Similar mitochondrial lesions have been shown in Leishmania major promastigotes treated with chalcones that inhibited the respiration and mitochondrial dehydrogenases (29). Interestingly, mitochondria are the target organelle in different species of Leishmania treated with pentamidine (5, 12), paramomycin (15), licochalcone A (4) and dihydroxy-methoxychalcone (21). It is interesting that ketoconazol, which is an inhibitor of ergosterol synthesis, also affects mitochondrion morphology (24). Ergosterol synthesis is an important chemotherapeutic target, as this lipid is the main component of Leishmania membranes. Although we still don't know the mechanism responsible for parasite killing, elucidation of this phenomenon is currently under investigation in our laboratory.

Our results reveal a novel pharmacological activity of coronaridine, besides its antiaddictive property (7). Laboratory synthesis and the possibility to modify coronaridine chemical structure constitute important advantages for development of new antileishmanial therapy.