Aspidosperma pyrifolium, a medicinal plant from the Brazilian caatinga, displays a high antiplasmodial activity and low cytotoxicity

Plant collection and access to genetic resources were approved by CNPq (Process N°010861/2013-0) and registered in the National System for the Management of Genetic Heritage and Associated Traditional Knowledge (SisGen, Process N°A61DDB0 and N°A646A52, respectively).

Parts of A. pyrifolium (Fig. 1) were collected in São José da Tapera (Alagoas, Brazil) in October 2001; the leaves were collected in 2016 at the margins of São Francisco River. The species was identified by José Elias de Paula, from the Department of Botany (Universidade de Brasília, UNB), where a plant voucher is deposited (JEP 3686-UnB).

Purification of A. pyrifolium crude stem bark extract was performed as previously described [33] (Fig. 2a). Briefly, the stem bark of the plant was air-dried at room temperature and ground to a course powder (2.5-mm mesh size) using a laboratory mill. The powdered material (3.0 kg) was extracted with 95% ethanol (5.5 l) in a Soxlet apparatus for 72 h, and then concentrated under reduced pressure in a rotary evaporator. The remaining water was removed using a freeze dryer yielding 147 g of crude stem bark extract (AP1) and kept at 4 °C until use. AP1 was dissolved in methanol, water was added (2:3) and the mixture was partitioned with ethyl acetate. The hydromethanolic phase was lyophilized producing 55 g of an alkaloid rich fraction (AP2), and the ethyl acetate phase was concentrated under pressure yielding 91 g of the AP3 fraction. The alkaloids were detected in silica gel TLC plates by spraying with Dragendorff reagent.

The fractionation procedures of the stem extract are summarized in Fig. 2b. The isolated plant material was air-dried as described above; the powder was extracted with 95% ethanol and concentrated under reduced pressure yielding 147 g of extract (AP4). This extract was dissolved in water: methanol (2:3), and the mixture partitioned with ethyl acetate. The organic fraction (AP5) was solubilized in chloroform and partitioned with hydrochloric acid 0.1 M. The aqueous acidic phase was separated and adjusted to pH 10 with 1 M NaOH. The free base alkaloids were extracted with chloroform, producing an aqueous (AP5F.AQ) and an alkaloid-rich (AP5F.ALC) fraction. The aqueous phase was partitioned with butanol yielding an organic (AP6) and an aqueous (AP7) fraction. Dragendorff reagent was used to reveal the presence of alkaloids in TLC plates. The extracts and fractions obtained were further used in biological assays. Fractionation of the crude leaves, root and root bark extracts was as described above for the stem.

The analyses were performed on a Nexera UHPLC-system (Shimadzu) hyphenated to a maXis ETD high-resolution ESI-QTOF mass spectrometer (Bruker) controlled by the Compass 1.5 software package (Bruker). Samples were diluted to final concentration of 5 mg/ml and 1 µl was injected on a Shimadzu Shim-Pack XR-ODS-III column (C18, 2.2 µm, 2.0 × 150 mm) at 40°C at a flow rate of 400 μl/min. An additional identical run was performed with 5 µl (50 µg) injected, with 200 µl fractions collected in each well of a microtitre plate. The mobile phases A and B (0.1% formic acid in distilled water and acetonitrile, respectively) were used to form an isocratic run of 5% B during the initial 5 min, followed by a linear gradient to 100% B in 40 min and a hold at 100% B for 5 min. The mass spectra were acquired in positive mode at a spectra rate of 2 Hz. Ion-source parameters were set to 500 V end plate offset, 4500 V capillary voltage, 3.0 bar nebulizer pressure, 8 l/min and 200 °C dry gas flow and temperature respectively. Data-dependent fragment spectra were recorded using a collision energy range between 15 and 60 eV. Ion cooler settings were optimized for an m/z 40–1000 range using a calibrant solution of 1 mM sodium formate in 50% 2-propanol. Mass calibration was achieved by initial ion-source infusion of 20 μl calibrant solution and post-acquisition recalibration of the raw data. Compound detection was performed by chromatographic peak dissection with subsequent formula determination according to exact mass and isotope pattern (MS1). Identification was based on comparison of compound fragment spectra (MS2) with reference spectra of an in-house database of standard compounds, the public spectra database MassBank [34], as well as in silico fragment spectra generated from the Universal Natural Product Database (UNPD-ISDB) [35]. The solvents were removed in a vacuum centrifuge at 45 °C before the P. falciparum assay described below.

The blood forms of P. falciparum (W2 clone, chloroquine-resistant and mefloquine-sensitive) [36] were cultured as described [37]. After synchronization with sorbitol, the ring forms parasitized erythrocytes were used in tests of anti-parasite activity using the immune enzymatic assay with specific monoclonal antibodies to a parasite protein rich in histidine and alanine (HRPII), as described [38]. The antibodies were commercially acquired (MPFM55A and MPFM55P, ICLLAB, Portland, OR, USA). Endotoxin-free sterile disposables labware were used in all experiments.

The half-maximal drug inhibitory concentration (IC₅₀) was estimated by curve fitting with the software Origin (OriginLab Corporation, USA) compared to parasite growth in drug-free medium. Fractions with IC₅₀ values bellow 10 µg/ml were considered active; between 10 e 20 µg/ml as partially active, and above 20 µg/ml as inactive as demonstrated before [30, 31]. The chloroquine was used as anti-malarial control in all assays.

The toxicity tests of plant extracts and fractions were performed against a monkey kidney cell line (BGM), a human hepatoma cell line (HepG2), and freshly isolated human peripheral blood mononuclear cells (PBMC) collected from healthy volunteers (approved by Ethics Committee, Centro de Pesquisas René Rachou-FIOCRUZ, CAAE 03209212.7.0000.5091 at 10/03/2012). Cytotoxicity was evaluated by the MTT assay [(3-(4,5-dimethylthiazol-2-yl)-2,2-diphenyltetrazolium)] as described [39]. The cell viability was expressed as the percentage of the control absorbance in the untreated cells after subtracting the background. The minimum lethal dose for 50% of the cells (MLD₅₀) was determined as previously described [40]. The ratio between drug cytotoxicity (MLD₅₀) and activity (IC₅₀) was used to estimate the selective index (SI), or therapeutic activity as described for A. nitidum [29]. A SI ≤ 10 was indicative of toxicity.

Swiss outbred adult female mice (20 ± 2 g) were each inoculated intraperitoneally with 1 × 10⁵ P. berghei (strain NK65)-infected red blood cells, as described before [29]. The mice were divided randomly in groups of 3–5 animals per cage and then treated by gavage for 3 consecutive days with 100 mg/kg of the fractions dissolved in DMSO 3% (v/v). Blood samples were collected from the mice tails on day 5 and 10 of infection, fixed with methanol, Giemsa stained and used for parasitaemia determination by microscopy. The extracts and fractions were evaluated in one test and the per cent reduction of parasitaemia calculated considering untreated mice parasitaemia as 100%, the leaves extract was tested three times. Drugs reducing parasitaemia by 40% or more were considered active; reductions between 20 and 40% partially active, and reductions below 20% inactive as demonstrated before [29]. Mice mortality was monitored daily until day 30 post-infection. The chloroquine was used as anti-malarial control at a sub-curative dose of 20 mg/kg. The protocol for animal use was approved by the Ethics Committee at FIOCRUZ (CEUA LW-23/13 at 05/20/2013).

The potential of the active alkaloid fraction from the plant stem to induce mutagenic and genotoxic effects in vitro was evaluated with Ames tests [41] performed at the Genotox-Royal Institute (Rio Grande do Sul, Brazil), contract number GT00748. Five strains of Salmonella typhimurium with several specially constructed mutants were tested in the presence of various drug concentrations in the absence and presence of the metabolizing rat liver fraction for the ability of a given extract or fraction to induce mutations.

The A. pyrifolium extracts and fractions were not toxic at the highest doses tested against the BGM cells (MLD₅₀ ≥ 1000 µg/ml), except for the crude extracts from root bark and leaves (MLD₅₀ = 287 to 486 µg/ml). Some toxicity to HepG2 cells and to PBMC freshly collected also occurred with most extracts and fractions of the crude stem and stem extracts (Table 1).

Among 5 crude extracts (from stem, roots, leaves) and 7 different fractions tested against P. falciparum chloroquine-resistant parasites, the best in vitro activity was shown by the crude stem bark extract (AP1) (IC₅₀ = 3 µg/ml). This was followed by the alkaloid fraction AP5F.ALC (IC₅₀ = 5 µg/ml) and the organic fraction (AP5) (IC₅₀ = 6 µg/ml) both from the crude stem extract. The stem (AP4), root bark (AP9), roots (AP12) and leaves extracts, and the aqueous fraction (AP2) from stem bark extract were all partially active; all other fractions were inactive (Table 1). Chloroquine was used as positive control in all assays.

None of the crude extracts or fractions was cytotoxic to HepG2, BGM or human peripheral blood mononuclear cells; the MDL₅₀ values of all the crude bark extracts and fractions were similar or better when tested on normal cells, with the exception of AP5 and AP5F.ALC fractions (Table 1).

The ratio between in vitro cytotoxicity and activity or selectivity index (SI) was highest for the crude stem bark (AP1) regardless of the source of cells used to evaluate toxicity in vitro, varying from 137 to 333 (Table 2). The organic (ethyl acetate) fractions from stem bark (AP3), stem (AP5), and the alkaloid-rich fraction (AP5F.ALC) from stem extract (AP4) showed toxicity to the BGM cell line only, with high specific activity to P. falciparum reflected in SI values of 111 (AP3), 167 (AP5) and 200 (AP5F.ALC).

The extracts from the root bark (AP9), root (AP12) and leaves, which were available in sufficient amounts, were also evaluated in mice for their anti-malarial activity. Oral administration (100 mg/kg) of the crude root extract (AP12) for 3 consecutive days reduced P. berghei parasitaemia by 75 and 52% on the 5th and 10th days of infection, respectively (Table 3), and by 79% when animals were treated by the crude root bark extract (AP9) on the 5th day. The leaves extract resulted on slightly reduced parasitaemia. The aqueous fraction AP2 (derived from AP1) caused a 93 and 57% reduction on 5th and 10th days of infection; the alkaloid fraction AP5F.ALC (derived from AP4) caused 79 and 57% reduction at both time points. Chloroquine, the standard anti-malarial, was tested in parallel each time at 20 mg/kg. Other samples that presented activity in vitro were not tested in vivo due to insufficient mass amounts available.

The remaining few milligrams of AP5F.ALC were analysed on an UPLC coupled with a HRMSMS equipment. The fractions collected after the injection of 50 mg in a reversed phase analytical column were assayed in vitro against P. falciparum. The two fractions that eluted between 19.5 and 20.5 min (Fig. 3a, b) were active in the bioassay. The m/z profile of this region (Fig. 3c) showed that these fractions were composed of two major components, A and B, exhibiting [M+H]⁺ ions with m/z of 555.3488 and 627.3698, compatible with molecular formulas C₃₈H₄₂N₄ and C₄₁H₄₆N₄O₂, respectively.