Anthelmintic acetogenin from Annona squamosa L. Seeds

Annona squamosa seeds extracts showed anthelmintic activity against Haemonchus contortus, the main nematode of sheep and goat in Northeastern Brazil. A compound 1 was isolated from ethyl acetate extract and inhibited the egg hatching of H. contortus at 25 mg ml−1. The structure of 1 was determined as a C37 trihydroxy adjacent bistetrahydrofuran acetogenin based on spectroscopic analysis.


INTRODUCTION
In Northeastern Brazil, gastrointestinal parasitism of sheep and goats is one of the leading causes of mortality, producing high economic losses (Pinheiro et al. 2000). Among gastrointestinal nematodes, Haemonchus contortus is the most frequent and pathogenic, being responsible for the high mortality rate in young animals during the rainy season (Menezes et al. 1992, Arosemena et al. 1999. Trying to reduce these losses, synthetic anthelmintics are routinely used, often indiscriminately, causing reduced efficacy besides environmental pollution and food residues (Waller et al. 1995, Herd 1995. To decrease these negative impacts, the phytotherapy is an alternative tool that has been studied by many researches nowadays. The study of plants from Annonacea family has demonstrated the presence of active substances with parasiticidal effects (Duret et al. 1998). Annona squamosa seed powder is popularly used against insects in Northeast of Brazil (Braga 2001). In the search for new anthelmintic agents to be used in the control of goat nematodes, the anthelmintic activity of extracts and isolated compounds of A. squamosa seeds were evaluated on egg hatching of H. contortus.

PREPARATION OF A. squamosa EXTRACTS
The seeds of A. squamosa were obtained by manual extraction from fruits bought in the central market of Fortaleza. The extracts were prepared by the methodology performed by Nonfon (1990) as follows: A. squamosa seeds (2 kg) were triturated to a powder that was mixed with methanol/water solution (90:10, 3 l) and left in contact for seven days. This mixture was filtered and an aliquot of this initial solution (300 ml) was evaporated to dryness for bioassay analysis. The aqueous-methanol solution was evaporated under reduced pressure using a rotatory evaporator. Methanol was eliminated and the remaining aqueous solution was transferred to a separatory funnel and washed up four times (4 × 30 ml) with ethyl acetate that was evaporated to obtain the correspondent extract (15.0 g). The aqueous layer was dried in a water bath to obtain the aqueous extract (44.0 g).

ISOLATION OF ACETOGENIN 1
The Ethyl acetate extract (EtOAc) (15 g) was submitted to a silica gel (0.063-0.2 mm) being eluted with chloroform, ethyl acetate and methanol in mixtures of increasing polarity. Fifty-two (10 ml) fractions were obtained and compared by TLC using silica gel kieselgel 60 from Merck, visualized by exposure to I 2 vapor. Compound 1 (302 mg) was isolated by elution the column with 100% ethyl acetate.
The acetyl derivative was obtained by reaction of compound 1 (100 mg) with acetic anhydride (3 ml) and pyridine (1 ml) mixture at room temperature for 24 hours. Then a hydrochloric acid solution (5%) was added (10 ml). The reactional mixture was transferred to a separatory funnel and extracted with chloroform (3× 5 ml). The organic layer, containing the acetyl derivative was washed with water (5 × 5 ml), dried with sodium sulfate and the solvent was evaporated leaving 1a (80 mg) (Furniss et al. 1989).

CHEMICAL ANALYSIS OF ACETOGENIN 1
Optical rotation was made in CHCl 3 (Perkin Elmer 341); IR spectra were recorded on Perkin Elmer FT-IR spectrum 100 spectrophotometer and the values are expressed in cm −1 . NMR spectra were recorded on a Brucker Avance DRX-500 spectrometer in CDCl 3 . 70 eV EI-Mass spectra was obtained using a Hewlett-Packard 5971 GC/MS instrument employing the following conditions: column: Dimethylpolysiloxane DB-1 coated fused silica capillary column (30 m × 0.25 mm × 0.25 mm); carrier gas: He (1 ml/min); capillary injector operating at 250 • C in the split mode (1:100); detector temperature: 200 • C; column temperature: 35-180 • C at 4 • C/min then 180-250 • C at 10 • C/min; mass spectra: electron impact. High-resolution mass spectra were obtained in an ultrOTOF Q -ESI-TOF Mass Spectrometer, Bruker Daltonics, Billerica, MA, USA. H. contortus eggs were recovered according to Hubert and Kerboeuf (1984). Briefly, 10 g of feces were collected directly from the rectum of sheep experimentally infected with H. contortus mixed with tepid water and filtered through 590; 149; 101 and 30 µm aperture sieves. The eggs retained on the 30 µm sieve were collected.
The in vitro egg hatch test was carried out by the method of Coles et al. (1992). To perform this test, a volume of 500 µl, with 250 µl of the eggs solution (about 120 eggs) and 250 µL of extract solution were placed in 5 ml tubes. The extract solutions for the bioassay were prepared using the concentrations of the extracts tested were 25; 5; 1; 0.2; 0.04 mg ml −1 , the negative control was the diluent Tween 80 and the positive control thiabendazole (0.1 µg ml −1 ). After incubation for 48 hours, a drop of Lugol was added. All larvae and eggs 1 H (500 MHz) and 13 C (100 MHz) NMR spectral data for 1 and 1a, including heteronuclear 2D shift-correlated obtained by 1 H and 13 C-COSY-n J C H (n=1, HMQC, n=2 and 3, HMBC) experiments, in CDCl 3 as solvent, chemical shifts (δ, ppm) and coupling constants (J , Hz, in parenthesis). a were counted using a microscope. Five replicates were performed for each concentration. Data were expressed as percentage of egg hatching inhibition. Statistical comparisons of the results from the egg hatching inhibition by the extracts on different concentrations were performed using Kruskal-Wallis test, with significance level of 5%. The effective concentration to inhibit half eggs hatching (CE 50 ) for each extract was calculated by the equation which showed the tendency line with minimum determination coefficient of 70%. The values of negative and positive controls are expressed as mean results. Table I displays the 1 H and 13 C-COSY 2D NMR (HMQC and HMBC) data of compound 1 and its acetyl derivative 1a. Compound 1 showed to be an acetogenin, a type of compound common in plants of Annonaceae family, with 37 carbons, three hydroxyl groups, an unsaturated lactone moiety (δ 6.99, H-35; δ 148.9, C-35; δ 134.2, C-2; δ 173.8, C-1) and two adjacent tetrahydrofuran (THF) rings ( Figure 2). The adjacent bis-THF rings with flanking OH groups in 1 were indicated by 1 HNMR resonances at δ 74.1 (C-15), 83.2 (C-16), 82.5 (C-19), 82.1 (C-20), 82.7 (C-23) and 71.4 (C-24). The relative stereochemistry of the bis-THF moiety of 1 was suggested to be threo-trans-threo-trans-erythro by careful comparison of the 1 H and 13 C NMR signals of 1 with stereochemically defined bis-THF acetogenins (Araya et al. 2002, Alali et al. 1999, Zhou et al. 2000, Hopp et al. 1998 -20, H-23), which integrated constitutes 6 protons. A third hydroxyl group was evidenced by 1 H NMR signal at δ 3.62 and a 13 C NMR signal at δ 71.7. Its position was predicted to be at C-5 based on the EIMS fragments ions at m/z 155, 111, 137, 111 and 97 ( Figure 1) common in other C-5 hydroxylated acetogenins (Alali et al. 1999) and associated signals of 1 H-1 H COSY, HMQC and HMBC of the acetyl derivative 1a. In the bidimensional homonuclear 1 H-1 H COSY spectrum of 1a, the sequence of couplings between H-3 (t, δ 2.32) and H-4 (m, δ 1.55) and H-4 with H-5 (m, δ 4.91) assures the presence of a hydroxyl group at C-5. The IR spectra of compound 1 showed as main peaks at 3418 cm −1 for hydroxyl groups and at 1748 and 1652 cm −1 for the C=O and the double bond of the lactone moiety respectively, that are compatible with data of similar acetogenins from seeds of A. squamosa (Rupprecht et al. 1990). The percentage of egg hatching inhibition of H. contortus using MeOH/H 2 O, aqueous, EtOAc extracts and compound 1 is shown in Table II.

RESULTS AND DISCUSSION
The highest inhibition of egg hatching was obtained with the EtOAc extract, in the concentrations of 5 and 25 mg ml −1 , which did not demonstrate significant difference from thiabendazole, at 0.1 µg ml −1 . The aqueous extract showed an inhibition percentage higher than the negative control, Tween 80, at 25 mg ml −1 , but at 5 mg ml −1 was not statistically different from the nega-  tive control. The EtOAc extract was submitted to a silica gel column chromatography and the main isolated compound 1 was evaluated for its anthelmintic activity. Compound 1 inhibited the egg hatching more than 90% at 5 and 25 mg ml −1 concentrations, similarly to the EtOAc extract. The concentration that is effective for killing half of larvae amount (CE 50 ) of the aqueous, methanol/water, ethyl acetate extracts and compound 1 was 10.02, 0.89, 0.76 and 0.78 mg ml −1 respectively. Probably these results are due to the low solubility of acetogenins in water.
The ovicidal effect obtained with ethyl acetate extract is lower than synthetic anthelmintics nevertheless it is important when compared with other plant extracts. Assis et al. (2003) reported 20% inhibition of H. contortus egg hatching with ethyl acetate extract of Spigelia anthelmia at 12.5 mg ml −1 and the EtOAc extract of A. squamosa showed 99% egg hatch inhibition at 5 mg ml −1 . The effect of eight acetogenins, isolated from Uvaria hookeri and U. narum, against H. contortus adult specimens showed a negative relationship between the death time in minutes and acetogenins concentration (Padmaja et al. 1993). In conclusion, the seeds of A. squamosa represent an alternative natural source for anthelmintic compounds.