- Email: [email protected]
Anticrotalic and antitumoral activities of gel filtration fractions of aqueous extract from Tabernaemontana catharinensis (Apocynaceae)
Comparative Biochemistry and Physiology Part C 137 (2004) 19–27
Anticrotalic and antitumoral activities of gel filtration fractions of aqueous extract from Tabernaemontana catharinensis (Apocynaceae) Lucilene de Almeidaa, Adelia C.O. Cintraa, Elen L.G. Veronesea, Auro Nomizoa, ˜ Jose´ Francoa, Eliane C. Arantesb, Jose´ Roberto Giglioc, Suely Vilela Sampaioa,* Joao a
´ ´ ´ ´ ˆ ˆ ˜ Preto–USP, Departamento de Analises Clınicas, Toxicologicas e Bromatologicas, Faculdade de Ciencias Farmaceuticas de Ribeirao ´ syn, 14040-903 Ribeirao ˜ Preto, SP, Brazil Avenida do Cafe, b ´ ´ ˆ ˆ ˜ Preto, Universidade de Sao ˜ Paulo, Departamento de Fısica e Quımica, Faculdade de Ciencias Farmaceuticas de Ribeirao ˜ Preto, SP, Brazil 14040-903 Ribeirao c ´ ˜ Preto, Universidade de Sao ˜ Paulo, Departamento de Bioquımica e Imunologia, Faculdade de Medicina de Ribeirao ˜ Preto, SP, Brazil 14049-900 Ribeirao Received 16 March 2003; received in revised form 21 October 2003; accepted 28 October 2003
Abstract The high mortality caused by Crotalus durissus terrificus snake venom is mainly due to crotoxin, which acts on the neuromuscular junction inhibiting the mechanism mediating acetylcholine release, thus leading to motor and respiratory paralysis and subsequently to animal death. We recently demonstrated that the aqueous extract (AE) of Tabernaemontana catharinensis can inhibit the lethal activity of C. d. terrificus venom. Eight fractions, PI to PVIII , were obtained by gel filtration of the extract on Sephadex G-10, and assayed for lethality and cytotoxicity. Fraction PVII w2.0 mgy100 g raty 50 ml saline solution (ss)x injected intramuscularly (i.m.) 20 s after the venom (240 mg) or crotoxin (200 mgy50 ml ss) neutralized the lethal activity of 2 LD50 of both. Fractions PI , PVI and PVIII (5.0 mgy100 g raty50 ml ss) presented potent antitumoral activity in vitro against cells from human breast carcinoma (SK-BR–3) after 24 h incubation, as measured by Mosmann colorimetric method. Fraction PVII contains 12-methoxy-4-methylvoachalotine as its major component. These results demonstrate that the antivenom and antitumoral activities of the AE of T. catharinensis are exerted by different substances present in fraction PVII and fractions PI , PVI and PVIII , respectively, whose characteristics are distinct in terms of staining and Rf when analyzed by thin layer chromatography. The results also show that a preliminary fractionation by Sephadex G-10 gel filtration is a good option as a first step for isolation of biologically active substances from T. catharinensis. 䊚 2003 Elsevier Inc. All rights reserved. Keywords: Anticrotalic activity; Antitumoral activity; Crotalus durissus terrificus venom; Crotoxin; Tabernaemontana catharinensis
1. Introduction The search for antidotes against snakebites obtained from several plants used in Brazilian folk *Corresponding author. Tel.: q55-16-6024287; fax: q5516-6331092. E-mail address: [email protected] (S.V. Sampaio).
medicine started with the species Curcuma sp. and with the aqueous-ethanolic extract from the root ¸ de negro’ (black man’s head) originatof ‘cabeca ing from northeastern Brazil. The ar-turmerone fraction isolated from the rhizome of Curcuma sp. inhibited the action of a neurotoxin from Naja naja siamesis venom (Cherdchu et al., 1978). Two
1532-0456/04/$ - see front matter 䊚 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.cca.2003.10.012
20
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27
components, cabenegrins A-I and A-II (pterocar¸ pans), in the aqueous-ethanolic extract of ‘cabeca de negro’ neutralized the toxic cardiovascular effect of the Bothrops atrox venom (Nakagawa and Nakanishi, 1982). Later, Rizzini et al. (1988) enumerated 83 plants species belonging to 34 families used in Brazilian folk medicine against animal venoms, especially snake venoms. Among these families is the Apocynaceae, which contains several species. Tabernaemontana catharinensis (Apocynaceae), popularly known as ‘leiteiro de vaca’ (cow’s dogbane), occurs in Argentina, Paraguay, Bolivia and southern Brazil. Studies conducted by Batina et al. (1997, 2000) demonstrated that the fresh and stabilized aqueous extracts of the root of Peschiera fuchsiaefolia, today reclassified as T. catharinensis, can inhibit the lethal and myotoxic activity of Crotalus durissus terrificus venom (South American Rattlesnake, cascavel). Batina et al. (2000) isolated from T. catharinensis a quaternary base alkaloid, 12-methoxy-4-methylvoachalotine (MMV), which proved to be able to inhibit the lethal activity of 2 LD50 of C. d. terrificus venom. Substances with trypanocidal and antitumoral activity were also isolated from this plant (Pereira et al., 1999). In addition, the aqueous extract of T. catharinensis was shown to have a potent cytotoxic action on human tumor lines such as SK-BR-3, MCF-7 and C-8161 in vitro (Meyer et al., 1997). On the basis of the results obtained with the species T. catharinensis and due to its diversity of alkaloids with pharmacological activity, the objective of the present study was to fractionate the aqueous extract (AE) of T. catharinensis and evaluate the action of the fractions obtained in terms of anticrotalic and antitumoral activity.
and voucher specimens were deposited in the herbarium (SPFR) of Faculdade de Filosofia, ˆ ˜ Preto, USP (Reg. n8 Ciencias e Letras de Ribeirao 02940). 2.2. Preparation and fractionation AEs were prepared as follows: after identification the plant root bark was removed, washed, ground in distilled water (5 gy10 ml water) and kept at 4 8C for 12 h, strained through a nylon fabric, kept at 4 8C for an additional 12 h, and then filtered under reduced pressure. The AE was lyophilized and stored dry at y10 8C. 2.3. Fractionation of T. catharinensis AE One gram of lyophilized T. catharinensis AE was dissolved in 10 ml 0.3 M ammonium bicarbonate buffer (AMBIC), pH 8.0, and centrifuged at 12 100=g for 10 min at 4 8C. An aliquot of the supernatant was removed for the determination of absorbance at 280 nm. The remainder of the supernatant was applied on a Sephadex G-10 column (2.0=99.0 cm) and eluted with the same buffer. Five-milliliter fractions were collected at a flow rate of 20.0 mlyh at room temperature. Absorbance at 280 nm was determined for each collected fraction and the fractions were pooled and lyophilized. This gel filtration was performed several times in order to produce enough material for subsequent experiments. 2.4. Fractionation of C. d. terrificus venom Crotoxin, the major constituent of C. d. terrificus venom, was obtained as described by Laure (1975).
2. Materials and methods
2.5. Determination of crotoxin i.m. LD50 in rats
2.1. Venom and plant material
Crotoxin was dissolved in 0.9% (myv) saline solution and clarified by centrifugation at 800=g for 5 min at room temperature. An aliquot of the supernatant was used for determination of protein concentration (Itzhaki and Gill, 1964). The solution was then diluted at a 1.15 geometric ratio. Rats were divided at random into groups of six. Each animal received two injections in the right hind thigh with a 20 s interval, for a final volume of 100 ml. The animals were then examined over a 24 h period and dead and alive animals were
C. d. terrificus venom was purchased from ˜ Paulo, SP. Albino male Instituto Butantan, Sao Wistar rats weighing 90–110 g were used. The animals were kept at 24"1 8C and fed ad libitum with commercial ration and water. All animals were carefully monitored and maintained in accordance with ethical recommendations of the Ethic Commission for the Use of Animals (CEUA). T. catharinensis A. DC. was collected in Assis, SP,
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27
counted. Rats were also examined for signs of hind leg paralysis, generalized paralysis, body rigidity and dyspnea. The LD50 was determined by the method of Finney (1964) using Graphpad Prism software. 2.6. Evaluation of the action of the fractions of T. catharinensis AE against the lethal activity of C. d. terrificus crude venom and crotoxin The lethality caused by 2 LD50 was used as a parameter to evaluate the effects of the total extract or fraction on C. d. terrificus envenomation. We adopted a protocol, which simulated an ophidian accident and followed the progress of local and systemic effects in animals that received only venom or toxin compared with those that, in addition, received the extract or fraction from T. catharinensis 20 s after. C. d. terrificus venom and crotoxin were dissolved in ss and clarified by centrifugation at 800=g for 5 min at room temperature. Protein concentration was determined in an aliquot of each supernatant. The dose of crude venom was fixed at 240 mg proteiny100 g raty50 ml ss, corresponding to 2 LD50 (Batina et al., 1997, 2000) and that of crotoxin was fixed at 200 mg crotoxiny100 g raty50 ml ss (also 2 LD50). Increasing doses of the fractions of T. catharinensis AE were administered. The fractions were dispersed in ss and clarified by centrifugation at 800=g for 5 min. The supernatant was separated and kept at room temperature until use. For biological assays the rats were divided at random into groups of six animals. Each animal received two i.m. injections, 20 s apart, in the right hind thigh. Rats were then submitted to the treatments described below. 2.7. Treatment with the T. catharinensis fractions 20 s after C. d. terrificus venom In group 1, rats received two injections of 50 ml ss and, in group 2, they received 2 LD50 of the crude venomy50 ml ss and then 50 ml ss. In the remaining groups, the animals received 2 LD50 of C. d. terrificus venom and increasing doses of the fractions of T. catharinensis AE, starting with 1.0 mgy100 g raty50 ml ss. A control was also prepared for each AE fraction (5.0 mgy100 g raty 50 ml ss). The animals were observed for 96 h, at
21
2 h intervals up to 12 h and at 24 h intervals thereafter. 2.8. Treatment with fraction PVII 20 s after crotoxin In group 1, the animals received two injections of 50 ml ss each. In group 2, the animals received 2 LD50 of crotoxin and 50 ml ss. In group 3, the animals received 50 ml ss and 4.0 mg of fraction PVII y100 g raty50 ml ss. In the remaining groups, the animals received 2 LD50 of crotoxin and increasing doses of fraction PVII (1.0–4.0 mgy100 g raty50 ml ss). The rats were then examined as described above. 2.9. Antitumoral activity For evaluation of the antitumoral activity we used the human breast adenocarcinoma line SKBR-3, obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA). The tumor cells were first expanded in cell culture flasks in complete RPMI medium (15 mlyflaskRPMI 1640 supplemented with 10% fetal calf serum, 20 mM L-glutamine, 7.5% Na2HCO3 and 10 mgyml gentamicin) at 37 8C in a humidified incubator containing 95% air and 5% CO2. Lyophilized fractions of T. catharinensis were first dissolved in complete RPMI medium, vortexed and then sterilized by filtration through 0.22 mm membranes. Different concentrations of the AE fractions were added to 96-well cell culture plates containing 5=105 SK-BR-3 cells and incubated for 24 h. The cytotoxic activity of the fractions of T. catharinensis AE against the tumor line was evaluated by the colorimetric method described by Mosmann (1983). 2.10. Analysis by thin layer chromatography (TLC) Fractions of T. catharinensis AE and the MMV alkaloid were analyzed by TLC on silica gel 60 F254 (Aldrich) plates using chloroform: methanol (9:1, vyv) and butanol: acetic acid: water (4:1:5, vyv) as the organic phases. Visualization of chromatographic bands was achieved by reaction with sulfuric vanillin and Dragendorff reagent (Wagner et al., 1984) and observation under UV light. 3. Results Fig. 1 shows the gel filtration profile of T. catharinensis AE on Sephadex G-10, which
22
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27
revealed eight fractions, PI to P VIII. The recovery in terms of weight for each fraction is showed in Table 1. LD50 of crotoxin was approximately 100 mgy100 g raty50 ml ss. In TLC, fractions PVI, PVII,PVIII and MMV showed an orange color when developed with the Dragendorff reagent and only fraction PVII and MMV showed a bluish color when developed with sulfuric vanillin. The Rf values for these compounds were 0.35 (PVI), 0.46 (PVII), 0.45 (PVIII) and 0.46 (MMV), respectively. An additional orange spot (Rfs0.23) was also present in fraction PVI (Fig. 2). Two LD50 of crude C. d. terrificus venom paralyzed the rats’ hind legs within 10 min. Generalized paralysis was observed within 30 min in all animals. Within 4 h, 83% of the animals manifested severe dyspnea, apnea and death. Muscle spasms were also observed. T. catharinensis AE fractions had no apparent effect on the control animals, all survived the experiments. Fraction PVII, 2.0 mgy100 g raty 50 ml ss, injected i.m. 20 s after the venom, completely neutralized the lethal action of 2 LD50 of the venom. Doses above 2.0 mg did not change the rate of inhibition of the lethality induced by 2 LD50, as shown in Fig. 3A. Fraction PVII did not inhibit hind leg paralysis, a characteristic effect of
Table 1 Recovery of fractions from T. catharinensis aqueous extract (AE) after gel filtration on Sephadex G-10 (Fig. 1) Fractions
% mass recovery
AE PI PII PIII PIV PV PVI PVII PVIII Total
100.0 3.6 30.7 41.9 2.4 0.7 0.8 7.6 0.3 88.0
crotamine from the crotalic venom, which disappeared within the next 48 h. The animals presented moderate dyspnea, which disappeared within 72 h. Fraction PVII, 2.0 mgy100 g raty50 ml ss, injected i.m. 20 s after crotoxin, completely abolished the lethality of 2 LD50 of crotoxin (Fig. 3b). During the first 4 h, the animals presented moderate dyspnea and generalized paralysis, which receded within the next 48 h. Several experiments gave consistent results, with S.D.s0. Fraction PI, 2.5 and 5.0 mgy100 g raty50 ml ss, inhibited the lethality of the venom by 50%. During the first 120 min, 50% of the animals presented paralysis of the hind legs and moderate
Fig. 1. Gel filtration profile of T. catharinensis aqueous extract (AE) on a Sephadex G-10 column (2.0=99.0 cm). 1 g sample of AE dissolved in 10 ml 0.3 M AMBIC, pH 8.0, was applied and eluted with the same buffer. Flow rate was 20 mlyh and 5 mlytube were collected at 25 8C.
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27
23
Fig. 2. Thin layer chromatography of the quaternary base 12 methoxy-4-methylvoachalotine (MMV) and fractions PI to PVIII from T. catharinensis. Mobile phase was: n-butanol: acetic acid: water (BAW 4:1:5). Test reagents were: (a) Dragendorff reagent and (b) sulfuric vanillin. Samples: 1, 2, 3: AE; 4:PI; 5: PII; 6: PIII; 7: PIV; 8: PV; 9: PVI; 10: PVII; 11: PVIII; 12: MMV.
dyspnea. After 8 h, dyspnea increased and the animals returned to normal within 72 h. Fractions PII to PV did not inhibit the lethality of the venom at the doses assayed. Fractions PVI and PVIII were not assayed because their yield after gel filtration on Sephadex G-10 was very low, 0.8% and 0.3%, respectively (Table 1). The LD50 (i.m., 24 h) of crotoxin was approximately 100 mgy100 g raty50 ml ss (92.31–107.92 mg proteiny100 g raty50 ml ss). 2 LD50 caused marked dyspnea during the first 2 h and generalized paralysis within 4 h. After 12 h, dyspnea increased drastically in 52.4% of the rats, causing death. In animals that did not die, the generalized paralysis and dyspnea receded during the first 96 h. AE and fractions PI, PVI and PVIII had potent antitumoral activity against cells from human breast carcinoma (SK-BR-3) at concentrations of 5 mgyml. In addition, fractions PVI and PVIII and the AE also had antitumoral activity at 2.5 mgyml (Table 2). These results are comparable to those obtained with the methotrexate control, a drug used for the clinical treatment of breast adenocarcinoma.
4. Discussion The usual phytochemical methods for micromolecule purification are cumbersome, involving several steps of extraction with various solvents of different polarity, and are not always efficient. Changes of solvents or a mixture of solvents, or even changes of the stationary phase are necessary for the purification of these micromolecules, but the use of these solvents hampers the solubilization of the fractions, possibly also causing injury to the animals and inactivation of the venom or of the fraction intended for study. In the present study, a single step gel filtration on Sephadex G-10 provided a partial resolution of the T. catharinensis AE into eight fractions, PI to PVIII, soluble in distilled water and in saline solution, four of which demonstrated anticrotalic or antitumoral activity. Since biologically active components were detected in some of these still heterogeneous fractions, this mild, simple and high yield classical procedure can be used as a first step of a more elaborate methodology for the ultimate isolation of specific active components. In PVII, it was possible to identify the quaternary
24
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27
Fig. 3. Percent survival. (a) Percent survival of rats that received 2 LD50 of C. d. terrificus venom (i.m.) and different doses of fraction PVII (mgy100 g raty50 ml ss) from T. catharinensis 20 s later. Each group contained six animals. Statistically identical groups were followed by identical letters (P-0.05). (b) Percent survival of rats that received 2 LD50 of crotoxin (i.m.) and different doses of fraction PVII (mgy100 g raty50 ml ss) of T. catharinensis 20 s later. Each group contained six animals. The dots represent the mean"S.E.M. for the groups. Statistically identical groups are followed by identical letters (P-0.05).
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27 Table 2 Percent cytotoxicity of the different fractions of T. catharinensis AE against SK-BR-3 cells Plant extract or fraction
Aqueous extract (AE) PI PII PIII PIV PV PVI PVII PVIII Methotrexate
Cytotoxicity activity on SKBR-3 tumor cells (%) 2.5 mgyml
5.0 mgyml
69.0"1.3 37.3"1.2* 33.9"0.9* 0* 0* 19.6"2.6* 50.4"0.5 28.1"0.8* 50.5"0.2 54.3"1.4
82.4"0.5 66.2"1.1 46.5"0.7* 0* 0* 25.8"1.3* 64.0"0.4 45.8"0.7* 68.1"1.6 64.8"0.8
% Cytotoxicity – Cytotoxicity of the fractions was analyzed by the colorimetric method described by Mosmann (1983). Results are shown as means"S.D. Statistical differences were analyzed by Mann–Whitney non-parametric test. Values of P0.005* were considered significantly different form the control (aqueous extract).
alkaloid MMV, previously isolated by Pereira et al. (1999) and Batina et al. (2000). The results of TLC (Fig. 2) showed that fractions PVI, PVII and PVIII and the MMV component stained orange with the Dragendorff reagent, a characteristic of nitrogenated substances such as alkaloids. Fraction PVII and MMV were also stained by sulfuric vanillin, a fact also indicating the presence of an alkaloid (Fig. 2b). However, fractions PVI, PVII and PVIII showed different Rf and staining, indicating that they are not the same substances. Staining and Rf of fraction PVII fit with those of MMV. Therefore we suggest that fraction PVII contains MMV as the major component. Fraction PI was not stained by the reagents used (Fig. 2). Its structure is unknown and should be investigated. Thus, fractions PI, PVI, PVII and PVIII are distinct, presenting different staining, Rf and elution volumes when submitted to gel filtration on Sephadex G-10 (Fig. 1). The i.m. route was chosen for the administration of C. d. terrificus venom and crotoxin because it simulates the inoculation of the venom by the snake, and was also chosen for the administration of the fractions of T. catharinensis AE because it is closer to the bandage application in popular use ¸ 1992). (Batina and Penco, The symptoms observed in the animals injected
25
with 2 LD50 of C. d. terrificus have been described by Vital Brazil (1980). Crotoxin, a presynaptic neurotoxin, inhibits the release of acetylcholine, whose absence causes neuromuscular blockade and consequently motor and respiratory paralysis (Chang and Dong Lee, 1977). Paralysis of the hind legs and temporary rigidity of skeletal muscle is provoked by the neurotoxin crotamin (Giglio, 1975; Vital Brazil et al., 1979). When injected 20 s after 2 LD50 of crude venom, fractions PI and PVII neutralized the progress of the generalized paralysis that caused respiratory arrest in 100% of the animals previously injected with the venom. However, it did not inhibit the paralysis of the hind legs, but this paralysis, which is induced by pure crotamine, receded during the observation period (96 h). The animals showed dyspnea during the first hours, but this symptom also ceased during the observation period. Fraction PVII also neutralized the progress of generalized paralysis caused by crotoxin. Extracts and components of other species of medicinal plants similarly neutralize or inhibit the lethality of C. d. terrificus venom. However, neutralization of the signs of progressive paralysis has not been recorded (Mors et al., 1989; Ferreira et al., 1992). The extract and fractions of these plants can neutralize lethality, but do not fully inhibit these signs. The alkaloid MMV isolated from T. catharinensis (Pereira et al., 1999), when injected i.m. at 1.7 mgy100 g raty50 ml ss 20 s after the administration of 2 LD50 (240 mgy100 g raty50 ml ss) of C. d. terrificus venom, was able to inhibit the lethal activity of the latter (Batina et al., 2000). By examining only the efficiency of each fraction in inhibiting lethality when administered 20 s after the venom, we observed that fraction PVII and the alkaloid MMV were equally effective and significantly more effective than the extracts studied by Batina et al. (1997), being 5 and 10 times more active, respectively. Fraction PVII also had an advantage over the MMV component because it was soluble in saline solution, the ideal solvent for in vivo experiments. Since this fraction apparently contains, but not pure MMV, its solubility in saline may be due to an associated anionic component able to promote it. The evaluation of the antitumoral activity of the different chromatographic fractions demonstrated that fractions PI, PVI and PVIII presented a high cytotoxic activity at 5.0 mgyml. Additionally, fractions PVI and PVIII had significant cytotoxic activity
26
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27
at the concentration of 2.5 mgyml, a fact that was not observed for fraction PI. However, the AE was more active, suggesting that its antitumoral effect may be due to a possible synergistic effect of cytotoxic substances predominantly present in PI, PVI and PVIII (Table 2), but not in fraction PVII which presented potent antivenom activity (Fig. 3). Fractions PVI and PVIII, which presented potent antitumoral activity in vitro, were stained by the Dragendorff reagent, suggesting that the cytotoxic substances are nitrogenated. Braga and Ruis (1987) identified the existence of ternary and quaternary alkaloids in P. fuchsiaefolia extracts. Additionally, Pereira et al. (1999) identified the presence of bis-indole alkaloids in T. catharinensis extracts. Thus, we may postulate that the cytotoxic activity of T. catharinensis AE against tumor cell lines observed in the present study and others (Meyer et al., 1997) may be due in part to alkaloid substances other than MMV (Batina et al., 2000). This hypothesis is supported by the fact that Kingston (1978) demonstrated the cytotoxic activity of bis-indole alkaloids of the voacamine type from various species of the genus Tabernaemontana on tumor cell lines such as P-388 lymphocytic anemia and Eagle nasopharynx carcinoma. However, only the evaluation of the antitumoral activity of the substances isolated could confirm this hypothesis. Taken together, these results clearly demonstrate that the antivenom and antitumoral activities of T. catharinensis AE are exerted by distinct substances respectively, present in fraction PVII and in fractions PVI and PVIII which present different characteristics in terms of staining and Rf when analyzed by TLC. The results also show that fractionation by gel filtration on Sephadex G-10 is a good start step for the isolation of biologically active substances present in the AE of T. catharinensis. Acknowledgments The authors are indebted to Capes, CNPq and Fapesp for financial support. References ´ ¸ C.C.F., 1992. Estudo antropologico Batina, M.F.C., Penco, e ´ ˆ ecologicoybotanico do uso de plantas na medicina popular
da cidade de Assis-SP. In: Simposio ´ de Plantas Medicinais do Brasil, 12. Resumos. Curitiba p.223. Batina, M.F.C, Giglio, J.R., Sampaio, S.V., 1997. Methodological care in the evaluation of the LD50 and the neutralization of the lethal effect of Crotalus durissus terrificus venom by the plant Peschiera fuchsiaefolia (Apocynaceae). J. Venom. Anim. Toxins 3, 23–29. Batina, M.F.C., Cintra, A.C.O., Veronese, E.L.G., Lavrador, M.A.S., Giglio, J.R., Pereira, P.S., et al., 2000. Inhibition of the lethal and myotoxic activities of Crotalus durissus terrificus venom by Tabernaemontana catharinensis A. DC. (Apocynaceae). Identification of one the active components. Planta Medica ´ 66, 424–428. Braga, R.M., Ruis, F.A.M., 1987. Quaternary alkaloids from P. fuchsiaefolia. Phytochemistry 26, 833–836. Chang, C.C., Dong Lee, J., 1977. The presynaptic neuromuscular blocking action of taipoxin. A comparison with B-bungarotoxin and crotoxin. Toxicon 15, 571–580. Cherdchu, C., Srisurawat, K., Ratanabangkoon, K., 1978. Snake neurotoxin inhibiting activity found in the extract of Curcuma sp (Zingiberaceae). J. Med. Assoc. Thailand 61, 544–550. Ferreira, L.A.F., Henriques, O.B., Andreoni, A.A.S., Vital, G.R.F., Campos, M.M.C., Habermehl, G.G., et al., 1992. Antivenom and biological effects of ar-turmerone isolated from Curcuma longa (Zingiberaceae). Toxicon 30, 1211–1218. Finney, D.J., 1964. Statistical Method in Biological Assay. 2nd ed. New York, Hafner Publishing Company. Giglio, J.R., 1975. Analytical studies on crotamine hydrochloride. Anal. Biochem. 69, 207–221. Itzhaki, R.F., Gill, D.M., 1964. A micro-biuret method for estimating proteins. Anal. Biochem. 9, 401–410. Kingston, D.G.I., 1978. Plant anticancer agents VII: Structural effects on cytotoxicity of bisindole alkaloids of voacamine type. J. Pharmacol. Sci. 67, 272–274. Laure, C.J., 1975. Die Primarstruktur ¨ des Crotamins. HoppeSeyler’s Z. Physiol. Chem. 365, 213–215. Meyer, A.V., Dias, D.A., Alencar, R.E., Barbuto, J.A.M., Nomizo, A., 1997. Antineoplastic activity of aqueous extracts of Peschiera fucsiaefolia, Gomphrena globosa, Alternantera brasiliana on human tumor cell lines in vitro. Bollettino Chimico Farmaceutico 136, 82 abstract. Mors, W.B., Nascimento, M.C., Parente, J.P., Silva, M.H., Melo, P.A., Suarez-Kurtz, G., 1989. Neutralization of lethal and myotoxic activities of South American rattlesnake venom by extracts and constituents of plant Eclipta prostrata (Asteraceae). Toxicon 27, 1003–1009. Mosmann, T., 1983. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55–63. Nakagawa, M., Nakanishi, K., 1982. Structures of cabenegrins A-I and A-II, potent anti-snake venoms. Tetrahedron Lett. 23, 3855–3858. ¸ S.C., Dias, D.A., 1999. Pereira, P.S., Sampaio, S.V., Franca, Indole alkaloids from Tabernaemontana catharinensis. Acta Horticulturae 501, 171–176.
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27 Rizzini, C.T., Mors, W.B., Pereira, N.A., 1988. Plantas brasi¸ leiras tidas como ativas contra peconhas animais, especialmente venenos de cobras. Rev. Bras. Farm. 69, 82–86. Vital Brazil, O., Prado-Franceschi, J., Laure, C.J., 1979. Repetitive muscle responses induced by crotamine. Toxicon 17, 61–65.
27
´ ´ Vital Brazil, O., 1980. Venenos ofıdicos neurotoxicos. Rev. Assoc. Med. Bras. 26, 212–218. Wagner, H., Bladt, S., Zgainski, E.M., 1984. Plant Drug Analysis. New York, Springer-Verlag, pp. 320.
Anticrotalic and antitumoral activities of gel filtration fractions of aqueous extract from Tabernaemontana catharinensis (Apocynaceae) Lucilene de Almeidaa, Adelia C.O. Cintraa, Elen L.G. Veronesea, Auro Nomizoa, ˜ Jose´ Francoa, Eliane C. Arantesb, Jose´ Roberto Giglioc, Suely Vilela Sampaioa,* Joao a
´ ´ ´ ´ ˆ ˆ ˜ Preto–USP, Departamento de Analises Clınicas, Toxicologicas e Bromatologicas, Faculdade de Ciencias Farmaceuticas de Ribeirao ´ syn, 14040-903 Ribeirao ˜ Preto, SP, Brazil Avenida do Cafe, b ´ ´ ˆ ˆ ˜ Preto, Universidade de Sao ˜ Paulo, Departamento de Fısica e Quımica, Faculdade de Ciencias Farmaceuticas de Ribeirao ˜ Preto, SP, Brazil 14040-903 Ribeirao c ´ ˜ Preto, Universidade de Sao ˜ Paulo, Departamento de Bioquımica e Imunologia, Faculdade de Medicina de Ribeirao ˜ Preto, SP, Brazil 14049-900 Ribeirao Received 16 March 2003; received in revised form 21 October 2003; accepted 28 October 2003
Abstract The high mortality caused by Crotalus durissus terrificus snake venom is mainly due to crotoxin, which acts on the neuromuscular junction inhibiting the mechanism mediating acetylcholine release, thus leading to motor and respiratory paralysis and subsequently to animal death. We recently demonstrated that the aqueous extract (AE) of Tabernaemontana catharinensis can inhibit the lethal activity of C. d. terrificus venom. Eight fractions, PI to PVIII , were obtained by gel filtration of the extract on Sephadex G-10, and assayed for lethality and cytotoxicity. Fraction PVII w2.0 mgy100 g raty 50 ml saline solution (ss)x injected intramuscularly (i.m.) 20 s after the venom (240 mg) or crotoxin (200 mgy50 ml ss) neutralized the lethal activity of 2 LD50 of both. Fractions PI , PVI and PVIII (5.0 mgy100 g raty50 ml ss) presented potent antitumoral activity in vitro against cells from human breast carcinoma (SK-BR–3) after 24 h incubation, as measured by Mosmann colorimetric method. Fraction PVII contains 12-methoxy-4-methylvoachalotine as its major component. These results demonstrate that the antivenom and antitumoral activities of the AE of T. catharinensis are exerted by different substances present in fraction PVII and fractions PI , PVI and PVIII , respectively, whose characteristics are distinct in terms of staining and Rf when analyzed by thin layer chromatography. The results also show that a preliminary fractionation by Sephadex G-10 gel filtration is a good option as a first step for isolation of biologically active substances from T. catharinensis. 䊚 2003 Elsevier Inc. All rights reserved. Keywords: Anticrotalic activity; Antitumoral activity; Crotalus durissus terrificus venom; Crotoxin; Tabernaemontana catharinensis
1. Introduction The search for antidotes against snakebites obtained from several plants used in Brazilian folk *Corresponding author. Tel.: q55-16-6024287; fax: q5516-6331092. E-mail address: [email protected] (S.V. Sampaio).
medicine started with the species Curcuma sp. and with the aqueous-ethanolic extract from the root ¸ de negro’ (black man’s head) originatof ‘cabeca ing from northeastern Brazil. The ar-turmerone fraction isolated from the rhizome of Curcuma sp. inhibited the action of a neurotoxin from Naja naja siamesis venom (Cherdchu et al., 1978). Two
1532-0456/04/$ - see front matter 䊚 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.cca.2003.10.012
20
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27
components, cabenegrins A-I and A-II (pterocar¸ pans), in the aqueous-ethanolic extract of ‘cabeca de negro’ neutralized the toxic cardiovascular effect of the Bothrops atrox venom (Nakagawa and Nakanishi, 1982). Later, Rizzini et al. (1988) enumerated 83 plants species belonging to 34 families used in Brazilian folk medicine against animal venoms, especially snake venoms. Among these families is the Apocynaceae, which contains several species. Tabernaemontana catharinensis (Apocynaceae), popularly known as ‘leiteiro de vaca’ (cow’s dogbane), occurs in Argentina, Paraguay, Bolivia and southern Brazil. Studies conducted by Batina et al. (1997, 2000) demonstrated that the fresh and stabilized aqueous extracts of the root of Peschiera fuchsiaefolia, today reclassified as T. catharinensis, can inhibit the lethal and myotoxic activity of Crotalus durissus terrificus venom (South American Rattlesnake, cascavel). Batina et al. (2000) isolated from T. catharinensis a quaternary base alkaloid, 12-methoxy-4-methylvoachalotine (MMV), which proved to be able to inhibit the lethal activity of 2 LD50 of C. d. terrificus venom. Substances with trypanocidal and antitumoral activity were also isolated from this plant (Pereira et al., 1999). In addition, the aqueous extract of T. catharinensis was shown to have a potent cytotoxic action on human tumor lines such as SK-BR-3, MCF-7 and C-8161 in vitro (Meyer et al., 1997). On the basis of the results obtained with the species T. catharinensis and due to its diversity of alkaloids with pharmacological activity, the objective of the present study was to fractionate the aqueous extract (AE) of T. catharinensis and evaluate the action of the fractions obtained in terms of anticrotalic and antitumoral activity.
and voucher specimens were deposited in the herbarium (SPFR) of Faculdade de Filosofia, ˆ ˜ Preto, USP (Reg. n8 Ciencias e Letras de Ribeirao 02940). 2.2. Preparation and fractionation AEs were prepared as follows: after identification the plant root bark was removed, washed, ground in distilled water (5 gy10 ml water) and kept at 4 8C for 12 h, strained through a nylon fabric, kept at 4 8C for an additional 12 h, and then filtered under reduced pressure. The AE was lyophilized and stored dry at y10 8C. 2.3. Fractionation of T. catharinensis AE One gram of lyophilized T. catharinensis AE was dissolved in 10 ml 0.3 M ammonium bicarbonate buffer (AMBIC), pH 8.0, and centrifuged at 12 100=g for 10 min at 4 8C. An aliquot of the supernatant was removed for the determination of absorbance at 280 nm. The remainder of the supernatant was applied on a Sephadex G-10 column (2.0=99.0 cm) and eluted with the same buffer. Five-milliliter fractions were collected at a flow rate of 20.0 mlyh at room temperature. Absorbance at 280 nm was determined for each collected fraction and the fractions were pooled and lyophilized. This gel filtration was performed several times in order to produce enough material for subsequent experiments. 2.4. Fractionation of C. d. terrificus venom Crotoxin, the major constituent of C. d. terrificus venom, was obtained as described by Laure (1975).
2. Materials and methods
2.5. Determination of crotoxin i.m. LD50 in rats
2.1. Venom and plant material
Crotoxin was dissolved in 0.9% (myv) saline solution and clarified by centrifugation at 800=g for 5 min at room temperature. An aliquot of the supernatant was used for determination of protein concentration (Itzhaki and Gill, 1964). The solution was then diluted at a 1.15 geometric ratio. Rats were divided at random into groups of six. Each animal received two injections in the right hind thigh with a 20 s interval, for a final volume of 100 ml. The animals were then examined over a 24 h period and dead and alive animals were
C. d. terrificus venom was purchased from ˜ Paulo, SP. Albino male Instituto Butantan, Sao Wistar rats weighing 90–110 g were used. The animals were kept at 24"1 8C and fed ad libitum with commercial ration and water. All animals were carefully monitored and maintained in accordance with ethical recommendations of the Ethic Commission for the Use of Animals (CEUA). T. catharinensis A. DC. was collected in Assis, SP,
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27
counted. Rats were also examined for signs of hind leg paralysis, generalized paralysis, body rigidity and dyspnea. The LD50 was determined by the method of Finney (1964) using Graphpad Prism software. 2.6. Evaluation of the action of the fractions of T. catharinensis AE against the lethal activity of C. d. terrificus crude venom and crotoxin The lethality caused by 2 LD50 was used as a parameter to evaluate the effects of the total extract or fraction on C. d. terrificus envenomation. We adopted a protocol, which simulated an ophidian accident and followed the progress of local and systemic effects in animals that received only venom or toxin compared with those that, in addition, received the extract or fraction from T. catharinensis 20 s after. C. d. terrificus venom and crotoxin were dissolved in ss and clarified by centrifugation at 800=g for 5 min at room temperature. Protein concentration was determined in an aliquot of each supernatant. The dose of crude venom was fixed at 240 mg proteiny100 g raty50 ml ss, corresponding to 2 LD50 (Batina et al., 1997, 2000) and that of crotoxin was fixed at 200 mg crotoxiny100 g raty50 ml ss (also 2 LD50). Increasing doses of the fractions of T. catharinensis AE were administered. The fractions were dispersed in ss and clarified by centrifugation at 800=g for 5 min. The supernatant was separated and kept at room temperature until use. For biological assays the rats were divided at random into groups of six animals. Each animal received two i.m. injections, 20 s apart, in the right hind thigh. Rats were then submitted to the treatments described below. 2.7. Treatment with the T. catharinensis fractions 20 s after C. d. terrificus venom In group 1, rats received two injections of 50 ml ss and, in group 2, they received 2 LD50 of the crude venomy50 ml ss and then 50 ml ss. In the remaining groups, the animals received 2 LD50 of C. d. terrificus venom and increasing doses of the fractions of T. catharinensis AE, starting with 1.0 mgy100 g raty50 ml ss. A control was also prepared for each AE fraction (5.0 mgy100 g raty 50 ml ss). The animals were observed for 96 h, at
21
2 h intervals up to 12 h and at 24 h intervals thereafter. 2.8. Treatment with fraction PVII 20 s after crotoxin In group 1, the animals received two injections of 50 ml ss each. In group 2, the animals received 2 LD50 of crotoxin and 50 ml ss. In group 3, the animals received 50 ml ss and 4.0 mg of fraction PVII y100 g raty50 ml ss. In the remaining groups, the animals received 2 LD50 of crotoxin and increasing doses of fraction PVII (1.0–4.0 mgy100 g raty50 ml ss). The rats were then examined as described above. 2.9. Antitumoral activity For evaluation of the antitumoral activity we used the human breast adenocarcinoma line SKBR-3, obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA). The tumor cells were first expanded in cell culture flasks in complete RPMI medium (15 mlyflaskRPMI 1640 supplemented with 10% fetal calf serum, 20 mM L-glutamine, 7.5% Na2HCO3 and 10 mgyml gentamicin) at 37 8C in a humidified incubator containing 95% air and 5% CO2. Lyophilized fractions of T. catharinensis were first dissolved in complete RPMI medium, vortexed and then sterilized by filtration through 0.22 mm membranes. Different concentrations of the AE fractions were added to 96-well cell culture plates containing 5=105 SK-BR-3 cells and incubated for 24 h. The cytotoxic activity of the fractions of T. catharinensis AE against the tumor line was evaluated by the colorimetric method described by Mosmann (1983). 2.10. Analysis by thin layer chromatography (TLC) Fractions of T. catharinensis AE and the MMV alkaloid were analyzed by TLC on silica gel 60 F254 (Aldrich) plates using chloroform: methanol (9:1, vyv) and butanol: acetic acid: water (4:1:5, vyv) as the organic phases. Visualization of chromatographic bands was achieved by reaction with sulfuric vanillin and Dragendorff reagent (Wagner et al., 1984) and observation under UV light. 3. Results Fig. 1 shows the gel filtration profile of T. catharinensis AE on Sephadex G-10, which
22
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27
revealed eight fractions, PI to P VIII. The recovery in terms of weight for each fraction is showed in Table 1. LD50 of crotoxin was approximately 100 mgy100 g raty50 ml ss. In TLC, fractions PVI, PVII,PVIII and MMV showed an orange color when developed with the Dragendorff reagent and only fraction PVII and MMV showed a bluish color when developed with sulfuric vanillin. The Rf values for these compounds were 0.35 (PVI), 0.46 (PVII), 0.45 (PVIII) and 0.46 (MMV), respectively. An additional orange spot (Rfs0.23) was also present in fraction PVI (Fig. 2). Two LD50 of crude C. d. terrificus venom paralyzed the rats’ hind legs within 10 min. Generalized paralysis was observed within 30 min in all animals. Within 4 h, 83% of the animals manifested severe dyspnea, apnea and death. Muscle spasms were also observed. T. catharinensis AE fractions had no apparent effect on the control animals, all survived the experiments. Fraction PVII, 2.0 mgy100 g raty 50 ml ss, injected i.m. 20 s after the venom, completely neutralized the lethal action of 2 LD50 of the venom. Doses above 2.0 mg did not change the rate of inhibition of the lethality induced by 2 LD50, as shown in Fig. 3A. Fraction PVII did not inhibit hind leg paralysis, a characteristic effect of
Table 1 Recovery of fractions from T. catharinensis aqueous extract (AE) after gel filtration on Sephadex G-10 (Fig. 1) Fractions
% mass recovery
AE PI PII PIII PIV PV PVI PVII PVIII Total
100.0 3.6 30.7 41.9 2.4 0.7 0.8 7.6 0.3 88.0
crotamine from the crotalic venom, which disappeared within the next 48 h. The animals presented moderate dyspnea, which disappeared within 72 h. Fraction PVII, 2.0 mgy100 g raty50 ml ss, injected i.m. 20 s after crotoxin, completely abolished the lethality of 2 LD50 of crotoxin (Fig. 3b). During the first 4 h, the animals presented moderate dyspnea and generalized paralysis, which receded within the next 48 h. Several experiments gave consistent results, with S.D.s0. Fraction PI, 2.5 and 5.0 mgy100 g raty50 ml ss, inhibited the lethality of the venom by 50%. During the first 120 min, 50% of the animals presented paralysis of the hind legs and moderate
Fig. 1. Gel filtration profile of T. catharinensis aqueous extract (AE) on a Sephadex G-10 column (2.0=99.0 cm). 1 g sample of AE dissolved in 10 ml 0.3 M AMBIC, pH 8.0, was applied and eluted with the same buffer. Flow rate was 20 mlyh and 5 mlytube were collected at 25 8C.
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27
23
Fig. 2. Thin layer chromatography of the quaternary base 12 methoxy-4-methylvoachalotine (MMV) and fractions PI to PVIII from T. catharinensis. Mobile phase was: n-butanol: acetic acid: water (BAW 4:1:5). Test reagents were: (a) Dragendorff reagent and (b) sulfuric vanillin. Samples: 1, 2, 3: AE; 4:PI; 5: PII; 6: PIII; 7: PIV; 8: PV; 9: PVI; 10: PVII; 11: PVIII; 12: MMV.
dyspnea. After 8 h, dyspnea increased and the animals returned to normal within 72 h. Fractions PII to PV did not inhibit the lethality of the venom at the doses assayed. Fractions PVI and PVIII were not assayed because their yield after gel filtration on Sephadex G-10 was very low, 0.8% and 0.3%, respectively (Table 1). The LD50 (i.m., 24 h) of crotoxin was approximately 100 mgy100 g raty50 ml ss (92.31–107.92 mg proteiny100 g raty50 ml ss). 2 LD50 caused marked dyspnea during the first 2 h and generalized paralysis within 4 h. After 12 h, dyspnea increased drastically in 52.4% of the rats, causing death. In animals that did not die, the generalized paralysis and dyspnea receded during the first 96 h. AE and fractions PI, PVI and PVIII had potent antitumoral activity against cells from human breast carcinoma (SK-BR-3) at concentrations of 5 mgyml. In addition, fractions PVI and PVIII and the AE also had antitumoral activity at 2.5 mgyml (Table 2). These results are comparable to those obtained with the methotrexate control, a drug used for the clinical treatment of breast adenocarcinoma.
4. Discussion The usual phytochemical methods for micromolecule purification are cumbersome, involving several steps of extraction with various solvents of different polarity, and are not always efficient. Changes of solvents or a mixture of solvents, or even changes of the stationary phase are necessary for the purification of these micromolecules, but the use of these solvents hampers the solubilization of the fractions, possibly also causing injury to the animals and inactivation of the venom or of the fraction intended for study. In the present study, a single step gel filtration on Sephadex G-10 provided a partial resolution of the T. catharinensis AE into eight fractions, PI to PVIII, soluble in distilled water and in saline solution, four of which demonstrated anticrotalic or antitumoral activity. Since biologically active components were detected in some of these still heterogeneous fractions, this mild, simple and high yield classical procedure can be used as a first step of a more elaborate methodology for the ultimate isolation of specific active components. In PVII, it was possible to identify the quaternary
24
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27
Fig. 3. Percent survival. (a) Percent survival of rats that received 2 LD50 of C. d. terrificus venom (i.m.) and different doses of fraction PVII (mgy100 g raty50 ml ss) from T. catharinensis 20 s later. Each group contained six animals. Statistically identical groups were followed by identical letters (P-0.05). (b) Percent survival of rats that received 2 LD50 of crotoxin (i.m.) and different doses of fraction PVII (mgy100 g raty50 ml ss) of T. catharinensis 20 s later. Each group contained six animals. The dots represent the mean"S.E.M. for the groups. Statistically identical groups are followed by identical letters (P-0.05).
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27 Table 2 Percent cytotoxicity of the different fractions of T. catharinensis AE against SK-BR-3 cells Plant extract or fraction
Aqueous extract (AE) PI PII PIII PIV PV PVI PVII PVIII Methotrexate
Cytotoxicity activity on SKBR-3 tumor cells (%) 2.5 mgyml
5.0 mgyml
69.0"1.3 37.3"1.2* 33.9"0.9* 0* 0* 19.6"2.6* 50.4"0.5 28.1"0.8* 50.5"0.2 54.3"1.4
82.4"0.5 66.2"1.1 46.5"0.7* 0* 0* 25.8"1.3* 64.0"0.4 45.8"0.7* 68.1"1.6 64.8"0.8
% Cytotoxicity – Cytotoxicity of the fractions was analyzed by the colorimetric method described by Mosmann (1983). Results are shown as means"S.D. Statistical differences were analyzed by Mann–Whitney non-parametric test. Values of P0.005* were considered significantly different form the control (aqueous extract).
alkaloid MMV, previously isolated by Pereira et al. (1999) and Batina et al. (2000). The results of TLC (Fig. 2) showed that fractions PVI, PVII and PVIII and the MMV component stained orange with the Dragendorff reagent, a characteristic of nitrogenated substances such as alkaloids. Fraction PVII and MMV were also stained by sulfuric vanillin, a fact also indicating the presence of an alkaloid (Fig. 2b). However, fractions PVI, PVII and PVIII showed different Rf and staining, indicating that they are not the same substances. Staining and Rf of fraction PVII fit with those of MMV. Therefore we suggest that fraction PVII contains MMV as the major component. Fraction PI was not stained by the reagents used (Fig. 2). Its structure is unknown and should be investigated. Thus, fractions PI, PVI, PVII and PVIII are distinct, presenting different staining, Rf and elution volumes when submitted to gel filtration on Sephadex G-10 (Fig. 1). The i.m. route was chosen for the administration of C. d. terrificus venom and crotoxin because it simulates the inoculation of the venom by the snake, and was also chosen for the administration of the fractions of T. catharinensis AE because it is closer to the bandage application in popular use ¸ 1992). (Batina and Penco, The symptoms observed in the animals injected
25
with 2 LD50 of C. d. terrificus have been described by Vital Brazil (1980). Crotoxin, a presynaptic neurotoxin, inhibits the release of acetylcholine, whose absence causes neuromuscular blockade and consequently motor and respiratory paralysis (Chang and Dong Lee, 1977). Paralysis of the hind legs and temporary rigidity of skeletal muscle is provoked by the neurotoxin crotamin (Giglio, 1975; Vital Brazil et al., 1979). When injected 20 s after 2 LD50 of crude venom, fractions PI and PVII neutralized the progress of the generalized paralysis that caused respiratory arrest in 100% of the animals previously injected with the venom. However, it did not inhibit the paralysis of the hind legs, but this paralysis, which is induced by pure crotamine, receded during the observation period (96 h). The animals showed dyspnea during the first hours, but this symptom also ceased during the observation period. Fraction PVII also neutralized the progress of generalized paralysis caused by crotoxin. Extracts and components of other species of medicinal plants similarly neutralize or inhibit the lethality of C. d. terrificus venom. However, neutralization of the signs of progressive paralysis has not been recorded (Mors et al., 1989; Ferreira et al., 1992). The extract and fractions of these plants can neutralize lethality, but do not fully inhibit these signs. The alkaloid MMV isolated from T. catharinensis (Pereira et al., 1999), when injected i.m. at 1.7 mgy100 g raty50 ml ss 20 s after the administration of 2 LD50 (240 mgy100 g raty50 ml ss) of C. d. terrificus venom, was able to inhibit the lethal activity of the latter (Batina et al., 2000). By examining only the efficiency of each fraction in inhibiting lethality when administered 20 s after the venom, we observed that fraction PVII and the alkaloid MMV were equally effective and significantly more effective than the extracts studied by Batina et al. (1997), being 5 and 10 times more active, respectively. Fraction PVII also had an advantage over the MMV component because it was soluble in saline solution, the ideal solvent for in vivo experiments. Since this fraction apparently contains, but not pure MMV, its solubility in saline may be due to an associated anionic component able to promote it. The evaluation of the antitumoral activity of the different chromatographic fractions demonstrated that fractions PI, PVI and PVIII presented a high cytotoxic activity at 5.0 mgyml. Additionally, fractions PVI and PVIII had significant cytotoxic activity
26
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27
at the concentration of 2.5 mgyml, a fact that was not observed for fraction PI. However, the AE was more active, suggesting that its antitumoral effect may be due to a possible synergistic effect of cytotoxic substances predominantly present in PI, PVI and PVIII (Table 2), but not in fraction PVII which presented potent antivenom activity (Fig. 3). Fractions PVI and PVIII, which presented potent antitumoral activity in vitro, were stained by the Dragendorff reagent, suggesting that the cytotoxic substances are nitrogenated. Braga and Ruis (1987) identified the existence of ternary and quaternary alkaloids in P. fuchsiaefolia extracts. Additionally, Pereira et al. (1999) identified the presence of bis-indole alkaloids in T. catharinensis extracts. Thus, we may postulate that the cytotoxic activity of T. catharinensis AE against tumor cell lines observed in the present study and others (Meyer et al., 1997) may be due in part to alkaloid substances other than MMV (Batina et al., 2000). This hypothesis is supported by the fact that Kingston (1978) demonstrated the cytotoxic activity of bis-indole alkaloids of the voacamine type from various species of the genus Tabernaemontana on tumor cell lines such as P-388 lymphocytic anemia and Eagle nasopharynx carcinoma. However, only the evaluation of the antitumoral activity of the substances isolated could confirm this hypothesis. Taken together, these results clearly demonstrate that the antivenom and antitumoral activities of T. catharinensis AE are exerted by distinct substances respectively, present in fraction PVII and in fractions PVI and PVIII which present different characteristics in terms of staining and Rf when analyzed by TLC. The results also show that fractionation by gel filtration on Sephadex G-10 is a good start step for the isolation of biologically active substances present in the AE of T. catharinensis. Acknowledgments The authors are indebted to Capes, CNPq and Fapesp for financial support. References ´ ¸ C.C.F., 1992. Estudo antropologico Batina, M.F.C., Penco, e ´ ˆ ecologicoybotanico do uso de plantas na medicina popular
da cidade de Assis-SP. In: Simposio ´ de Plantas Medicinais do Brasil, 12. Resumos. Curitiba p.223. Batina, M.F.C, Giglio, J.R., Sampaio, S.V., 1997. Methodological care in the evaluation of the LD50 and the neutralization of the lethal effect of Crotalus durissus terrificus venom by the plant Peschiera fuchsiaefolia (Apocynaceae). J. Venom. Anim. Toxins 3, 23–29. Batina, M.F.C., Cintra, A.C.O., Veronese, E.L.G., Lavrador, M.A.S., Giglio, J.R., Pereira, P.S., et al., 2000. Inhibition of the lethal and myotoxic activities of Crotalus durissus terrificus venom by Tabernaemontana catharinensis A. DC. (Apocynaceae). Identification of one the active components. Planta Medica ´ 66, 424–428. Braga, R.M., Ruis, F.A.M., 1987. Quaternary alkaloids from P. fuchsiaefolia. Phytochemistry 26, 833–836. Chang, C.C., Dong Lee, J., 1977. The presynaptic neuromuscular blocking action of taipoxin. A comparison with B-bungarotoxin and crotoxin. Toxicon 15, 571–580. Cherdchu, C., Srisurawat, K., Ratanabangkoon, K., 1978. Snake neurotoxin inhibiting activity found in the extract of Curcuma sp (Zingiberaceae). J. Med. Assoc. Thailand 61, 544–550. Ferreira, L.A.F., Henriques, O.B., Andreoni, A.A.S., Vital, G.R.F., Campos, M.M.C., Habermehl, G.G., et al., 1992. Antivenom and biological effects of ar-turmerone isolated from Curcuma longa (Zingiberaceae). Toxicon 30, 1211–1218. Finney, D.J., 1964. Statistical Method in Biological Assay. 2nd ed. New York, Hafner Publishing Company. Giglio, J.R., 1975. Analytical studies on crotamine hydrochloride. Anal. Biochem. 69, 207–221. Itzhaki, R.F., Gill, D.M., 1964. A micro-biuret method for estimating proteins. Anal. Biochem. 9, 401–410. Kingston, D.G.I., 1978. Plant anticancer agents VII: Structural effects on cytotoxicity of bisindole alkaloids of voacamine type. J. Pharmacol. Sci. 67, 272–274. Laure, C.J., 1975. Die Primarstruktur ¨ des Crotamins. HoppeSeyler’s Z. Physiol. Chem. 365, 213–215. Meyer, A.V., Dias, D.A., Alencar, R.E., Barbuto, J.A.M., Nomizo, A., 1997. Antineoplastic activity of aqueous extracts of Peschiera fucsiaefolia, Gomphrena globosa, Alternantera brasiliana on human tumor cell lines in vitro. Bollettino Chimico Farmaceutico 136, 82 abstract. Mors, W.B., Nascimento, M.C., Parente, J.P., Silva, M.H., Melo, P.A., Suarez-Kurtz, G., 1989. Neutralization of lethal and myotoxic activities of South American rattlesnake venom by extracts and constituents of plant Eclipta prostrata (Asteraceae). Toxicon 27, 1003–1009. Mosmann, T., 1983. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55–63. Nakagawa, M., Nakanishi, K., 1982. Structures of cabenegrins A-I and A-II, potent anti-snake venoms. Tetrahedron Lett. 23, 3855–3858. ¸ S.C., Dias, D.A., 1999. Pereira, P.S., Sampaio, S.V., Franca, Indole alkaloids from Tabernaemontana catharinensis. Acta Horticulturae 501, 171–176.
L. de Almeida et al. / Comparative Biochemistry and Physiology Part C 137 (2004) 19–27 Rizzini, C.T., Mors, W.B., Pereira, N.A., 1988. Plantas brasi¸ leiras tidas como ativas contra peconhas animais, especialmente venenos de cobras. Rev. Bras. Farm. 69, 82–86. Vital Brazil, O., Prado-Franceschi, J., Laure, C.J., 1979. Repetitive muscle responses induced by crotamine. Toxicon 17, 61–65.
27
´ ´ Vital Brazil, O., 1980. Venenos ofıdicos neurotoxicos. Rev. Assoc. Med. Bras. 26, 212–218. Wagner, H., Bladt, S., Zgainski, E.M., 1984. Plant Drug Analysis. New York, Springer-Verlag, pp. 320.