Mem Inst Oswaldo Cruz, Rio de Janeiro, 95(3) May/Jun 2000
Biological Screening of Brazilian Medicinal Plants
Laboratório de Química de Produtos Naturais, Centro de Pesquisas René Rachou-Fiocruz, Av. Augusto de Lima 1715, 30190-002 Belo Horizonte, MG, Brasil
IEmpresa de Pesquisa Agropecuária de Minas Gerais, Belo Horizonte, MG, Brasil
IIFundação Zoobotânica de Belo Horizonte, Belo Horizonte, MG, Brasil
IIIDepartamento de Microbiologia e Parasitologia, UFSC, Florianópolis, SC, Brasil
In this study, we screened sixty medicinal plant species from the Brazilian savanna ("cerrado") that could contain useful compounds for the control of tropical diseases. The plant selection was based on existing ethnobotanic information and interviews with local healers. Plant extracts were screened for: (a) molluscicidal activity againstu00a0Biomphalaria glabrata, (b) toxicity to brine shrimpu00a0(Artemia salina L.), (c) antifungal activity in the bioautographic assay withu00a0Cladosporium sphaerospermumu00a0and (d) antibacterial activity in the agar diffusion assay againstu00a0Staphylococcus aureus,u00a0Escherichia coli,u00a0Bacillus cereusu00a0andu00a0Pseudomonas aeruginosa. Forty-two species afforded extracts that showed some degree of activity in one or more of these bioassays.
The Brazilian savanna, known as "cerrado", comprises a very rich and characteristic flora (Burman 1991) that covers more than 2 million square kilometers of Brazilian inland (Ferri 1973). Many of these plants are used as natural medicines by the people living in the "cerrado" area to treat several tropical diseases including schistosomiasis, leishmaniasis, malaria, fungal and bacterial infections, among others (Ferreira 1980, Corrêa 1984, Grandi et al. 1989, Di Stasi 1989, Hirschmann & Arias 1990, Brandão 1991, Caribé & Campos 1991, Martins et al. 1994, Matos 1994).
Schistosomiasis, caused by the parasite Schistosoma mansoni, is an important endemic disease in Brazil and in many other tropical countries (Davis 1996). The life cycle of this parasite involves an intermediate host, represented in Brazil by snails of the genus Biomphalaria. Thus, besides chemotherapy of infected people, one of the strategies to combat this disease is to interrupt the parasite's life cycle in endemic areas via control of the snail's population. The search for molluscicidal compounds derived from renewable parts of plants that grow easily in endemic areas could improve the access of poor communities to molluscicidal agents to treat their water collections. This could enhance their chances to control schistosomiasis (Mott 1987).
Fungi and bacteria cause important human diseases, especially in tropical regions and in immunocompromised or immunodeficient patients. Despite the existence of potent antibiotic and antifungal agents, resistant or multi-resistant strains are continuously appearing, imposing the need for a permanent search and development of new drugs (Silver & Bostian 1993).
In an effort to discover new lead compounds, many research groups screen plant extracts to detect secondary metabolites with relevant biological activities. In this regard, several simple bioassays were developed for screening purposes (Hostettmann 1991), some of them were used in this screening. Thus, Artemia salina larvae have been used as a target organism to detect bioactive compounds in plant extracts and toxicity to this crustacean has a good correlation with anti-tumor (McLaughlin 1991) and anti-Trypanosoma cruzi (Zani et al. 1995) activities. T. cruzi is a protozoan that causes Chagas disease (American trypanosomiasis), an illness that affects approximately 16 million people in tropical and sub-tropical Americas (WHO 1993). The drugs currently in use against this disease, nifurtimox and benznidazole, present several side effects and have limited efficacy, especially in the chronic phase of the disease.
The aim of this study was to screen for medicinal plant extracts that could be useful for the development of new tools for the control of infectious diseases. While pursuing this goal, we initiated a systematic evaluation of extracts from the "cerrado" plant species in bioassays such as (a) the brine shrimp (Artemia salina Leach) lethality assay (BSLA), (b) the assay with the snails B. glabrata, (c) the bioautographic antifungal assay with spores of Cladosporium sphaerospermum and, (d) the agar diffusion assay using the bacteria Bacillus cereus,Escherichia coli, Pseudomonas aeruginosa and Staphylococcus aureus.
MATERIALS AND METHODS
Plant collection - The plants were collected in the vicinities of Belo Horizonte, Minas Gerais, Brazil. They were identified by M Brandão and TMS Grandi. Exsiccates of each species were deposited in the herbaria (Table I) of the following institutions: Fundação Zoobotânica de Belo Horizonte, Empresa de Pesquisas Agropecuária de Minas Gerais and Universidade Federal de Viçosa.
Extraction - The plant parts were dried at 35°C in a convection oven and powdered in a knife mill. The powder was macerated at room temperature for 24 h in CH2Cl2 followed by MeOH or in a mixture of CH2Cl2_MeOH (1:1, v/v), as indicated in Table II. After filtration, the solvents were removed by rotary evaporation under reduced pressure and at temperatures below 45°C. The yield of the crude extracts are indicated in Table II. In some cases, part of the powder was separated, extracted with H2O, centrifuged and the supernatant freeze-dried. The extracts were kept at -20°C and the residual solvent of an aliquot removed overnight under high vacuum prior to the bioassays.
Molluscicidal activity - The assay with snails B. glabrata was run as described by Alves et al. (1996). Briefly, the crude extracts were dissolved in 500 ml of dechlorinated tap water at an initial concentration of 100 ppm. Ten snails were put into 250 ml of this solution and maintained submerged by a nylon screen during 24 h. After this period, the surviving snails were removed to a flask containing 250 ml of dechlorinated tap water, fed with lettuce, and observed for three days. If all mollusks were killed dilutions were prepared to determine the minimum lethal concentration (LC100/24 h). Controls with and without niclosamide (1 ppm) were run in parallel.
Toxicity to Artemia salina - The protocol established by McLaughlin (1991) was employed. Briefly, each extract (2 mg) was dissolved in 2 ml of solvent. From these solutions, 500, 50 and 5 µl were transferred to vials in triplicate. The solvent was removed under high vacuum and seawater (5 ml) was added to each vial, resulting in final concentrations of 100, 10 and 1 µg/ml, respectively. Second instar larvae of A. salina (ten per vial) were added. After 24 h contact, the survivors were counted and the LC50 calculated using the probit method. Extracts with LC50 £ 100 ppm were considered active. Controls with and without thymol (10 ppm) were run simultaneously.
Antifungal activity - The bioautographic assay with C. sphaerospermum, developed by Homans and Fuchs (1970), was adopted. Briefly, the extracts (1 mg) were dissolved in a volatile solvent (100 µl) and an aliquot (10 µl) of each solution spotted on TLC plates. After complete solvent removal, a spore suspension of the fungus in nutrient medium was sprayed and the TLC plates incubated in a humid atmosphere for three days at room temperature. The appearance of a growth inhibition zone with the size of the spot (active) or larger than the spot (very active) indicated the presence of fungitoxic substances in the extract. Thymol (50 mg/spot) and solvent (10 µl) were used as controls.
Antimicrobial assay - The agar diffusion assay described by Smânia et al. (1995) was adopted. Briefly, B. cereus, E. coli(ATCC 25922), P. aeruginosa (ATCC 27853), and S. aureus (ATCC 25923) were grown in Mueller-Hinton agar and broth (Difco Laboratories). The strains were incubated at 36oC for 18 h, and were diluted to a final concentration of approximately 106 CFU/ml. Each bacterial suspension was spread over the surface of Mueller-Hinton agar containing five wells of 7 mm diameter. The wells were filled with 5 mg of extract dissolved in the medium. The plates were incubated at 36oC for 20 h. Penicillin (30 mg) was used as a positive control. The results were expressed in terms of the diameter of the inhibition zone: < 9 mm, inactive; 9-12 mm, partially active; 13-18 mm, active; >18 mm, very active.
RESULTS AND DISCUSSION
The sixty plant species evaluated in this screening are listed in Table I. They are distributed among 53 genera and 36 families, with Leguminosae being the most represented within the collection. Table II summarizes the results of the bioassays, listing only the species that presented some activity in at least one bioassay.
Eight species (13%) displayed LC100/72h £ 100 ppm in the assay with B. glabrata. Species from genus Byrsonima were especially active and the methanol extract from the leaves of B. intermedia Juss, was the most active among all extracts tested, presenting LC100 of 20 ppm. According to the World Health Organization guidelines (WHO 1993), this extract can be regarded as a promising candidate for a plant-derived molluscicide. The bark of B. intermedia is rich in tannins and its infusion is popularly used as febrifuge (Corrêa 1984). Several classes of compounds have been isolated from this genus, including triterpenes, flavonoids, benzenoids and steroids (Schultes & Raffauf 1990). Extracts from leaves of B. verbascifolia and bark of Hymenaeae stignocarpa Mart showed LC100 of 40 ppm. The bark of the latter species is reputed as a vermifuge (Caribé & Campos 1991).
From the 60 species evaluated in the BSLA, six (10 %) produced extracts that displayed LC50 £ 100 ppm. The CH2Cl2_MeOH extracts from the aerial parts of Leonurus sibiricus L and from the bark of Xylopia aromatica (Lam) Mart were the most active, with LC50 of 12 and 18 ppm, respectively. L. sibiricus, is popularly used in Minas Gerais for cold, diarrhea and digestive complaints (Grandi et al. 1989). Investigations of its chemistry and pharmacological properties by Chinese groups resulted in the isolation of bioactive alkaloids (Luo et al. 1985). Cytotoxic acetogenins, venezinin and asimicin, were isolated from the bark of X. aromatica using a BSLA-guided fractionation protocol (Colman-Saizaritoria et al. 1994). Some species from this genus are rich in kaurenoic acid (Alves et al. 1995), known for its activity in the BSLA and against trypomastigotes of T. cruzi (Zani et al. 1995). Although less potent, extracts from fruits of Cabralea polythrica, leaves of Caryocar brasiliensis and Copaifera langsdorffii and roots of Lantana camara were also active in the BSLA. All these plants are used to treat numerous ailments in Brazil (Ferreira 1980, Martins et al. 1994) and were already subjected to phytochemical studies by other groups. The observed activity of L. camara in the BSLA could be due to the presence of camaraside and lantanoside, cytotoxic components already isolated from this species (Mahato et al. 1994).
When tested against the spores of phytopathogenic fungus C. sphaerospermum, extracts from eight species (13%) were able to inhibit fungal development. The most active were the CH2Cl2 extracts from the roots of Brosimum gaudichaudiiand L. camara. The first is widely used in Brazil for treating skin depigmentation (vitiligo) (Corrêa 1984), an activity attributed to the presence of bergapten and psolaren. These compounds could also account for the observed fungicidal activity of this plant. Previous works showed that the essential oil of L. camara is active against fungi (Mahato et al. 1994) and bacteria (P. aeruginosa and S. aureus) (Ross et al. 1980).
In the agar diffusion assay using four different bacterial strains, more than 60% of plant species afforded extracts with some degree of activity, in particularly against the Gram-positive bacteria S. aureus or B. cereus. On the other hand, only four plant species (Dalbergia nigra, Lafoensia pacari, Miconia albicans, and Plathymenia foliolosa) afforded extracts that were highly active against the more resistant gram negative bacteria P. aeruginosa. Moreover, the extracts fromAustroplenckia populnea, D. nigra, Eugenia pitanga, Emmotum nitens, Miconia albicans, P. foliolosa, Pouteria torta,Roupala montana, Stryphnodendron adstringens and X. aromatica, presented some degree of activity against all bacterial strains, including E. coli. However, only extracts from D. nigra, M. albicans, P. foliolosa, R. montana and S. adstringens caused significant inhibition (13-18 mm inhibition zone) of E. coli growth at the dose used (5 mg/well).
Overall, extracts from 70% of the species collected were active in at least one of the bioassays adopted in this screening. If we consider only the most active extracts (bold face in Table II: bgl £ 40 ppm, asa £ 50 ppm, csp = 2, pae = 3 and eco = 2) and exclude the highly sensitive gram positive bacteria, this proportion is reduced to 23%. This can be considered a very good hit rate, reinforcing the importance of the ethnomedical information in the search of bioactive extracts. Furthermore, many of the highly active extracts presented some specificity against one of the organisms tested and not a generic toxicity. We observed that extraction with CH2Cl2 generated two thirds of the fungicidal extracts while none of the CH2Cl2 extracts was able to kill snails. Although 10% of the aqueous extracts were active against B. glabrata, none of them was fungicidal or toxic to A. salina. Finally, the use of methanol provided the majority of the highly active extracts in the bioassay with bacteria. These observations can be rationalized in terms of the polarity of the compounds being extracted by each solvent and, in addition to their intrinsic bioactivity, by their ability to dissolve or diffuse in the different media used in the assays. The more active extracts were prioritized for the identification of the active components, a work that is already under way.
To the University of Illinois at Chicago for the permission to access the NAPRALERT database. To Larisa Alonso for reviewing the manuscript.
Alves TMA, Chaves PPG, Nagem TJ, Murta SMF, Ceravolo IP, Romanha AJ, Zani CL1995. A diterpene fromMikania obtusata activeon Trypanosoma cruzi. Planta Med 61: 85-86.
Alves TMA, Ribeiro A, Mendes NM, Nagem TJ, Hamburguer M, Hostettmann K, Zani CL 1996. Molluscicidal saponins from Guaiacum officinale L. (Zygophyllaceae). Int J Pharmacog 34: 81-86.
Brandão M 1991. Plantas medicamentosas do cerrado mineiro. Inf Agropec 15: 15-20.
Burman A 1991. Saving Brazil's savannas. New Scientist 1758: 30-34.
Caribé J, Campos JM 1991. Plantas que Ajudam o Homem: Guia Prático para a Época Atual, Pensamento, São Paulo, p. 6
Colman-Saizarbitoria T, Zambrano J, Ferrigni NR, Gu ZM, Smith DL, McLaughlin JL 1994. Bioactive annonaceous acetogenins from the bark of Xylopia aromatica. J Nat Prod 57: 486-493.
Corrêa MP 1984. Dicionário das Plantas Úteis do Brasil e das Exóticas Cultivadas, Ministério da Agricultura, IBDF, Rio de Janeiro, Vols I-VI.
Davis A 1996. Schistosomiasis. In GC Cook, Manson's Tropical Diseases, 20th ed., WB Saunders Company Ltd, London, p. 1413-1456 .
Di Stasi LC 1989. Plantas Medicinais da Amazônia, Unesp, Sã Paulo, p. 39-40.
Ferreira MB 1980. Plantas portadoras de substâncias medicamentosas, de uso popular, nos cerrados de Minas Gerais. Inf Agropec 6: 19-23.
Ferri, MG 1973. A vegetação de cerrados brasileiros. In E Warming, Lagoa Santa, Universidade de São Paulo, São Paulo, p. 287-288.
Grandi TSM, Trindade JA, Pinto MJF, Ferreira LL, Catella AC 1989. Plantas medicinais de Minas Gerais. Acta Bot Bras 3: 185-224.
Hirschmann GS, Arias AR 1990. A survey of medicinal plants of Minas Gerais, Brazil. J Ethnopharmacol 29: 159-172.
Homans AL, Fuchs A 1970. Directed bioautography on thin layer chromatograms as a method for detecting fungitoxic substances. J Chromatogr 51: 327-329.
Hostettmann K 1991. Assays for Bioactivity. Methods in Plant Biochemistry, vol. 6, Academic Press, San Diego, 360 pp.
Luo S, Mai L, Zhu Z 1985. Separation and determination of alkaloids of Leonurus sibiricus. Chung Yao T'Ung Pao10: 32-35.
Martins ER, Castro DM, Castellani DC 1994. Plantas Medicinais, Imprensa Universitária, UFV Viçosa, 220 pp.
Mahato SB, Sahu NP, Roy SK, Sharma OP 1994. Potential antitumor agents from Lantana camara: structures of flavonois and phenylpropanois glycosides. Tetrahedron 50: 9439-9446.
Matos FJA 1994. Farmácias Vivas, EUFC, Fortaleza, 179 pp.
McLaughlin JM 1991. Crown gall tumours on potato discs and brine shrimp lethality: two simple bioassays for higher plant screening and fractionation. In K Hostettmann, Assays for Bioactivity, Academic Press, San Diego, p. 2-32.
Mott K 1987. Plant moluscicides, UNDP/World Bank/WHO/TDR, John Wiley and Sons Ltd., 326 pp.
Ross SA, El-Keltawi NE, Megalla SE 1980. Antimicrobial activity of some Egyptian aromatic plants. Fitoterapia51: 201-205.
Schultes RE, Raffauf RF 1990. The Healing Forest. Medicinal and Toxic Plants of the Northwest Amazônia, Dioscorides Press, Portland, Oregon, 484 pp.
Silver LL, Bostian KA 1993. Discovery and development of new antibiotics: the problem of antibiotic resistance.Antimicrob Agents Chemother 37: 377-383
Smânia A, Delle Monache F, Smânia EFA, Gil ML, Benchetrit LC, Cruz FS 1995. Antibacterial activity of substance produced by the fungus Pycnoporus sanguineus (Fr.) merr. J Ethopharmacol 45: 177-181.
WHO-World Health Organization 1993. Tropical Disease Research, Geneva, 134 pp.
Zani CL, Chaves PPG, Queiroz R, Mendes NM, Oliveira AB, Cardoso JE, Anjos AMG, Grandi TS 1995. Brine shrimp lethality assay as a prescreening system for anti-Trypanosoma cruzi activity. Phytomedicine 2: 47-50.