- Email: [email protected]
Toxicological evaluation of a tea from leaves of Vernonia condensata
Journal of Ethnopharmacology 74 (2001) 149 – 157 www.elsevier.com/locate/jethpharm
Toxicological evaluation of a tea from leaves of Vernonia condensata M.H.D. Monteiro a, M.R. Gomes-Carneiro a, I. Felzenszwalb b, I. Chahoud c, F.J.R. Paumgartten a,* a
Laborato´rio de Toxicologia Ambiental, Escola Nacional de Sau´de Pu´blica, Fundac¸ao Oswaldo Cruz, Rio de Janeiro, A6. Brasil 4365, Rio de Janeiro 21045 -900, Brazil b Departmento de Biofisica e Biometria, Instituto de Biologia Roberto Alcaˆntara Gomes, Uni6ersidade do Estado do Rio de Janeiro, Rio de Janeiro, Brazil c Abteilung fuer Toxikologie, Institut fuer Klinische Pharmakologie und Toxikologie, Freie Uni6ersita¨t Berlin, Garystraße 5, 14195, Berlin, Germany Received 24 January 2000; received in revised form 10 October 2000; accepted 17 October 2000
Abstract Teas of Vernonia condensata Baker (Asteraceae) are widely used in Brazil for gastro-intestinal disorders and to treat several other diseases. In this study, we evaluated the acute toxicity, embryotoxicity and mutagenicity of a lyophilized aqueous extract (LAE) from V. condensata leaves. Single doses of LAE, up to 5000 mg/kg body weight, were given orally or intraperitoneally to male and female Swiss albino mice. No toxicity was observed after oral administration. The ‘Approximate Lethal Dose’ after intraperitoneal injections was 3400 mg/kg for males and 5000 mg/kg for females. Embryotoxicity was investigated in Han:NMRI mice. LAE (0, 500 and 2000 mg/kg/day) was given by gavage on days 10, 11 and 12, and dams were submitted to caesarean sections on day 18 of pregnancy. Fetuses were weighed, examined for externally visible malformations, and evaluated for skeletal anomalies. Except for a slight reduction of fetal body weight accompanied by signs of delayed ossification at the highest dose, no other embryotoxic effect was noted in the exposed offspring. LAE-induced mutagenicity was evaluated in the Salmonella/microsome assay without and with S9 mixture. LAE, tested up to 5000 mg/plate, was not mutagenic to tester strains TA97a, TA98 and TA100. Results therefore suggest that V. condensata aqueous extracts present low acute toxicity and pose neither teratogenic nor mutagenic risks. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Medicinal plants; Genotoxicity; Teratogenicity; Toxicity; Analgesia
1. Introduction
* Corresponding author. E-mail address: [email protected] (F.J.R. Paumgartten).
Several species of the genus Vernonia (Asteraceae= Compositae) are used in traditional medicine mainly in South and Central America and in Africa. In Brazil, Vernonia condensata
0378-8741/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 7 8 - 8 7 4 1 ( 0 0 ) 0 0 3 6 3 - 9
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
150
Baker seems to be one of the species of Vernonia mostly often used in folk medicine for indications such as gastro-intestinal disorders, headache, diarrhea, and protection against snakebites (Table 1). A survey performed by Rizzo et al. (1985) found that 75 out of 249 families living in Goiania (GO, Brazil) made frequent use of infusions of V. condensata leaves to treat headache and disorders of the liver and stomach. The plant is grown in different regions of the country where it is known as ‘assa-peixe’, ‘boldo’, ‘boldo grande’, ‘macela˜o’ and ‘aluma˜’, although these popular names are also shared by other species of the same genus. Some of the traditional uses of V. condensata seem to be supported by data provided by pharmacological studies. Recently, it was demonstrated that aqueous extracts of V. condensata leaves possess analgesic properties and inhibit dose-dependently abdominal writhings induced by 0.6% acetic acid in mice (Frutuoso et al., 1994). The aqueous extract was also shown to potentiate aspirin- and indomethacin-induced analgesia in mice and to protect gastric mucosa against ulcerogenic effects of these nonsteroidal antiinflammatory drugs (NSADs) (Frutuoso et al., 1994). Furthermore, substances isolated from V. condensata, such as caffeic acid derivatives and chloro-
genic acid, given orally to mice, proved to inhibit lethality caused by the venom of the snake Bothrops jararaca (Pereira et al., 1994). Notwithstanding the widespread use of infusions of V. condensata in the Brazilian folk medicine, no study of the toxic effects of these preparations has been performed so far. It should be emphasized that tradition in use by no means warrants that the use is safe, particularly with regard to mutagenicity/carcinogenicity and embryotoxicity, where cause–effect relationships are rather complex and not easily recognized by the population. Moreover, pronounced mammalian toxicity has been found in some species belonging to the same genus. Vernonia molissima Don, for instance, was reported to cause deaths of cattle, sheep and goats in Mato Grosso do Sul state, Brazil (Do¨bereiner et al., 1976). The dried and powdered sprouts of V. molissima, administered as a single oral dose (0.25–6.0 g/kg body weight) to rabbits, proved to be highly nephrotoxic and hepatotoxic, thereby killing most of the treated animals (Tokarnia et al., 1986). The present study was undertaken to provide a minimum set of data on the safety of the aqueous extract of V. condensata leaves, focusing on its acute toxicity, embryotoxicity and mutagenicity.
Table 1 The uses of V. condensata Baker in the Brazilian folk medicine Medicinal use
Part used
Mode(s) of preparationa
Reference
Liver problems
Leaves
Tea, decoction
Stomach problems Headache Protection against snakebites Diarrhea
Leaves
Tea, decoction
Leaves Leaves
Tea, maceration Tea
Rizzo et al., 1985; Grandi et al., 1988; Magalha˜es et al., 1990; Teno´rio et al., 1991 Rizzo et al., 1985; Grandi et al., 1988; Magalha˜es et al., 1990; Teno´rio et al., 1991 Rizzo et al., 1985 Pereira et al., 1994, Houghton and Osibogun, 1993
Leaves
Begossi et al., 1993
Intestinal problems Cholagogue
Leaves
Pounded with water Infusion
Leaves
Infusion
Grandi et al., 1988
a
Grandi et al., 1988
Modes of preparation are mentioned here as reported in the cited references.
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
151
Table 2 Acute toxicity of aqueous extracts from leaves of V. condensata Baker to micea Dose (mg/kg body weight)
Sex
Route of administration Oral
0 300 450 670 1000 1500 2250 3400 5000
M F M F M F M F M F M F M F M F M F
Intraperitoneal
D/T
Symptoms
D/T
Latency
Symptoms
0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1
No No No No No No No No No No No No No No No No No No
0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 1/1 0/1 1/1 1/1
– – – – – – – – – – – – – – \24 h,B36 h – \12 h,B24 h \12 h,B24 h
No No No No No No No No No No No No No No Hipoactivity Hipoactivity Hipoactivity Hipoactivity
a The lyophilized extract dissolved in distilled water was administered orally (gavage) or intraperitoneally and all treated animals were carefully examined during 14 days for any sign of toxicity. D/T, Dead/treated mice; no, no symptoms during the observation period; latency, time to death.
2. Materials and methods
2.1. Plant material V. condensata Baker (Asteraceae) was identified by Dr Roberto Esteves from the Department of Botany, State University of Rio de Janeiro, and a voucher specimen was deposited (Reference 294825) in the herbarium of the Botanical Garden of Rio de Janeiro. The leaves were collected from plants grown in the county of Sa˜o Joa˜o do Meriti, Rio de Janeiro, Brazil, between December 1994 and September 1995.
2.2. Preparation of extracts Fresh leaves were washed in running water and then dried and powdered. Powdered leaves were extracted with distilled water (1 g powder/10 ml water) at 60°C for 30 min. The aqueous crude extract obtained was filtered and lyophilized
(yield, approximately 3% w/w). The lyophilized aqueous extract was stored at − 20°C until further use. The aqueous extract of V. condensata leaves was prepared essentially as described by Frutuoso et al. (1994). This aqueous crude extract was previously demonstrated to possess analgesic properties (Frutuoso et al., 1994).
2.3. Acute toxicity Albino Swiss mice of either sex (females, 2692 g; and males, 2992 g) from the Oswaldo Cruz Foundation Central Animal House breeding stock were used. The mice were housed individually in plastic cages with stainless steel cover lids and white pinus shavings as bedding. Photoperiod (lights on from 06:30 to 18:30 h), air changes and room temperature (25°C) were controlled. All animals had free access to tap water and were fed ad libitum (Nuvital®; Nuvilab Ltd, Curitiba, PR, Brazil) except for a short fasting period starting 2
152
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
h before and lasting until 2 h after the treatment with a single dose of the lyophilized extract of V. condensata. Lyophilized extract was diluted in distilled water and was given either by the oral (gavage) or the intraperitoneal route. Doses were increased progressively so that each dose was 50% higher than the preceding one, as suggested by Kennedy et al. (1986) for the determination of the Approximate Lethal Dose (ALD). The mice were observed for 14 days following treatment, and all signs of toxicity and deaths and their latencies were recorded. All animals that died during the
observation period and all the surviving ones (sacrificed by cervical dislocation) were subject to necropsy.
2.4. Embryotoxicity Male and female Han:NMRI mice (Zentralinstitut fuer Versuchstierzucht, Hannover, Germany) were kept under spf conditions at 219 1°C, 70% humidity and controlled photoperiod (lights on from 09:00 to 21:00 h). A pelleted diet (Altromin® 1324) and tap water were pro-
Table 3 Maternal weight gain and caesarean section data of mice treated orally with lyophilized aqueous extract of V. condensata (0, 500 and 2000 mg/kg body weight/day) on days 10, 11 and 12 of pregnancya Treatment
V. condensata aqueous extract (mg/kg body weight/day) 0
Treated females Pregnant females (%)
500
2000
29 23 (79.3)
20 18 (90.0)
25 23 (92.0)
Maternal weight (g) Day 1 Day 18
34.419 2.29 56.809 7.34
36.59 9 2.59 56.68 9 3.93
35.69 92.82 59.63 95.78
Pregnant uterus weight (g)
18.619 5.38
17.97 9 3.10
18.52 95.03
Pregnancy weight gain (g) Dday 1–18 Dday 1–18 (minus uterus weight)
24.489 5.76 5.96 9 1.52
22.09 9 2.74 4.13 91.31
23.97 9 4.66 4.64 93.36
Implantation sites Total (N) Per litter
304 13.293.6
240 13.3 92.2
315 13.7 9 2.3
Resorptions Early Intermediate Late Total (N) Per implantations (%) Per litter
0 22 5 27 8.8 1.179 1.15
0 21 5 26 10.8 1.419 1.50
0 17 6 23 7.3 1.00 9 1.04
Dead fetuses (N) Li6e fetuses Total (N) Per implantations Per litter Fetal body weight (litter mean) (g) a
2
4
275 90.5 11.9693.76
210 87.5 11.67 9 2.42
1.189 0.10
1.18 90.07
1 291 92.4 12.65 9 2.37 1.12 9 0.07*
Proportions were analysed by the chi-square test. Maternal and pregnant uterus weights, number of implantations, resorptions and live fetuses per litter as well as fetal body weight are shown as means 9 S.D. and were evaluated by one-way analysis of variance followed by the Student t-test. * Values significantly different (PB0.05) from controls.
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
153
Table 4 Occurrence of fetal skeleton anomalies in the offspring of mice treated orally with lyophilized aqueous extract of V. condensata (0, 500 and 2000 mg/kg body weight/day) on days 10, 11 and 12 of pregnancya Treatment
Fetuses examined Litters examined
Historical control
5551 486
V condensata aqueous extract (mg/kg body weight/day) 0
500
2000
275 23
210 18
291 23
Percentage of fetuses showing anomalies Skull Palate (cleft) Os basioccipitalis (smaller) Os maxill. (irregularly shaped) Os maxill. (smaller) Os squamos. (shorter or smaller) Pr. ang. mandib. (missing)
0.61 0 0 0 0 0
0.36 0 0 0 0 0
0.95 0 0 0 0 0.48
0.69 0.34 1.03 0.69 1.03 0
Atlas Bifurcated Irregularly shaped Irregularly positioned Narrower Wider
0.61 0.10 0.01 0 0.21
2.18 0.73 0 0 1.82
2.86 1.43 0.48 0.48 0.95
3.09 0.34 0 0.34 0.69
Axis Irregularly shaped Irregularly positioned Narrower Wider Additional ossif. center
0 0.01 0 0.01 0.14
0.36 0 0.36 0 0.36
0.48 2.38 0.48 3.33 1.43
0.69 0 0 0 0
Vertebral column Thor. vert. (o.c. dumb-bell shaped)
0.01
0
0.48
0.34
Ribs Fused
0.12
0.73
0
0.34
Sternum Irregularly shaped Ossif. center 1 (irregularly shaped) Ossif. center 5 (smaller)
0.37 0.03 0.41
0 0.36 0
0 0 0
0.69 0.34 0.69
0 0 0.05 0 0
0 0 0 0 0
0 0 0 0 0
1.03 0.34 0.69 1.72 1.03
0.05 0
3.27 9.82
4.29 2.38
9.97* 15.81*
0.01 0
0.36 0
1.43 0
6.19* 0.34
Forelimbs Clavicula (Bent) (Broken bone) (Irregularly shaped) Scapula (smaller) Os humerus (shorter) Fingers (Ossif. Center not ossified) (Poorly ossified) Processus deltoideus (missing) Hindlimbs Metatars. 1 (ossif. center not ossified) Os pubis (bent)
a Proportions were analysed by the chi-square test or, alternatively, by the Fisher exact test. * Values significantly different (PB0.05) from controls are indicated by (*). Historical control: data of untreated control Han:NMRI mice evaluated at the Institute of Clinical Pharmacology and Toxicology of FU-Berlin. Os maxill., Os maxillare; Os squamos., Os squamosum; Pr. ang. mandib., Processus angularis of mandibula; ossif. center, Ossification center; Thor. vert. Thoracic vertebrae; oc dumb-bell shaped, ossification center dumb bell shaped; Metatars., Metatarsus
154
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
vided ad libitum. Mating was performed as reported by Chahoud and Kwasigroch (1977). Five females were placed into the cage of one male for 3 h at the end of the dark period (06:00 to 09:00 h). Immediately after the mating period, the females were examined for the presence of a vaginal plug. The day on which a vaginal plug was found was considered day 0 of pregnancy. The lyophilized extract dissolved in distilled water was administered to the mice by gavage (0, 500 or 2000 mg/kg body weight/day) on days 10, 11 and 12 of pregnancy. All mice were weighed on days 1, 10, 11, 12 and 18 of pregnancy. On day 18 of pregnancy, dams were killed by cervical dislocation. The gravid uterus was weighed and the number of implantations, living and dead fetuses, and resorptions were recorded. The living fetuses were immediately weighed, numbered with a marker pen, examined for externally visible anomalies and fixed in a 5% formalin solution. The fetuses were subsequently cleared with KOH and stained with alizarin red S for further evaluation of skeletal abnormalities.
2.5. Mutagenicity (Ames test) The Salmonella typhimurium tester strains (TA97a, TA98 and TA100) were kindly supplied
by Dr B.N. Ames (University of California, Berkeley, CA, USA). The bacterial strains were checked for their genetic markers at the beginning of the study. For all tests, overnight fresh cultures were prepared from frozen permanents (200 ml culture to 20 ml Oxoid Nutrient Broth No.2) and incubated at 37°C with shaking until a concentration of approximately 1.2× 109 bacteria/ml was obtained. The Salmonella mutagenicity test was performed by the plate incorporation method essentially as described by Maron and Ames (1983). Briefly, top-agar (2 ml) was mixed with 100 ml overnight grown culture of S. typhimurium, 100 ml solution of lyophilized extract in distilled water, the negative control or the positive control (PC), and 500 ml phosphate buffer or S9 mixture. Lyophilized rat liver S9 fraction induced by Aroclor 1254 was purchased from Moltox® (Molecular Toxicology, Annapolis, USA). The S9 mixture (21 ml) was prepared for use as described elsewhere (Gomes-Carneiro et al., 1998). Distilled water served as the negative control, while the positive control substances were: sodium azide (SA) (0.5 mg/plate), 4-nitroquinolone-n-oxide (4NQNO) (1 mg/plate), 2-aminofluorene (2AF) (10 mg/plate), 2-aminoanthracene (2AA) (1 mg/plate), 2-nitrofluorene (2NF) (1 mg/plate). SA was dissolved in distilled water and dimethyl sulfoxide
Table 5 Mutagenicity testing of the lyophilized aqueous extract from leaves of V. condensata Baker in the Salmonella/microsome assay (TA97a, TA98 and TA100 tester strains)a Dose (mg/plate)
Number of his+ revertants per plate (mean 9S.D.) TA97a
5000 4000 3000 2000 1000 0 PC
TA98
TA100
−S9
+S9
−S9
+S9
−S9
+S9
176918 173910 149923 160925 1599 11 1409 20 639956
2799 20 234939 257933 252929 228 9 16 2459 5 1090 936
38 96 30 96 48 913) 32 9 8 42 99 43 98 488 914
67 91 64 9 10 54 9 4 58 911 48 9 4 47 9 7 763 945
158 912 135 921 175 9 11 182 920 195 941 164 9 19 626 9 31
248 9 36 224 9 26 230 9 16 239 9 5 230 9 22 204 9 7 812 9 32
a Dose 0, Negative control (100 ml distilled water); PC (positive control), TA97a/−S9; 4-NQNO (1 mg/plate), TA97a/+S9; 2AF (10 mg/plate), TA98/−S9; 2NF (1 mg/plate), TA98/+S9; 2AA (1 mg/plate), TA100/−S9; SA (0.5 mg/plate), TA100/+S9; 2AA (1 mg/plate). Assays were carried out in the presence (+S9) and in the absence (−S9) of an extrinsic metabolic activation system (S9 mixture).
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
was the solvent used for the other positive controls. Plates were incubated at 37°C for 72 h in the dark and then scored for revertant his + bacte- ria colonies. Every determination was made in triplicate and experiments were repeated at least once in order to check the reproducibility of the results.
3. Results and discussion
3.1. Acute toxicity As shown in Table 2, acute oral toxicity of the aqueous extract was very low in mice. No deaths and no other signs of toxicity were observed up to the highest dose tested (5000 mg/kg/body weight). Necropsy of these mice 14 days after oral treatment with the lyophilized extract did not reveal any gross pathological alteration either. Since the extract had been previously shown to be pharmacologically active by the oral route (Frutuoso et al., 1994), one can conclude that the active substance(s) presents a rather low acute toxicity as well. Taking into account that the ED50 value for analgesia in the same species was 241 mg/kg body weight (Frutuoso et al., 1994) and that no deaths were observed up to the highest dose tested in this study (ALD \ 5000 mg/kg body weight), the safety margin (ALD/ED50) seems to be wide (\20.7) for the aqueous extract. Lethality was noted only after intraperitoneal (i.p.) administration, with an ALD as high as 3400 mg/kg body weight for males and as high as 5000 mg/kg body weight for females (Table 2). Except for hypoactivity, no other adverse effect preceding death was noted. Since these mice died during the night, post-mortem examination was impaired by a pronounced degree of autolysis and the cause of death remained obscure. The necropsy of surviving mice 14 days after treatment did not reveal any salient finding either. Anyhow, besides the fact that ALD values following i.p. administration were also high, human exposure to aqueous extracts of V. condensata leaves is very unlikely to occur by parenteral routes.
155
condensata aqueous extract treated mice is presented in Table 3. No difference in maternal weight gain was noted at term (day 18) between controls and groups treated with 500 or 2000 mg/kg body weight per day on days 10, 11 and 12 of pregnancy. Moreover, no treatment-related reduction of gravid uterus weight was observed, and no difference between the control and aqueous extract treated groups was found after deduction of uterine weight at term (Table 3). These results seem to indicate that the aqueous extract was not maternally toxic over the dose range tested. Except for a bump head in one fetus of the highest dose group, no externally visible malformation was detected either in control or in treated mice. Neither an increase in the resorption rate or a decrease in the number of live fetuses was found (Table 3). These findings indicated that the aqueous extract did not cause embryo deaths. Nonetheless, a slight, but statistically significant, reduction of fetal body weight was noted at the highest dose tested (2000 mg/kg) (Table 3). The occurrence of skeletal abnormalities in fetuses exposed to V. condensata aqueous extract on pregnancy days 10, 11 and 12 is presented in Table 4. No treatment-induced increase in the proportion of major structural anomalies (malformations) was noted. The skeletal changes, the frequency of which were clearly increased at the highest dose tested (2000 mg/kg body weight), were also found in control fetuses and could be better classified as signs of delayed ossification (e.g. fingers ‘ossification centers not ossified’ or ‘poorly ossified’, and metatarsus 1 ‘ossification centers not ossified’). These few signs of retarded ossification and the lower fetal body weight indicated that doses of V. condensata aqueous extract as high as 2000 mg/kg body weight caused a prenatal growth retardation. The foregoing results suggested that, except for a slight retardation of fetal growth, no other embryotoxic effect was caused by the aqueous extract of V. condensata leaves administered orally to mice on days 10–12 of pregnancy.
3.3. Mutagenicity 3.2. Embryotoxicity The pregnancy weight gain of control and V.
Table 5 shows the results of the Salmonella mutagenicity assay (tester strains TA97a, TA98
156
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
and TA100) with the aqueous crude extract of V. condensata leaves, without and with the addition of an extrinsic metabolic activation system (S9 mixture). A preliminary dose range finding test (not shown) was carried out with TA98 tester strain, and neither reductions of the number of the revertants or alterations of the auxotrophic bacteria grown (i.e. the background lawn) were apparent up to the highest dose tested (5000 mg/ plate). In the absence of toxicity signs, 5000 mg/ plate has been recommended by different authors as the upper limit of the dose interval to be tested in the Salmonella/microsome assay (Gatehouse et al., 1994). As shown in Table 5, V. condensata aqueous extract did not induce any consistent or dose-related increase in the number of revertant his+ colonies over negative control values obtained for strains TA97a, TA98 and TA100, either in the presence or in the absence of metabolic activation (S9 mixture). These three tester strains detect frameshift mutations (TA97a, TA98) as well as base pair substitutions (TA100) and most of the known mutagens (Maron and Ames, 1983). Only a few mutagens, such as some oxidants and cross-linking agents (e.g. mitomycin C, psoralens), seem not easily detectable by the tester strain battery used in the present study. These results thus indicate that the aqueous extract of V. condensata is not mutagenic in the Ames test, thereby suggesting that its folk medicinal use is unlikely to pose genotoxic risks. 4. Conclusions In conclusion, oral acute toxicity of the aqueous extract from leaves of V. condensata is very low and no evidence was found that it poses teratogenic or mutagenic risks. The only toxic effect noted following oral treatment was a slight fetal growth retardation at doses of the lyophilized aqueous extract as high as 2000 mg/kg body weight/day. This dose of the lyophilized extract corresponds to approximately 67 g dried leaves/kg body weight. Human exposure to comparable levels (i.e. 4.69 kg dried leaves for an adult weighing 70 kg) is unlikely to result from the medicinal use of teas from leaves of V. condensata.
Acknowledgements This study was supported by grants from the Brazilian National Research Council (CNPq) and PAPES-FIOCRUZ. F.J.R.P. and I.F. are recipients of research fellowships from CNPq.
References Begossi, A., Leita˜o-Filho, H.F., Richerson, P.J., 1993. Plant uses in a Brazilian coastal fishing community (Bu´zios island). Journal of Ethnobiology 13 (2), 233 – 256. Chahoud, I., Kwasigroch, T.E., 1977. Controlled breeding of laboratory animals. In: Neubert, D., Merker, H.J., Kwasigroch, T.E. (Eds.), Methods in Prenatal Toxicology. George Thieme, Stuttgart, pp. 78 – 91. Do¨bereiner, J., Tokarnia, C.H., Purisco, E., 1976. Vernonia molissima, planta to´xica responsa´vel por mortandades de bovinos no sul de Mato Grosso. Pesquisa Agropecua´ria Brasileira, Se´rie Veterina´ria 11, 49 – 58. Frutuoso, V.S., Gurja˜o, M.R., Cordeiro, R.S., Martins, M.A., 1994. Analgesic and anti-ulcerogenic effects of a polar extract from leaves of Vernonia condensata. Planta Medica 60, 21 – 25. Gatehouse, D., Haworth, S., Cebula, T., Gocke, E., Kier, L., Matsushima, T., Melcion, C., Nohmi, T., Ohta, T., Venitt, S., Zeiger, E., 1994. Recommendations for the performance of bacterial mutation assays. Mutation Research 312, 217 – 233. Gomes-Carneiro, M.R., Felzenszwalb, I., Paumgartten, F.J.R., 1998. Mutagenicity testing ( 9 )-camphor, 1,8-cineole, citral, citronellal, ( −)-menthol and terpineol with the Salmonella/microsome assay. Mutation Research 416, 129 – 136. Grandi, T.S.M., Da-Trindade, J.A., Pinto, M.J.F., Ferreira, L.L., Catella, A.C., 1988. Plantas medicinais de Minas Gerais, Brasil. Acta Botaˆnica Brasileira 3 (Suppl.), 185 – 224. Houghton, P.J., Osibogun, I.M., 1993. Flowering plants used against snakebites. Journal of Ethnopharmacology 39, 1 – 29. Kennedy, G.L., Jr, Ferenz, R.L., Burgess, B.A., 1986. Estimation of acute oral toxicity in rats by determination of the approximate lethal dose rather than the LD50. Journal of Applied Toxicology 6, 145 – 148. Magalha˜es, R.A., Machado, M.I.L., Matos, F.J.A., 1990. Constituintes quı´micos macromoleculares do geˆnero Vernonia (Revisa˜o). Proceedings of the XI Simpo´sio de Plantas Medicinais do Brasil, Joa˜o Pessoa, PA, 12 – 14 September, Abstract 2.24. Maron, D.M., Ames, B.N., 1983. Revised methods for Salmonella mutagenicity rest. Mutation Research 113, 173 – 215.
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157 Pereira, N.A., Pereira, B.M.R., Nascimento, M.C., Parente, J.P., Mors, W.B., 1994. Folk medicine as snake venom antidotes; VI — protection against Jararaca venom by isolated constituents. Planta Medica 60, 99–100. Rizzo, J.A., Monteiro, M.S.R., Bitencourt, C., 1985. Utilizac¸a˜o de plantas medicinais em Goia˜nia. Anais do XXXVI Congresso Brasileiro de Botaˆnica, pp. 691–714. Teno´rio, M.A.R.O., van den Berg, M.E., Menezes, O.F., Sales, P., 1991. Fitoterapia: uma estrate´gia terapeˆutica
157
natural do Amapa´. In: Buchillet, D. (Ed.), Medicinas Tradicionais e Medicina Ocidental na Amazoˆnia, contribuic¸o˜es cientificas apresentadas no encontro de Bele´m, 27 November – 01 December 1989, Edic¸o˜es Cejup, Belem, PA, Brasil. Tokarnia, C.B., Do¨bereiner, J., Amorin, P.R., Gava, A., Consorte, L.B., 1986. Intoxicac¸a˜o experimental por Vernonia molissima (Compositae) em coelhos. Pesquisa Veterina´ria Brasileira 6 (1), 5 – 10.
.
Toxicological evaluation of a tea from leaves of Vernonia condensata M.H.D. Monteiro a, M.R. Gomes-Carneiro a, I. Felzenszwalb b, I. Chahoud c, F.J.R. Paumgartten a,* a
Laborato´rio de Toxicologia Ambiental, Escola Nacional de Sau´de Pu´blica, Fundac¸ao Oswaldo Cruz, Rio de Janeiro, A6. Brasil 4365, Rio de Janeiro 21045 -900, Brazil b Departmento de Biofisica e Biometria, Instituto de Biologia Roberto Alcaˆntara Gomes, Uni6ersidade do Estado do Rio de Janeiro, Rio de Janeiro, Brazil c Abteilung fuer Toxikologie, Institut fuer Klinische Pharmakologie und Toxikologie, Freie Uni6ersita¨t Berlin, Garystraße 5, 14195, Berlin, Germany Received 24 January 2000; received in revised form 10 October 2000; accepted 17 October 2000
Abstract Teas of Vernonia condensata Baker (Asteraceae) are widely used in Brazil for gastro-intestinal disorders and to treat several other diseases. In this study, we evaluated the acute toxicity, embryotoxicity and mutagenicity of a lyophilized aqueous extract (LAE) from V. condensata leaves. Single doses of LAE, up to 5000 mg/kg body weight, were given orally or intraperitoneally to male and female Swiss albino mice. No toxicity was observed after oral administration. The ‘Approximate Lethal Dose’ after intraperitoneal injections was 3400 mg/kg for males and 5000 mg/kg for females. Embryotoxicity was investigated in Han:NMRI mice. LAE (0, 500 and 2000 mg/kg/day) was given by gavage on days 10, 11 and 12, and dams were submitted to caesarean sections on day 18 of pregnancy. Fetuses were weighed, examined for externally visible malformations, and evaluated for skeletal anomalies. Except for a slight reduction of fetal body weight accompanied by signs of delayed ossification at the highest dose, no other embryotoxic effect was noted in the exposed offspring. LAE-induced mutagenicity was evaluated in the Salmonella/microsome assay without and with S9 mixture. LAE, tested up to 5000 mg/plate, was not mutagenic to tester strains TA97a, TA98 and TA100. Results therefore suggest that V. condensata aqueous extracts present low acute toxicity and pose neither teratogenic nor mutagenic risks. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Medicinal plants; Genotoxicity; Teratogenicity; Toxicity; Analgesia
1. Introduction
* Corresponding author. E-mail address: [email protected] (F.J.R. Paumgartten).
Several species of the genus Vernonia (Asteraceae= Compositae) are used in traditional medicine mainly in South and Central America and in Africa. In Brazil, Vernonia condensata
0378-8741/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 7 8 - 8 7 4 1 ( 0 0 ) 0 0 3 6 3 - 9
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
150
Baker seems to be one of the species of Vernonia mostly often used in folk medicine for indications such as gastro-intestinal disorders, headache, diarrhea, and protection against snakebites (Table 1). A survey performed by Rizzo et al. (1985) found that 75 out of 249 families living in Goiania (GO, Brazil) made frequent use of infusions of V. condensata leaves to treat headache and disorders of the liver and stomach. The plant is grown in different regions of the country where it is known as ‘assa-peixe’, ‘boldo’, ‘boldo grande’, ‘macela˜o’ and ‘aluma˜’, although these popular names are also shared by other species of the same genus. Some of the traditional uses of V. condensata seem to be supported by data provided by pharmacological studies. Recently, it was demonstrated that aqueous extracts of V. condensata leaves possess analgesic properties and inhibit dose-dependently abdominal writhings induced by 0.6% acetic acid in mice (Frutuoso et al., 1994). The aqueous extract was also shown to potentiate aspirin- and indomethacin-induced analgesia in mice and to protect gastric mucosa against ulcerogenic effects of these nonsteroidal antiinflammatory drugs (NSADs) (Frutuoso et al., 1994). Furthermore, substances isolated from V. condensata, such as caffeic acid derivatives and chloro-
genic acid, given orally to mice, proved to inhibit lethality caused by the venom of the snake Bothrops jararaca (Pereira et al., 1994). Notwithstanding the widespread use of infusions of V. condensata in the Brazilian folk medicine, no study of the toxic effects of these preparations has been performed so far. It should be emphasized that tradition in use by no means warrants that the use is safe, particularly with regard to mutagenicity/carcinogenicity and embryotoxicity, where cause–effect relationships are rather complex and not easily recognized by the population. Moreover, pronounced mammalian toxicity has been found in some species belonging to the same genus. Vernonia molissima Don, for instance, was reported to cause deaths of cattle, sheep and goats in Mato Grosso do Sul state, Brazil (Do¨bereiner et al., 1976). The dried and powdered sprouts of V. molissima, administered as a single oral dose (0.25–6.0 g/kg body weight) to rabbits, proved to be highly nephrotoxic and hepatotoxic, thereby killing most of the treated animals (Tokarnia et al., 1986). The present study was undertaken to provide a minimum set of data on the safety of the aqueous extract of V. condensata leaves, focusing on its acute toxicity, embryotoxicity and mutagenicity.
Table 1 The uses of V. condensata Baker in the Brazilian folk medicine Medicinal use
Part used
Mode(s) of preparationa
Reference
Liver problems
Leaves
Tea, decoction
Stomach problems Headache Protection against snakebites Diarrhea
Leaves
Tea, decoction
Leaves Leaves
Tea, maceration Tea
Rizzo et al., 1985; Grandi et al., 1988; Magalha˜es et al., 1990; Teno´rio et al., 1991 Rizzo et al., 1985; Grandi et al., 1988; Magalha˜es et al., 1990; Teno´rio et al., 1991 Rizzo et al., 1985 Pereira et al., 1994, Houghton and Osibogun, 1993
Leaves
Begossi et al., 1993
Intestinal problems Cholagogue
Leaves
Pounded with water Infusion
Leaves
Infusion
Grandi et al., 1988
a
Grandi et al., 1988
Modes of preparation are mentioned here as reported in the cited references.
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
151
Table 2 Acute toxicity of aqueous extracts from leaves of V. condensata Baker to micea Dose (mg/kg body weight)
Sex
Route of administration Oral
0 300 450 670 1000 1500 2250 3400 5000
M F M F M F M F M F M F M F M F M F
Intraperitoneal
D/T
Symptoms
D/T
Latency
Symptoms
0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1
No No No No No No No No No No No No No No No No No No
0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 0/1 1/1 0/1 1/1 1/1
– – – – – – – – – – – – – – \24 h,B36 h – \12 h,B24 h \12 h,B24 h
No No No No No No No No No No No No No No Hipoactivity Hipoactivity Hipoactivity Hipoactivity
a The lyophilized extract dissolved in distilled water was administered orally (gavage) or intraperitoneally and all treated animals were carefully examined during 14 days for any sign of toxicity. D/T, Dead/treated mice; no, no symptoms during the observation period; latency, time to death.
2. Materials and methods
2.1. Plant material V. condensata Baker (Asteraceae) was identified by Dr Roberto Esteves from the Department of Botany, State University of Rio de Janeiro, and a voucher specimen was deposited (Reference 294825) in the herbarium of the Botanical Garden of Rio de Janeiro. The leaves were collected from plants grown in the county of Sa˜o Joa˜o do Meriti, Rio de Janeiro, Brazil, between December 1994 and September 1995.
2.2. Preparation of extracts Fresh leaves were washed in running water and then dried and powdered. Powdered leaves were extracted with distilled water (1 g powder/10 ml water) at 60°C for 30 min. The aqueous crude extract obtained was filtered and lyophilized
(yield, approximately 3% w/w). The lyophilized aqueous extract was stored at − 20°C until further use. The aqueous extract of V. condensata leaves was prepared essentially as described by Frutuoso et al. (1994). This aqueous crude extract was previously demonstrated to possess analgesic properties (Frutuoso et al., 1994).
2.3. Acute toxicity Albino Swiss mice of either sex (females, 2692 g; and males, 2992 g) from the Oswaldo Cruz Foundation Central Animal House breeding stock were used. The mice were housed individually in plastic cages with stainless steel cover lids and white pinus shavings as bedding. Photoperiod (lights on from 06:30 to 18:30 h), air changes and room temperature (25°C) were controlled. All animals had free access to tap water and were fed ad libitum (Nuvital®; Nuvilab Ltd, Curitiba, PR, Brazil) except for a short fasting period starting 2
152
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
h before and lasting until 2 h after the treatment with a single dose of the lyophilized extract of V. condensata. Lyophilized extract was diluted in distilled water and was given either by the oral (gavage) or the intraperitoneal route. Doses were increased progressively so that each dose was 50% higher than the preceding one, as suggested by Kennedy et al. (1986) for the determination of the Approximate Lethal Dose (ALD). The mice were observed for 14 days following treatment, and all signs of toxicity and deaths and their latencies were recorded. All animals that died during the
observation period and all the surviving ones (sacrificed by cervical dislocation) were subject to necropsy.
2.4. Embryotoxicity Male and female Han:NMRI mice (Zentralinstitut fuer Versuchstierzucht, Hannover, Germany) were kept under spf conditions at 219 1°C, 70% humidity and controlled photoperiod (lights on from 09:00 to 21:00 h). A pelleted diet (Altromin® 1324) and tap water were pro-
Table 3 Maternal weight gain and caesarean section data of mice treated orally with lyophilized aqueous extract of V. condensata (0, 500 and 2000 mg/kg body weight/day) on days 10, 11 and 12 of pregnancya Treatment
V. condensata aqueous extract (mg/kg body weight/day) 0
Treated females Pregnant females (%)
500
2000
29 23 (79.3)
20 18 (90.0)
25 23 (92.0)
Maternal weight (g) Day 1 Day 18
34.419 2.29 56.809 7.34
36.59 9 2.59 56.68 9 3.93
35.69 92.82 59.63 95.78
Pregnant uterus weight (g)
18.619 5.38
17.97 9 3.10
18.52 95.03
Pregnancy weight gain (g) Dday 1–18 Dday 1–18 (minus uterus weight)
24.489 5.76 5.96 9 1.52
22.09 9 2.74 4.13 91.31
23.97 9 4.66 4.64 93.36
Implantation sites Total (N) Per litter
304 13.293.6
240 13.3 92.2
315 13.7 9 2.3
Resorptions Early Intermediate Late Total (N) Per implantations (%) Per litter
0 22 5 27 8.8 1.179 1.15
0 21 5 26 10.8 1.419 1.50
0 17 6 23 7.3 1.00 9 1.04
Dead fetuses (N) Li6e fetuses Total (N) Per implantations Per litter Fetal body weight (litter mean) (g) a
2
4
275 90.5 11.9693.76
210 87.5 11.67 9 2.42
1.189 0.10
1.18 90.07
1 291 92.4 12.65 9 2.37 1.12 9 0.07*
Proportions were analysed by the chi-square test. Maternal and pregnant uterus weights, number of implantations, resorptions and live fetuses per litter as well as fetal body weight are shown as means 9 S.D. and were evaluated by one-way analysis of variance followed by the Student t-test. * Values significantly different (PB0.05) from controls.
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
153
Table 4 Occurrence of fetal skeleton anomalies in the offspring of mice treated orally with lyophilized aqueous extract of V. condensata (0, 500 and 2000 mg/kg body weight/day) on days 10, 11 and 12 of pregnancya Treatment
Fetuses examined Litters examined
Historical control
5551 486
V condensata aqueous extract (mg/kg body weight/day) 0
500
2000
275 23
210 18
291 23
Percentage of fetuses showing anomalies Skull Palate (cleft) Os basioccipitalis (smaller) Os maxill. (irregularly shaped) Os maxill. (smaller) Os squamos. (shorter or smaller) Pr. ang. mandib. (missing)
0.61 0 0 0 0 0
0.36 0 0 0 0 0
0.95 0 0 0 0 0.48
0.69 0.34 1.03 0.69 1.03 0
Atlas Bifurcated Irregularly shaped Irregularly positioned Narrower Wider
0.61 0.10 0.01 0 0.21
2.18 0.73 0 0 1.82
2.86 1.43 0.48 0.48 0.95
3.09 0.34 0 0.34 0.69
Axis Irregularly shaped Irregularly positioned Narrower Wider Additional ossif. center
0 0.01 0 0.01 0.14
0.36 0 0.36 0 0.36
0.48 2.38 0.48 3.33 1.43
0.69 0 0 0 0
Vertebral column Thor. vert. (o.c. dumb-bell shaped)
0.01
0
0.48
0.34
Ribs Fused
0.12
0.73
0
0.34
Sternum Irregularly shaped Ossif. center 1 (irregularly shaped) Ossif. center 5 (smaller)
0.37 0.03 0.41
0 0.36 0
0 0 0
0.69 0.34 0.69
0 0 0.05 0 0
0 0 0 0 0
0 0 0 0 0
1.03 0.34 0.69 1.72 1.03
0.05 0
3.27 9.82
4.29 2.38
9.97* 15.81*
0.01 0
0.36 0
1.43 0
6.19* 0.34
Forelimbs Clavicula (Bent) (Broken bone) (Irregularly shaped) Scapula (smaller) Os humerus (shorter) Fingers (Ossif. Center not ossified) (Poorly ossified) Processus deltoideus (missing) Hindlimbs Metatars. 1 (ossif. center not ossified) Os pubis (bent)
a Proportions were analysed by the chi-square test or, alternatively, by the Fisher exact test. * Values significantly different (PB0.05) from controls are indicated by (*). Historical control: data of untreated control Han:NMRI mice evaluated at the Institute of Clinical Pharmacology and Toxicology of FU-Berlin. Os maxill., Os maxillare; Os squamos., Os squamosum; Pr. ang. mandib., Processus angularis of mandibula; ossif. center, Ossification center; Thor. vert. Thoracic vertebrae; oc dumb-bell shaped, ossification center dumb bell shaped; Metatars., Metatarsus
154
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
vided ad libitum. Mating was performed as reported by Chahoud and Kwasigroch (1977). Five females were placed into the cage of one male for 3 h at the end of the dark period (06:00 to 09:00 h). Immediately after the mating period, the females were examined for the presence of a vaginal plug. The day on which a vaginal plug was found was considered day 0 of pregnancy. The lyophilized extract dissolved in distilled water was administered to the mice by gavage (0, 500 or 2000 mg/kg body weight/day) on days 10, 11 and 12 of pregnancy. All mice were weighed on days 1, 10, 11, 12 and 18 of pregnancy. On day 18 of pregnancy, dams were killed by cervical dislocation. The gravid uterus was weighed and the number of implantations, living and dead fetuses, and resorptions were recorded. The living fetuses were immediately weighed, numbered with a marker pen, examined for externally visible anomalies and fixed in a 5% formalin solution. The fetuses were subsequently cleared with KOH and stained with alizarin red S for further evaluation of skeletal abnormalities.
2.5. Mutagenicity (Ames test) The Salmonella typhimurium tester strains (TA97a, TA98 and TA100) were kindly supplied
by Dr B.N. Ames (University of California, Berkeley, CA, USA). The bacterial strains were checked for their genetic markers at the beginning of the study. For all tests, overnight fresh cultures were prepared from frozen permanents (200 ml culture to 20 ml Oxoid Nutrient Broth No.2) and incubated at 37°C with shaking until a concentration of approximately 1.2× 109 bacteria/ml was obtained. The Salmonella mutagenicity test was performed by the plate incorporation method essentially as described by Maron and Ames (1983). Briefly, top-agar (2 ml) was mixed with 100 ml overnight grown culture of S. typhimurium, 100 ml solution of lyophilized extract in distilled water, the negative control or the positive control (PC), and 500 ml phosphate buffer or S9 mixture. Lyophilized rat liver S9 fraction induced by Aroclor 1254 was purchased from Moltox® (Molecular Toxicology, Annapolis, USA). The S9 mixture (21 ml) was prepared for use as described elsewhere (Gomes-Carneiro et al., 1998). Distilled water served as the negative control, while the positive control substances were: sodium azide (SA) (0.5 mg/plate), 4-nitroquinolone-n-oxide (4NQNO) (1 mg/plate), 2-aminofluorene (2AF) (10 mg/plate), 2-aminoanthracene (2AA) (1 mg/plate), 2-nitrofluorene (2NF) (1 mg/plate). SA was dissolved in distilled water and dimethyl sulfoxide
Table 5 Mutagenicity testing of the lyophilized aqueous extract from leaves of V. condensata Baker in the Salmonella/microsome assay (TA97a, TA98 and TA100 tester strains)a Dose (mg/plate)
Number of his+ revertants per plate (mean 9S.D.) TA97a
5000 4000 3000 2000 1000 0 PC
TA98
TA100
−S9
+S9
−S9
+S9
−S9
+S9
176918 173910 149923 160925 1599 11 1409 20 639956
2799 20 234939 257933 252929 228 9 16 2459 5 1090 936
38 96 30 96 48 913) 32 9 8 42 99 43 98 488 914
67 91 64 9 10 54 9 4 58 911 48 9 4 47 9 7 763 945
158 912 135 921 175 9 11 182 920 195 941 164 9 19 626 9 31
248 9 36 224 9 26 230 9 16 239 9 5 230 9 22 204 9 7 812 9 32
a Dose 0, Negative control (100 ml distilled water); PC (positive control), TA97a/−S9; 4-NQNO (1 mg/plate), TA97a/+S9; 2AF (10 mg/plate), TA98/−S9; 2NF (1 mg/plate), TA98/+S9; 2AA (1 mg/plate), TA100/−S9; SA (0.5 mg/plate), TA100/+S9; 2AA (1 mg/plate). Assays were carried out in the presence (+S9) and in the absence (−S9) of an extrinsic metabolic activation system (S9 mixture).
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
was the solvent used for the other positive controls. Plates were incubated at 37°C for 72 h in the dark and then scored for revertant his + bacte- ria colonies. Every determination was made in triplicate and experiments were repeated at least once in order to check the reproducibility of the results.
3. Results and discussion
3.1. Acute toxicity As shown in Table 2, acute oral toxicity of the aqueous extract was very low in mice. No deaths and no other signs of toxicity were observed up to the highest dose tested (5000 mg/kg/body weight). Necropsy of these mice 14 days after oral treatment with the lyophilized extract did not reveal any gross pathological alteration either. Since the extract had been previously shown to be pharmacologically active by the oral route (Frutuoso et al., 1994), one can conclude that the active substance(s) presents a rather low acute toxicity as well. Taking into account that the ED50 value for analgesia in the same species was 241 mg/kg body weight (Frutuoso et al., 1994) and that no deaths were observed up to the highest dose tested in this study (ALD \ 5000 mg/kg body weight), the safety margin (ALD/ED50) seems to be wide (\20.7) for the aqueous extract. Lethality was noted only after intraperitoneal (i.p.) administration, with an ALD as high as 3400 mg/kg body weight for males and as high as 5000 mg/kg body weight for females (Table 2). Except for hypoactivity, no other adverse effect preceding death was noted. Since these mice died during the night, post-mortem examination was impaired by a pronounced degree of autolysis and the cause of death remained obscure. The necropsy of surviving mice 14 days after treatment did not reveal any salient finding either. Anyhow, besides the fact that ALD values following i.p. administration were also high, human exposure to aqueous extracts of V. condensata leaves is very unlikely to occur by parenteral routes.
155
condensata aqueous extract treated mice is presented in Table 3. No difference in maternal weight gain was noted at term (day 18) between controls and groups treated with 500 or 2000 mg/kg body weight per day on days 10, 11 and 12 of pregnancy. Moreover, no treatment-related reduction of gravid uterus weight was observed, and no difference between the control and aqueous extract treated groups was found after deduction of uterine weight at term (Table 3). These results seem to indicate that the aqueous extract was not maternally toxic over the dose range tested. Except for a bump head in one fetus of the highest dose group, no externally visible malformation was detected either in control or in treated mice. Neither an increase in the resorption rate or a decrease in the number of live fetuses was found (Table 3). These findings indicated that the aqueous extract did not cause embryo deaths. Nonetheless, a slight, but statistically significant, reduction of fetal body weight was noted at the highest dose tested (2000 mg/kg) (Table 3). The occurrence of skeletal abnormalities in fetuses exposed to V. condensata aqueous extract on pregnancy days 10, 11 and 12 is presented in Table 4. No treatment-induced increase in the proportion of major structural anomalies (malformations) was noted. The skeletal changes, the frequency of which were clearly increased at the highest dose tested (2000 mg/kg body weight), were also found in control fetuses and could be better classified as signs of delayed ossification (e.g. fingers ‘ossification centers not ossified’ or ‘poorly ossified’, and metatarsus 1 ‘ossification centers not ossified’). These few signs of retarded ossification and the lower fetal body weight indicated that doses of V. condensata aqueous extract as high as 2000 mg/kg body weight caused a prenatal growth retardation. The foregoing results suggested that, except for a slight retardation of fetal growth, no other embryotoxic effect was caused by the aqueous extract of V. condensata leaves administered orally to mice on days 10–12 of pregnancy.
3.3. Mutagenicity 3.2. Embryotoxicity The pregnancy weight gain of control and V.
Table 5 shows the results of the Salmonella mutagenicity assay (tester strains TA97a, TA98
156
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157
and TA100) with the aqueous crude extract of V. condensata leaves, without and with the addition of an extrinsic metabolic activation system (S9 mixture). A preliminary dose range finding test (not shown) was carried out with TA98 tester strain, and neither reductions of the number of the revertants or alterations of the auxotrophic bacteria grown (i.e. the background lawn) were apparent up to the highest dose tested (5000 mg/ plate). In the absence of toxicity signs, 5000 mg/ plate has been recommended by different authors as the upper limit of the dose interval to be tested in the Salmonella/microsome assay (Gatehouse et al., 1994). As shown in Table 5, V. condensata aqueous extract did not induce any consistent or dose-related increase in the number of revertant his+ colonies over negative control values obtained for strains TA97a, TA98 and TA100, either in the presence or in the absence of metabolic activation (S9 mixture). These three tester strains detect frameshift mutations (TA97a, TA98) as well as base pair substitutions (TA100) and most of the known mutagens (Maron and Ames, 1983). Only a few mutagens, such as some oxidants and cross-linking agents (e.g. mitomycin C, psoralens), seem not easily detectable by the tester strain battery used in the present study. These results thus indicate that the aqueous extract of V. condensata is not mutagenic in the Ames test, thereby suggesting that its folk medicinal use is unlikely to pose genotoxic risks. 4. Conclusions In conclusion, oral acute toxicity of the aqueous extract from leaves of V. condensata is very low and no evidence was found that it poses teratogenic or mutagenic risks. The only toxic effect noted following oral treatment was a slight fetal growth retardation at doses of the lyophilized aqueous extract as high as 2000 mg/kg body weight/day. This dose of the lyophilized extract corresponds to approximately 67 g dried leaves/kg body weight. Human exposure to comparable levels (i.e. 4.69 kg dried leaves for an adult weighing 70 kg) is unlikely to result from the medicinal use of teas from leaves of V. condensata.
Acknowledgements This study was supported by grants from the Brazilian National Research Council (CNPq) and PAPES-FIOCRUZ. F.J.R.P. and I.F. are recipients of research fellowships from CNPq.
References Begossi, A., Leita˜o-Filho, H.F., Richerson, P.J., 1993. Plant uses in a Brazilian coastal fishing community (Bu´zios island). Journal of Ethnobiology 13 (2), 233 – 256. Chahoud, I., Kwasigroch, T.E., 1977. Controlled breeding of laboratory animals. In: Neubert, D., Merker, H.J., Kwasigroch, T.E. (Eds.), Methods in Prenatal Toxicology. George Thieme, Stuttgart, pp. 78 – 91. Do¨bereiner, J., Tokarnia, C.H., Purisco, E., 1976. Vernonia molissima, planta to´xica responsa´vel por mortandades de bovinos no sul de Mato Grosso. Pesquisa Agropecua´ria Brasileira, Se´rie Veterina´ria 11, 49 – 58. Frutuoso, V.S., Gurja˜o, M.R., Cordeiro, R.S., Martins, M.A., 1994. Analgesic and anti-ulcerogenic effects of a polar extract from leaves of Vernonia condensata. Planta Medica 60, 21 – 25. Gatehouse, D., Haworth, S., Cebula, T., Gocke, E., Kier, L., Matsushima, T., Melcion, C., Nohmi, T., Ohta, T., Venitt, S., Zeiger, E., 1994. Recommendations for the performance of bacterial mutation assays. Mutation Research 312, 217 – 233. Gomes-Carneiro, M.R., Felzenszwalb, I., Paumgartten, F.J.R., 1998. Mutagenicity testing ( 9 )-camphor, 1,8-cineole, citral, citronellal, ( −)-menthol and terpineol with the Salmonella/microsome assay. Mutation Research 416, 129 – 136. Grandi, T.S.M., Da-Trindade, J.A., Pinto, M.J.F., Ferreira, L.L., Catella, A.C., 1988. Plantas medicinais de Minas Gerais, Brasil. Acta Botaˆnica Brasileira 3 (Suppl.), 185 – 224. Houghton, P.J., Osibogun, I.M., 1993. Flowering plants used against snakebites. Journal of Ethnopharmacology 39, 1 – 29. Kennedy, G.L., Jr, Ferenz, R.L., Burgess, B.A., 1986. Estimation of acute oral toxicity in rats by determination of the approximate lethal dose rather than the LD50. Journal of Applied Toxicology 6, 145 – 148. Magalha˜es, R.A., Machado, M.I.L., Matos, F.J.A., 1990. Constituintes quı´micos macromoleculares do geˆnero Vernonia (Revisa˜o). Proceedings of the XI Simpo´sio de Plantas Medicinais do Brasil, Joa˜o Pessoa, PA, 12 – 14 September, Abstract 2.24. Maron, D.M., Ames, B.N., 1983. Revised methods for Salmonella mutagenicity rest. Mutation Research 113, 173 – 215.
M.H.D. Monteiro et al. / Journal of Ethnopharmacology 74 (2001) 149–157 Pereira, N.A., Pereira, B.M.R., Nascimento, M.C., Parente, J.P., Mors, W.B., 1994. Folk medicine as snake venom antidotes; VI — protection against Jararaca venom by isolated constituents. Planta Medica 60, 99–100. Rizzo, J.A., Monteiro, M.S.R., Bitencourt, C., 1985. Utilizac¸a˜o de plantas medicinais em Goia˜nia. Anais do XXXVI Congresso Brasileiro de Botaˆnica, pp. 691–714. Teno´rio, M.A.R.O., van den Berg, M.E., Menezes, O.F., Sales, P., 1991. Fitoterapia: uma estrate´gia terapeˆutica
157
natural do Amapa´. In: Buchillet, D. (Ed.), Medicinas Tradicionais e Medicina Ocidental na Amazoˆnia, contribuic¸o˜es cientificas apresentadas no encontro de Bele´m, 27 November – 01 December 1989, Edic¸o˜es Cejup, Belem, PA, Brasil. Tokarnia, C.B., Do¨bereiner, J., Amorin, P.R., Gava, A., Consorte, L.B., 1986. Intoxicac¸a˜o experimental por Vernonia molissima (Compositae) em coelhos. Pesquisa Veterina´ria Brasileira 6 (1), 5 – 10.
.