**6. Toxicological studies**

Guaco species have been widely used by the South America population; thus, several studies, although insufficient, have been done to evaluate the toxicity of the extracts, phytomedicines, and isolated compounds.

The aqueous extract of *M. laevigata* was screened for anti-mutagenic activity using the Salmonella/microsome assay. The infusions was negative for mutagenic activity, showing high percentages of inhibition of mutagenesis induced by mutagens 2-aminofluorene (2AF), in the presence of exogenous metabolism (S9 fraction), for frameshift (TA98) and base pair substitution (TA100) lesions. In addition, these inhibitions were observed against mutagen

Fig. 2. Pathways of coumarin metabolism. Dihydrocoumarin (DHC);

HDHC-GSH).

**6. Toxicological studies** 

phytomedicines, and isolated compounds.

*o*-hydroxyphenylpropionic acid (o-HPPA); *o*-coumaric acid (o-CA); 3, 4, 5, 6, 7 and 8-

*o*-hydroxyphenylethanol (o-HPE); 4-hydroxydihydrocoumarin-glutathione-conjugated (4-

Guaco species have been widely used by the South America population; thus, several studies, although insufficient, have been done to evaluate the toxicity of the extracts,

The aqueous extract of *M. laevigata* was screened for anti-mutagenic activity using the Salmonella/microsome assay. The infusions was negative for mutagenic activity, showing high percentages of inhibition of mutagenesis induced by mutagens 2-aminofluorene (2AF), in the presence of exogenous metabolism (S9 fraction), for frameshift (TA98) and base pair substitution (TA100) lesions. In addition, these inhibitions were observed against mutagen

hydroxycoumarin (HC); 7-hydroxycoumarin glucoronide (7-HC-GLUC); 6,7 dihydroxycoumarin (6,7-diHC); *o*-hydroxyphenylacetic acid (o-HPLA);

o-hydroxyphenylacetic acid (o-HPAA); *o*-hydroxyphenylacetaldehyde (o-HPA);

sodium azide in assays with the TA100 strain, without exogenous metabolism (S9 fraction). A synergistic effect was also observed in frameshift mutagenic events, with direct action in the presence of 4-oxide-1-nitroquinoline and a tendency to a low percentage of action enhancement in the presence of the 2AF mutagen (Fernandes & Vargas, 2003).

In contrast to the outcomes from the Salmonella/microsome trials, studies conducted by the comet assay revealed that guaco extracts have deleterious effects. DNA damage was observed in rat hepatoma cells treated with hydroalcoholic maceration (10 and 20 μL/mL) and infusion (20 and 40 μL/mL) of the leaves. The genotoxic potential of the infusion was also observed by the micronucleous test at a very high concentration (40 μL/mL), suggesting a limitation in the phytotherapeutic use of guaco species (Costa et al., 2008).

Caution is recommended for patients who use lyophilized extracts or medicines containing isolated compounds, such as coumarin and *o*-coumaric acid. Hemorrhaging lung tissue was observed in mice treated with these substances and with the extract. However, this effect was not observed in animals treated with the whole hydroalcoholic extract, leading to the conclusion that some protective effect of the whole extract can be lost during the lyophilization process (Santos et al., 2006).

Because guaco showed an effect against *L. amazonensis* and *T. cruzi*, it was important to assess its toxic effects on mammalian host cells to determine the ratio of selectivity to biological activity. For this purpose, a test of cytotoxicity in sheep erythrocytes was performed using hydroalcoholic extracts of leaves at different concentrations and times of incubation. At 100, 500 and 1000 μg/mL, guaco extracts caused, respectively, 25, 50 and 75% hemolysis in erythrocytes incubated at 120 min. However, the hydroalcoholic extract was not considered cytotoxic to sheep erythrocytes because no significant hemolytic effect was observed at 100 and 500 μg/mL after 60 minutes of incubation (Luize et al., 2005).

The hydroalcoholic extract did not impair the fertility of rats following 52 days of oral treatment with a chronic dose of 3.3 g/kg of animal. In females, no changes in mating, gestation, preimplantation loss, the number of implanted embryos or offspring, weaning and the implantation and resorption indexes were observed using this kind of extract (SÁ et al., 2006). In males, the treatment did not alter body and organ weights and did not interfere in gamete production, serum testosterone levels or food intake (SÁ et al., 2003). Following 90 days of treatment, no significant change was observed in body and organ weights, gamete concentration on the epididymis cauda, serum testosterone level or food consumption, suggesting the absence of toxicity or antifertility activity of the hydroalcoholic extract (SÁ et al., 2010).

The absence of any effect on body weight gain or behavioral patterns in mice subjected to a repeated-dose over 14-, 28- or 60-day treatments (3 mg/kg) indicated that the *M. laevigata*  ethanolic extract does not induce significant toxicity. The lack of alterations in hematological parameters, liver cell injury and serum aminotransferases (AST and ALT) was indicative of normal hepatic and biliary function. In addition, there was no change in urea levels, indicating the absence of alterations in the kidney. Additionally, the LD50 was found to be almost 75-fold higher than the pharmacological dose tested (Alves et al., 2009).

The potential genotoxicity of the dichloromethane fraction of the hydroalcoholic extract was evaluated on plasmid DNA using an alkaline lysis procedure, in which plasmid DNA was treated with SnCl2 and the *M. glomerata* extract fraction. The role of reactive oxygen species in DNA damage was also evaluated by incubating the extract fraction with sodium benzoate, a hydroxyl radical scavenger. The results showed that the dichloromethane

*Mikania glomerata* and *M. laevigata*: Clinical and Toxicological Advances 309

At 30 and 60 μg/mL, kaurenoic acid also induces DNA breaks and cytogenetic abnormalities in human peripheral blood leukocytes (PBLs), as evaluated by comet, cytokinesis-block micronucleus and chromosomal aberration assays. Using a yeast cell model, cytotoxic and mutagenic effects of kaurenoic acid were also observed in the XV185- 14c strain: there was an increase in the frequencies of point, frameshift, and forward mutations in the stationary phase at high concentrations (0.5–2 μg/mL). However, these effects were more pronounced when cells were treated in the exponential phase than in

Positive genotoxic effects have also been described testing kaurenoic acid *in vivo* in multiple organs, such as the liver, kidney and spleen of mice (alkaline comet assay). DNA migration in liver cells was considerable at all tested doses (25, 50 and 100 mg/kg, i.p.) and at higher doses (50 and 100 mg/kg) in kidney cells. No DNA breaks were observed after the treatment in spleen cells (Cavalcanti et al., 2010). Finally, genotoxicity in Chinese hamster lung fibroblast cells was also observed using the comet and the micronucleus assays. However, lower concentrations (2.5, 5, and 10 μg/mL) failed to induce significant effects, whereas higher concentrations (30 and 60 μg/mL) lead to an increase in cell damage index and frequency. These data indicated that kaurenoic acid induces dose-dependent

Dihydrocoumarin is one of the most studied guaco metabolites in regards to its toxic effects. In the human TK6 lymphoblastoid cell line, dihydrocoumarin caused an increase in p53 acetylation and cytotoxicity. Flow cytometric analysis to detect annexin V binding to phosphatidylserine demonstrated that dihydrocoumarin also increased apoptosis more than 3-fold over controls. In addition, dihydrocoumarin disrupted epigenetic processes in the yeast *Saccharomyces cerevisiae* and also inhibited several human sirtuin deacetylases (SIRT1 and SIRT2), a class of proteins that control some epigenetic processes and has, interestingly, been implicated in extending the longevity of several organisms (Olaharski et al., 2005). Toxicity and carcinogenicity studies were also conducted by administering 99% pure dihydrocoumarin to groups of rats and mice in short (16 days), 13-week, and long (2 years) exposures. The short exposure lead to the death all male and female rats treated with 3000 mg/kg of dihydrocoumarin. At 1500 mg/kg, half of the animals died and a gain of body weight was observed; however, there were no clinical findings of organ-specific toxicity or evidence of impaired blood coagulation. A similar finding was also observed in mice groups, however with total mortality observed at a lower body/weight concentration than

Following 13 weeks of administration, groups of 10 male and 10 female rats were studied, and a difference of exposure sensitivity was observed between the groups. In this case, two male and five female rats died after the administration of 1200 mg/kg dihydrocoumarin. The platelet counts were diminished in males receiving 600 mg/kg and in the female groups receiving 300 mg/kg dihydrocoumarin. Hemoglobin and hematocrit values were significantly lower in males that received 300 mg/kg dihydrocoumarin; this dose caused hepatocellular hypertrophy in both sexes. Additionally, the absolute and relative liver and kidney weights were significantly greater than those of the controls following a treatment of 600 mg/kg dihydrocoumarin. In mice groups, mortality was 80% in male and 50% in female receiving 1600 mg/kg dihydrocoumarin. With this exposure, the absolute and relative liver weight in both sexes and the relative kidney weight in males were significantly greater than those of the controls. However, no variation in body weight or changes in hematologic

growth or non-growth conditions (Cavalcanti et al., 2010).

genotoxicity (Cavalcanti et al., 2006).

the rat groups (2250 mg/kg).

parameters were observed in either sex.

fraction was not genotoxic because this fraction did not damage DNA directly or by producing the hydroxyl radical reactive oxygen species (Moura et al., 2002)

The pharmaceutical preparation of guaco syrup did not produce any disturbances in the hematological or biochemical parameters in rodents following 90 days of treatment with subchronic and chronic doses (75, 150 and 300 mg/kg). Additionally, no evidence of toxicity in the hepatic, renal or pancreatic systems was reported. At reproductive endpoints, no alterations in body and organ weights, sperm, spermatid number, testosterone levels, or sperm morphology were observed after exposure to guaco syrup (Graca et al., 2007a, 2007b). In humans, only two phase I clinical studies have been conducted to evaluate the clinical safety of guaco syrup. The volunteers (n= 24 – 26) received an oral dose of 15 mL phytomedicine four times a day over 21 to 28 days; after the treatment, any clinically significant changes in coagulation parameters were observed. In some cases, low variations in biochemical, hematological and serological analysis were observed, but none of the volunteers had values out of the established normality limits. Among them, only two volunteers reported mild drowsiness during the treatment, and one reported diarrhea and nausea. However, it is unclear if these effects were caused by guaco ingestion and in addition, clinical, electrocardiographic and laboratory tests did not show any evidence of toxicity. Nevertheless, more conclusive studies should be made because only phytomedicines containing low amounts of guaco extract associated with other plants were evaluated (Soares et al., 2006; Tavares et al., 2006).

The toxicity of the main isolated compounds has also been assessed. For example, kaurenoic acid has been shown to kill sea urchin embryos by inhibiting the first cleavage of the fertilized eggs (IC50 = 84.2 μM). Additionally, this compound progressively induced the destruction of embryos in other development stages (IC50 = 44.7 μM for blastulae stages and < 10 μM for larvae stages) (Costa-Lotufo et al., 2002).

Kaurenoic acid has been shown to have a weak to negligible capacity for killing human sperm. The estimated ED50 for sperm immobilization was 374.1 μg/mL, using 15 × 106 sperm/500 μL (VALENCIA et al., 1986). This compound has also been shown to induce dose-dependent hemolysis of mouse and human erythrocytes with an EC50 of 74.0 and 56.4 μM, respectively (Costa-Lotufo et al., 2002).

By the microculture tetrazolium test (MTT) assay, 78 μM kaurenoic acid causes cytotoxicity in CEM leukemic cells, leading to a 95% growth inhibition. This effect was also observed in MCF-7 breast and HCT-8 colon cancer cells, with a growth inhibition of 45% (Costa-lotufo et al., 2002). Moderate antiproliferative effects were also observed in K562, HL60, MDA-MB435 and SF295 human cancer cell lines (IC50 = 9.1 – 14.3 μg/mL). Fluorescence microscopy using acridine orange/ethidium bromide staining indicated that kaurenoic acid induced apoptosis and necrosis in HL-60 cell cultures, consistent with the findings described in the MTT assay. However, the antiproliferative effects were not selective to cancer cells because inhibition of lymphocyte proliferation also occurred (IC50 = 12.6 μg/mL) (Cavalcanti et al., 2009).

The cytotoxic effects of kaurenoic acid have been partly associated with its partial inhibitory effect on human topo-isomerase (topo) I activity. In contrast, 14-hydroxy-kaurane, xylopic acid, and semi-synthetic derivatives of kaurenoic acid [16*a*-methoxy-(−)-kauran-19-oic acid, 16*a*-methoxy-(−)-kauran-19-oic methyl ester and 16*a*-hydroxy-(−)-kauran-19-oic acid] lack genotoxic and mutagenic effects. This result suggests that the exocyclic double bond (C16) moiety may be the active pharmacophore for the genetic toxicity of kaurenoic acid (Cavalcanti et al., 2009).

fraction was not genotoxic because this fraction did not damage DNA directly or by

The pharmaceutical preparation of guaco syrup did not produce any disturbances in the hematological or biochemical parameters in rodents following 90 days of treatment with subchronic and chronic doses (75, 150 and 300 mg/kg). Additionally, no evidence of toxicity in the hepatic, renal or pancreatic systems was reported. At reproductive endpoints, no alterations in body and organ weights, sperm, spermatid number, testosterone levels, or sperm morphology were observed after exposure to guaco syrup (Graca et al., 2007a, 2007b). In humans, only two phase I clinical studies have been conducted to evaluate the clinical safety of guaco syrup. The volunteers (n= 24 – 26) received an oral dose of 15 mL phytomedicine four times a day over 21 to 28 days; after the treatment, any clinically significant changes in coagulation parameters were observed. In some cases, low variations in biochemical, hematological and serological analysis were observed, but none of the volunteers had values out of the established normality limits. Among them, only two volunteers reported mild drowsiness during the treatment, and one reported diarrhea and nausea. However, it is unclear if these effects were caused by guaco ingestion and in addition, clinical, electrocardiographic and laboratory tests did not show any evidence of toxicity. Nevertheless, more conclusive studies should be made because only phytomedicines containing low amounts of guaco extract associated with other plants were

The toxicity of the main isolated compounds has also been assessed. For example, kaurenoic acid has been shown to kill sea urchin embryos by inhibiting the first cleavage of the fertilized eggs (IC50 = 84.2 μM). Additionally, this compound progressively induced the destruction of embryos in other development stages (IC50 = 44.7 μM for blastulae stages and

Kaurenoic acid has been shown to have a weak to negligible capacity for killing human sperm. The estimated ED50 for sperm immobilization was 374.1 μg/mL, using 15 × 106 sperm/500 μL (VALENCIA et al., 1986). This compound has also been shown to induce dose-dependent hemolysis of mouse and human erythrocytes with an EC50 of 74.0 and 56.4

By the microculture tetrazolium test (MTT) assay, 78 μM kaurenoic acid causes cytotoxicity in CEM leukemic cells, leading to a 95% growth inhibition. This effect was also observed in MCF-7 breast and HCT-8 colon cancer cells, with a growth inhibition of 45% (Costa-lotufo et al., 2002). Moderate antiproliferative effects were also observed in K562, HL60, MDA-MB435 and SF295 human cancer cell lines (IC50 = 9.1 – 14.3 μg/mL). Fluorescence microscopy using acridine orange/ethidium bromide staining indicated that kaurenoic acid induced apoptosis and necrosis in HL-60 cell cultures, consistent with the findings described in the MTT assay. However, the antiproliferative effects were not selective to cancer cells because inhibition of

lymphocyte proliferation also occurred (IC50 = 12.6 μg/mL) (Cavalcanti et al., 2009).

The cytotoxic effects of kaurenoic acid have been partly associated with its partial inhibitory effect on human topo-isomerase (topo) I activity. In contrast, 14-hydroxy-kaurane, xylopic acid, and semi-synthetic derivatives of kaurenoic acid [16*a*-methoxy-(−)-kauran-19-oic acid, 16*a*-methoxy-(−)-kauran-19-oic methyl ester and 16*a*-hydroxy-(−)-kauran-19-oic acid] lack genotoxic and mutagenic effects. This result suggests that the exocyclic double bond (C16) moiety may be the active pharmacophore for the genetic toxicity of kaurenoic acid

producing the hydroxyl radical reactive oxygen species (Moura et al., 2002)

evaluated (Soares et al., 2006; Tavares et al., 2006).

< 10 μM for larvae stages) (Costa-Lotufo et al., 2002).

μM, respectively (Costa-Lotufo et al., 2002).

(Cavalcanti et al., 2009).

At 30 and 60 μg/mL, kaurenoic acid also induces DNA breaks and cytogenetic abnormalities in human peripheral blood leukocytes (PBLs), as evaluated by comet, cytokinesis-block micronucleus and chromosomal aberration assays. Using a yeast cell model, cytotoxic and mutagenic effects of kaurenoic acid were also observed in the XV185- 14c strain: there was an increase in the frequencies of point, frameshift, and forward mutations in the stationary phase at high concentrations (0.5–2 μg/mL). However, these effects were more pronounced when cells were treated in the exponential phase than in growth or non-growth conditions (Cavalcanti et al., 2010).

Positive genotoxic effects have also been described testing kaurenoic acid *in vivo* in multiple organs, such as the liver, kidney and spleen of mice (alkaline comet assay). DNA migration in liver cells was considerable at all tested doses (25, 50 and 100 mg/kg, i.p.) and at higher doses (50 and 100 mg/kg) in kidney cells. No DNA breaks were observed after the treatment in spleen cells (Cavalcanti et al., 2010). Finally, genotoxicity in Chinese hamster lung fibroblast cells was also observed using the comet and the micronucleus assays. However, lower concentrations (2.5, 5, and 10 μg/mL) failed to induce significant effects, whereas higher concentrations (30 and 60 μg/mL) lead to an increase in cell damage index and frequency. These data indicated that kaurenoic acid induces dose-dependent genotoxicity (Cavalcanti et al., 2006).

Dihydrocoumarin is one of the most studied guaco metabolites in regards to its toxic effects. In the human TK6 lymphoblastoid cell line, dihydrocoumarin caused an increase in p53 acetylation and cytotoxicity. Flow cytometric analysis to detect annexin V binding to phosphatidylserine demonstrated that dihydrocoumarin also increased apoptosis more than 3-fold over controls. In addition, dihydrocoumarin disrupted epigenetic processes in the yeast *Saccharomyces cerevisiae* and also inhibited several human sirtuin deacetylases (SIRT1 and SIRT2), a class of proteins that control some epigenetic processes and has, interestingly, been implicated in extending the longevity of several organisms (Olaharski et al., 2005).

Toxicity and carcinogenicity studies were also conducted by administering 99% pure dihydrocoumarin to groups of rats and mice in short (16 days), 13-week, and long (2 years) exposures. The short exposure lead to the death all male and female rats treated with 3000 mg/kg of dihydrocoumarin. At 1500 mg/kg, half of the animals died and a gain of body weight was observed; however, there were no clinical findings of organ-specific toxicity or evidence of impaired blood coagulation. A similar finding was also observed in mice groups, however with total mortality observed at a lower body/weight concentration than the rat groups (2250 mg/kg).

Following 13 weeks of administration, groups of 10 male and 10 female rats were studied, and a difference of exposure sensitivity was observed between the groups. In this case, two male and five female rats died after the administration of 1200 mg/kg dihydrocoumarin. The platelet counts were diminished in males receiving 600 mg/kg and in the female groups receiving 300 mg/kg dihydrocoumarin. Hemoglobin and hematocrit values were significantly lower in males that received 300 mg/kg dihydrocoumarin; this dose caused hepatocellular hypertrophy in both sexes. Additionally, the absolute and relative liver and kidney weights were significantly greater than those of the controls following a treatment of 600 mg/kg dihydrocoumarin. In mice groups, mortality was 80% in male and 50% in female receiving 1600 mg/kg dihydrocoumarin. With this exposure, the absolute and relative liver weight in both sexes and the relative kidney weight in males were significantly greater than those of the controls. However, no variation in body weight or changes in hematologic parameters were observed in either sex.

*Mikania glomerata* and *M. laevigata*: Clinical and Toxicological Advances 311

of platelet counts was also observed. With this dose, the values of alanine aminotransferase, sorbitol dehydrogenase, and g-glutamyltransferase significantly increased in males, whereas these effects were observed only from 100 mg/kg in females. Additionally, lesions associated with the administration of coumarin were also observed during the long exposure, which include an increase of the severity of nephropathy, increase of incidences of bile duct and parathyroid gland hyperplasias, increase of the incidences of ulcers, and necrosis, fibrosis, and cytologic alterations in the liver (National Toxicology Program, 1993b). A carcinogenic potential has also been described for coumarin, especially in the liver and lungs of rats and mice. However, the dose–response relationships are nonlinear with tumor formation and hepatic and pulmonary toxicity are associated only with high doses

Regarding the mutagenic and genotoxic potential of coumarin, it showed weak clastogenic activity in Chinese Hamster ovary cells *in vitro*. However, this response was observed only at a very high concentration (10.95 μM). Negative responses were reported in the *Salmonella typhimurium* assay in the TA98, TA1535, TA1537 and TA1538 strains, either with or without metabolic activation. However, gene mutations have been described in the TA100 strain in

Using the Ames genotoxic assay, coumarin has not been shown to be a mutagenic agent in the TA100 strain assessed without metabolic activation (liver S9 fraction). With metabolic activation, coumarin produces only a weak positive effect at a high concentration. However, this effect has been widely discussed because a greater response was achieved in the presence of liver S9 fraction from untreated Syrian hamsters than from rats treated with Aroclor 1254, a substance used to stimulate coumarin metabolism by the 3,4-epoxidation pathway in rat hepatic microsomes. This result does not correlate with the extent of coumarin metabolism and coumarin-induced liver injury in these species. Because of the differences in their metabolic pathways, a chronic dose of coumarin induces liver lesions and tumor in the rat and not in the Syrian hamster, which appears to be refractory for

In addition*, in vivo* studies have shown that coumarin is unable to induce sex-linked recessive lethal mutations in germ cells of male *Drosophila melanogaster*. Furthermore, no evidence for coumarin-induced genotoxicity has been observed in the *in vivo* micronucleus test in mouse peripheral blood cells. The conclusion is that coumarin is not DNA-reactive and that the induction of tumors at high doses in rodents is attributed to cytotoxicity and

In humans, the majority of tests for mutagenicity and genotoxicity also suggest that coumarin is not a toxic agent (Lake, 1999). The lack of toxicity has been associated with the detoxification mechanism of coumarin, which in humans involves the 7-hydroxylation pathway, a minor route in rats and mice; these rodents use the 3,4-epoxidation pathway instead, which results in toxic metabolite formation (Gasparetto et al., 2011). Thus, the species-specific target organ toxicity has been attributed to the pharmacokinetics of coumarin metabolism, causing rats to be susceptible to liver effects and mice to have toxicity particularly in the lung. Therefore, it is possible to conclude that the use of rats and mice is not an adequate model to compare the metabolism and toxicity of coumarin with humans, due to their particular metabolism. Because *in vitro* genotoxicity studies demonstrated toxicity only at very high doses and no evidence for *in vivo* studies was observed, it is possible to conclude that there is no human health risk from coumarin exposure in natural

the presence of a metabolic activating system (S9) (Lake, 1999).

coumarin-induced hepatotoxicity (Lake, 1999).

regenerative hyperplasia (Felter et al., 2006).

dietary sources, such guaco species.

(Lake, 1999).

Under long dihydrocoumarin exposure (2 years), carcinogenic activity in male rats was evident based on the increased of incidence of renal tubule adenoma and focal hyperplasia. The transitional cell carcinomas in two males were chemical related. No evidence of carcinogenic activity was observed in female rats receiving 150, 300, or 600 mg/kg dihydrocoumarin. In mice, no evidence of carcinogenic activity was observed in male groups receiving 200, 400 or 800 mg/kg dihydrocoumarin; however, these doses led to an increase in the incidence of hepatocellular adenoma and carcinoma (combined) in females. In addition, ulcers, forestomach inflammation, parathyroid gland hyperplasia, and increased nephropathy were observed in these groups of rodents (National Toxicology Program, 1993a).

Coumarin, a main compound of guaco extracts, is a substance known to cause hepatotoxicity in liver rats. Prior to the existence of any available carcinogenicity and mutagenicity data, it was classified as a toxic substance by the Food and Drug Administration. Thus, it was banned in the USA in 1954 and in the UK in 1965 (Lake, 1999).

Various tests have been conducted to evaluate the toxicity and health effects of coumarin in laboratory animals. For example, doses of 25 mg/kg or higher were reported to produce liver damage in dogs (Felter et al., 2006). In primates (baboons) that received dietary coumarin for 2 years (0 to 67.5 mg/kg/day), no evidence of toxicity from biochemical and histochemical analyses was observed. However, an increase in the relative liver weight occurred at a high dosage, with additional dilatation of the endoplasmic reticulum observed after 10 months of treatment (Felter et al., 2006). The Syrian hamster has also been found to be resistant to coumarin-induced toxicity (Lake, 1999)

In groups of rats and mice, 97% pure coumarin administered orally has toxic effects with a short exposure (16 days), after 13 weeks, and a long (2 years) exposure. All groups of male and female rats died following 16 days of treatment with 400 mg/kg of coumarin. Increases in mean body weight also occurred, but no clinical signs of organ-specific toxicity were observed. Additionally, coagulation parameters were not impaired. In mice, groups of 5 male and 5 female rats were studied, and all 10 mice receiving 600 mg/kg, two male mice receiving 300 mg/kg, and one male mouse receiving 75 mg/kg died. With a short exposure, coagulation parameters were not impaired; however, an increase in the mean body weight and excessive lacrimation, piloerection, bradypnea and ataxia were observed for the 300 mg/kg dose in the first hours of administration (National Toxicology Program, 1993b).

Following 13 weeks of exposure to 300 mg/kg of coumarin, 30% of rats in the male and female groups died. Both groups had increased erythrocyte counts and decreased hemoglobin and erythrocyte mean volumes. Serum levels of total bilirubin and one or more cytoplasmic enzymes were higher than those of control groups. The absolute and relative liver weights also increased significantly following the administration of 150 mg/kg coumarin, and centrilobular hepatocellular degeneration and necrosis, chronic active inflammation, and bile duct hyperplasia were also observed in the liver. In the mice groups, 20% of male and female groups receiving 300 mg/kg coumarin died; similar to rats, coumarin decreased the erythrocyte volume and hemoglobin. Centrilobular hepatocellular hypertrophy was observed in both sexes at 300 mg/kg, and the absolute and relative liver weights increased following treatment with 150 mg/kg coumarin.

During the long (2 years) exposure, groups of 60 male and 60 female rats were treated with coumarin at different dosages, and after 15 months, 10 animals from each group were evaluated. Treatment with 50 mg/kg led to a significant reduction in the activated partial thromboplastin times and the erythrocyte volume and hemoglobin values, and an increase

Under long dihydrocoumarin exposure (2 years), carcinogenic activity in male rats was evident based on the increased of incidence of renal tubule adenoma and focal hyperplasia. The transitional cell carcinomas in two males were chemical related. No evidence of carcinogenic activity was observed in female rats receiving 150, 300, or 600 mg/kg dihydrocoumarin. In mice, no evidence of carcinogenic activity was observed in male groups receiving 200, 400 or 800 mg/kg dihydrocoumarin; however, these doses led to an increase in the incidence of hepatocellular adenoma and carcinoma (combined) in females. In addition, ulcers, forestomach inflammation, parathyroid gland hyperplasia, and increased nephropathy were observed in these groups of rodents (National Toxicology

Coumarin, a main compound of guaco extracts, is a substance known to cause hepatotoxicity in liver rats. Prior to the existence of any available carcinogenicity and mutagenicity data, it was classified as a toxic substance by the Food and Drug Administration. Thus, it was banned in the USA in 1954 and in the UK in 1965 (Lake, 1999). Various tests have been conducted to evaluate the toxicity and health effects of coumarin in laboratory animals. For example, doses of 25 mg/kg or higher were reported to produce liver damage in dogs (Felter et al., 2006). In primates (baboons) that received dietary coumarin for 2 years (0 to 67.5 mg/kg/day), no evidence of toxicity from biochemical and histochemical analyses was observed. However, an increase in the relative liver weight occurred at a high dosage, with additional dilatation of the endoplasmic reticulum observed after 10 months of treatment (Felter et al., 2006). The Syrian hamster has also been found to

In groups of rats and mice, 97% pure coumarin administered orally has toxic effects with a short exposure (16 days), after 13 weeks, and a long (2 years) exposure. All groups of male and female rats died following 16 days of treatment with 400 mg/kg of coumarin. Increases in mean body weight also occurred, but no clinical signs of organ-specific toxicity were observed. Additionally, coagulation parameters were not impaired. In mice, groups of 5 male and 5 female rats were studied, and all 10 mice receiving 600 mg/kg, two male mice receiving 300 mg/kg, and one male mouse receiving 75 mg/kg died. With a short exposure, coagulation parameters were not impaired; however, an increase in the mean body weight and excessive lacrimation, piloerection, bradypnea and ataxia were observed for the 300 mg/kg dose in the first hours of administration (National Toxicology Program, 1993b). Following 13 weeks of exposure to 300 mg/kg of coumarin, 30% of rats in the male and female groups died. Both groups had increased erythrocyte counts and decreased hemoglobin and erythrocyte mean volumes. Serum levels of total bilirubin and one or more cytoplasmic enzymes were higher than those of control groups. The absolute and relative liver weights also increased significantly following the administration of 150 mg/kg coumarin, and centrilobular hepatocellular degeneration and necrosis, chronic active inflammation, and bile duct hyperplasia were also observed in the liver. In the mice groups, 20% of male and female groups receiving 300 mg/kg coumarin died; similar to rats, coumarin decreased the erythrocyte volume and hemoglobin. Centrilobular hepatocellular hypertrophy was observed in both sexes at 300 mg/kg, and the absolute and relative liver

During the long (2 years) exposure, groups of 60 male and 60 female rats were treated with coumarin at different dosages, and after 15 months, 10 animals from each group were evaluated. Treatment with 50 mg/kg led to a significant reduction in the activated partial thromboplastin times and the erythrocyte volume and hemoglobin values, and an increase

Program, 1993a).

be resistant to coumarin-induced toxicity (Lake, 1999)

weights increased following treatment with 150 mg/kg coumarin.

of platelet counts was also observed. With this dose, the values of alanine aminotransferase, sorbitol dehydrogenase, and g-glutamyltransferase significantly increased in males, whereas these effects were observed only from 100 mg/kg in females. Additionally, lesions associated with the administration of coumarin were also observed during the long exposure, which include an increase of the severity of nephropathy, increase of incidences of bile duct and parathyroid gland hyperplasias, increase of the incidences of ulcers, and necrosis, fibrosis, and cytologic alterations in the liver (National Toxicology Program, 1993b). A carcinogenic potential has also been described for coumarin, especially in the liver and lungs of rats and mice. However, the dose–response relationships are nonlinear with tumor formation and hepatic and pulmonary toxicity are associated only with high doses (Lake, 1999).

Regarding the mutagenic and genotoxic potential of coumarin, it showed weak clastogenic activity in Chinese Hamster ovary cells *in vitro*. However, this response was observed only at a very high concentration (10.95 μM). Negative responses were reported in the *Salmonella typhimurium* assay in the TA98, TA1535, TA1537 and TA1538 strains, either with or without metabolic activation. However, gene mutations have been described in the TA100 strain in the presence of a metabolic activating system (S9) (Lake, 1999).

Using the Ames genotoxic assay, coumarin has not been shown to be a mutagenic agent in the TA100 strain assessed without metabolic activation (liver S9 fraction). With metabolic activation, coumarin produces only a weak positive effect at a high concentration. However, this effect has been widely discussed because a greater response was achieved in the presence of liver S9 fraction from untreated Syrian hamsters than from rats treated with Aroclor 1254, a substance used to stimulate coumarin metabolism by the 3,4-epoxidation pathway in rat hepatic microsomes. This result does not correlate with the extent of coumarin metabolism and coumarin-induced liver injury in these species. Because of the differences in their metabolic pathways, a chronic dose of coumarin induces liver lesions and tumor in the rat and not in the Syrian hamster, which appears to be refractory for coumarin-induced hepatotoxicity (Lake, 1999).

In addition*, in vivo* studies have shown that coumarin is unable to induce sex-linked recessive lethal mutations in germ cells of male *Drosophila melanogaster*. Furthermore, no evidence for coumarin-induced genotoxicity has been observed in the *in vivo* micronucleus test in mouse peripheral blood cells. The conclusion is that coumarin is not DNA-reactive and that the induction of tumors at high doses in rodents is attributed to cytotoxicity and regenerative hyperplasia (Felter et al., 2006).

In humans, the majority of tests for mutagenicity and genotoxicity also suggest that coumarin is not a toxic agent (Lake, 1999). The lack of toxicity has been associated with the detoxification mechanism of coumarin, which in humans involves the 7-hydroxylation pathway, a minor route in rats and mice; these rodents use the 3,4-epoxidation pathway instead, which results in toxic metabolite formation (Gasparetto et al., 2011). Thus, the species-specific target organ toxicity has been attributed to the pharmacokinetics of coumarin metabolism, causing rats to be susceptible to liver effects and mice to have toxicity particularly in the lung. Therefore, it is possible to conclude that the use of rats and mice is not an adequate model to compare the metabolism and toxicity of coumarin with humans, due to their particular metabolism. Because *in vitro* genotoxicity studies demonstrated toxicity only at very high doses and no evidence for *in vivo* studies was observed, it is possible to conclude that there is no human health risk from coumarin exposure in natural dietary sources, such guaco species.

*Mikania glomerata* and *M. laevigata*: Clinical and Toxicological Advances 313

Bertolucci, S. K.; Pereira, A. B.; Pinto, J. E.; Aquino Ribeiro, J. A.; Oliveira, A. B. & Braga, F.

Biavatti, M. W.; Koerich, C. A.; Henck, C. H.; Zucatelli, E.; Martineli, F. H.; Bresolin, T. B. &

Bighetti, A. E.; Antonio, M. A.; Kohn, L. K.; Rehder, V. L.; Foglio, M. A.; Possenti, A.; Vilela,

Block, L. C.; Scheidt, C.; Quintao, N. L.; Santos, A. R. & Cechinel-Filho, V. (1998).

Bolina, R. C.; Garcia, E. F. & Duarte, M. G. R. (2009). Estudo Comparativo da Composição

Bortolomeazzi, R.; Sebastianutto, N.; Toniolo, R. & Pizzariello, A. (2007). Comparative

*Chemistry*, Vol.100, No.4, (November, 2005), pp. 1481-1489, ISSN 0308-8146 Botsaris, A. S. (2007). Plants Used Traditionally to Treat Malaria in Brazil: The Archives of

Brasil (2008). Agência Nacional de Vigilância Sanitária - ANVISA. Instrução Normativa no 5

Bueno, P. C. P. & Bastos, J. K. (2009). A Validated Capillary Gas Chromatography Method

Cabral, L. M.; Santos, T. C. & Alhaique, F. (2001). Development of a Profitable Procedure for

Cavalcanti, B. C.; Bezerra, D. P.; Magalhaes, H. I. F.; Moraes, M. O.; Lima, M. A. S.; Silveira,

Fitoterápicos no Brasil. *Diário Oficial da União*, Brasília - DF.

(January-March, 2009), pp. 218-223, ISSN 0102-695X

No.8, (August, 2009), pp.1097-1104, ISSN 0022-3573

Vol.12, No.1, (January, 2005), pp. 72-77, ISSN 0944-7113

(January/March, 2009), pp. 294-298, ISSN 0102-695X

ISSN 0032-0943

0031-7144

1520-5762

0260-437X

pp. 18, ISSN 1746-4269

pp. 197-200, ISSN 0939-5035

Carrageenan-induced Peritonitis. *Journal of Pharmacy and Pharmacology*, Vol.61,

C. (2008). Development and Validation of an RP-HPLC Method for Quantification of Cinnamic Acid Derivatives and Kaurane-Type Diterpenes in *Mikania laevigata* and *Mikania glomerata*. *Planta Medica*, Vol.75, No.3, (December, 2008), pp. 280-285

Leite, S. N. (2004). Coumarin Content and Physicochemical Profile of *Mikania laevigata* Extracts. *Zeitschrift für Naturforschung*, Vol.59, No.3, (March-April, 2004),

L. & Carvalho, J. E. (2005). Antiulcerogenic Activity of a Crude Hydroalcoholic Extract and Coumarin Isolated from *Mikania laevigata* Schultz Bip. *Phytomedicine*,

Phytochemical and Pharmacological Analysis of Different Parts of *Wedelia paludosa* DC. (Compositae). *Pharmazie,* Vol.53, No.10, (October, 1998), pp. 716-718, ISSN

Química das Espécies Vegetais *Mikania glomerata* Sprengel e *Mikania laevigata* Schultz Bip. ex Baker. *Revista Brasileira de Farmacognosia*, Vol.19, No.1,

Evaluation of the Antioxidant Capacity of Smoke Flavouring Phenols by Crocin Bleaching Inhibition, DPPH Radical Scavenging and Oxidation Potencial. *Food* 

Flora Medicinal. *Journal of Ethnobiology and Ethnomedicine*, Vol.3, No.1, (May, 2007),

de 11 de dezembro de 2008 - Determina a Lista de Registro Simplificado de

for Guaco (*Mikania glomerata* S.) Quality Control and Rastreability: From Plant Biomass to Phytomedicines. *Revista Brasileira de Farmacognosia*, Vol.19, No.1,

the Extraction of 2-H-1-benzopyran-2-one (coumarin) from *Mikania glomerata*. *Drug Development and Industrial Pharmacy*, Vol.27, No.1, (July, 2009), pp. 103-106, ISSN

E. R.; Camara, C. A. G.; Rao, V. S.; Pessoa, C. & Costa-Lotufo, L. V. (2009). Kauren-19-oic Acid Induces DNA Damage Followed by Apoptosis in Human Leukemia Cells. *Journal of Applied Toxicology,* Vol.29, No.7, (October, 2009), pp. 560-568, ISSN
