Toxicology Letters 151 (2004) 481–487

Genotoxic potentiality of aqueous extract prepared from

Chrysobalanus icaco L. leaves
S.C. Ferreira-Machado a , M.P. Rodrigues a , A.P.M. Nunes a ,
F.J.S. Dantas a , J.C.P. De Mattos a , C.R. Silva a , E.G. Moura b ,
R.J.A.C. Bezerra a , A. Caldeira-de-Araujo a,∗
a Departamento de Biofı́sica e Biometria, Universidade do Estado do Rio de Janeiro, Instituto de Biologia Roberto Alcantara Gomes,
Av. 28 de setembro, 87, Rio de Janeiro 20551-030, Brazil
b Departamento de Ciências Fisiológicas, Universidade do Estado do Rio de Janeiro, Instituto de Biologia Roberto Alcantara Gomes,
Av. 28 de setembro, 87, Rio de Janeiro 20551-030, Brazil

Received 23 March 2004; received in revised form 23 March 2004; accepted 24 March 2004

Available online 24 April 2004

Abstract

Plants have been related to our lives, being used as medicine, regardless of scientific evidence of side effects. This work analyses
the toxicological effects of Chrysobalanus icaco L. aqueous extract, used in different pathologies. It was studied through: (i)
alteration of plasmid pUC 9.1 topology; (ii) survival of bacterial strains submitted, or not, to previous treatment with SnCl2 ; (iii)
transformation efficiency of E. coli strain by the treatment with the plasmid pUC 9.1. In (i), the treatment of the plasmid resulted
in DNA single-strand breaks (SSB). A decrease of the lethal effect induced by SnCl2 in presence of the extract was found, while
no C. icaco bacterial survival reduction was observed. The transformation efficiency of the plasmid was also reduced. Results
suggest that the extract could present a potential genotoxic effect, as demonstrated either by the induction of SSB in plasmid or
in transformation efficiency experiments. Finally, it presents an antioxidant action.
© 2004 Published by Elsevier Ireland Ltd.

Keywords: Chrysobalanus icaco; Plasmid pUC 9.1; Escherichia coli; Genotoxic potentiality

1. Introduction side effects. Various practices of traditional medicine

During the last decade, plants have been employed
in the prevention and treatment of diseases. Their use
has considerably increased among populations, as they
are believed to be beneficial and having few significant
∗ Corresponding author. Tel.: +55-21-25876391;
fax: +55-21-22543532.
E-mail address: adriano@uerj.br (A. Caldeira-de-Araujo).
have been developed in different cultures and regions,
based on indigenous experiences, philosophy and
personal attitudes without a parallel development
of international standards and appropriate methods
for evaluating traditional medicine (Rates, 2001;
Ankli et al., 2002; Steenkamp, 2003). The concept
that these products are “natural” is neither ade-
quate nor sufficient to assure safety and efficacy.
In addition, the amount of available information on

0378-4274/$ – see front matter © 2004 Published by Elsevier Ireland Ltd.

doi:10.1016/j.toxlet.2004.03.014
482 S.C. Ferreira-Machado et al. / Toxicology Letters 151 (2004) 481–487
the toxicity and therapeutic properties in the hu-
man organism about several medicinal herbs is still
quite limited. Most of the information does not
have scientific support. Medicinal herbs have their
use as medicines, based only on traditional folk
use, which has been passed on from generation to
generation (Beyerstein, 2001; Talalay, 2001; Ernst,
2002).
Recently, interest has increased on the research of
antioxidant properties of several plants against free
radicals or reactive oxygen species (ROS), to pre-
vent cellular injuries in the human organism. ROS
are generated from both endogenous and exogenous
sources and may be involved in the aetiologies of
several human diseases. ROS, such as hydroxyl rad-
icals (OH• ), produced in the cells, are capable of
damaging important cellular targets, such as mem-
branes and deoxyribonucleic acid (DNA) (Reiniger
et al., 1998; Nakagawa and Yokozawa, 2002; Auddy
et al., 2003; Lee et al., 2003; Sroka and Cisowski,
2003). Dantas et al. (2002) related the toxic effects
of stannous chloride, a powerful reducing agent used
for packing canned food, in dental amalgams and for
preparing 99m Technetium-radiopharmaceuticals, on
the generation of ROS in K562 cell line.
Emphasis is given, in this paper, to the poorly
studied medicinal plant species Chrysobalanus
icaco Linnaeus (C. icaco), also known as abajeru,
coco-plum and many other common names. This
is a medium-sized shrub, native to coastal areas of
the American continent, belonging to the Chrysobal-
anaceae family (Dahlgren, 1980). Consumption of
C. icaco as a tea prepared with various parts of the
plant has been thoroughly used, in folk medicine, for
the control of several pathologies such as leucorrhea,
haemorrhages and chronic diarrhea. The species is
also known to bring forth diuretic, hypoglycemic and
antiangiogenic effects (Costa, 1977; Vargas-Simon
et al., 1997; Paulo et al., 2000). In the literature, there
is at least one publication showing the induction of
apoptosis by triterpenoids obtained from methanol
extract of C. icaco leaves (Fernandes et al., 2003).
The abusive use of the extract of C. icaco leaves and
the poor knowledge of its therapeutic and side effects
emphasize the need for a deeper study, including the
evaluation of its toxicity in several biological systems.
The purpose of this work was to examine the relative
antioxidant activity and the genotoxic potential from
C. icaco aqueous extract, as it is frequently used by
the population. This research was performed in three
experimental models: (i) agarose gel electrophoresis
analysis of structural conformation changes of plasmid
pUC 9.1, previously treated with different concentra-
tions of the extract; (ii) survival of Escherichia coli
(E. coli) strains treated with aqueous extract, alone or
simultaneously with stannous chloride; and (iii) the
capacity of the treated plasmids in transforming com-
petent E. coli AB1157 (wild-type) cells.
2. Material and methods
2.1. Plant material and reagents
C. icaco was collected at a sand bank area in Cabo
Frio, Rio de Janeiro state, Brazil, from October to
November 2002, between 9 and 10 a.m., the leaves
being processed all at once to obtain all the extract
used during this work. The plant was classified and
deposited in the Herbarium Bradeanum, at the Rio
de Janeiro State University under registration num-
ber 85422. Decoctions were prepared in the traditional
way: 10 g of leaves were boiled in 1 L of water on
slow heat for 5 min and cooled at room temperature
(25 ◦ C). Then, the supernatant was filtered with a pa-
per filter. Finally, the filtered was lyophilized, being
stored at −20 ◦ C, until further use. The dry extract
obtained was suspended either in Mili-Q water (Milli-
pore Corp., Bedford, MA, USA), in order to proceed
with DNA electrophoresis and bacterial transforma-
tion experiments, or solved in 0.9% NaCl solution in
order to run the bacterial inactivation assays. After-
wards, it was sterilized by filtration through 0.22 ␮m
Millipore membrane.
Stannous chloride (SnCl2 ·2H2 O) obtained from
Sigma Chemicals Co., USA, was employed as
positive control to induce DNA strand breaks
(Caldeira-de-Araújo et al., 1996; Dantas et al.,
1999; De Mattos et al., 2000). The SnCl2 solution
(200 ␮g/mL) was freshly prepared in Mili-Q water
before each experiment.
2.2. Bacterial strains and plasmid
DNA repair wild-type E. coli K12 AB1157
(Howard-Flanders and Theriot, 1966), PQ35 (rfa)
 S.C. Ferreira-Machado et al. / Toxicology Letters 151 (2004) 481–487
and PQ37 (rfa and uvrA) (Quillardet and Hofnung,
1985) strains were used. E. coli DH5␣F’Iq (rec−)
(Hannanah, 1983) strain hosted plasmid pUC 9.1.
This plasmid, carrying an ampicillin resistance gene,
was obtained through alkaline lysis method, as de-
scribed in Sambrook et al. (1989). Plasmid samples
were further purified from high molecular weight
RNA contaminants, performing LiCl precipitation
(2.5 M final concentration), while the residual RNA
contaminants were digested by RNAse (20 ␮g/mL)
treatment for 30 min at room temperature (25 ◦ C).
2.3. DNA treatment
To evaluate the action of C. icaco aqueous ex-
tract in DNA topology, plasmid pUC 9.1 (200 ng per
tube) was incubated either with increasing concen-
trations of the extract (0.7, 1.75, 3.5 and 7 mg/mL)
or with SnCl2 solution 200 ␮g/mL (positive control)
in 10 mM Tris–HCl buffer at pH 7.4. In all cases,
the reaction mixture was incubated at room temper-
ature for 40 min. After incubation, each sample was
mixed with loading buffer (0.25% xylene cyanol FF;
0.25% bromophenol blue; 30% glycerol in water),
and applied in 0.8% agarose horizontal gel elec-
trophoresis chamber in Tris acetate–EDTA buffer at
pH 8.0 and run at 6 V/cm. Here, the electrophoresis
assay was performed to separate different structural
conformations of plant extract-treated plasmid DNA:
form I supercoiled (SC) native conformation, form II
open circle (OC), resulting from single-strand breaks,
and form III linear (L), resulting from double-strand
breaks. The gel was stained with ethidium bromide
(0.5 ␮g/mL) and the DNA bands were visualized
by fluorescence under an ultraviolet (UV) transil-
luminator system. The assay was repeated, at least
five times, the best result being digitalized (Kodak
Digital Science 1d, EDAS 120), and the bands quan-
tified through the Gel-Pro Analyzer 3.0 computer
program.
2.4. Statistical analysis of DNA strand breaks
ANOVA followed by the Student Newman Keuls
multiple comparison test through the statistical pro-
gram InStat version 3.01 (GraphPad Software, San
Diego, CA, USA) was used to compare the results ob-
tained from the different extract concentrations against
483
the control (not treated). A significance level of 5%
was adopted to compare data.
The data collected by densitometry provided us with
the null events percentage (no breaks = p(0; µ)) for
each one of the aqueous extract concentrations tested.
Thus, the mean values of breaks per genome for each
one of the concentrations, using Poisson distribution
were obtained as follows (Remington and Shor, 1985):
µ = −ln p(0; µ)
2.5. Bacterial inactivation
E. coli AB1157, PQ35 and PQ37 exponentially
growing cultures in LB medium (Miller, 1992) were
centrifuged, washed twice in 0.9% NaCl solution
and suspended in the same solution. Samples of this
suspension were incubated with the aqueous extract
(0.7, 3.5 and 7 mg/mL), or SnCl2 (25 ␮g/mL) or
both agents, for 60 min at 37 ◦ C. At 20 min intervals,
aliquots were diluted and spread onto glass Petri
dishes containing solidified LB medium (1.5% agar).
Colonies were counted after overnight incubation at
37 ◦ C, and the survival fractions (N/No) determined
as the mean of three isolated experiments. Standard
deviations did not exceed 10%.
2.6. Bacterial transformation
Competent bacteria (E. coli AB1157), sensitive to
ampicillin, were obtained as described by Sambrook
et al. (1989). The transformation efficiency of plasmid
DNA, previously treated with different aqueous ex-
tract C. icaco concentrations (0.7, 3.5 and 7 mg/mL),
was then analysed. Following, 200 ␮L of competent
bacterial suspensions (∼109 cells/mL) were mixed
with pretreated plasmid DNA (5 ng) and incubated
on ice for 30 min. Next, the tubes were placed in
water-bath at 42 ◦ C for 2 min and immediately in-
cubated on ice for 2 min. Afterwards, 500 ␮L of
pre-warmed LB (37 ◦ C) were added. The samples
were then incubated for 1 h at 37 ◦ C and aliquots
were withdrawn and plated onto LB-agar, supple-
mented with 60 ␮g ampicillin/mL. Plates were incu-
bated at 37 ◦ C for 24 h and the number of colonies
was counted. A mean value from each C. icaco ex-
tract concentration employed was obtained, and the
transformation efficiency was expressed as the per-
484 S.C. Ferreira-Machado et al. / Toxicology Letters 151 (2004) 481–487

Fig. 1. A percentage of 0.8 agarose gel electrophoresis and densitometric analysis of plasmid pUC 9.1 (200 ng) treated with increasing
concentrations of C. icaco or with SnCl2 as a positive control. Aliquots of pUC 9.1 plasmid DNA (200 ng) were incubated with different
concentrations of C. icaco extract for 40 min at room temperature (25 ◦ C). Each sample was mixed with loading buffer and submitted to 0.8%
agarose gel electrophoresis. (A) The experiment was performed, at least, five times and the best photo is presented here. (B) Densitometric
measures were obtained from the gel, through Gel Pro Analyser computer program. Lanes: (1) pUC 9.1; (2) SnCl2 (200 ␮g/mL); (3)
0.7 mg/mL; (4) 1.75 mg/mL; (5) 3.5 mg/mL and (6) 7 mg/mL of the extract plant.

centage of the maximal value (100%), attributed to
the control (not treated). This experiment was carried
out in triplicate and the results presented are the mean
average. The standard deviations did not exceed 15%
of the respective mean values. Data were analysed
by using ANOVA, followed by the Student Newman
Keuls multiple comparison test through the statisti-
cal program InStat version 3.01 (GraphPad Software,
San Diego, CA, USA). To evaluate the transformation
efficiency significance, a significant level of 5% was
adopted to compare these data.
0.7–7.0 mg/mL. This result could classify the C. icaco
extract as a weak strand break inducer. This genotoxic
effect was confirmed by densitometry analysis. The
open circle and supercoiled forms were analysed and
the result suggests that, regarding the control, there
was a significative (P < 0.001) increase in the num-
ber of lesions when plasmid DNA was treated with C.
icaco, without be observed the linear form (III).
0,45
0,4
0,35
SSB/Genome

0,3
0,25
3. Results and discussion 0,2
0,15
0,1
The conformational changes in pretreated plasmid
0,05
DNA were analysed through agarose gel electrophore- 0
sis. Our results showed that plasmid DNA had their 0 0,5 1 1,5 2 2,5 3 3,5 4 4,5 5 5,5 6 6,5 7 7,5 8

[mg/mL] C. icaco extract

supercoiled form (I) changed to relaxed open circular
form (II) after treatment with increasing concentra-
tions of extract plant (from 0.7 to 7 mg/mL) (Fig. 1).
Thus, this extract was able to induce single-strand
breaks, as can be seen in Fig. 2. On the other hand,
a dose–response pattern was not obtained (Fig. 2),
because there was no significative difference (P >
0.05) in the number of breaks induced at the interval
Fig. 2. Number of single-strand breaks/genome in plasmid DNA
treated with C. icaco extract. Analysis of densitometric data per-
formed by the Gel Pro Analyser computer program, gave the null
events percentage (no breaks) for each one of the C. icaco extract
concentrations tested. Using the Poisson distribution, the mean
value of breaks for each one of the concentrations, was obtained as
follows µ = −ln p(0; µ). Number of single-strand breaks/genome
values are the means of five isolated experiments.
 S.C. Ferreira-Machado et al. / Toxicology Letters 151 (2004) 481–487
10
Survival Fraction
1 0,01
0,1
0 20 40
treatment time (min)
Fig. 3. Effect of C. icaco extract on the survival of E. coli AB1157.
Exponential growing cells, suspended in 0.9% NaCl solution, were
treated with C. icaco for different incubation times. The survival
fractions were determined: (䉬) 0.9% NaCl; (䊐) 0.7 mg/mL C.
icaco extract; ( ) 3.5 mg/mL C. icaco extract; (䊊) 7 mg/mL C.
icaco extract.
The inactivation induced in different DNA repair
proficient E. coli strains by various concentrations of
the aqueous extract was also evaluated. As it can be
observed (Figs. 3–5), there was no important decrease
in the survival fractions, either in the wild-type or in
the repair mutant tested (uvrA). Also, an important
decrease of the survival levels was not observed in
other repair mutant E. coli strains (AB2463, recA;
BW9091, xthA; BW544, xthA, nfo; BW535, nth, nfo,
xthA; GY5022, envA; and GY5027, envA, uvrB) (data
not shown).
These results could initially suggest the absence of
toxicity of this extract, but we cannot discard the fact
that the substances, present in extract composition,
did not enter in sufficient quantity into the bacterial
cells to cause damage to the genetic material. To elu-
cidate this point, 5 ng of the treated plasmid fraction
10
Survival Fraction
1
(N/No)
0,1
0,01
0,001
0,0001
0 30
Treatment Time (min)
Fig. 4. Effect of C. icaco extract on the inactivation induced by
stannous chloride on E. coli PQ35 (rfa). Exponential growth cells
suspended in 0.9% were treated with stannous chloride (25 ␮g/mL)
for different incubation times in the presence or absence of the
extract (7 mg/mL). The survival fraction was determined: (䉬) 0.9%
NaCl; (䊐) SnCl2 ; ( ) C. icaco extract; (䊊) SnCl2 + C. icaco
extract.
485
Survival Fraction
10
1
0,1
0,001
0,0001
0,00001
60 0 30 60
Time (min)
Fig. 5. Effect of C. icaco extract on the inactivation induced
by stannous chloride on E. coli PQ37 (rfa and uvrA). Expo-
nential growth cells suspended in 0.9% were treated with SnCl2
(25 ␮g/mL) for different incubation times in the presence or ab-
sence of the extract (7 mg/mL). The survival fraction were deter-
mined: (䉬) 0.9% NaCl; (䊐) SnCl2 ; ( ) C. icaco extract; (䊊)
SnCl2 + C. icaco extract.
was used to transform wild-type E. coli AB1157 com-
petent cells (ampR), as described in Section 2. Thus,
the genotoxic ability of this extract was observed in
this assay (Fig. 6). In fact, there was a decrease in
the transformation efficiency of plasmid DNA treated
with the extract, when compared to the control and
among them (P < 0.05). In this system, the ability of
a transformed wild-type bacterial cell to grow in se-
lective media is a result of its capacity to repair plas-
mid DNA induced lesions. Thus, transformed bacteria
were incubated for 1 h in LB medium, without ampi-
cillin. So, bacteria were allowed to recover the plasmid
DNA from the lesions to further encode ␤-lactamase
plasmid gene that would enable bacteria to replicate
in solid LB containing ampicillin medium (Hannanah,
120
Transformation
100
efficiency %
80
60
40
20
0
60 1 2 3 4 5
Treatment
Fig. 6. Transformation efficiency of plasmid pUC 9.1 onto
wild-type E. coli AB1157. Aliquots of plasmid pUC 9.1 treated
with increasing concentrations of C. icaco extract plant and SnCl2
solution 200 ␮g/mL (positive control) were mixed to the competent
bacterial suspension and submitted to transformation. Values are
the mean of three isolated experiments: (1) pUC 9.1; (2) SnCl2 ; (3)
0.7 mg/mL; (4) 3.5 mg/mL and (5) 7.0 mg/mL of C. icaco extract.
486 S.C. Ferreira-Machado et al. / Toxicology Letters 151 (2004) 481–487
1983). From these results, we could suggest that the
decrease in transformation efficiency could be due to
the inability of the bacteria to repair plasmid DNA
whenever augmented concentrations of C. icaco aque-
ous extract are employed. On the other hand, hidden
lesions should also be generated, but could not be de-
tected to the extent of the methodology employed in
this work. This could explain why a dose response pat-
tern in the transformation efficiency experiments was
observed (Fig. 6), but not in the electrophoresis as-
says, in which there was only the DNA being injured
(Figs. 1–2). However, these results may reflect differ-
ent transformation efficiencies, due to the difficulty of
the damaged plasmids enter into the cell, rather than
decreasing in repair ability of lesioned plasmids.
Natural products have been used as antioxidants
against reactive oxygen species generated by our own
metabolism. To test whether C. icaco had an antioxi-
dant effect, the strains PQ35 (rfa) and PQ37 (rfa and
uvrA) were incubated with SnCl2 (25 mg/mL) in the
presence, or not, of the extract (0.7 mg/mL). The strain
PQ35, when compared to the deficient in the excision
repair (PQ37), presented a lower inactivation, when
treated only with SnCl2 (Figs. 4–5). The association
of the extract with SnCl2 promoted a reduction in the
lethality in both strains, when compared to the treat-
ment with SnCl2 alone. This cellular protection was
more important in the wild-type, demonstrating that
the product of the gene uvrA, absent in the PQ37
strain, is important to repair the damage directly or
indirectly induced by SnCl2 . These results lead us to
verify if the extract could also avoid the breaks in-
duction, by stannous chloride, on the plasmid DNA.
Unexpectedly, there was the formation of a molecular
aggregated, blocking the DNA migration through the
agarose gel (data not shown).
The results, at the first sight, seem contrasting since
some indicate a toxic potential, while another suggests
a protective effect against the damage induced by the
metal. The lethal effect can be associated with a frac-
tion, or a combination of fractions, of the plant. In this
way, it is now known from the literature (Fernandes
et al., 2003) that leaves from C. icaco contain some
triterpenoids, capable of inducing apoptosis in K562
erythroleukemia cells. On the other hand, the com-
pounds in the crude extract could: (i) chelate stannous
ions, protecting them against oxidation and preventing
the generation of ROS; (ii) acting as scavengers of free
radicals generated by the SnCl2 . Similarly, Melo et al.
(2001) demonstrated that the lethal effect of stannous
chloride was reduced by the extract of Cymbopogon
citratus, Maytenus ilicifolia and Baccharis genistel-
loides. Thus, these data indicate that DNA is an im-
portant target to the C. icaco extract effects, while the
protective action induced against the metal lethal ef-
fect is probably due to unspecific activities such as:
scavenging of stannous ions prior to it uptake in E.
coli.
Finally, the results presented in this work ratify the
need for further toxicological studies of substances
taken for granted as innocuous, which have been used
by the population without medical prescriptions.
Acknowledgements
This research was supported by grants from Capes,
UERJ, CNPq and FAPERJ.
References
Ankli, A., Heinrich, M., Bork, P., Wolfram, L., Bauerfeind, P.,
Brun, R., Schmid, C., Weiss, C., Bruggisser, R., Gertsch, J.,
Wasescha, M., Sticher, O., 2002. Yucatec Mayan medicinal
plants: evaluation based on indigenous uses. J. Ethnopharmacol.
79, 43–52.
Auddy, B., Ferreira, M., Blasina, F., Lafon, L., Arredondo,
F., Dajas, F., Tripathi, P.C., Seal, T., Mukherjee, B.,
2003. Screening of antioxidant activity of three Indian
medicinal plants, traditionally used for the management of
neurodegenerative diseases. J. Ethnopharmacol. 84, 131–138.
Beyerstein, B.L., 2001. Alternative medicine and common errors
of reasoning. Acad. Med. 76 (3), 230–237.
Caldeira-de-Araújo, A., Dantas, F.J.S., Moraes, M.O.,
Felzenszwalb, I., Bernardo-Filho, M., 1996. Stannous chloride
participates in the generation of reactive oxygen species. J.
Brazil. Assoc. Adv. Sci. 48, 109–113.
Costa, O.A., 1977. Plantas hipoglicemiantes brasileiras II. Leandra
6–7 (7), 63–75.
Dahlgren, R.M.T., 1980. A revised system of classification of the
angiosperms. Bot. J. Linn. Soc. 80 (2), 91–124.
Dantas, F.J.S., Moraes, M.O., De Mattos, J.C.P., Bezerra, R.J.A.C.,
Carvalho, E.F., Bernardo-Filho, M., Caldeira-de-Araújo, A.,
1999. Stannous chloride mediates single strand breaks in
plasmid DNA. Toxicol. Lett. 110, 129–136.
Dantas, F.J.S., De Mattos, J.C.P., Viana, M.E., Lage, C.A.S.,
Cabral-Neto, J.B., Leitão, A.C., Bernardo-Filho, M., Bezerra,
R.J.A.C., Carvalho, J.J., Caldeira-de-Araújo, A., 2002.
Genotoxic effects of stannous chloride (SnCl2 ) in K562 cell
line. Food Chem. Toxicol. 40, 1493–1498.
 S.C. Ferreira-Machado et al. / Toxicology Letters 151 (2004) 481–487
De Mattos, J.C.P., Dantas, F.J.S., Bezerra, R.J.A.C., Bernardo-
Filho, M., Cabral-Neto, J.B., Lage, C.A.S., Leitão, A.C.,
Caldeira-de-Araújo, A., 2000. Damage induced by stannous
chloride in plasmid DNA. Toxicol. Lett. 116, 159–
163.
Ernst, E., 2002. The risk-benefit profile of commonly used herbal
therapies: ginkgo, St. John’s Wort, ginseng, echinacea, saw
palmetto, and kava. Am. College Phys.: Am. Soc. Intern. Med.
136, 42–53.
Fernandes, J., Castilho, R.O., Costa, M.R., Wagner-Souza, K.,
Kaplan, M.A.C., Gattass, C.R., 2003. Pentacyclic triterpenes
from Chrysobalanaceae species: cytotoxicity on multidrug
resistant and sensitive leukemia cell lines. Cancer Lett. 190,
165–169.
Hannanah, D., 1983. Study on transformation of Escherichia coli
with plasmids. J. Mol. Biol. 166, 557–567.
Howard-Flanders, P., Theriot, L., 1966. Mutants of Escherichia
coli K12 defective in DNA repair and genetic recombination.
Genetics 53, 1137–1150.
Lee, S.E., Hwang, H.J., Ha, J., Jeong, H., Kim, J.H., 2003.
Screening of medicinal plant extracts for antioxidant activity.
Life Sci. 73, 167–179.
Melo, S.F., Soares, S.F., Costa, R.F., Silva, C.R., Oliveira, M.B.N.,
Bezerra, R.J.A.C., Caldeira-de-Araujo, A., Bernardo-Filho, M.,
2001. Effect of the Cymbopogon citratus, Maytenus ilicifolia
and Baccharis genistelloides extracts against the stannous
chloride oxidative damage in Escherichia coli. Mutat. Res.
496, 33–38.
Miller, J.H., 1992. A short course in bacterial genetics. A
Laboratory Manual and Handbook for Escherichia coli and
Related Bacteria. Cold Spring Harbor Laboratory Press, Cold
Spring Habor, NY.
Nakagawa, T., Yokozawa, T., 2002. Direct scavenging of nitric
oxide and superoxide by green tea. Food Chem. Toxicol. 40,
1745–1750.
487
Paulo, S.A., Balassiano, I.T., Silva, N.H., Castilho, R.O., Kaplan,
M.A.C., Cabral, M.C., Carvalho, M.G.C., 2000. Chrysobalanus
icaco L. extract for antiangiogenic potential observation. Int.
J. Mol. Med. 5 (6), 667–669.
Quillardet, P., Hofnung, M., 1985. The SOS Chromotest, a
colorimetric bacterial assay for genotoxins: procedures. Mutat.
Res. 147, 65–78.
Rates, S.M.K., 2001. Plants as source of drugs. Toxicon 39, 603–
613.
Reiniger, I.W., Silva, C.R., Felzenszwalb, L., de Mattos, J.C.P.,
Oliveira, J.F., Dantas, F.J.S., Bezerra, R.J.A., Caldeira-de-
Araújo, A., Bernardo-Filho, M., 1998. Boldine action against
the stannous chloride effect. J. Ethnopharmacol. 68, 345–348.
Remington, R.D., Shor, M.A., 1985. Some important distributions.
In: Johnson, R.A., Wichern, D.W. (Eds.), Statistics with
Applications to the Biological and Health Sciences. Prentice-
Hall, Englewood Cliffs, NJ.
Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Extraction and
purification of plasmid DNA. In: Molecular Cloning. A
Laboratory Manual. Cold Spring Harbour Laboratory, Cold
Spring Harbour, New York.
Sroka, Z., Cisowski, W., 2003. Hydrogen peroxide scavenging,
antioxidant and anti-radical activity of some phenolic acids.
Food Chem. Toxicol. 41, 753–758.
Steenkamp, V., 2003. Traditional herbal remedies used by
South African women for gynaecological complaints. J.
Ethnopharmacol. 86, 97–108.
Talalay, P., 2001. The importance of using scientific principles in
the development of medicinal agents from plants. Acad. Med.
76 (3), 238–247.
Vargas-Simon, G., Arellano-Ostoa, G., Garcia-Villanueva, E.,1997.
Propagación por estacas com hojas de icaco (Chrysobalanus
icaco) y anatomı́a del enraizamiento. In: Proceedings of
the International Society for Tropical Horticulture, vol. 41,
pp. 264–269.