Journal of Ethnopharmacology 155 (2014) 1156–1163

Contents lists available at ScienceDirect

Journal of Ethnopharmacology
journal homepage: www.elsevier.com/locate/jep

Anti-Helicobacter pylori, gastroprotective, anti-inflammatory, and

cytotoxic activities of methanolic extracts of five different populations
of Hippocratea celastroides collected in Mexico
Wendy Itzel Escobedo Hinojosa a, Macdiel Acevedo Quiróz a, Irma Romero Álvarez b,
Patricia Escobar Castañeda c, María Luisa Villarreal a,n, Alexandre Cardoso Taketa a,n
a
Centro de Investigación en Biotecnología. Universidad Autónoma del Estado de Morelos. Av. Universidad 1001. Col. Chamilpa. Cuernavaca 62209,
Morelos, México
b
Departamento de Bioquímica. Facultad de Medicina. Universidad Nacional Autónoma de México. Ciudad Universitaria, México 04510, D.F. México
c
Facultad de Ciencias Biológicas. Universidad Autónoma del Estado de Morelos, Morelos, México

art ic l e i nf o a b s t r a c t

Article history: Ethnopharmacological relevance: The root barks of Hippocratea celastroides have been used for decades in
Received 1 April 2014 Mexican traditional medicine to treat gastritis and ulcers. To investigate the anti-Helicobacter pylori,
Received in revised form gastroprotective, anti-inflammatory, and cytotoxic activities of methanolic extracts obtained from the
16 June 2014
leaves, stems, and root bark of Hippocratea celastroides collected in five different localities in Mexico,
Accepted 17 June 2014
Available online 24 June 2014
during the winter of 2009, in order to establish differences in biological activities in terms of plant
organs, as well as places of collection.
Keywords: Materials and methods: Whole individuals were collected in five separate localities in Mexico: La Mancha,
Hippocratea celastroides Veracruz (VL), Yautepec, Morelos (MY), Jojutla, Morelos (MJ), Temalac, Guerrero (GT), and Landa de Matamoros,
Helicobacter pylori
Querétaro (QL). Methanolic crude extracts from wild plant specimens were tested using in vivo ethanol-
Inflammation
induced mice gastric ulcer model, and 12-O-tetradecanoylphorbol-13-acetate (TPA) induced ear edema in mice
Gastroprotective
Cytotoxicity assay, and in vitro anti-Helicobacter pylori model, and carcinoma cell line cytotoxic assays.
Mexican traditional medicine Results and conclusions: The leaves, stems, and root bark from MY specimens, as well as the leaves and root
bark of materials from VL, presented the highest activity against Helicobacter pylori (MIC values ranging from
7.81 to 31.25 μg/ml). Most gastroprotective effects were displayed by the leaves of plants collected in MY, with
89.8571.91% of protection (300 mg/kg) and an ED50 ¼ 27 mg/kg, which was corroborated by histological
analysis. The root bark extracts from MY achieved the highest edema inhibition values (ED50 ¼0.18 mg/ear),
which were comparable to indomethacin (ED50 ¼0.16 mg/ear). Finally, all extracts from MY (three plant parts)
were cytotoxic against nasopharyngeal (KB), breast (MCF-7), and colon (HCT-116) carcinoma cell lines with IC50
values between 1.18 and 9.77 μg/ml, except that no activity was detected for root bark extracts against HCT-116
normal fibroblasts. The activities of methanolic extracts from leaves, stems and root bark of plants collected in
five different locations varied considerably, representing a notable problem facing the quality control of the
plant material from Hippocratea celastroides used for medicinal purposes. The ethnomedical information of this
plant in regards to treating gastritis and ulcers was strongly evidenced by the findings of the experimental
models employed in this study.
& 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction drink in place of ordinary water (Monroy-Ortiz and Castillo-

Hippocratea celastroides HBK (Hippocrateaceae) is used in
Mexican traditional medicine for the treatment of gastric disor-
ders, inflammation or infection. To prepare the medication, a small
handful of the root bark is boiled in 1.5 liter of water and then
n
Corresponding authors. Tel.: þ 52 777 329 7057; fax: þ52 777 329 7030.
E-mail addresses: luisav@uaem.mx (M.L. Villarreal),
ataketa@uaem.mx (A.C. Taketa).
España, 2007). This plant is known popularly as “hierba del piojo”
(louse´s herb), “matapiojo” (louse killer), as well as “ixcate-
blanco,” which is a compound name derived from the Náhuatl
and Spanish languages that means “cotton-white” and refers to
the cobweb aspect of the crushed bark of the congeneric species
Hippocratea excelsa (Carranza-González, 2001). Hippocratea celas-
troides is a liana that has a widespread distribution in the country
and has a range in Central America and in South America up to the
northern part of Argentina (Castillo-Campos and Medina-Abreo,
2005). In Mexico, the decoction of the root bark, combined with

http://dx.doi.org/10.1016/j.jep.2014.06.044
0378-8741/& 2014 Elsevier Ireland Ltd. All rights reserved.
 W.I. Escobedo Hinojosa et al. / Journal of Ethnopharmacology 155 (2014) 1156–1163 1157

Malva sylvestris, is used to release stomach infections (Monroy-
Ortiz and Castillo-España, 2007). In the state of Guerrero, Mexico,
the traditional use of this plant is to treat peptic ulcers. The
ethnopharmacological uses of Hippocratea celastroides as an insec-
ticide have also been documented, and in one study, dimers of the
triterpene celastroide A inhibited the feeding of 88.7% of the
stored grain pest insect Sitophyllus zeamay but only increased
the mortality in 2%, indicating low insect toxicity (Jiménez-Estrada
et al., 2000; Reyes-Chilpa et al., 2003). In addition, triterpenes
derived from friedelin have been reported in the leaves, and the
alditol galactitol has been detected in the roots of this plant
(González et al., 1989).
Because of the high demand for new products for the treatment
of ulcers, especially against the persistent Helicobacter pylori, we
performed a pharmacological screening study to validate the
ethnomedical uses of the root bark and other organs, such as the
stems and leaves of H. celastroides. Furthermore, in this study, we
analyzed and compared the biological activities of five different
plant populations with the aim of determining the pharmacologi-
cal variability among these populations, which could allow proper
standardization processes for Hippocratea celastroides in the near
future.
2. Material and methods
in small pieces using gardening shears. The material was left to dry
in the shade at room temperature until constant weight. After this,
the dried material was ground in an electric mill (Krups, F408) to
afford a powder. Pulverized dried plant material (4 g), which
included five populations with their three plant parts (leaves,
stems and root bark), was separately processed (a total of 15
samples). Each sample was mixed with 80 ml of MeOH (J.T. Baker),
agitated, sonicated for 15 min. The mixture was filtered at room
temperature through a Whatman filter paper, grade 2 (Kent, UK).
Each sample was processed to complete a series of five consecu-
tive extractions. All five extracts were combined and evaporated in
a rotavapor (Büchi, R-124) at 40 1C until dryness.
2.3. Pharmacological assays
All in vivo pharmacological experiments were conducted in
mice according to the Mexican Guidelines for Animal Welfare
NOM-Z00-062-1999 (2013) and were approved by the Ethics and
Security Committee from the Centro de Investigación en Biotec-
nología, UAEM. Male ICR mice were housed in groups of 5 per
cage, for at least 1 week prior to the experiments. Food and water
were available ad libitum, and the animals were conditioned to a
12 h light/dark cycle, at a temperature of 227 1 1C.

2.1. Plant material

Fresh plant samples were collected in five different locations in
Mexico during the winter of 2009 (January–April). A total of six
individuals were collected per locality (Table 1). The abbreviations
“VL, MY, MJ, GT and QL” were used to identify those locations.
Different anatomical parts of the plants, leaves (l), stems (s), and
root bark (r), were subjected to pharmacological assays. The
abbreviations used in this work to identify the plant samples are
given by three letters. The first two letters express the locality and
the third the plant part, e.g., “VLl” indicates a leaf sample collected
in La Mancha, Veracruz. Voucher specimens were authenticated by
Rolando Ramírez and then deposited at the HUMO Herbarium,
CEAMISH (Centro de Educación Ambiental e Investigación Sierra
de Huautla) UAEM.
2.2. Preparation of the methanolic crude extracts for the
pharmacological assays
Plant parts were separated in leaves, stems and root bark. In
order to facilitate the drying process, stems and root bark were cut
2.3.1. Gastroprotective assay (ethanol-induced ulcer)
The gastroprotective activity of the methanolic crude extracts
from all localities was evaluated using an ethanol-induced gastric
ulcer model in ICR mice (30–35 g). Food was withdrawn 24 h
before the experiment, and the mice were randomly divided into
three groups that received the following solutions through an
orogastric tube: group 1, 1 ml of vehicle (10% Tween 80; Sigma, in
saline solution); group 2, 10 mg/kg omeprazole (Liomont Lab.);
and group 3, methanolic crude extracts at a dose of 300 mg/kg.
After 45 min, all groups were orally treated with absolute EtOH
(J.T. Baker) (5 ml/kg) for gastric lesion induction. After 45 min, the
animals were sacrificed by cervical dislocation. The stomachs were
excised and inflated by injection of saline solution (1 ml), fixed in
5% formalin for 30 min, and opened along the greater curvature.
An ulcer index (UI), based on the measurement of the area (mm2)
of the gastric lesion produced, was adopted to express the results
(Skliar and Bucciarelli, 2007). All treatments were performed in
triplicate. The ED50 value for the most active extract was obtained
through a curve constructed with four concentrations (10, 30, 100
and 300 mg/kg) and with five replicates for each concentration.

Table 1
General data for collections, and methanolic extracts yields of leaves, stems and root bark from different populations of Hippocratea celastroides.

Locality GPS coordinates Date of collection Voucher number Plant part Key MeOH extract yields
(2009) (%, w/w)

Jojutla, Morelos state 18136´N, 99110´W January 20 26446 Leaves MJl 9.0
Stems MJs 6.3
Root bark MJr 4.0
La Mancha, Veracruz state 19136´N, 96122´W March 3 26568 Leaves VLl 14.0
Stems VLs 7.3
Root bark VLr 3.8
Yautepec, Morelos state 18151´N, 9915´W March 13 26447 Leaves MYl 8.1
Stems MYs 7.0
Root bark MYr 7.1
Temalac, Guerrero state 18106´N, 98157´O March 26 26445 Leaves GTl 21.4
Stems GTs 6.9
Root bark GTr 6.0
Landa de Matamoros, Querétaro state 21110´N, 99119´O April 17 26444 Leaves QLl 21.0
Stems QLs 12.7
Root bark QLr 15.6
1158 W.I. Escobedo Hinojosa et al. / Journal of Ethnopharmacology 155 (2014) 1156–1163
The gastroprotection percentage was calculated by the follow-
ing formula:
 
UI control  UI treated
Gastroprotection ð%Þ ¼ x100:
UI control
2.3.2. 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced ear
edema in mice
The topical anti-inflammatory activity of the methanolic crude
extracts was studied using mice ear edema produced by TPA
(Sigma) as a phlogistic agent. Edema was induced on the right
ear by application of 10 μl of TPA at 0.25 mg/ml. The left ear was
selected as negative control and received 20 μl of absolute EtOH.
After 10 min, a volume of 20 μl containing 0.25 mg of crude
extract was applied to the right ear. A punch biopsy of 7 mm
diameter from the right auditory pinna was weighed four hours
after induction of inflammation. The edema was measured by the
weight difference between both ears. The crude extract ear edema
inhibition is expressed as the percentage of inhibition determined
by comparing the phlogogen-applied ears without and with
extract treatment (Puigneró and Queralt, 1997) and using the
following equation: % Inhibition ¼ [(A  B)/A]  100, where A ¼
edema from group treated with TPA, and B ¼ edema from the
group treated with TPA and crude extract. Each crude extract was
assayed in triplicate, as well as the positive control indomethacin
(0.25 mg/ml). The determination of the ED50 for the most active
extract was obtained through a curve constructed with four
concentrations (0.05, 0.1, 0.5 and 1 mg/ear), with five replicates
for each concentration.
2.3.3. in vitro anti-Helicobacter pylori assay
Helicobacter pylori standard strain ATCC 43504 was used. This
strain has the virulence factors vacA s1a/m1 (vacuolating toxin), and
cagA (cytotoxin-associated gene product). Stock cultures were stored
at  70 1C in Brucella broth with 10% fetal bovine serum (GIBCO BRL)
and 10% glycerol. The bacteria were grown on a Casman agar base
(BBL) Petri dishes supplemented with 5% defibrinated sheep blood
and 10 μg/ml vancomycin (Fluka) for 24 h at 37 1C under micro-
aerophilic conditions (10% CO2). At the end of each experiment
viable counts were performed, and after 5 days of incubation, the
purity and uniformity of the colonies morphology was always
checked (small 0.5–1 mm translucent colonies).
Helicobacter pylori was routinely identified using standard
criteria: Gram staining (Gram negative), biochemical tests (urease,
oxidase and catalase positive), and PCR amplification (by using
ureA, and cagA primers).
2.3.3.1. MIC determinations. MIC was determined in the methanolic
extracts of broth cultures by using the broth dilution method. The
extracts were dissolved in DMSO (Reasol) to obtain final
concentrations of 7.8, 15.6, 31.2, 62.5, 125, 250, and 500 μg/ml.
These dilutions (in a volume of 10 μl DSMO) were added to 1.5 ml of
Helicobacter pylori broth culture at the beginning of the exponential
growth phase ( 108 CFU/ml, determined by viable count). ΔA660
was determined after maintaining the tubes for approximately 24 h
at 37 1C in 10% CO2 with gentle shaking (150 rpm). ΔA660 was used
to calculate the growth percentage of inhibition with respect to a
control that was grown only with DMSO.
Liquid cultures were carried out in 1.5 ml of Mueller Hinton
broth (DIFCO) plus 0.2% β-cyclodextrin (Sigma) and vancomycin
(10 mg/l), trimethoprim (Fluka) (5 mg/l), amphotericin B (Sigma)
(2 mg/l), and polymyxin B (Sigma) (2.5 mg/l); and maintained
under gentle shaking. The DMSO concentration never exceeded
0.66% (v/v) in the medium and did not have any effect in the
bacterial growth.
All the experiments were performed in triplicate, and in two
different sets. Amoxicillin (Sigma) and metronidazole (Fluka) were
used as reference antibiotics for validation of results.
2.3.4. Cytotoxic assay
Methanolic crude extracts were subjected to a cytotoxic eva-
luation using nasopharyngeal (KB), breast (MCF-7), colon (HCT-
116), and prostate (PC-3) human cancer cell lines from ATCC, along
with normal human fibroblast cells (HSF-30), passage number
45. KB cells are epidermal carcinoma of the mouth established via
contamination by Hela cells. The cells were maintained in RPMI
medium (Sigma) supplemented with 10% fetal bovine serum
(PAA), and cultured at 37 1C in an atmosphere of 5% CO2 in air
(100% humidity). The cells in a log phase of their growth cycle
were treated in triplicate with various concentrations of the test
samples (0.16–20 μg/ml) and incubated for 72 h in the conditions
described above. In order to guarantee that the cells are in
exponential growth, the criteria of confluence between 60–70%
was adopted. The cell concentration was determined by the NCI
sulforhodamine method (Skehan et al., 1990). The results are
expressed as the dose that inhibits 50% control growth after the
incubation period (IC50). The values were estimated from a log10
plot of the drug concentration against the percentage of viable
cells. According to NCI protocols, an extract with IC50 r 20 μg/ml
is considered active (Suffness and Pezzuto, 1991). Campthotecin
(Sigma), etoposide (Sigma), and podophyllotoxin (Sigma) were
included as positive controls.
2.4. Thin layer chromatography analysis (TLC)
TLC was carried out using Macherey-Nagel pre-coated aluminum
sheets 20  20 cm2, Si 60 F254. A 50 μl solution of the methanolic
extract (2 mg/400 μl) was applied directly in the TLC sheets. The
mobile phase were CHCl3:MeOH (4:1 v/v) and CHCl3:MeOH:CH3COOH
(10:4:1 v/v). The sheets were stained with a spray solution containing
vanillin (1 g of vanillin in 100 ml of H2SO4), and heated until the colors
appeared, and also with a solution containing Dragendorff reagent.
3. Results
3.1. General extract yields
The methanolic crude extracts yields exhibited large variations
depending on the locality of collection, although all populations
were harvested in the winter. These variations could be attributed
to different factors including ontogenic, genetic, or environmental.
These must be defined in further studies. The Temalac leaves (GTl)
produced a crude extract yield of 21.4% (w/w), whereas leaves of
the Yautepec population (MYl) provided just 8.1%. In general, the
leaves provided a higher percentage of crude extracts than other
plant parts. For example, the Temalac leaves (GTl) exhibited yield
values of 21.4%, which were three times higher than the values of
the stems (6.9%) and root bark (6.0%) (Table 1).
3.2. Gastroprotective assay
The ability of the methanolic crude extracts to promote
gastroprotection in the ethanol-induced ulcer model was evalu-
ated. This model is widely used in research because ethanol can
easily and quickly penetrate the gastric mucosa, causing cell
necrosis and resulting in ulcerative lesions. The results, expressed
as gastroprotection percent of each extract, were evaluated using
the Shapiro–Wilkinson statistical test to check the normality.
However, the raw data did not display a normal distribution.
Because of this, we used an arcsine transformation to obtain a
 W.I. Escobedo Hinojosa et al. / Journal of Ethnopharmacology 155 (2014) 1156–1163 1159

more normally distributed data with a linear trend (p 4 0.05). Table 3

Next, the data were analyzed using parametric statistics, applying Dose response values (ED50 mg/kg, p.o.) for Hippocratea celastroides (MY1) and
omeprazole in the gastroprotective model.
an analysis of variance following a clustering of Duncan´s multiple
comparisons of means to display the statistically homogeneous MYl crude extract % Omeprazole % Gastroprotection 7 S.E.
groups together under a system of completely random analysis. Gastroprotection 7 S.E.M. M.
Our results indicated that all of the 15 crude extracts displayed
statistically significant differences in comparison with the negative Dose (mg/kg) 4.1.1 Dose (mg/kg) 4.1.2
10 29.737 18.71 5 57.077 7.50
control. The Duncan´s test differentiated the MYl treatment from 30 58.65 7 10.65 10 83.25 7 5.36
the rest of the samples, which proved to be the most active extract, 100 69.12 77.76 50 88.337 3.13
even surpassing the positive control. GTr, MYr, VLl, GTs, GTl, 300 90.167 1.93 100 94.78 7 4.98
QLs, QLl, QLr, MJs, and MJl extracts displayed a very good gastro- R2 equation 0.963 R2 equation 0.772
ED50 (P, confidence 27.00 (13, ED50 (P, confidence 1.18 (1.13,
protective effect in the same order of the reference control
limits) 101) limits) 1.29)
(omeprazole). MYr, VLs, VLr, and MYs extracts were the less active
treatments (Table 2A).
Dose-response curves from both YHl methanolic crude extracts
and omeprazole were constructed. The tested doses for the YHl
methanolic crude extracts were 10, 30, 100 and 300 mg/kg mice, percentage of gastroprotection. This treatment had a positive
and for omeprazole, doses of 5, 10, 50 and 100 mg/kg mice, using protective effect on mouse gastric mucosa as indicated by the
an n¼ 6 for each dose, were used. The results from the evaluations ulcer index, i.e., the area in mm2 of damaged tissue decreased
are shown in Table 3, where it can be observed that the YHl extract significantly compared with the negative control, and the repaired
exerts a gastroprotective effect by a linear dose-response relation- area was comparable to the gastroprotection level conferred by the
ship, resulting in a media effective dose (ED50) of 27 mg/kg with a positive control omeprazole. These results highlight that the
regression confidence (R2) equal to 0.96. On the other hand, the extract prevents lesion formation and favors or promotes the
dose-response curve for the positive control omeprazole was protective mechanisms of the gastric mucosa.
ED50 ¼ 0.18 mg/kg (R2 0.80). Although the YHl extract was less To strengthen the information obtained in the macroscopic
active than omeprazole, a pure compound, its gastroprotection observations, an histological study with hematoxylin/eosin stain-
action is considered relevant when compared with other plant ing was performed with the gastric mucosal tissues (Fig. 1, right).
extracts, e.g., the hydroalcoholic extract from barks of Persea major Histological studies revealed that the ulcer control group (negative
displayed an ED50 ¼49.6 mg/kg in a previous study (Marques et al., control) exhibited severe damage of the gastric mucosa. Photo-
2007). micrography of the antral gastric mucosa revealed that the ulcer
Macroscopic comparisons of mice stomachs evaluated for the control group exhibited severe damage of gastric mucosa, along
gastroprotective effect (Fig. 1, left) indicated that the administra- with edema and leukocytes infiltration, characterized by the
tion of the MYl methanolic crude extract resulted in the highest marked presence of polymorphonuclear leukocytes, as well as
signs of exocytosis and cell disruption as a result of nuclear
irregularity, hyperchromatic, slight alteration in the nucleus/cyto-
plasm ratio, discontinuity in the epithelium and intense intrae-
Table 2 pithelial inflammatory infiltrate (Fig. 1B). In contrast to these
in vivo assays: Gastroprotective effect, and anti-inflammatory assay of the metha- observations, the vehicle, omeprazole and MYl treatment dis-
nolic crude extracts of leaves, stems and root bark of Hippocratea celastroides from played a gastric epithelium with cytoplasmic mucus secretion
five different localities.
preserved, with high epithelium, slight hyperchromasia, and an
A B abundant and finely vacuolated cytoplasm (Fig. 1A, C and D).

MeOH crude % MeOH crude % Edema

extract Gastroprotection 7 S. extract Inhibition 7S.E.M.b 3.3. Anti-inflammatory assay
E.M.a

MJl 66.847 0.32(B) MJl 68.11 78.78(A) The anti-inflammatory results were expressed as % edema
MJs 74.94 7 7.05(A,B) MJs 21.21 74.78(E,F) inhibition (% EI), and data were subjected to a normal test
MJr 83.517 9.63(A,B) MJr 6.7771.31(G,H) following an analysis of variance and Duncan grouping for multi-
VLl 83.197 8.72(A,B) VLl 34.16 75.63(D,E,F)
ple comparison of means. The data indicated that 12 out 15
VLs 40.02 7 9.33(C) VLs 60.42 77.31(A,B)
VLr 37.63 7 2.98(C) VLr 55.17 74.87(A,B,C) treatments displayed significant differences with respect to the
MYl 89.85 7 1.91(A) MYl 55.17 75.54(A,B,C) negative control (absolute ethanol), see Table 2B. The Duncan test
MYs 12.09 7 1.64(D) MYs 15.78 71.31(F,G,H) indicated that MYr and MJl were the most active extracts, with
MYr 43.79 7 1.27(C) MYr 70.92 73.38(A) EI¼70.92% and 68.11%, respectively, which is statistically compar-
GTl 81.03 7 6.04(A,B) GTl 34.16 78.30(D,E,F)
GTs 82.747 6.80(A,B) GTs 58.36 76.71(A,B)
able to the positive control indomethacin (EI ¼71.11%).
GTr 64.057 6.04(B) GTr 35.47 76.69(D,E,F) The statistical study classified MYr and MJl as the best extracts
QLl 78.94 7 7.02(A,B) QLl 29.66 74.78(D,E,F) with anti-inflammatory potential. MYr and the control indometha-
QLs 79.767 7.69(A,B) QLs 45.41 78.50(B,C,D) cin were chosen to determine the average effective dose (ED50)
QLr 75.58 7 1.87((A,B) QLr 37.16 76.95(C,D,E)
through a dose-response curve, employing the doses of 0.05, 0.1,
VEHI 07 2.67(E) VEHI 0 75.07(H)
OMEP 85.127 7.68(A,B) INDO 71.11 74.73(A) 0.5 and 1 mg/mouse ear (n ¼6). These results are shown in Table 4,
where it can be observed that the anti-inflammatory effect of the
VEHI (vehicle) OMEP (omeprazole) Same letters indicate no significant differences MYr extract displays a trend of dose-response type, obtaining an
by Duncan´s multiple-range test with Po 0.05. ED50 ¼0.18 mg/ear, with a confidence interval (CI) ranging from
INDO (indomethacin).
a
0.05–0.70 mg/ear and a R2 ¼ 0.99. On the other hand, for the
Comparing treatments performed at 300 mg/kg with the positive control
omeprazole (10 mg/kg).
positive control indomethacin in the dose-response curve, an
b
Treatments were performed with 0.25 mg of crude extract or indomethacin/ ED50 ¼0.16 mg/ear was found, with CI¼ 0.05–0.64 mg/ear-mouse,
ear. and R2 ¼0.96.
1160 W.I. Escobedo Hinojosa et al. / Journal of Ethnopharmacology 155 (2014) 1156–1163

200X 630X

630X 630X

Fig. 1. Macroscopic (left) and microscopic (right) appearances of the stomachs of mice subjected to the gastroprotective assay. A) stomach of a healthy mouse treated with the
vehicle; B) negative control, the stomach of a mouse to which 10% Tween in saline (5 ml/kg mouse) was initially administered, followed by the injurious agent of absolute ethanol
(5 ml/kg mouse); C) positive control, the stomach of a mouse to which omeprazole (10 mg/kg mouse) was initially administered, followed by the injurious agent of absolute
ethanol (5 ml/kg mouse); D) MYl crude extract, stomach of a mouse to which MYl extract (300 mg/kg mouse) was initially administered, followed by the injurious agent of absolute
ethanol (5 ml/kg mouse). The microscopic analysis used the hematoxylin and eosin technique, 200X and 630X magnifications, and the sagittal plane with a 5-μm viewing area.
 W.I. Escobedo Hinojosa et al. / Journal of Ethnopharmacology 155 (2014) 1156–1163 1161

Table 4 (HFS-30). In this context, we determined the mean effective dose

Dose-response values for the ED50 calculation for MYr crude extract and indo- of all 15 methanolic extracts for inhibition of the growth of each
methacin in the TPA-induced ear mouse edema model.
cell line (Table 5B), and we considered those values with an IC50
Dose (mg/ear) % Edema Inhibition 7S.E.M. o20 μg/ml as active extracts. MCF-7 cells were slightly more
sensitive than KB cells to the Hippocratea celastroides methanolic
MYr extract Indomethacin crude extracts. HCT-116 cells were less sensitive than KB cells,
being affected by 5 active extracts out of 15 assayed samples.
0.05 25.047 3.88 17.95 7 3.94
0.1 36.30 7 6.29 47.44 78.02
Prostate PC-3 cells were the most resistant line, being affected by
0.5 72.88 7 5.02 78.84 7 4.90 2 extracts. This overview highlights the great variability in the
1 82.22 7 6.26 87.96 7 3.00 chemical profiles of the extracts. The MYs sample was the most
R2 equation 0.994 0.961 active against KB cells (IC50 ¼1.18 μg/ml), as well as MCF-7 cells
ED50 (P, confidence limits) 0.18 (0.05, 0.70) 0.16 (0.05, 0.64)
(IC50 ¼1.51 μg/ml). However, VLl extract was the most active
for the HCT-116 cells (IC50 ¼1.19 μg/ml), followed by the MYl
(IC50 ¼6.62 μg/ml), MYs (IC50 ¼6.76 μg/ml), and MJl (IC50 ¼
9.77 μg/ml) extracts. In contrast to the active extracts, GTr and
Table 5 QLl treatments did not inhibit the growth of any of the cell lines. In
in vitro assays: (A) Anti-Helicobacter pylori activity, and (B) Cytotoxicity action of of addition, the VLr and VLs were the only extracts that inhibited the
the methanolic crude extracts from leaves, stems and root bark of Hippocratea
growth of PC-3 cells, with IC50 ¼7.20 and 12.15 μg/ml, respectively.
celastroides from five different localities against KB, MCF-7, HCT-116, and PC-3
cancer cell lines, and HFS-30 normal fibroblast cells. None of the extracts exhibited toxic activity against the normal
fibroblast cells line HFS-30.
A B

MeOH Anti-Helicobacter MeOH Cytotoxicitya

crude pylori MIC (μg/ml) crude Cell lines IC50 (μg/ml) 4. Discussion
extract extract
KB MCF-7 HCT- PC-3 In the multifactorial etiology of gastroduodenal ulcer diseases,
116 Helicobacter pylori plays an important role, and the presence of
MJl 31.25 MJl 2.19 420 9.77 420
certain strains is found to be more common in patients with peptic
MJs 15.63 MJs 4.68 420 4 20 420 ulcer and gastric cancer (Blaser and Crabtree, 1996). Stomach cancer
MJr 15.63 MJr 6.17 420 4 20 420 is the fourth most commonly occurring cancer in the world, and its
VLl 7.81 VLl 2.02 2.29 1.19 420 high malignancy is associated with its ability to disseminate to
VLs 31.25 VLs 420 2.57 4 20 12.15
other organs, especially the esophagus, lungs and liver. This tumor
VLr 7.81 VLr 420 2.81 4 20 7.20
MYl 7.81 MYl 9.77 1.62 6.62 420 occupies the second position as the main type of cancer, accounting
MYs 7.81 MYs 1.18 1.51 6.76 420 for 736, 000 deaths in 2008, only after lung cancer, which accounts
MYr 7.81 MYr 3.31 1.86 4 20 420 for 1.37 million deaths (WHO, 2008). Currently, there is no single
GTl 125 GTl 420 1.87 4 20 420 therapy to eradicate Helicobacter pylori. In the United States, a
GTs 31.25 GTs 420 12.3 4 20 420
GTr 7.81 GTr 420 420 4 20 420
proton pump inhibitor (PPI), clarithromycin, and amoxicillin, or
QLl 125 QLl 420 420 4 20 420 metronidazole (clarithromycin-based triple therapy), or a PPI or
QLs 31.25 QLs 17.78 13.00 4 20 420 histamine-2 receptor antagonists, bismuth, metronidazole, and
QLr 62.5 QLr 1.44 13.80 7.41 420 tetracycline (bismuth quadruple therapy) are recommended treat-
AMOX 0.01 CAMP 0.0010 0.0095 0.0054 0.0007
ment regimens, whose eradication rates reach 70–85% (Chey and
CLAR 0.5 ETOP 0.0002 0.1104 0.0664 0.0004
METRO 300 PODO 0.0016 0.0033 0.0093 0.0084 Wong, 2007). The two major reasons for treatment failure are
antibiotic resistance and patient noncompliance (CDC, 2006).
AMOX (amoxicillin). In this regard, medicinal plants can be a source of new drugs
CLAR (clarithromycin) METRO (metronidazole). with antibacterial activity or even as resistance modifiers acting
CAMP (camptothecin), ETOP (etoposide), PODO (podophyllotoxin).
a
as bacterial efflux pump inhibitors (Okoh and Sibanda, 2007).
All extracts presented IC50 420 μg/ml for HFS-30 cell line using as positive
controls camptothecin (IC50 ¼ 0.5821 μg/ml), etoposide (IC50 ¼ 0.0910 μg/ml), and
Hippocratea celastroides has the strong potential as a new option
podophyllotoxin (IC50 ¼0.009 μg/ml). for the treatment of ulcers and gastric disorders because of its broad
spectrum of activities, including anti-Helicobacter pylori, gastropro-
tective and anti-inflammatory activities, which work together to
3.4. Anti-Helicobacter pylori assay address the same problem.
It is important to note that a wide range of yields was obtained
Thirteen out of 15 methanolic crude extracts inhibited 100% of for the crude extracts from individuals of the same species
the growth of Helicobacter pylori at a concentration lower than growing in different conditions and environmental pressures,
125 μg/ml, indicating strong activity (Table 5A). VLs, VLr, MYr, MJl, which may influence their chemical profiles, and in last instance,
GTs, GTr, and QLl extracts inhibited 90% of the bacteria growth their pharmacological responses. On the other hand, the great
(MIC90) when tested at the next lower concentration for each case, variability exhibited by the results of the anti-Helicobacter pylori,
unlike the rest of the extracts where the inhibitory activity dropped gastroprotective and anti-inflammatory assays allow us to create a
dramatically at the tested lower concentrations. All samples from bioactive index by reducing the data dimensionality to facilitate
Yautepec, Morelos state, as well as the VLl, VLr, and GTr extracts, the acquisition of an overall view. Depending on the level of
displayed the highest MIC range values (7.81 μg/ml). activity, each extract received a bioactive index (BI) between 0–4,
with 0 indicating the least active extract, and 4 indicating the most
3.5. Cytotoxic assay active extract. This classification criterion was established follow-
ing the Duncan´s multiple-range test for the crude extracts
Cytotoxicity of the crude extracts was assessed against KB, activities (Fig. 2).
MCF-7, HCT-116 and PC-3 human cancer cell lines from the The most active extracts from Hippocratea celastroides inhibited
American Typing, as well as toward normal human fibroblast cells the growth of Helicobacter pylori at much higher concentrations
1162 W.I. Escobedo Hinojosa et al. / Journal of Ethnopharmacology 155 (2014) 1156–1163
Fig. 2. Bioactive Index (BI) of the methanolic crude extracts of Hippocratea
celastroides populations for the anti-Helicobacter pylori, gastroprotective, and
anti-inflammatory activities. BI ¼ 4 (strong activity), BI ¼ 3 (activity), BI ¼ 2 (moder-
ate activity), BI ¼1 (low activity), and BI ¼ 0 (no activity).
than the reference antibiotics amoxicillin and clarithromycin
(MICs ¼0.01 and 0.5 μg/ml, respectively), but compared with
metronidazole, all the methanolic crude extracts with MIC values
between 7.81–250 μg/ml were more active than metronidazole
(MIC ¼300 μg/ml). It should be considered that we tested extracts
and not isolated compounds. Thus, the isolation of different and
potent compounds from the most active extracts is encouraging.
The methanolic extracts obtained from all organs of Hippocratea
celastroides collected in Yautepec, Morelos (MY), displayed the best
anti-Helicobacter pylori values, with a BI ¼4. In addition, the root
bark and leaves of samples from La Mancha, Veracruz (VL),
together with the root bark of Temalac, Guerrero (GTr), were the
most active with a BI ¼4. The less active extracts belong to the
individuals collected in Landa de Matamoros, with a BI ¼1
observed with stems (QLs), leaves (QLl), and root bark (QLr)
extracts. In a screening study of anti-Helicobacter pylori activity
of 53 plants used in Mexican traditional medicine for gastrointest-
inal disorders, only the leaf methanolic extract of Persea americana
Mill. (avocado) displayed a MIC o7.5 μg/ml (Romero et al., 2009),
indicating how difficult it is to overcome the resistance of
Helicobacter pylori as well as the importance of finding new
natural sources of active drugs against this pathogen. In this
context, our results confirm the popular uses of Hippocratea
celastroides to treat gastric problems and highlight their great
potential in the development of anti-Helicobacter pylori therapies.
The only treatment that displayed a BI ¼ 4 for gastroprotection
was the extract obtained from Yautepec leaves (MYl), with a
percentage protection of 89.85%. However, Yautepec root bark
extracts (MYr), which was prepared using the same plant parts
employed in ethnomedicine to treat ulcers, was the most active in
the anti-inflammatory test, displaying 70.92% of edema inhibition
(BI¼ 4) but presented one of the lowest values of gastroprotection
(43.79%), corresponding to BI ¼1. This evidence indicates that the
leaves of Hippocratea celastroides from Yautepec have greater
potential as a gastroprotective, compared with its root bark, whose
activity is toward to act as anti-inflammatory. This evidence
suggests that the entire plant and not only the root bark may be
used in the treatment of gastritis. In the case of the Jojutla locality,
a large contrast appears when we compare the activities of leaves
(MJl) and root bark (MJr) extracts. At this time, the leaves extract
possesses one of the highest anti-inflammatory values (BI¼ 4), but
the root bark extract displayed no activity (BI¼ 0). Such variation
in activity is certainly linked to the metabolite variability.
MYr and MJl extracts presented the highest bioactive index
(BI¼ 4) of the 15 methanolic extracts of Hippocratea celastroides in
the anti-inflammatory model. The ED50 ¼ 0.18 mg/ear of the MYr
was similar to the positive control indomethacin, with an
ED50 ¼0.16 mg/ear. These data highlight that these specimens, in
particular the root bark of the samples from Yautepec, are ideal
candidates for further in-depth studies aimed at the identification
of the anti-inflammatory molecules. In addition, in this case, the
obtained results corroborate the uses of this plant for the treat-
ment of inflammation associated with ulcers, which could be
related to the presence of triterpenes derivatives from friedeline
and lupane structures (González et al., 1989). However, it is
necessary to perform further experiments in order to corroborate
the anti-inflammatory properties of the plant.
It appears that triterpenes and flavonoids from Hippocratea
celastroides possibly contribute to a gastroprotective action, such
as in the case of Hippocratea excelsa, whose root bark contain
metabolites responsible for the gastroprotective activity when
evaluated in the ethanol-induced gastric lesions model, including
β-sitosterol, a mixture of α- and β-amyrins, and epicatechin, the
most active compound (Navarrete et al., 2002). The gastroprotec-
tive activity of flavonoids has been established for different plants,
as is the case of Maytenus ilicifolia, a very effective plant used in
Brazil to treat dispepsy and gastric ulcers, which possesses
flavonoid glycosides, as well as epicatechin, which elevates the
volume of gastric secretion and pH, acting as gastroprotective
(Leite et al., 2010).
The macroscopic observations of the ulcer index matched with
the microscopic results in the gastroprotective assay. When compar-
ing a healthy mouse stomach with an injured stomach, it is possible
to observe the strong exocytosis of polymorphonuclear leukocytes in
the gastric antral mucosa as well as cell disruption signs, character-
ized by nuclear irregularity, slight alteration in polarity and changes
in the nucleus position, discontinuity in the epithelium and intense
intraepithelial inflammatory infiltrate (Fig. 1B). MYl crude extract
protected the integrity of the mucosa comparable to the positive
control omeprazole, displaying moderate exocytosis of polymorpho-
nuclear leukocytes, intraepithelial inflammatory infiltrate, epithelial
alteration, and hyperchromasia (Fig. 1C and D).
All plant parts from the Yautepec population displayed high
anti-Helicobacter levels (BI¼ 4), suggesting that the whole plant
could be used to control this bacteria. Although the root bark and
leaves of this population possess anti-inflammatory activity, the
stems did not. This contrast in activities is also notable in the
gastroprotection assay, where the leaves displayed a BI ¼4 and the
root bark poor activity, with BI ¼ 1.
The cytotoxic assay of the methanolic crude extracts of Hippo-
cratea celastroides using the sulforhodamine B method against KB,
MCF-7, HCT-116 and PC-3 human cancer lines produced quite
variable results. Extracts from Queretaro, QLr and QLl did not
inhibit the growth of any of the four tested cancer lines, although
the rest of the treatments inhibited the growth of at least one
carcinoma cell line. Interestingly, the extracts from Veracruz VLs
and VLr presented selective action by inhibiting only the growth of
the prostate PC-3 line. The cytotoxic extracts were selective in
action, affecting almost all the cancer cell lines, whereas the
normal fibroblast cells HFS-30 were unaffected (IC50 420 μg/ml).
Despite the apparently non-harmful effects to the non-cancer cells,
it is important to perform toxicological studies to guarantee the
safety of the extracts for further clinical studies.
In the attempt to find a relationship between the biological
activity described here and the chemical profiling of the crude
extracts, thin layer chromatography (TLC) using vanillin/H2SO4 and
Dragendorff reagents to reveal the metabolites, was performed. It
was possible to observe some similarities in the metabolite content
among extracts of the same plant part from different populations.
However, some marked differences are also observed in the
chemical profile of the extracts. All leaf extracts displayed a
 W.I. Escobedo Hinojosa et al. / Journal of Ethnopharmacology 155 (2014) 1156–1163
pink–red spot with Rf 0.75, with exception of the MYl extract.
Similarly, only in the VLL and QLL extracts a spot of the same
pink–red color with Rf 0.26 was observed. Also, only in the QLS
extract a gray spot appeared at Rf 0.17 (see S.1). For the first time,
we describe the presence of alkaloids identified by TLC. These
metabolites appeared in good concentrations in the VLr, MYr, GTs,
GTr, and QLr extracts (see S.2). A bioguided strategy must be applied
to this plant in order to find out if the metabolites observed on the
TLCs are responsible for the promising activities described here.
This is the first time that a screening study comprising four
pharmacological activities, i.e., anti-Helicobacter pylori, gastropro-
tective, anti-inflammatory, and cytotoxic assays, has been per-
formed to test the activities of leaves, steams, and root bark
methanolic extracts from five different populations of Hippocratea
celastroides collected in Mexico. The huge variability observed in
the pharmacological response of all the extracts is noteworthy. In
general, very good activities for all the four assays were found.
There is a consensus on the need to find new drugs with anti-
Helicobacter activity, as this bacterium is a risk factor for the develop-
ment of stomach cancer. In view of the low MIC required to exert their
bactericidal action, it would be promising to initiate in vivo studies to
confirm their bioactivity, as well as in combined therapies, which
include proton pump inhibitors and/or other antibiotics.
In conclusion, Hippocratea celastroides represents a new and
important alternative for the treatment of gastric disorders.
A relevant related species, Hippocratea excelsa, which is used in
Mexican ethnopharmacology to treat cancer and gastric ailments
(Mata et al., 1990), has been studied in different ways and is one of
the plants included in the Mexican Pharmacopoeia. It is necessary
to conduct further studies on Hippocratea celastroides to charac-
terize its bioactive compounds and to use them effectively as
markers for standardization purposes, especially in such countries
as Mexico, where most herbs collected for industry or self-
consumption are still wild-harvested. In addition, the variation
of chemical profiling must be well-defined in terms of plant
seasonality, ontogenetic, and genetic variability within popula-
tions for the establishment of chemotypes.
Acknowledgments
This work was supported by the Consejo Nacional de Ciencia y
Tecnología (CONACYT Grant No. 156276). W. Escobedo is indebted
to Conacyt for their support of her master's fellowship and to the
Santander-Universia program for the training in the anti-Helico-
bacter assay in Dr. Irma Romero´s lab. We also thank Dr. Blanca
Nader for the plant collection in Veracruz. We are greatly indebted
to Dr. Fernando Romero of the Instituto Nacional de Salud Pública
for having donated the mice used in this study.
Appendix A. Supporting information
Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.jep.2014.06.044.
References
Blaser, M.J., Crabtree, J.E., 1996. CagA and the outcome of Helicobacter pylori
infection. American Journal of Clinical Pathology 106, 565–567.
Carranza-González, E., 2001. Flora del Bajío y de regiones adyacentes. Hippocratea-
ceae, 98. Instituto de Ecología A.C. Pátzcuaro, Michoacán, Mexico.
Castillo-Campos, G., Medina-Abreo, M.E., 2005. Flora de Veracruz – Hippocratea-
ceae, 137. Instituto de Ecología A.C., Xalapa, Veracruz, Mexico, pp. 8–11.
CDC, 2006. Centers for Disease Control and Prevention. 〈http://www.cdc.gov/ulcer/
keytocure.htm#cdc〉. (accessed on 30.01.13).
Chey, W.D., Wong, B.C.Y., 2007. The practice parameters committee of the American
College of Gastroenterology. American College of Gastroenterology guideline on
1163
the management of Helicobacter pylori infection. The American Journal of
Gastroenterology 102, 1808–1825, http://dx.doi.org/10.1111/j.1572-0241.2007.
01393.x.
González, A.G., Bazzocchi, I.L., Ravelo, A.G., Luis, J.G., 1989. Triterpenos de Hippo-
cratea celastroides (Celastraceae). Revista Latinoamericana de Química 20, 17.
Jiménez-Estrada, M., Reyes-Chilpa, R., Hernández-Ortega, S., Cristóbal-Telésforo, E.,
Torres-Colín, L., Jankowski, C.K., Aumelas, A., Van Calsteren, M.R., 2000. Two
novel Diels-Alder adducts from Hippocratea celastroides roots and their insecti-
cidal activity. Canadian Journal of Chemistry 78, 248  254.
Leite, J.P.V., Braga, F.C., Romussi, G., Persoli, R.M., Tabach, R., Carlini, E.A., Oliveira, A.
B., 2010. Constituents from Maytenus ilicifolia leaves and bioguided fractiona-
tion for gastroprotective activity. Journal of the Brazilian Chemical Society 21,
248–254, http://dx.doi.org/10.1590/S0103-50532010000200009.
Marques, M.C.A., Cosmo, S.A., Mayer, B., Freitas, C.S., Baggio, C.H., 2007. Gastro-
protective effect of hydroalcoholic extract from barks of Persea major Kopp
(Lauraceae) in rats. Brazilian Journal of Pharmacognosy 17, 533–537.
Mata, R., Toscano, R.A., Calzada, F., 1990. Chemical studies on Mexican plants used
in traditional medicine, XV. Sesquiterpene evoninoate alkaloids from Hippo-
cratea excelsa. Journal of Natural Products 53, 1212–1219, http://dx.doi.org/
10.1021/np50071a012.
Monroy-Ortiz, C., Castillo-España, P., 2007. Plantas Medicinales en el Estado de
Morelos. Universidad Autónoma del Estado de Morelos, Mexico p. 147.
Navarrete, A., Trejo-Miranda, J.L., Reyes- Trejo, L., 2002. Principles of root bark of
Hippocratea excelsa with gastroprotective activity. Journal of Ethnopharmacol-
ogy 79, 383–388, http://dx.doi.org/10.1016/S0378-8741(01)00414-7.
NOM-062-ZOO-1999. 〈http://www.sagarpa.gob.mx/normateca/Normateca/SENA
SICA%20NORM%20143.pdf〉. (accessed on 30.01.13)2013.
Okoh, A.I., Sibanda, T., 2007. The challenges of overcoming antibiotic resistance:
plant extracts as potential sources of antimicrobial and resistance modifying
agents. African Journal of Biotechnology 6, 2886–2896.
Puigneró, V., Queralt, J., 1997. Effect of topically applied cyclosporine A on
arachidonic acid (AA)- and tetradecanoylphorbol acetate (TPA)-induced dermal
inflammation in mouse ear. Inflammation 21, 357–369, http://dx.doi.org/
10.1023/A:1027358102096.
Reyes-Chilpa, R., Jiménez-Estrada, M., Cristóbal-Telésforo, E., Torres-Colín, L.,
Villavicencio, M.A., Pérez-Escandón, B.E., Mercado-González, R., 2003. Natural
insecticides from Hippocratea excelsa and Hippocratea celastroides. Economic
Botany 57, 54–64, http://dx.doi.org/10.1663/0013-0001(2003)057%5B0054:
NIFHEA%5D2.0.CO;2.
Romero, I., Castillo-Juárez, I., González, V., Jaime-Aguilar, H., Martínez, G., Linares,
E., Bye, R., 2009. Anti-Helicobacter pylori activity of plants used in Mexican
traditional medicine for gastrointestinal disorders. Journal of Ethnopharmacol-
ogy 122, 402–405, http://dx.doi.org/10.1016/j.jep.2008.12.021.
Skehan, P., Storeng, R., Scudiero, D., Monks, A., McMahon, J., Vistica, D., Warren, J.T.,
Bokesch, H., Kenney, S., Boyd, M.R., 1990. New colorimetric cytotoxicity assay
for anticancer-drug screening. Journal of the National Cancer Institute 82,
1107–1112, http://dx.doi.org/10.1093/jnci/82.13.1107.
Skliar, M.I., Bucciarelli, A., 2007. Medicinal plants form Argentina with gastro
protective activity. Ars Pharmaceutica 48, 361–369.
Suffness, M., Pezzuto, J., 1991. Assays related to cancer drug discovery. In: Dey, P.,
Harborne, J. (Eds.), Methods in Plant Biochemistry. Academic Press, London,
pp. 71–133.
WHO, 2008. World Health Organization. 〈http://www.who.int/gho/ncd/mortality_
morbidity/cancer_text/en/〉. a(ccessed on 30.01.13).
Glossary
ATCC: American Type Culture Collection;
BI: bioactive index;
CEAMISH: Centro de Educación Ambiental e Investigación Sierra de Huautla;
CEIB: Centro de Investigación en Biotecnología;
DMSO: dimethyl sulfoxide;
EI: edema inhibition;
ED50: median effective dose;
EtOH: ethanol;
GT: Temalac, Guerrero;
Helicobacter pylori: Helicobacter pylori;
HCT-116: colon carcinoma;
IC50: half maximal inhibitory concentration;
ICR mice: Imprinting Control Region mice;
HUMO: Herbarium of the University of Morelos;
KB: nasopharyngeal carcinoma;
MCF-7: breast carcinoma;
MeOH: methanolic;
MIC: minimum inhibitory concentration;
MJ: Jojutla, Morelos;
MY: Yautepec, Morelos;
NCI: National Cancer Institut;
QL: Landa de Matamoros, Querétaro;
S.E.M: standard error of the mean;
TPA: 12-O-tetradecanoylphorbol-13-acetate;
UAEM: Universidad Autónoma del Estado de Morelos;
VL: La Mancha, Veracruz; w/w: weight/weight.