Hamelia patens JAQC. (Rubiaceae) is a perennial bushgrowing in the tropical areas of the American continent. Itcan reach up to 3 m in height, bearing bright reddish-orangetubular flowers that give rise to red berries. It is commonlyknown as “bayetilla,” “trompetilla,” “coralillo,” or “hierbacoral.” These two last names are related to the red color ofsome of its organs. A number of traditional medicinal appli-cations are known for this species along Latin America, fromMexico to Argentina. In Mexico it is used for 42 differentmedicinal purposes, especially to stop bleeding, healingsores, and in menstrual disorders (Mendoza, 2000; Arguetaet al., 1994).1,2) Regarding its use in menses, habitants of twodifferent regions of Veracruz, Mexico, use H. patens to stopexcessive menstruation (Cano-Asseleith, 1997).3) Contrast-ingly, the peasants of Zapotitlan, State of Puebla, Mexico,use it as a remedy for amenorrhea. To prepare the infusion ahandful of leaves is poured in 4 l of water, and boiled untilonly half the volume is remaining. The patient must drink theinfusion “constantly” in the next days and follow a diet lowin fat. This is also administered against flux and dark bleed-ing after menstruation, or child delivering (Martínez, et al.,1995).4) Hamelia patens is a well known medicinal plant forcertain ethnic groups in Mexico. For instance, it is the mostfrequently recorded among 614 medicinal plant speciesknown by the Popoluca people inhabiting the Sierra de SantaMarta in the State of Veracruz, Mexico. In this region, thefresh leaf sap is applied to stop the bleeding of a wound;while the root is used to treat inflamed uterus (Leonti et al.,2001).5)

Scientific research that could sustain H. patens purportedfolk medicinal uses is scant. However, recent experimentalevidence indicates that the topical application of an ointmentprepared with the ethanol extract of H. patens aerial parts in-

creases the breaking strength in wounds induced in rats(Gomez-Beloz et al., 2003).6) To the best of our knowledge,the effect of H. patens extracts on myometrium contractilityhas not been investigated yet.

Hamelia patens has been studied chemically. It is knownto contain pentacyclic oxindole alkaloids (Borges del Castilloet al., 1979a, 1979b, 1981).7—9) Some of them includingpteropodine and isopteropodine, have also been found in Un-caria tomentosa (cat’s claw). These alkaloids have been re-ported to stimulate production of lymphocytes (Keplinger etal., 1999),10) and also modulate positively muscarinic M1and 5-HT-2 receptors (Kang et al., 2002).11) On the otherhand, tetracyclic oxindole alkaloids, such as isorhynchophy-line, are the active components of the hooks and stems ofUncaria sinensis, which protect against glutamate-inducedneuronal death (Shimada et al., 1999).12)

Ethnopharmacological information suggests that H. patensmay exhibit activity on myometrium contractility. Neverthe-less, to date there are no studies focused to systematicallyevaluate this property. In addition, due to the wide geograph-ical range of distribution of this species, it is likely that itscontent of alkaloids may suffer quantitative variations andthat physiological effects might differ due to quantitativevariations in the content of alkaloids. Therefore, the aims ofthe present study were: 1) to determine the content of alka-loids of five samples of Hamelia patens collected at differentplaces and times of the year, and 2) to investigate whether H.patens extracts have any effect on myometrium contractility.As a first approach to the effects of the methanol extracts ob-tained from H. patens, we will evaluate the effects of the ex-tracts on the contraction induced by high KCl (60 mM) in es-trogen-primed rat myometrium.

October 2004 Biol. Pharm. Bull. 27(10) 1617—1620 (2004) 1617

* To whom correspondence should be addressed. e-mail: mariagcampos@yahoo.com © 2004 Pharmaceutical Society of Japan

Methanol Extracts of Hamelia patens Containing Oxindole AlkaloidsRelax KCl-Induced Contraction in Rat Myometrium

Ricardo REYES-CHILPA,a Jesús RIVERA,b Martha OROPEZA,b Pilar MENDOZA,c Badia AMEKRAZ,d

Christopher JANKOWSKI,d and Maria CAMPOS*,b

a Instituto de Química, Universidad Nacional Autónoma de México. Circuito Exterior Ciudad Universitaria; México04510, D.F. México: b Unidad de Investigación Médica en Farmacología, Hospital de Especialidades, Centro MédicoNacional Siglo XXI, Instituto Mexicano del Seguro Social; San Francisco 350–502 Col. Del Valle. México 03100, D.F.México: c Estación de Biología Tropical “Los Tuxtlas,” Instituto de Biología, Universidad Nacional Autónoma de México;Apartado Postal 94, San Andrés Tuxtla, Veracruz, 95701, México: and d Departement de Chimie et Biochemie, Universitéde Moncton; Moncton NB, EIA 3E9, Canada. Received May 6, 2004; accepted July 12, 2004

Hamelia patens JAQC. (Rubiaceae) is a medicinal bush widely distributed in tropical areas of the Americancontinent. It is used in Mexican Traditional Medicine for the treatment of menstrual disorders, therefore suggest-ing that its chemical constituents may have some effect on myometrium contractility. Physiological effects mightdiffer due to quantitative variations in the content of alkaloids arisisng from its wide geographical distribution.To test this hypothesis, the content of oxindole alkaloids in methanol extracts of five different samples collected inMexico was quantified by GC-MS. Each extract was assayed on contractility of estrogen-primed rat my-ometrium. Variations in the content of alkaloids were observed among the different samples. All samples relaxedin a concentration-dependent manner the high KCl-induced contraction in rat myometrium. Those which lackrumberine and/or maruquine displayed a higher relaxant effect than samples containing them, suggesting thatthese alkaloids might counteract the effects of isopteropodine. However, in contrast with verapamil, Hameliapatens metanol extracts are poor relaxants.

Key words Hamelia patens; Rubiaceae; medicinal plant; oxindole alkaloid; rat myometrium; Mexico

MATERIALS AND METHODS

Plant Material Hamelia patens was collected in threedifferent locations in Mexico. Samples A, B and D were ob-tained at the “Estación de Biologia Tropical Los Tuxtlas,”State of Veracruz, during October 1998, April 1999, andSeptember 1999, respectively. Sample C was obtained at“Ejido Benigno Mendoza,” State of Veracruz, during Sep-tember 1999. Sample E was obtained at “Huautla” Ecologi-cal Reserve, State of Morelos, during November 1999. Biol.Miguel Sinaca Colín and Mr. Fracisco Ramos from the Insti-tute of Biology of the National University of Mexico(UNAM) authenticated the samples. Voucher specimens weredeposited in herbarium IMSMM in Mexico city.

Extraction of Alkaloids The leaves were dried at roomtemperature, in the shadow, and powdered to prepare ex-tracts. A methanol extract rich in alkaloids was prepared foreach sample following the method described by Wagner etal. (1984).13) The powdered leaves (1 g) were extracted with1 ml 10% NH4OH and 15 ml of MeOH (15�, 60 °C). Each ex-tract was analyzed by TLC (Silica Gel GF254, 0.25 mm,10�10 cm, Merck), using as mobile phase EtOAc–MeOH–H2O (100 : 13.5 : 10) and Dragendorff reagent to visualize alkaloids. The extracts were analyzed with a Nermag Riber R10-10 GC-MS system equipped with aDelsi DI-700 chromatograph with a SE-30 column(Chrompak, 25 m, 0.3f) in DCI-PI mode. The chemical ion-ization spectra were obtained using NH3 as a reagent gas onthe sample introduced in CH2Cl2. Following the desorptionspecific protonated (M��1)� and ammonium (M�NH4)

�1

ions were recorded for the five major alkaloids. The semi-quantification was done from the ratio of their specific cur-rent recorded for these alkaloids. The calibration was per-formed using the closely related indole alkaloid carboline,and its keto-spiro-derivative obtained from lead acetate oxi-dation of the carboline according to Barton procedure (Finchet al., 1963).14) A mixture of 90% H. patens alkaloids to-gether with 10% (w/w) of carboline derivative enabled toidentify the ratio of the alkaloids at 10�1% directly from theion current.

Relaxation Experiments. Animals Animals were adultfemale Sprague-Dawley rats (230—260 g body weight), bredin our laboratory with free access to food and water. Ratswere treated s.c. with 10 mg of 3-benzoate,17-b estradiol(BE2) dissolved in 0.1 ml corn oil, 48 and 24 h before experi-ments.

Since this is the first approach to investigate the effects ofH. patens extracts on myometrium contractility, the relaxanteffect of methanol extracts was assayed on the KCl-inducedcontraction of myometrium from estrogen-primed rats.

Relaxation Assays Rats were killed by cervical disloca-tion and the uterine horns were removed and cleaned of ad-herences. Rings approximately 3—5 mm long were dissected,three from each uterine horn. Uterine rings were placed in a5 ml organ bath containing Krebs-Ringer-Bicarbonate solu-tion with the following composition (mM): NaCl, 120; KCl,4.6; KH2PO4, 1.2; MgSO4, 1.2; CaCl2, 1.5; NaHCO3, 20, andglucose 11. KRB solution, pH 7.4, was maintained at 37 °Cand gassed continuously with a mixture of 95% O2–5% CO2.Each uterine segment was placed under optimum restingforce of 1 g and allowed to equilibrate for 1 h before initiate

experiments; during this equilibration period tissues werewashed with fresh KRB every 10 min. The contractile re-sponses of the segments were recorded isometrically with atension transducer FT03 Grass, connected to a polygraphGrass model 7B. Tissues were twice stimulated with a highKCl (60 mM) solution prepared by equimolar substitution ofNaCl for KCl, only uterine rings with two similar consecu-tive responses to KCl were included in the study. A thirdstimulus with KCl (60 mM) was given and when the responsereach a platteau methanol extracts of samples A, B, C, D, orE of H. patens (0, 10, 30, 100, 300 mg/ml) were added to thebath. Extracts were dissolved in a mixture of ethanol : di-methylsulfoxide (5 : 2, v/v), in such a way that the final vol-ume added to the bath did not modify KCl-induced contrac-tile response. To avoid exceeding the innocuous volume ofvehicle added to the bath only a dose of the correspondingextract was tested in each uterine ring. Verapamil (0.001,0.01, 0.1, 1 or 10 mg/ml) was assayed as positive control.Concentrations are expressed as mg/ml for comparison withH. patens extracts, and correspond to 2.04�10�9 to2.04�10�6

M; only one dose of verapamil was tested in eachuterine ring. These procedures have been approved by the Po-lices on Animal Experimentation of the Scientific Committeeof the Instituto Mexicano del Seguro Social.

Drugs Three-benzoate, 17-b estradiol, verapamil andsalts used in the preparation of KRB or high KCl solutionswere from Sigma (St. Louis, MO, U.S.A.).

Data Analysis Results of relaxation experiments are ex-pressed as a percentage of relaxation of the contraction pro-duced by KCl (60 mM) and are the mean�S.E.M. of 8—10experiments. The 100% relaxation was achieved when thebaseline of recording prior to KCl-induced contraction wasreached. The responses to the five samples were assessed bymaximal response and EC50 (concentration of methanol ex-tracs that relaxed 50% the prior contraction induced by KCl60 mM), values were calculated from linear regression analy-sis with Sigma Plot ver 4.01. Differences between the relax-ation induced by the methanol extracts from the differentsamples of H. patens were evaluated by one-way analysis ofvariance (ANOVA) followed by multiple comparison withBonferroni test. p�0.05 was considered statistically signifi-cant.

RESULTS

Identification of Hamelia patens Alkaloids Themethanol extracts analyzed by GC/MS of the five differentsamples of H. patens (collects A to E) contained some or allof the following oxindole alkaloids: isopteropodine, rumber-ine, palmirine, maruquine and alkaloid A (Fig. 1). Quantita-tive differences were observed among the five samples.Isopteropodine was the most abundant alkaloid in all sam-ples, representing 30.5 to 98%. The next one in abundancewas alkaloid “A” (30.5 to 43%). Palmirine, rumberine, andmaruquine were less abundant, only 0 to 17.5% (Fig. 2).

Relaxant Activity The extracts of the five samples re-laxed in a concentration-dependent manner the KCl-inducedcontraction in estrogen-treated rat myometrium. Maximal re-sponse of samples A and C was higher than that of sample D(p�0.05). Maximal response of verapamil was higher thanthose of samples D and E (p�0.01), (Fig. 3, Table 1). Only

1618 Vol. 27, No. 10

the EC50 of sample C was lower than that of sample D(p�0.05), (Table 1). On the other hand, verapamil displayeda lower EC50 than samples A, B, D, and E (p�0.05, Table 1).The EC50 of verapamil was three orders of magnitude lowerthan samples A, B, D, and E and two orders of magnitudelower than sample C. These differences indicate the minor re-laxant potency of extracts as compared with verapamil.

DISCUSSION

Ethnopharmacological information suggests that Hameliapatens may exhibit activity on uterine tissues. To our knowl-edge this is the first in vitro study concerning the relaxant ef-fect of methanol extracts obtained from Hamelia patens.This is also the first comparative screening of the alkaloidcontent in samples of this species, collected in different geo-graphic areas and times of the year. The methanol extracts ofthe five different samples contained some or all of the oxin-dole alkaloids previously reported for H. patens (Gary et al.,1986; Borges del Castillo et al., 1979a, b, 1981).7—9,15) Dif-ferences in time and place of collection resulted in variationsin the content of alkaloids. This finding support the hypothe-sis that differences in phenological and geographical condi-tions might be correlated with different alkaloid profiles. Forinstance, collections from the same month (September 1999)but from different places, samples C (Ejido Benigno Men-doza) and D (Los Tuxtlas) showed differences in their alka-loid content.

The methanol extracts of the five samples relaxed, in aconcentration-dependent manner, the high KCl-induced con-traction in rat myometrium. Samples A and C showed thehighest relaxant effect, sample B exhibited a middle effect,and samples D and E, low effect. Only differences betweensamples A and D as well as C and D were statistically signifi-cant. The highest effect of samples A and C coincides with

October 2004 1619

Table 1. Effects of Hamelia patens on the High KCl-Induced Contractionin Uterine Rings Isolated from Estrogen-Treated Rats

Effective concentration (50) Maximal response(mg/ml) (%)

HPA 170.55�37.43b) 84.13�7.08a)

HPB 143.94�21.22b) 75.46�4.88HPC 76.92�19.20a) 87.94�3.57a)

HPD 293.17�46.71b) 50.97�6.15HPE 145.78�41.64b) 65.56�7.73c)

Verapamil 0.36�0.08 99.17�0.83c)

Mean�standard error (n�8—10). a) p�0.05 as compared with sample D. b)p�0.05. c) p�0.01 as compared with verapamil.

Fig. 1. Oxindole Alkaloids from Hamelia patens Methanol Extract of theLeaves Determined by GC-MS

Fig. 2. Content of Oxindole Alkaloids in the Different Samples ofHamelia patens Which Were Collected in: Los Tuxtlas, State of Veracruz,México (A: 10/1998. B: 04/1999. D: 09/1999); Ejido Benigno Mendoza,State of Veracruz, México (C: 09/1999); and Huautla, State of Morelos,México (E: 11/1999)

Fig. 3. Maximal Relaxant Response of Samples A, B, C, D, and E of Hamelia patens (Panel A), and Verapamil (Panel B) on the High KCl-Induced Con-traction in Uterine Smooth Muscle of Estrogen-Primed Rats

Bars represent the mean of 8—10 observations, vertical lines represent the S.E.M. * p�0.05 as compared with HPD.

the fact that they lack rumberine, and sample A also lacks al-kaloid A and palmirine, and maruquine level is near to zero,whereas sample C lacks maruquine. These alkaloids besidesof isopteropodine were all found in the other three sampleswhich exerted a middle or low relaxant effect. In the light ofthese findings it is likely that rumberine or maruquine hadsuch an effect that counteracts the relaxation induced byisopteropodine which quantity is the highest in A, B, C, andD samples. Studies with the isolated compounds will be use-ful to elucidate this notion.

When the relaxant effect of methanol extracts is comparedto verapamil, it is clear that the extracts are poor relaxants,this might be related to the mechanism selected to inducecontraction (electromechanical coupling) and/or to the factthat the extracts assayed were not pure compounds and so theeffect observed was the sum of possible effects of all com-pounds. Also, the variability observed in the EC50 valuesmust be mentioned. In particular sample E was not fully dis-solved in the vehicle used (ethanol : dimethylsulfoxide, 5 : 2,v/v) and this fact might be related to its variable effects.

In calcium-cointaining solution, iso-osmotic addition ofKCl induce a rapid, phasic contraction followed by a pro-longed sustained plateau (tonic component) of smaller ampli-tude; this has been proposed to occur through membrane de-polarization induced by calcium influx through voltage-de-pendent calcium channels (Bolton, 1979).16) Nevertheless,Trujillo et al. (2000)17) found that KCl (60 mM) produced a76�16.2% increase in total [3H]inositol-phosphates abovebasal levels and increased the intracellular levels of freearachidonic acid. These results suggested that in the estro-gen-primed rat uterus, iso-osmotic high K� solutions, in ad-dition to their well-known effect on Ca2� influx, activateother cellular processes leading to an increase in the calciumsensitivity of the contractile machinery by a mechanism in-dependent of extracellular calcium. In this respect, the relax-ant effect of methanol extracts of H. patens observed in thepresent study might be due either to a direct inhibitory effecton the calcium channels or to some other indirect effect onthe contractile machinery. More studies are required in orderto clarify the mechanism of action of these methanol ex-tracts.

The relaxant effect observed in the present study is in con-trast with the use of H. patens in traditional medicine to alle-viate menstrual disorders as amenorrhea or to stop flux andbleeding after menstruation and child delivery. Because ofthese uses, a contractile but not a relaxant effect on my-ometrium might be expected. The reasons for this contro-versy might be: 1) the use of methanol and not aqueous ex-tracts, since the chemical constituents could be different; 2)the model used for the evaluation, it would be necessary toevaluate the effect of extracts on the pharmacomechanical

coupling of the contraction, and 3) differences betweenspecies (human and rat).

Finally, since the aim of ethnopharmacological studies isto elucidate the rationale behind the use of plants (Leonti etal., 2001),5) it would be interesting to further test the abovehypotheses, or alternatively to search on cultural and anthro-pological grounds an explanation. For instance, it seemslikely that a morphological plant character, such as the redcolor of some organs of Hamelia patens, might explain sev-eral of its folk medicinal uses, many of them related withblood.

Acknowledgements Authors thank to Dirección Generalde Intercambio Académico, UNAM for supporting Dr.Jankowski visits to Mexico.

REFERENCES

1) Mendoza-Marquez P., “Tesis de Licenciatura en Biología. Facultad deCiencias,” Universidad Nacional Autonoma de México, 2000.

2) Argueta A., Cano L., Rodarte M. (eds.), “Atlas de las Plantas de laMedicina Tradicional Mexicana,” Instituto Nacional Indigenista,México D.F., México, 1994.

3) Cano-Asseleith L., “Flora Medicinal de Veracruz,” Universidad Ver-acruzana, Xalapa, México, 1997, p. 166.

4) Martínez M. A., Evangelista V., Mendoza M., Morales G., Toledo G.,Wong A., “Catalogo de plantas útiles de la sierra norte de Puebla,México,” Cuadernos del Instituto de Biología No. 27. UNAM,México, D. F., Mexico, 1995.

5) Leonti M., Vibrans H., Sticher O., Heinrich M., J. Pharm. Pharmacol.,53, 1653—1669 (2001).

6) Gomez-Beloz A., Rucinski J., Balick M., Tipton C., J. Ethnopharma-col., 88, 169—173 (2003).

7) Borges del Castillo J., Manresa M., Martin J., Pascual C., Rumbero A.,Tetrahedron Lett., 34, 3197—3200 (1979a).

8) Borges del Castillo J., Manresa M., Martín J., Rodríguez F., VázquezP., Anales de Química (Serie C), 76, 294 (1979b).

9) Borges del Castillo J., Manresa M., Martín J., Rodríguez F., VázquezP., Anales Química (Serie. C), 78, 180 (1981).

10) Keplinger K., Laus G., Wurm M., Dierich M., Teppner H., J.Ethnopharmacol., 64, 23—34 (1999).

11) Kang T. H., Matsumoto K., Tohda M., Murakami Y., Takayama H., Ki-tajima M., Aimi N., Watanabe H., Eur. J. Pharmacol., 444, 39—45(2002).

12) Shimada Y., Goto H., Sakakibara I., Kubo M., Sasaki H., Terasawa K.,J. Pharmacol., 51, 715—722 (1999).

13) Wagner H., Bladt S., Zgainski E., “Plant Drug Analysis. A Thin LayerChromatography Atlas,” Springer Verlag, New York, 1984.

14) Finch N., Gemenden C. W., Hsiu-Chu I., Taylor W. I., J. Am. Chem.Soc., 85, 1520—1523 (1963).

15) Gary E. M., Redhika S., Maktoob A., J. Nat. Prod., 49, 406—411(1986).

16) Bolton T. B., Physiol. Rev., 59, 606—718 (1979).17) Trujillo M. M., Ausina P., Savineau J. P., Marthan R., Strippoli G., Ad-

venier C., Pinto F. M., Cárdenas M. L., Life Sci., 66, 2441—2453(2000).

1620 Vol. 27, No. 10