In vivo anti-inflammatory and anti-ulcerogenic activities of extractsfrom wild growing and in vitro plants of Castilleja tenuiflora Benth.(Orobanchaceae)

Paul Mauricio Sanchez a, María Luisa Villarreal b, Maribel Herrera-Ruiz c,Alejandro Zamilpa c, Enrique Jiménez-Ferrer c, Gabriela Trejo-Tapia a,n

a Departamento de Biotecnología, Centro de Desarrollo de Productos Bióticos, Instituto Politécnico Nacional, PO Box 24, Yautepec, Morelos C.P. 62730, Mexicob Centro de Investigación en Biotecnología, Universidad Autónoma del Estado de Morelos, Av. Universidad 1001, Col. Chamilpa, Cuernavaca, Morelos C.P.62209, Mexicoc Centro de Investigación Biomédica del Sur, Instituto Mexicano del Seguro Social, Argentina No. 1, Col. Centro, Xochitepec, Morelos C.P. 62790, Mexico

a r t i c l e i n f o

Article history:Received 30 April 2013Received in revised form13 September 2013Accepted 2 October 2013Available online 18 October 2013

Keywords:Castilleja tenuifloraCytotoxicityIridoid glycosidesGastric ulcerPhenylethanoid glycosidesTPA-inflammation

a b s t r a c t

Ethnopharmacological relevance: Castilleja tenuiflora Benth. (Orobanchaceae) is a perennial shrub usedsince the 16th century in Mexican traditional medicine for the treatment of a number of health disordersincluding inflammation, stomach pain and tumors. The aim of the present study was to evaluate thecytotoxic, anti-inflammatory and anti-ulcerogenic activities of ethyl acetate (EaE), methanol (ME) andaqueous extracts (AE) of Castilleja tenuiflora wild grown (CtW) and in vitro plants (CtIv).Material and method: Phytochemical analysis of the phenylethanoid glycoside (PhG) and iridoid glyco-side (IG) components was carried out by chromatographic methods. In vitro cytotoxic activity of theextracts was evaluated in the following four carcinoma cell lines: colon (HF-6), breast (MCF-7), prostate(PC-3), and nasopharyngeal (KB). The topical anti-inflammatory activity was evaluated in mouse earedema induced by 12-O-tetradecanoylphorbol-13-acetate (TPA). Anti-ulcerogenic activity was evaluatedin rats using an absolute ethanol-induced acute gastric ulcer model.Results: The main compounds in the extracts were isoverbascoside, verbascoside and aucubin and theirconcentration depended both on the solvent used and on the plant material origin. None of the extractsshowed cytotoxicity against the tested cell lines. In contrast, CtWEaE, CtWAE and CtIvEaE (1.6 mg/ear)showed moderate anti-inflammatory activity similar to dexamethasone (1 mg/ear) with a 38.2, 39.3 and49.1% decrease of inflammation, respectively. CtWEaE and CtIvEaE (100 mg/kg) showed high anti-ulcerogenic activity with 88.3 and 83.1% inhibition, respectively, compared to famotidine (20 mg/kg,32.8% inhibition).Conclusion: Castilleja tenuiflora extracts provided significant gastric protection in an acute ulcerinduction model and topical anti-inflammatory activity in a mouse ear edema model. These activitiesare related to verbascoside and may explain the traditional use of Castilleja tenuiflora in the treatment ofanti-inflammatory and gastrointestinal disorders. Cultured Castilleja tenuiflora plants (in vitro) exhibitedpharmacological activities and also have the potential to produce bioactive compounds.

& 2013 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

Medicinal plants exhibiting a variety of therapeutic properties areof great interest as starting materials for isolation and chemicalstructure identification studies (Ankli et al., 2002). The selection andevaluation of medicinal plants is important for their ethno-medicinaluse, reducing their empirical use and enhancing the probability ofsuccess in the discovery of new drugs (Cos et al., 2006). Castillejatenuiflora Benth. is a small perennial shrub belonging to the

Orobanchaceae family and is distributed in mountainous areas of theSouthern United States and Mexico (Holmgren, 1976). In “The Historyof the Plants of New Spain” (16th Century), this species was recordedunder the Náhuatl name “Atzoyatl”; it is described as a plant of “hotnature”whose root preparations heal colics originated from fecal massretention (Béjar et al., 2000). In Mexican folk medicine, the plant isused in the form of decoctions or infusions to treat conditionsassociated with cancer symptomatology (Graham et al., 2000) as wellas for the treatment of cough, dysentery, nerves, nausea, inflammation,vomiting, hepatic and gastrointestinal disorders (Biblioteca Digital dela Medicina Tradicional Mexicana, 2009). In the state of Veracruz(southern region of Mexico), the flower infusion is used to treatstomach pain and for “good digestion” (Cano, 1997).

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/jep

Journal of Ethnopharmacology

0378-8741/$ - see front matter & 2013 Elsevier Ireland Ltd. All rights reserved.http://dx.doi.org/10.1016/j.jep.2013.10.002

n Corresponding author. Tel.: þ52 735 394 2020; fax: þ52 735 394 1896.E-mail addresses: gttapia@ipn.mx, gabriela.trejo@yahoo.com (G. Trejo-Tapia).

Journal of Ethnopharmacology 150 (2013) 1032–1037

Phytochemical investigations indicate that the plant containsiridoid glycosides (IGs) such as aucubin in its aerial parts and roots(Gómez-Aguirre et al., 2012). Iridoids have proven antitumoractivity as they inhibit the enzyme topoisomerase I which isinvolved in tumoral processes (Gálvez et al., 2005) and they alsohave cytotoxic activity toward lung cancer cell lines (Hung et al.,2008). In addition to IGs, the phenylethanoid glycosides (PhGs)verbascoside and isoverbascoside were isolated from Castillejatenuiflora (Gómez-Aguirre et al., 2012). PhGs have a wide rangeof biological activities including free radical scavenging activity(Hanson et al., 2011). Other reports indicate that verbascosideinhibits cholinesterase and acts as a potent inhibitor of geneexpression and de novo synthesis of inflammatory chemokines(Georgiev et al., 2012). Isoverbascoside is known to inhibit the cellproliferation of the human gastric cancer cell line MGC803 andreverse its malignant phenotypic characteristics (Chen et al.,2002). The flavonoids apigenin and quercetin glycoside have alsobeen isolated from this species (López-Laredo et al., 2012). Wholeplant extracts from Castilleja tenuiflora showed cytotoxic activityagainst breast and cervical human cancer cell lines (Moreno-Escobar et al., 2011) and methanol extracts of wild and cultivatedplants showed free-radical scavenging properties and reducingcapabilities (López-Laredo et al., 2012). To our knowledge, scarcescientific studies have been conducted to confirm its therapeuticproperties.

Castilleja tenuiflora is wild harvested for traditional use, aprocess which may have some disadvantages such as risk ofadulteration, misidentification or sustainability. Alternatively,in vitro propagated plants may be a source of biologically activecompounds such as IGs (Martínez-Bonfíl et al., 2011) and PhGs andmay prove useful for germplasm conservation and biodiversityprotection (Lubbe and Verpoorte, 2011). The present study wasconducted to evaluate the cytotoxic, anti-inflammatory and anti-ulcerogenic activities of ethyl acetate (EaE), methanol (ME) andaqueous extracts (AE) from Castilleja tenuiflora wild grown (CtW)and in vitro plants (CtIv). Furthermore, phytochemical analysisindicated the presence of potentially active compounds of greatinterest in the field of biomedicine.

2. Materials and methods

2.1. Plant material

Castilleja tenuiflora was collected at Juchitepec, State of Mexico,Mexico in December 2009. This place is situated at 2800 m.a.s.l.(latitude 19110 N, longitude 98192 W). Plants were positivelyidentified as Castilleja tenuiflora by Rolando Ramírez, M. Sc., HeadCurator of the HUMO Herbarium of the Universidad Autónoma delEstado de Morelos (UAEM). A voucher specimen (HUMO25205)has been deposited in the Herbarium. In vitro culture plants wereobtained as described elsewhere (Martínez-Bonfíl et al., 2011). Forextracts preparation, 40 days-old plants were used.

2.2. Preparation of extracts

The whole plant material was air-dried and pulverized with agrinder and extracted in sequence with ethyl acetate, methanoland water (1 g in 50 mL, w/v) by maceration individually at roomtemperature for 24 h. All extracts were filtered using Whatmanpaper No. 1 and concentrated in a Büchi rotary evaporator(Büchi-490; Büchi, Switzerland) at 40 1C under vacuum and storedat 4 1C for later use.

2.3. Phytochemical screening and HPLC quantificationof PhGs and IGs

For TLC analysis, aliquots of extracts were spotted onto silica gel60 F254 aluminum plates (Merck) and eluted with a CHCl3/MeOH(7:3) solvent system. The plates were examined by UV fluores-cence (254 and 365 nm) and then were sprayed with ethanol, 0.5%p-anisaldehyde, and 10% H2SO4, and then heated at 105 1C for1–2 min (Wagner et al., 1983).

For analysis of PhGs by HPLC we followed the method developedby Gómez-Aguirre et al. (2012), using a Waters 2695 separationmodule HPLC system equipped with a Waters 996 photodiode arraydetector and Empower Pro software (Waters Corporation, USA). ALiChrospher 100 RP-18 column (4 mm�250mm, 5 mm) (Merck,Darmstadt, Germany) connected to a guard column was used. Forelution of the PhGs, two solvents denoted as water (solvent A) (pH 3.5,containing 0.5% TFA) and acetonitrile (solvent B) were used. Thegradient systemwas as follows: 0–2min, 100–0% B; 2–10min, 77–23%B; 12–13min, 70–30% B; 14–15min, 100% B; 16–17min, 0% B. Theflow rate was 1 mL/min and the injection volume was 20 mL. Theabsorbance was measured at 330 nm. The PhGs peaks were identifiedby comparison of the retention times and UV spectra with those ofreference compounds which were previously obtained from roots ofCastilleja tenuiflora (Gómez-Aguirre et al., 2012). The retention timesfor isoverbascoside and verbascoside were 7.5 min (λ¼218, 247, 292and 331 nm) and 7.9 min (λ¼218, 247, 290 and 330 nm), respectively.The amounts of PhGs were estimated by interpolation of the peakareas and comparison with a calibration curve constructed forverbascoside. The calibration curve was linear in the range of 62.5 to500 mg verbascoside/mL in methanol (y¼22144x�161443,R2¼0.999). All analyses were repeated in triplicate. Results areexpressed as the mean from three determinations in mg/g of extract.

For aucubin quantification, a different chromatographic methodwas established. The mobile phase consisted of a water–acetonitrilemixture (97:3 ratio) (solvent A) and acetonitrile (solvent B). Thegradient system was as follows: 0–8min, 100–0% B; 9–10min, 80–20% B; 11–15min, 70–30% B; 16 min, 40–60% B; 23–25min, 20–80%B; 26–28min, 0–100% B; 29–32min, 100–0% B. The flow rate was1 mL/min and the injection volume was 20 mL. The absorbance wasmeasured at 205 nm. The retention times and UV spectra of majorpeaks were analyzed. The concentration of iridoids was estimated byinterpolation of the peak areas and comparison with a calibrationcurve constructed for aucubin (Sigma–Aldrich, Inc. St. Louis, MO). Thecalibration curve was linear in the range of 112–900 mg/mL inmethanol (y¼7575.9xþ248079, R2¼0.995). Results were expressedas the mean from three determinations in mg/g of extract.

2.4. Cytotoxicity assay

The in vitro cytotoxicity was measured by the sulphorhodamine B(SRB) (MP Biomedicals, LLC) protein staining assay (Houghton et al.,2007). The following cell lines were used: KB (nasopharyngealcarcinoma), HF-6 (colon carcinoma), MCF-7 (breast carcinoma) andPC-3 (prostate carcinoma). The cell cultures were maintained in RPMI-1640 medium supplemented with 10% fetal bovine serum, 5000 units/mL penicillin, 5 mg/mL streptomycin, 7.5% NaHCO3, and cultured in a96-well microliter plate (104 cells/mL, 190 μL/well) at 37 1C in a 5%CO2-air atmosphere (100% humidity). The cells at the log phase ofgrowth were treated in triplicate (n¼3) with different concentrationsof the extracts (0.16, 0.8, 4, and 20 μg/mL), and incubated for 72 h. Thecell concentration was determined by protein analysis. The opticaldensity was measured at 590 nm with an ELISA-Reader (MolecularDevices, SPECTRA max plus 384). Results are expressed as theconcentration that inhibits 50% of control growth after the incubationperiod (IC50). The values were estimated from a semilog plot of theextract concentration (μg/mL) against the percentage of viable cells.

P. Mauricio Sanchez et al. / Journal of Ethnopharmacology 150 (2013) 1032–1037 1033

Camptothecin and etoposide were included as positive standards.Extracts with IC50r20 μg/mL were considered active according to theNational Cancer Institute (NCI) guidelines described in the literature(Suffness and Pezzuto, 1991; Moreno-Escobar et al., 2011).

2.5. Anti-inflammatory activity in mice with TPA

ICR mice of either sex with weights between 28 and 30 g wereused. The experiments were performed in strict adherence to theofficial requirements of the Mexican Regulations of ExperimentalAnimal Care (NOM-062-ZOO-1999) and international ethicalguidelines for the care and use of experimental animals. Theexperimental protocol was authorized by the Local HealthResearch Committee (IMSS, Register number: R-2012-1701-58).Mice were maintained at a temperature of 2473 1C, 7075% ofhumidity with a 12-h light/dark cycle and food/water ad libitum. Acontrol group received acetone as vehicle and dexamethasone wasused as the anti-inflammatory positive control.

Animal inflammation was induced using 12-O-tetradecanoylphor-bol-13-acetate (TPA) by the method previously described by Payá et al.(1993). Mice were grouped (seven individuals) and TPA (2.5 μg)dissolved in acetone (20 mL) was topically applied on the internaland external surface of the right ear to cause edema. Doses of 1.6 mg/ear of each treatment were applied on the ear of each individual. Thereference anti-inflammatory drug was administered at 1 mg/ear. Alltreatments were dissolved in acetone and applied topically on bothears after the administration of TPA. Four hours after administration ofthe inflammatory agent, the animals were sacrificed by cervicaldislocation. Circular sections of 6 mm in diameter were taken fromboth the treated (t) and the non-treated (nt) ears, which wereimmediately weighted to determine the extent of the inflammation.The percentage of inhibitionwas obtained using the expression below:

Inhibition ð%Þ ¼ Δw control � Δw treatment� �� 100

where Δw¼wt—wnt; wt is the weight of the section of the treatedear; wnt is the weight of the section of the non-treated ear.

2.6. Anti-ulcerogenic activity in rats with absolute alcohol-inducedulcer (ETH) model

Sprague Dawley albino rats of either sex weighting between350 and 400 g were housed in acrylic boxes at a temperature of2473 1C, 7075% of humidity with a 12-h light/dark cycle. Theanimals were deprived of food for 48 h prior to experimentation,but had free access to drinking water. The experiment was carriedout according to Samonina et al. (2004). The animals were dividedinto groups, consisting of five rats each. After 48 h of fasting, therats were orally pre-treated with the various extracts of Castillejatenuiflora (100 mg/kg), a vehicle (water 5 mL/500 g body weight)or famotidine (20 mg/kg), 30 min before administration of theulcerogenic agent. Gastric lesions were induced using absoluteethanol (1 mL/200 g body weight, ETH) (Noor et al., 2006). Onehour later, the rats were anesthetized using urethane (0.9 g/kgbody weight) and then sacrificed by cervical dislocation. Thestomachs were removed and opened along the greater curvatureand then were gently rinsed with water to remove gastriccontents. Ethanol induced ulcers appeared as elongated bands ofhemorrhagic lesions in the glandular region. After identification ofthe ulcers (damaged area), their area was measured by planimetry(Shuai et al., 2011); the area of each ulcer lesion was measured bycounting the number of small squares, 1 mm�1 mm, covering thelength and width of each ulcer. The sum of the areas of all lesionsfor each stomach was used for the calculation of the ulcerated area(UA, mm2). The ulcer index (UI) was estimated by dividing the UAbetween the total stomach area (mm2).

The protection percentage was calculated using the followingformula:

Protectionð%Þ ¼ UI control� UI pre� treated groupUI control

� �� 100

2.7. Statistical analysis

Statistical analysis for anti-inflammatory and antiulcer activ-ities was performed using Sigma Stat for Windows version 11.0(Systat Software Inc., San Jose, CA, USA). Using one-way analysis ofvariance (ANOVA) followed by Dunnett's multiple comparison test,po0.05 was considered to be statistically significant. IC50 valueswere obtained from regression lines with coefficient factorsbetween R2¼0.804 and 0.978.

3. Results

3.1. Phytochemical screening and HPLC quantificationof PhGs and IGs

HPLC analysis was carried out for the detection and quantificationof the major constituents of the tested extracts. Fig. 1 presents theHPLC profile of the extracts recorded at 330 nm (PhGs) and 205 nm(IGs). The main compounds were isoverbascoside (tR¼7.5 min), ver-bascoside (tR¼8.0 min) and aucubin (tR¼12.5 min). Table 1 indicatesthe contents of these compounds in the tested extracts of Castillejatenuiflora. Compared to the methanol and aqueous extracts, the ethylacetate extract showed the highest level of verbascoside and aucubin,and did not contained isoverbascoside. Ethyl acetate and aqueous

Fig. 1. HPLC chromatograms of active extracts from wild and in vitro plants ofCastilleja tenuiflora. (A) Detection of phenylethanoids glycosides at 330 nm and(B) detection of aucubin at 205 nm. Ct¼Castilleja tenuiflora; W¼Wild; Iv¼ In vitro;EaE¼Ethyl acetate extract; ME; Methanol extract; AE¼Aqueous extract.

P. Mauricio Sanchez et al. / Journal of Ethnopharmacology 150 (2013) 1032–10371034

extracts from wild harvested plants contained higher levels of thesechemical compounds than in vitro plants.

3.2. Cytotoxicity assays

The different extracts of Castilleja tenuiflora had IC50 values420 μg/mL against all the cancer cells lines tested: KB, HF-6, MCF-7 and PC-3. Based on the National Cancer Institute (NCI) guidelines

(Suffness and Pezzuto, 1991), these extracts did not show cytotoxicactivity.

3.3. Anti-inflammatory activity

Topical application of TPA resulted in a marked increase in earthickness with clear evidence of edema in all animals of thenegative control group (11.0070.61 mg). Topical application ofCtIvEaE, CtWAE and CtWEaE at a dose of 1.6 mg/ear inhibitedsignificantly (po0.001) edema formation above 38%, similar toreference drug dexamethasone (1 mg/ear, used as a positivecontrol) (see Table 2). The extract CtIvME did not significantlyinhibit the edema and the remaining extracts did not show anyanti-inflammatory activity.

3.4. Anti-ulcerogenic activity

The administration of ethanol to the control rat group pro-duced extensive ulcerations in form of hemorrhagic lesions onglandular area (Fig. 2A). In a 48-h period evaluation, rats given thetested extracts at 200 mg/kg had no mortality. CtWEaE, CtIvEaE,CtWAE and CtIvME at a dose of 100 mg/kg exhibited a significantreduction of gastric lesions (po0.001) when compared to thecontrol group (see Table 3). Remarkably, these four extractsprovided better protection from gastric lesions than the referencedrug famotidine (32.8%) (Table 3, Fig. 2). CtWME and CtIvAEextracts did not exhibit an anti-ulcerogenic effect. Based on theseresults, a dose–response curve of CtWEaE extract was also eval-uated (12.5 to 100 mg/kg). It was observed that the 25, 50 and100 mg/kg doses were effective, exhibiting significant differencesfrom the control group; the highest percentage of anti-ulcerogeniceffect was at 100 mg/kg (88.3%) with a calculated ED50 value of35.7 mg/kg (Table 3).

The gastric ulcer healing process involves a variety of mechan-isms including nitric oxide (NO) mediated repair (Tarnawski,2005). To explore if NO mediated the anti-ulcerogenic effect ofCastilleja tenuiflora extracts, an additional experiment was carriedout using isosorbide dinitrate (IDN), a NO-donor drug. It wasobserved that the active extract CtWEaE plus IDN, both at

Table 1Concentration of verbascoside, isoverbascoside and aucubin in Castilleja tenuifloraextracts (in mg/g extract).

Extract Verbascoside Isoverbascoside Aucubin

CtWEaE 262.43733.02 ND 32.54710.80CtIvEaE 142.40762.40 ND 22.9972.87CtWME 87.60724.79 22.9071.46 67.10714.80CtIvME 99.04710.60 40.9076.86 137.24716.00CtWAE 64.40731.63 6.7271.08 43.9575.01CtIvAE 11.1474.52 6.0571.66 69.5077.91

Values are mean7S.E.M (n¼3). ND¼Not Detected. Ct¼Castilleja tenuiflora; W¼Wild;Iv¼ In vitro; EaE¼Ethyl acetate extract; ME; Methanol extract; AE¼Aqueous extract.

Table 2Anti-inflammatory activity of Castilleja tenuiflora extracts on TPA-induced edemamodel in ICR mice.

Treatment Dose (mg/ear) Edema (mg) Edema inhibition (%)

TPA 2.5 mg 11.0070.61 –

CtWEaE 1.6 6.8070.36nnn 38.2CtWME 1.6 – NICtWAE 1.6 6.6870.70nnn 39.3CtIvEaE 1.6 5.6071.07nnn 49.1CtIvME 1.6 9.7070.71 11.8CtIvAE 1.6 — NIDexamethasone 1.0 4.6570.86nnn 57.7

Values are mean7S.E.M. (n¼7). ANOVA post-test of Dunnett with nnnpo0.001 incomparison with TPA group. NI¼Not Inhibits. TPA¼12-O-tetradecanoylphorbol-13-acetate; Ct¼Castilleja tenuiflora; W¼Wild; Iv¼ In vitro; EaE¼Ethyl acetate extract;ME; Methanol extract; AE¼Aqueous extract.

Fig. 2. Evaluation of anti-ulcerogenic activity of Castilleja tenuiflora extracts against ethanol-induced gastric lesions in rats. Stomachs of rats with different experimentalconditions: (A) Ulcer control rat; (B) 20 mg/kg of famotidine; (C) 100 mg/kg of CtWEaE; (D) 100 mg/kg CtWME; (E) 100 mg/kg of CtWAE; (F) 100 mg/kg of CtIvEaE;(G) 100 mg/kg of CtIvME; (H) of 100 mg/kg CtIvAE. Ct¼Castilleja tenuiflora; W¼Wild; Iv¼ In vitro; EaE¼Ethyl acetate extract; ME; Methanol extract; AE¼Aqueous extract.

P. Mauricio Sanchez et al. / Journal of Ethnopharmacology 150 (2013) 1032–1037 1035

ineffective doses (12.5 and 2.5 mg/kg, respectively, both sepa-rately), showed anti-ulcerogenic activity with 65.2% inhibition(Table 4), likely due to the synergistic effect of compounds presentin the extract with IDN.

4. Discussion

In spite of the long recognized therapeutic use of Castillejatenuiflora, very little pharmacological research has been carriedout. The present study was undertaken to evaluate the cytotoxic,anti-inflammatory and anti-ulcerogenic activities of the extractsfrom Castilleja tenuiflora, a plant that has long been associatedwith traditional medicine. We tested not only extracts from wild-growing plants, but also from plants grown in vitro as analternative source of plant material. Pharmacological results wereaccompanied by a quantitative chemical analysis of the extracts.HPLC analysis confirmed the chemical complexity of the testedextracts which contained both PhGs and IGs bioactive compoundsbut showed qualitative and quantitative differences in chemicalprofiles depending both on the solvent used for the extraction andon the plant material origin. The level of verbascoside was 20times higher in the ethyl acetate extract from wild-grown plants(CtWEaE) than in the aqueous extract from in vitro plants (CtIvAE).Isoverbascoside was only present in the extracts prepared withmethanol or water regardless of the plant material source. Aucu-bin contents were 2-times higher in methanol extracts fromin vitro plants (CtIvME) than in those from wild grown plants.

None of the extracts from Castilleja tenuiflora had cytotoxic effectson the tested cell lines. This result was unexpected because in anotherstudy, methanol extracts from this species (collected from Mountainsin Zempoala Lagoons National Park, Morelos) were selectively

cytotoxic against two of the six carcinoma cell lines tested: a breastcell line (MCF-7) with an IC50¼9.53 μg/mL, the same cell line tested inthe present study, and a cervical cell line (Ca Ski, IC50¼8.72 μg/mL)(Moreno-Escobar et al., 2011). The differences in the cytotoxic evalua-tion of Castilleja tenuiflora methanol extracts on the breast carcinomacell (MCF-7) line may be attributed to the difference in the origin ofthe plant material (geographic location), a parameter which has beenpreviously documented to influence the chemical composition of theplant and consequently its biological activity (Filippini et al., 2010;Sharma et al., 2012). Verbascoside and isoverbascoside exhibitedstrong cytotoxic activity against lymphocytic leukemia P-388 cells(IC50¼2.6 mg/mL and IC50¼8 mg/mL, respectively) (Pettit et al., 1990)while aucubin was potent against lung cancer A-549 (IC50¼4.33 mg/mL) (Hung et al., 2008). Nevertheless, verbascoside has documentedIC50 values of 49.3 and 55.6 mg/mL against MCF-7 and HEP-2 cancercells, respectively (Harput et al., 2012), confirming the selectivity ofthis compound against leukemia cancer cell lines.

Ethyl acetate extracts from wild-grown and in vitro plants(CtWEaE and CtIvEaE) and aqueous extracts from wild-grownplants (CtWAE) showed moderate topical anti-inflammatory activ-ity similar to the reference drug dexamethasone. These extractscontained verbascoside as the major constituent followed byaucubin. However, the concentration of verbascoside amongstthe extracts differed by almost four fold. Moreover, as mentionedabove, isoverbascoside is absent from the ethyl acetate activeextracts. Aucubin, in its aglycone form, inhibits the activity of theenzyme cyclooxygenase-2 (COX-2), which is one of the keyenzymes responsible for the conversion of arachidonic acid toprostaglandins as part of the inflammatory process (Park et al.,2007). In addition, verbascoside has shown anti-inflammatoryactivity in a carrageenan-induced hind paw edema model in mice(Akdemir et al., 2011) and in cultures of human keratinocytes(Georgiev et al., 2012). The potent anti-inflammatory action ofverbascoside is attributed to the fact that it inhibits pro-inflammatory chemokines at both transcriptional and translationallevels (Georgiev et al., 2012), COX-2 in a mouse peritonealmacrophage model (Díaz et al., 2004) and the enzyme calcineurin,an important regulator of T-cell mediated inflammation (Prescottet al., 2011). From the evaluation of the anti-inflammatory effectsof Verbascum xanthophoeniceum crude extracts, its fractions andisolated compounds in cultures of human keratinocytes, it seemsthat verbascoside inhibits inflammatory chemokines more effec-tively than aucubin (Georgiev et al., 2012).

Thus, both Castilleja tenuiflora wild grown and in vitro plants haveanti-inflammatory effects that may be related to its therapeutic use totreat inflammation. All extracts also were evaluated for their acuteanti-ulcerogenic activity in an absolute alcohol-induced ulcer (ETH),which is a reliable model of acute gastric mucosal ulceration, involvingthe stimulation of acid secretion, mucosal edema, sub-epithelialhemorrhage and oxygen free-radicals (Lee et al., 2012). The resultsshowed that the extracts CtWEaE, CtIvEaE and CtWAE exerted a highprotective effect (480%) against gastric lesions in contrast to thereference drug famotidine (32.8%). Chemical analysis demonstratedthat the major constituent of the active extracts is verbascosidefollowed by aucubin. Verbascoside has been reported to protect fromgastric ulcers because it reduces gastric acid secretion; its anti-secretory activity is related to the inhibition of Hþ , Kþ-ATPase activityand suppression of the hormone gastrin which modulates gastric acidsecretion (Singh et al., 2010). Although verbascoside and isoverbasco-side are effective radical scavengers, the observed effects may also belinked to their antioxidant capacity (Hanson et al., 2011). Aucubin orother related IGs have not been reported as anti-ulcerogenic agents.

To explore if the anti-ulcerogenic effect of Castilleja tenuifloraextracts is mediated by NO, the best anti-ulcerogenic extract(CtWEaE) was evaluated with IDN, a NO-donor drug. The combi-nation of CtWEaEþ IDN (12.50þ2.50 mg/kg) exhibited a 65.2% of

Table 3Effect of Castilleja tenuiflora extracts on ethanol-induced gastric lesion in rats.

Pre-treatment Dose (mg/kg) Ulcer index Protection (%)

Vehicle – 0.189870.0140 –

CtWEaEa 12.50 0.324570.0372 NI25 0.105170.0459nnn 44.650 0.065570.0083nnn 65.5100 0.022270.0099nnn 88.3

CtWME 100 0.156270.0085 17.7CtWAE 100 0.034870.0076nnn 81.6CtIvEaE 100 0.032070.0048nnn 83.1CtIvME 100 0.094270.0218nnn 50.3CtIvAE 100 0.165470.0194 12.8Famotidine 20 0.127470.0143nnn 32.8

Values are mean7S.E.M. (n¼5). ANOVA post-test of Dunnett with nnnpo0.001 incomparison with control group (Vehicle).

a Dose–response curve of the most active extract (CtWEaE). NI¼Not Inhibits.Ct¼Castilleja tenuiflora; W¼Wild; Iv¼ In vitro; EaE¼Ethyl acetate extract; ME;Methanol extract; AE¼Aqueous extract.

Table 4Effect of the co-administration of CtWEaE plus Isosorbide dinitrate (IDN), atineffective dose, on ethanol-induced gastric lesion in rats.

Pre-treatment Dose (mg/kg) Ulcer index Protection (%)

Vehicle – 0.189870.0140 –

CtWEaE 12.50 0.324570.0372 NIIDN 2.50 0.291570.0360 NICtWEaEþ IDN 12.50þ2.50 0.066070.0170nnn 65.20Famotidine 20 0.127470.0143nnn 32.80

Values are mean7S.E.M. (n¼5). ANOVA post-test of Dunnett with nnnpo0.001 incomparison with control group (Vehicle). NI¼Not Inhibits. Ct¼Castilleja tenuiflora;W¼Wild; EaE¼Ethyl acetate extract.

P. Mauricio Sanchez et al. / Journal of Ethnopharmacology 150 (2013) 1032–10371036

gastric ulcer inhibition (Table 4). Consequently, we suggest thatextracts of Castilleja tenuiflora possess anti-inflammatory and anti-ulcerogenic effects mediated by NO. NO can influence the gastro-intestinal mucosal integrity through many mechanisms includingreduction of mast cell degranulation and stimulating the release ofpro-inflammatory substances from macrophages, platelets, andneutrophils. NO will also reduce neutrophil adherence to thevascular endothelium and stimulate epithelial secretion. NO isknown to dilate blood vessels through relaxation of vascularsmooth muscle. NO also appears to be a key mediator of gastro-intestinal mucosal defense, and this has been exploited in thedevelopment of a novel class of anti-inflammatory drugs, called“NO-NSAIDs” (Nitric Oxide-Non Steroidal Anti-inflammatoryDrugs), which exhibit little if any gastrointestinal toxicity(Wallace and Miller, 2000).

In the present study, we also demonstrated that CtWEaE, CtIvEaEand CtWAE displayed anti-inflammatory activity in a TPA-inducededema model in ICR mice and anti-ulcer activities toward a ETH-induced gastric lesion in rats as evidenced by the significant protectionafforded to the indications in these models. Based on our preliminaryresearch, these anti-inflammatory and anti-ulcer activities may beattributed to the presence of compounds belonging to the PhGs andiridoids. As shown herein, both wild and in vitro plants of Castillejatenuiflora are important reservoirs of bioactive components. Thein vitro plants are a viable source of homogeneous material that isfree of pathogens for drug production while having the advantage oflimiting the harvest of the wild plant. Further work is ongoing into thepharmacological effects of the components of Castilleja tenuiflora.

Acknowledgments

This research was financed by the Secretaría de Investigación yPosgrado del Instituto Politécnico Nacional (SIP-IPN, Grants20110153, 20120745 and 20131786) and Fondo de InvestigaciónBásica del Consejo Nacional de Ciencia y Tecnología-México(CONACYT, Grant 100202). PMSO is indebted to CONACyT andPrograma Institucional de Formación de Investigadores (PIFI-IPN)for the awarded doctoral fellowship.

References

Akdemir, Z., Kahraman, C., Tatli, I.I., Küpeli Akkol, E., Süntar, I., Keles, H., 2011.Bioassay-guided isolation of anti-inflammatory, antinociceptive and woundhealer glycosides from the flowers of Verbascum mucronatum Lam. J. Ethno-pharmacol. 136, 436–443.

Ankli, A., Heinrich, M., Wolfram, B.P., Bauerfeind, P., Brun, R., Schmid, C., Weiaa, C.,Gertsch, J.B.R., Wasescha, M., Sticher, O., 2002. Yucatec mayan medicinal plant.Evaluation based on indigenous uses. J. Ethnopharmacol. 79, 43–52.

Béjar, E., Reyes-Chilpa, R., Jiménez-Estrada, M., 2000. In: Atta, R. (Ed.), BioactiveCompounds from Selected Plants Used in the XVI Century Mexican TraditionalMedicine, 24. Studies in Natural Products Chemistry Elsevier, pp. 799–844.

Biblioteca Digital de la Medicina Tradicional Mexicana. 2009. Web Page at: ⟨http://www.medicinatradicionalmexicana.unam.mx/index.php⟩ (accessed April 2013).

Cano, A.L.M., 1997. Flora Medicinal de Veracruz. I. Inventario Etnobotánico, PrimeraEdición. Universidad Veracruzana, Xalapa, Veracruz.

Chen, R.C., Su, J.H., Yang, S.M., Li, J., Wang, T.J., Zhou, H., 2002. Effect of isoverbasco-side, a phenylpropanoid glycoside antioxidant, on proliferation and differentia-tion of human gastric cancer cell. Acta Pharmacol. Sin. 23, 997–1001.

Cos, P., Vlietinck, A.J., Berghe, D.V., Maes, L., 2006. Anti-infective potential of naturalproducts: how to develop a stronger in vitro ‘proof-of-concept’. J. Ethnophar-macol. 106, 290–302.

Díaz, A.M., Abad, M.J., Fernández, L., Silván, A.M., De Santos, J., Bermejo, P., 2004.Phenylethanoid glycosides from Scrophularia scorodonia: in vitro anti-inflammatory activity. Life Sci. 75, 1149–1157.

Filippini, R., Piovan, A., Borsarini, A., Caniato, R., 2010. Study of dynamic accumula-tion of secondary metabolites in three subspecies of Hypericum perforatum.Fitoterapia 81, 115–119.

Gálvez, M., Martín-Cordero, C., Jesús-Ayuso, M., 2005. Iridoids as DNA topoisome-rase I poisons. J. Enzyme Inhib. Med. Chem. 20, 389–392.

Georgiev, M., Pastore, S., Lulli, D., Alipieva, K., Kostyuk, V., Potapovich, A., Panetta, M.,Korkina, L., 2012. Verbascum xanthophoeniceum-derived phenylethanoid glycosidesare potent inhibitors of inflammatory chemokines in dormant and interferon-gamma-stimulated human keratinocytes. J. Ethnopharmacol. 144, 754–760.

Gómez-Aguirre, Y.A., Zamilpa, A., Gonzales-Cortazar, M., Trejo-Tapia, G., 2012.Adventitious root cultures of Castilleja tenuiflora Benth. as a source of pheny-lethanoid glycosides. Ind. Crops Prod. 36, 188–195.

Graham, J.G., Quinn, M.L., Fabricant, D.S., Farnsworth, N.R., 2000. Plants usedagainst cancer-an extension of the work of Jonathan Hartwell. J. Ethnopharma-col. 73, 347–377.

Hanson, A., Joubert, E., De Beer, D., Malherbe, C.J., Witthuhn, R.C., 2011. Pheny-lethanoid glycosides as major antioxidants in Lippia multiflora herbal infusionand their stability during steam pasteurization of plant material. Food Chem.127, 581–588.

Harput, U.S., Genc, Y., Saracoglu, I., 2012. Cytotoxic and antioxidative activities ofPlantago lagopus L. and characterization of its bioactive compounds. FoodChem. Toxicol. 50, 1554–1559.

Holmgren, N.H., 1976. Four new species of mexican Castilleja (subgenus Castilleja,Scrophulariaceae) and their relatives. Brittonia 28, 195–208.

Houghton, P., Fang, R., Techatanawat, I., Steventon, G., Hylands, P.J., Lee, C.C., 2007.The sulphorhodamine (SRB) assay and other approaches to testing plantextracts and derived compounds for activities related to reputed anticanceractivity. Methods 42, 377–387.

Hung, J.Y., Yang, C.J., Tsai, Y.M., Huang, H.W., Huang, M.S., 2008. Antiproliferativeactivity of aucubin is through cell cycle arrest and apoptosis in human non-small cell lung cancer A549 cells. Clin. Exp. Pharmacol. Physiol. 35, 995–1001.

Lee, M-Y., Shin, I-S., Jeon, W-Y., Seo, C-S., Ha, H., Huh, J-I., Shin, H-K., 2012. Protectiveeffect of Bojungikki-tang, a traditional herbal formula, against alcohol-inducedgastric injury in rats. J. Ethnopharmacol. 142, 346–353.

López-Laredo, A.R., Gómez-Aguirre, Y.A., Medina-Pérez, V., Salcedo-Morales, G.,Sepúlveda-Jiménez, G., Trejo-Tapia, G., 2012. Variation in antioxidant propertiesand phenolics concentration in different organs of wild growing and greenhousecultivated Castilleja tenuiflora Benth. Acta Physiologiae Plant. 34, 2435–2442.

Lubbe, A., Verpoorte, R., 2011. Cultivation of medicinal and aromatic plants forspecialty industrial materials. Ind. Crops Prod. 34, 785–801.

Martínez-Bonfíl, B.P., Salcedo-Morales, G., López-Laredo, A.R., Ventura-Zapata, E.,Evangelista-Lozano, S., Trejo-Tapia, G., 2011. Shoot regeneration and determi-nation of iridoid levels in the medicinal plant Castilleja tenuiflora Benth. PlantCell Tissue Organ Cult. 107, 195–203.

Moreno-Escobar, J.A., Bazaldúa, S., Villarreal, M.L., Bonilla-Barbosa, J.R., Mendoza, S.,Rodríguez-López, V., 2011. Cytotoxic and antioxidant activities of selectedLamiales species from Mexico. Pharm. Biol. 49, 1243–1248.

Noor, S.M., Mahmood, A.A., Salma, I., Philip, K., 2006. Prevention of acute gastricmucosal lesions by R. hasseltii in rats. JAVA 5, 161–164.

Park, K.S., Kim, B.H., Chang, I.M., 2007. Inhibitory potencies of several iridoids oncyclooxygenase-1, cyclooxygenase-2 enzymes activities, tumor necrosis factor-α and nitric oxide production in vitro. Evidence Based Complement. Altern.Med. 7, 41–45.

Payá, M., Ferrándiz, M.L., Sanz, M.J., Bustos, G., Blasco, R., Rios, J.L., 1993. Study of theantioedema activity of some seaweed and sponge extracts from the Mediterra-nean coast in mice. Phytother. Res. 7, 159–162.

Pettit, R.G., Numata, A., Takemura, T., Ode, R.H., Nurula, A.S., Schmidt, J.M., Cragg, G.M., Pase, C.P., 1990. Antineoplastic agents, 107. Isolation of acteoside andisoacteoside from Castilleja linariaefolia. J. Nat. Prod. 53, 456–458.

Prescott, T.A.K., Veitch, N.C., Simmonds, M.S.J., 2011. Direct inhibition of calcineurinby caffeoyl phenylethanoid glycosides from Teucrium chamaedrys and Nepetacataria. J. Ethnopharmacol. 137, 1306–1310.

Samonina, G.E., Kopylova, G.N., Lukjanzeva, G.V., Zhuykova, S.E., Smirnova, E.A.,German, S.V., Guseva, A.A., 2004. Antiulcer effects of amylin: a review.Pathophysiology 11, 1–6.

Sharma, N., Pathania, V., Singh, B., Gupta, R.C., 2012. Intraspecific variability of mainphytochemical compounds in Picrorhiza kurroa Royle ex Benth. from NorthIndian higher altitude Himalayas using reversed-phase high-performanceliquid chromatography. J. Med. Plants Res. 6, 3181–3187.

Shuai, W., Yong-rui, B., Yun-Peng, D., Xian-Sheng, M., Ting-Guo, K., 2011. Evaluation ofgastric ulcer model based on gray-scale image analysis. Afr. J. Microbiol. Res. 5,1285–1290.

Singh, N., Shukla, N., Singh, P., Sharma, R., Rajendran, S.M., Maurya, R., Palit, G.,2010. Verbascoside isolated from Tectona grandis mediates gastric protection inrats via inhibiting proton pump activity. Fitoterapia 81, 755–761.

Suffness, M., Pezzuto, J.M., 1991. Assays related to cancer drug discovery. In: Hostett-mann, K. (Ed.), Methods in Plant Biochemistry. Academic Press, London, pp. 71–133.

Tarnawski, A.S., 2005. Cellular and molecular mechanisms of gastrointestinal ulcerhealing. Dig. Dis. Sci. 50, S24–S33.

Wagner, H., Bladt, S., Zgainski, E.M., 1983. Plant Drug Analysis: A Thin LayerChromatography Atlas, second ed. Springer-Verlag, Germany.

Wallace, J.L., Miller, M.J.S., 2000. Nitric oxide in mucosal defense: a little goes a longway. Gastroenterology 119, 512–520.

P. Mauricio Sanchez et al. / Journal of Ethnopharmacology 150 (2013) 1032–1037 1037