Anti-lipase and antioxidant properties of 30 medicinal plants used in Oaxaca , México

We report the results of in vitro anti-lipase and antioxidant assays using crude ethanolic extracts from 30 plants grown in Oaxaca, México. Anti-lipase tests were performed by using porcine pancreatic lipase (PPL) [EC 3.1.1.3] from Aff ymetrix/USB. The extracts of Solanum erianthum, Salvia microphylla, Brungmansia suaveolens and Cuphea aequipetala showed up to 60% PPL inhibition. The eff ect of these extracts on the kinetic parameters of PPL (Km= 0.36 mM, and Vmax=0.085 mM min -1) revealed that the alcoholic preparations of S. erianthum and C. aequipetala engendered a non-competitive inhibition (Vmax=0.055 mM min -1; Vmax= 0.053 mM min -1), whereas those of S. microphylla and B. suaveolens produced a mixed inhibition (Km= 0.567 mM, Vmax=0.051 mM min -1; Km=0.643 mM, Vmax= 0.042 mM min -1). In addition to these fi ndings, seven extracts from diff erent plants were able to inhibit PPL in the range of 30-50%. Antioxidant tests against 2,2-Diphenyl-1-picryl hydrazyl (DPPH) confi rmed that Arctostaphylos pungens, Gnaphalium roseum, Crotalaria pumila, Cuphea aequipetala, Rhus chondroloma, and Satureja laevigata possess relevant antioxidant activity (IC50=50-80 μg mL -1). The general composition of the most eff ective ethanolic extracts was obtained in order to confi rm their known chemistry reported by previous works. Comprehensive chemical analysis of the ethanolic extracts and their poisoning eff ects suggests that S. microphylla, C. aequipetala and A. pungens could be considered as the best sources with both desired properties. Key terms: Phytotherapy, anti-obesity, antioxidant, Oaxacan ethnobotany. 6 Villa-Ruano BR 46-2.indd 153 28-06-13 17:53 VILLA-RUANO ET AL. Biol Res 46, 2013, 153-160 154 2005). 50 g of distinct plant organs (Table I) were excised and immediately extracted with 100 mL of absolute ethanol for 15 days at room temperature in darkness. The crude extracts were fi ltered with Ahlstrom grade 642 fi lter paper and concentrated using a Buchi R-200 rotovaporator to approximately 10 mL. These preparations were completely dried in a centrifuge concentrator (RVC 2-25 CD) for 3 h. The resultant powder was stored in amber glass fl asks at 4 °C until used. 3% DMSO was employed for resuspending the powder and to prepare diff erent stocks (this solution did not show anti-lipase activity by itself).


INTRODUCTION
High prevalence illnesses such as diabetes mellitus, cancer and neurodegenerative diseases are induced by oxidative stress and obesity.These physiological disorders are commonly associated with improper diets and sedentary lifestyle.Current statistical studies demonstrate an alarming increase in obesity and neurodegenerative diseases in the past five years in Mexico (Rojas-Martínez et al., 2012).This fact has turned into a serious health problem that requires special attention.Scientifi c evidence suggests that neurological disorders, respiratory diseases and premature ageing are the result of continuous oxidative stress which is aggravated by alcoholism, depression, intensive work, smoking and being overweight (Reuter et al., 2010).Being overweight or obese should be prevented with constant exercise and dietary routines.Nevertheless when the latter techniques fail in obtaining 10% weight loss, pharmacological treatment is required.Despite the fact that this alternative is highly recommended, there are controversial results on the use of some pharmacological agents such as sibutramine (Villa-Ruano et al., 2011).Synthetic derivatives of lipstatin that inhibit triglyceride hydrolysis are strongly recommended for controlling weight disorders.Unfortunately, Orlistat ® and related competitive inhibitors have side eff ects; however, they are relatively less aggressive compared with those of sibutramine.To date, the pharmaceutical market is limited to a short list of anti-obesity drugs.As a consequence of the need for new and more effi cient anti-lipase agents, exhaustive plant screenings from Asiatic, African and European ethnobotanical sources have been performed (Birari and Bhutani 2007;Ekanem et al., 2007;Slanc et al., 2009;Gholamhoseinian et al., 2010;Zheng et al., 2010).Nonetheless, there is no information about the effi cacy of Latin American medicinal plants on this topic.Mexico is a privileged country in possessing extensive information about the use and therapeutic properties of many plants.Within México, Oaxaca is considered one of the most biodiverse and multicultural states, but also highly marginalized (Caballero et al., 2004).Currently, people from rural communities of this state still depend on traditional folk medicine to cure their illnesses.In these places, the latter alternative is in many cases more accepted than allopathic medicine by native people.Here we report the results of in vitro anti-lipase and antioxidant assays using ethanolic extracts from plants of Oaxacan ethnomedicine in order to reveal some species with potential use in preventing obesity and cellular oxidative stress.

Source of plants
The plant material was collected in San Andres Paxtlán, Oaxaca, México (16°13.081'N 096°30.331'W; 1996 m elevation) in April 2012 (Spring 2012).The identity of 30 species was corroborated at the Herbarium of the FCME-UNAM México, or by dichotomous keys (Martinez, 1979;Calderón y Rzedowski, 2005).50 g of distinct plant organs (Table I) were excised and immediately extracted with 100 mL of absolute ethanol for 15 days at room temperature in darkness.The crude extracts were fi ltered with Ahlstrom grade 642 fi lter paper and concentrated using a Buchi R-200 rotovaporator to approximately 10 mL.These preparations were completely dried in a centrifuge concentrator (RVC 2-25 CD) for 3 h.The resultant powder was stored in amber glass fl asks at 4 °C until used.3% DMSO was employed for resuspending the powder and to prepare diff erent stocks (this solution did not show anti-lipase activity by itself).

Anti-lipase activity
The turbidimetric method proposed by Shihabi and Bishop (1971) was used to determine the triolein hydrolysis under controlled conditions.The preparation of triolein emulsion was achieved as described by the same authors.Porcine pancreatic lipase (PPL) [EC 3.1.1.3]was purchased from Aff ymetrix/ USB and adjusted to a fi nal concentration of 250 μg mL -1 (~ 6 units mL -1 ) in 100 mM Tris buff er pH 9.0, containing triolein (0.3 mM) sodium deoxycholate (20 mM), colipase (3 mg L -1 ) and calcium chloride (0.02 mM).Enzymatic reactions had a fi nal volume of 1 mL and were initiated by its incubation at 37 °C for 30 min with two diff erent concentrations of ethanolic extracts (50 and 100 μg mL -1 ).Changes in turbidity were measured at 340 nm by spectrophotometry (Beckman DU 7400).The percentage of inhibition was calculated according to Gholamhoseinian et al., (2010) as a semi-quantitative parameter to estimate the eff ectiveness of each extract.Orlistat ® (Medimart Labs) was employed as the reference standard of inhibition.Kinetic parameters of PPL were determined by saturation curves using triolein as a substrate (0.2, 0.4, 0.8, 1.6, 2.0 and 2.2 mM) and adjusted by non-linear regression using the Solver® software as described by Villa-Ruano et al. (2009).The specifi c inhibitory eff ect of ethanolic extracts on PPL kinetics was determined from different substrate concentrations (0.2-2.2 mM) and two extract concentrations (50 and 100 μg mL -1 ).Double-reciprocal Lineweaver-Burk plot regressions were performed with SigmaPlot 10.0 (Systat Software, Inc).All the experiments were assayed in quintuplicate.

Comparative phytochemical screening
To obtain the general chemical composition of the most eff ective plant extracts in both biological activities ( 22), a comparative phytochemical screening was performed.These extracts were directly processed to determine flavonoids, tannin/phenols, terpenes, steroids, saponins, alkaloids and cardiac glycosides according to general methods developed by Harborne (1998) and Háuad-Marroquín (2010).Semiquantitative content was estimated for some metabolites in order to get an approach to their abundance in ethanolic extracts.Flavonoid content was determined at 450 nm by calibration curves using quercetin as a fl avonoid standard.Total tannin/phenolic content was determined using gallic acid as internal standard and using the reagent of Folin-Ciocalteau (750 nm) according to Nugroho et al. (2013).Alkaloid content was calculated with Dragendorff 's reagent at 435 nm according to Villa-Ruano et al. (2012).Total saponin content was determined at 408 nm considering digitonin as reference standard.Cardiac glycosides were determined at 486 nm based on their reaction with picric acid in alkaline media using digitoxine as reference standard (Solich et al, 1992).Results were expressed as abundant for the case of ≥500 μg gFW -1 (fresh weight), low for 50-400 μg gFW -1 or absent (undetectable).Terpene and steroid determination was only qualitative.All the reference standards were obtained from Sigma-Aldrich Co. and the experiments were performed in triplicate.

Statistical analysis
The anti-lipase and antioxidant data were processed by an ANOVA coupled to a Tukey Test (p<0.05) using SigmaPlot 10.0 software.All results were expressed as mean ± standard deviation (n=5).In order to compare the two biological parameters, the results were plotted in a stacked bar graph.

RESULTS
Table I  ) but only some of these species showed signifi cant antilipase activity (>40%).Figure 1 shows that anti-lipase activity is not necessarily associated with antioxidant activity, but there were interesting coincidences for the case of S. erianthum, C. aequipetala and A. pungens.According to our results, Lippia alba, Crotalaria pumila, Marrubium vulgare and Chrysanthemum parthenium could be considered as alternatives to antioxidants, but not as anti-lipase sources (Figure 1).The ANOVA coupled to a Tukey test revealed significant differences among the effectiveness of ethanolic extracts (p<0.05) in the two biological activities (Table I).The comparative phytochemical screening of the most relevant extracts (22) showed diverse components dissolved in each ethanolic preparation (Table II).Nevertheless, the presence and high abundance of fl avonoids and/or phenolic compounds was evident in all the samples.Alkaloid content was abundant mainly for S.  conditions, the K m value of triolein for this enzyme was 0.36 mM and V max was 0.085 mM min -1 (Table III).Lineweaver-Burk plot analysis performed for the S. erianthum extract showed a non-competitive inhibition where only V max was signifi cantly decreased by both assayed concentrations (Fig. 2, Table III).The S. microphylla extract revealed a mixed inhibition on PPL (Fig. 3, Table III).Similar results were obtained for the alcoholic extract of fl owers from B. suaveolens (Fig. 4, Table III).The eff ect of these extracts was clearly refl ected in the modifi cation of the K m and V max values (Table III).The corresponding extract of C. aequipetala had a non-competitive eff ect on kinetics of PPL, like S. erianthum (Figure 5, Table III).

DISCUSSION
According to our results, the kinetic parameters of PPL (Aff ymetrix/USB) were as follows: K m = 0.36 mM and V max = 0.085 mM min -1 .The absolute values of these parameters are    similar to those reported by Gholamhoseinian et al. (2010) but slight variations in K m and V max could be caused by the assay buffer (pH 9.0) and its composition.Considering the biochemical evidence, four plants exhibited very strong in vitro anti-lipase activity (>60%), S. erianthum being the most eff ective.S. erianthum is currently used in Oaxaca State as an oral analgesic to treat stomach ache and as a cutaneous antimicrobial agent.The chemistry of S. erianthum consists of sesquiterpenes as well as high endogenous levels of steroidalalkaloid glycosides, tropane-alkaloids and tannins in the aerial parts (Essien et al., 2012;Maurya et al., 2013).Some strong antilipase activities are apparently associated with pentacyclictriterpene glycosides as is the case with teasaponins, which share an association in chemical structure with S. erianthum glycosides (Birari and Bhutani 2007; Tucci et al., 2010).The results of phytochemical screening support the presence of tannins, terpenes, steroids and saponins which could be involved in those effects (Table II).However, the noncompetitive inhibition by the crude ethanolic extract suggests that many compounds may be involved in the observed biochemical behavior.Despite the aerial parts of this plant having evident anti-lipase properties, the toxic endogenous + 50-400 µg gFW -1 ++ ≥500 µg gFW -1 ; -Undetectable under those conditions.* Triterpene saponins.√ Both steroid aglycones and terpene occurrences were qualitatively determined.
levels of tropane and Solanum alkaloids (Table II) must be considered before performing in vivo trials (Huang et al., 2009).Further chromatographic isolation of the bioactive molecules and their spectroscopic confirmations are necessary to determine the anti-lipase molecules of S. erianthum.S. microphylla was the second-best anti-lipase source.Phenolic compounds, sesquiterpenes, and 12-methoxycarnosic acid, as well as triterpene aglycones such as oleanolic acid, lupeol and β-sitosterol are the main constituents of this species (Aydogmuş et al., 2006).Our phytochemical tests confi rm the incidence of flavonoids tannins, terpenes, steroids and saponins that could include the latter compounds (Table II).Carnosic acid isolated from Salvia off cinalis exhibits a strong in vitro and in vivo PPL inhibition (Tucci et al., 2010).According to our results a mixed inhibition was observed by the ethanolic extract of S. microphylla.This fact suggests the putative involvement of 12-carnosic acid (as a possible competitive inhibitor) and other compounds such as pentacyclic diterpene glycosides and phenolic compounds as non-competitive inhibitors that bind to the enzyme-substrate complex.Carnosic acid and its derivatives seem to be chemo-taxonomic markers of the Salvia genus.Nevertheless, previous reports on anti-lipase properties of Salvia rhytidea and Salvia miltiorrhiza suggest that only some Salvia species possess worthwhile levels of effi cacy (Gholamhoseinian et al., 2010, Zheng et al., 2010).Geographical location, altitude, soil composition, weather and specifi c biotic elicitors may be involved in the biosynthesis of terpene and phenolic compounds in these plants.The high abundance of phenolic metabolites (Table II) is probably related to the signifi cant antioxidant activity (IC 50 = 99.2μg mL -1 ) of S. microphylla.The ethanolic extract of fl owers from B. suaveolens had a mixed inhibition eff ect on PPL.This species produces classical tropane alkaloids with medicinal uses whose hyoscine is hydrolyzed by this enzyme under buffering conditions (Henry, 1949).Our phytochemical studies also confi rm high endogenous levels of alkaloids and saponins as well as a low level of phenolic compounds in B. suaveolens flowers (Table II).This evidence could support the participation of alkaloids and saponins as the main anti-lipase agents.The mixed inhibitory eff ect on the kinetics of PPL suggests that some tropane alkaloids could interact with both the active site of the enzyme and the enzyme-substrate complex.According to our results, the fourth most effi cient species with anti-lipase activity was C. aequipetala, a native Mexican medicinal plant.The plant is commonly known as "hierba del cancer" and has been traditionally administered for centuries without formal reports of poisoning (Waizel-Bucay et al., 2003).According to previous phytochemical studies the aerial parts contain high endogenous levels of fatty acids, mainly miristic acid (in oil fraction), as well as fl avonoids and triterpene glycosides in the alcoholic extracts (Hirsinger, 1980;Waizel-Bucay et al., 2003).Detection of phenolics, terpenes, steroids and saponins in the phytochemical screening confi rm the possible incidence and involvement of the latter compounds in the eff ects here reported (Table II).Considering the results of this study and the chemistry of C. aequipetala, the signifi cant antioxidant properties (IC 50 = 76.5 μg mL -1 ) should be related to phenolic compounds and the strong noncompetitive inhibition on PPL (Figure 5, Table III) could be associated with triterpene glycosides and steroids.Further experiments are required to verify this hypothesis.Due to C. aequipetala demonstrating interesting nutritional properties and containing low levels of alkaloids (Table II), this plant could be considered a good candidate for in vivo trials.Another plant with strong anti-lipase activity is S. scandens.Currently this species is being studied by our research group and some of the chemical compounds from the stem bark are being isolated, characterized and quantified by chromatographic and spectroscopic methods (data not shown).Interestingly, the phytochemical screening reveals an alkaloid content that could be linked to the presence of specifi c alkylamides that contain nitrogen and probably interact with Dragendorff 's reagent.A. pungens represents another desirable source with anti-lipase and antioxidant properties.This study confi rms the previous fi ndings of Slanc et al. (2009) where this Ericaceae showed a similar inhibition degree on PPL.The intrinsic chemistry of A. pungens reveals the presence of phenolic compounds such as quercetin, catechins and tannins (Winkelman, 1989) which are strongly involved in both biological activities (Zheng et al., 2010).The general composition here found corresponds to high amounts of fl avonoids and related phenolics but also with some steroids (Table II).Further chromatographic separations are required to determine the anti-lipase agents of this plant.R. chrondoloma and C. anagyris showed good anti-lipase and acceptable antioxidant activities; nevertheless, their chemistry have been poorly studied (Young, 1979).According to our results R. chondroloma showed high levels of fl avonoids and tannins as well as terpenes and steroids (Table II).The moderate antioxidant activity of its ethanolic extract is probably linked to the high fl avonoid and phenolic content (Table II) whereas the anti-lipase effects could be also associated with the latter compounds but possibly with other metabolites such as steroids.Additionally, R. chondroloma sowed an interesting presence of cardiac glycosides.A similar condition is observed for the case of C. anagyris (except for cardiac glycosides and alkaloids).Phaseolus coccineus contains a great variety of flavonoids that could participate in the observed anti-lipase activity (Williams et al., 1995); the general composition of its ethanolic extract sustains this fi nding (Table II).Chiranthodendron pentadactylon ("Flor de Manita") is one of the most relevant plants in Mexican ethnobotany because of its anti-secretory properties (Velázquez et al., 2012).The anti-lipase eff ect reported in the present work contributes to the discovery of novel uses for this native plant that are strongly related to its antioxidant catechins and fl avonoids (Table II).The ethanolic extract of leaves from Sambucus mexicana showed similar antilipase eff ects to those of C. pentadactylon.High endogenous levels of anthocyanins and other phenolic compounds have been reported in this species (Lee et al., 2007).The general phytochemical composition of those extracts endorses the occurrence of phenolic metabolites, steroids and low levels of alkaloids (Table II) but detailed experimentation is required to reveal the anti-lipase agents of S. mexicana.Additionally, the presence of cardiac glycosides was detected in the leaf extracts of this plant; this finding could be considered as novel phytochemical information for the plant (Table II).Gholamhoseinian et al. (2010) reported a moderate anti-lipase activity of Artemisia santolina as we do for A. ludoviciana.This latter plant contains high fl avonoid content (Table II) that should be participating in the moderate antioxidant activity reported in this study.The terpene content of A. ludoviciana (Table II) could include small quantities of sesquiterpene lactones that perhaps interact with the active site of PPL.It is known that the lactone moiety of Orlistat plays an important function in the irreversible inhibition of PPL.There is no information about the possible involvement of sesquiterpene lactones on this matter so far.Also, the same authors report the same low anti-lipase activity for Marrubium anisodon as we do for M. vulgare.Although this species showed good phenolic content (Table II) and antioxiadant activity (Table I), its antilipase activity was non-signifi cant compared to other plant extracts.The ethanolic preparations of M. vulgare and C. parthenium clearly show that high phenolic content is not necessarily connected with relevant anti-lipase activity.Loeselia mexicana, Microsechium helleri, S. laevigata, I. wolcottiana and Physalis philadelphica showed a moderate anti-lipase activity and almost all of them except I. wolcottiana and L. mexicana demonstrated an acceptable antioxidant activity (Table I).Current phytochemical studies in M. helleri revealed the presence of many triterpene saponins (Hernández-Carlos et al., 2012), some of them perhaps with anti-lipase properties.
Presence and high abundance of saponins in this plant is also demonstrated in our phytochemical screening (Table II).C. pumila is an edible plant used to prepare traditional foods; the high phenolic content of this species (Table II) is strongly linked to its characteristic antioxidant activity (Table I).The presence of cardiac glycosides in the ethanolic extract of C. pumila could be a new chemical fi nding for this plant (Table II).
As a fi nal conclusion, we report on some medicinal plants used in the state of Oaxaca-Mexico that possess noteworthy anti-lipase and antioxidant properties.Some of those plant species might be considered as a cheap herbal therapy for the prevention of high prevalence diseases originated by oxidative stress and obesity.The known chemistry and general composition of some species was considered to explain their respective biological activities.Chemical composition and the correlation of the two biological activities demonstrate that plants with anti-lipase activity are not necessarily good antioxidant agents and vice versa.In addition, this study remarks on the need for performing accurate phytochemical experimentation in less-studied plants in order to identify and isolate their bioactive compounds.Comprehensive analyses of the chemistry and toxic risks of these plants leads us to consider S. microphylla, C. aequipetala and A. pungens as the best medicinal plants with both desired properties.

Figure 1 .
Figure 1.Effect of 30 ethanolic extracts on PPL activity and DPPH stability.Numbers correspond to the order of Table I.Number with asterisks indicate the reference standards used to normalize both biological activities.Bars represent the standard deviation from fi ve repetitions.

TABLE I
Anti-lipase and antioxidant activities of 30 plants from Oaxacan ethnnomedicine.

TABLE II
Comparative phytochemical screening of the most effective anti-lipase and antioxidant ethanolic extracts

TABLE III
Effect of the four best alcoholic extracts on the kinetic parameters of PPL Means with standard deviation from experiments performed in quintuplicate.*Results from saturation curves performed under laboratory conditions without any inhibitor