NPC Natural Product Communications Antifeedant and Phagostimulant Activity of Extracts and Pure Compounds from Hymenoxys robusta on Spodoptera exigua (Lepidoptera: Noctuidae) Larvae 2014 Vol. 9 No. 7 895 - 898 Zaida N. Juáreza, Antonio M. Fortunab, Eugenio Sánchez-Arreolac, Jesús F. López-Olguínd, Horacio Bache and Luis R. Hernándezc,* a Departamento de Ciencias Biológicas, Facultad Biotecnoambiental, Universidad Popular Autónoma del Estado de Puebla, 72410, Puebla, México b Facultad de Agronomía y Zootecnia, Universidad Nacional de Tucumán, Av. Roca 1900, 4000, San Miguel de Tucumán, Argentina c Departamento de Ciencias Químico Biológicas, Universidad de las Américas Puebla, Ex Hacienda Santa Catarina Mártir s/n, 72810, Cholula, Puebla, México d Centro de Agroecología, Instituto de Ciencias, Benemérita Universidad Autónoma de Puebla, 14 Sur 6301 C.U., 72570, Puebla, Puebla, México e Department of Medicine, Division of Infectious Diseases, University of British Columbia, 2733 Heather St. BC V5Z 3J5, Vancouver, Canada luisr.hernandez@udlap.mx Received: February 24th, 2014; Accepted: May 21st, 2014 In this study, the insecticidal and antifeedant activities of compounds from the leaves of Hymenoxys robusta on Spodoptera exigua, also known as armyworm, are reported. The methanol extract possessed suppressive activity, as well as a high antifeedant activity, suggesting that this extract has toxic effects on larvae. Compounds isolated from the plant show that a derivative of inositol and hymenolides stimulate the feeding, while vermeerin shows a high antifeedant effect. In addition, the methanol extract inhibited oviposition, whereas consumption of the n-hexane extract and hymenolides produced infertile eggs. These findings suggest that compounds extracted from H. robusta have potential for the development of products for pest control. Keywords: Hymenoxys robusta, Asteraceae, Antifeedant activity, Toxicity, Vermeerin, Spodoptera exigua. A major challenge in agriculture is the persistent attack of insects on crops causing onerous losses to farmers. Currently, most pest control is mediated by the use of chemical insecticides that are strictly regulated due to their environmental impact, selectivity, emerging parasite resistance, and efficiency. Therefore, there is a need to develop alternative methods of pest control that are environmentally safe and compatible with the principles of integrated control and sustainability of agriculture. One such alternative is the use of natural compounds extracted from plants, and it is expected that in the future biologically active plant-derived chemicals will play a growing role in the development of new formulations for crop protection [1]. The genus Hymenoxys, family Asteraceae, is widely distributed from Canada to Argentina. Members of this genus are known as poisons to livestock [2a-c] and their toxicity is mainly due to the secohelenanolide, 10-methylsecopseudoguaianolide (hymenoxon), a dihemiacetal sesquiterpene lactone [2a,3], which has been shown to possess cytotoxic effects [2a]. The main symptoms include a bleeding heart, fatty degeneration of the liver, lung congestion and digestive system problems. Hymenoxys robusta (Rusby) Parker (named kellu-kellu in Bolivia) is a poisonous weed found in a restricted area of South America, from southern Peru to northern Argentina, an area of scarce resources that is difficult for agriculture. Studies conducted in Bolivia reported that this plant kills livestock with a mortality rate of 30% [4a]. It is an invasive plant and contaminates alfalfa pastures and other grazing ground for livestock. This forces the inhabitants of these regions to leave in search of better land. In Bolivia, a plan has been implemented by the Economic Commission for Latin America (ECLA) of the United Nations to eradicate this plant [4b]. Given the detrimental toxic effects of the plant on cattle, we investigated the possibility of finding some use for it as an insecticide to be available for farmers, bringing implicit and straightforward benefits, such as (a) no need to eradicate this plant, as proposed in a Bolivian project [4b]; and (b) no major ecological impact as a result of the eradication of a species that can directly affect the balance of the ecosystem of the area. Following a program to find new natural products with insecticidal activity, the insecticidal activities of extracts and pure compounds (1-6) isolated from H. robusta were assessed on the growth of the armyworm, Spodoptera exigua (Hübner) (Lepidoptera: Noctuidae). This insect was chosen because it is an economically important pest that occurs in a cyclical manner. It feeds from the foliage, stems, and sometimes roots, while its larval stage attacks a range of different crops [5]. Our results indicate that compounds extracted from H. robusta effectively decrease armyworm feeding behavior in a no preference test and show important post-treatment effects. The antifeedant activities of three extracts of H. robusta on larvae of S. exigua are shown in Table 1 (only the best concentration in the assayed range is shown). Neither antifeedant nor deterrent activities, but a phagostimulating activity was observed when the larvae were fed on the n-hexane extract (HE) treated discs, in a position of 896 Natural Product Communications Vol. 9 (7) 2014 Juarez et al. choice. The chloroform extract (CE) showed a moderate effect on suppression and low antifeedant effects. In this case we conclude that larvae prefer treated discs in a preference test, and the consumption is reduced in a no preference test. The methanol extract (ME) showed high suppressive and important antifeedant activity, but not a deterrent effect. We conclude that larvae show an acute toxic effect from the SI value, while low consumption was observed in a no preference test. The high suppression and antifeedant indexes suggest that larvae stop to feed after consuming a very little amount of treated discs, even when untreated discs are available (preference test). This behavior indicates that the treatment has toxicity. Table 2: Indexes of pure compounds from Hymenoxys robusta. Compound (300 ppm) AIa Inositol derivative (1) 17.1 (8, 29)d Hymenolides (4-5) 1.3 (-9, 7)d Vermeerin (6) 63.5 (46, 72)e DIb 13.5 (0.2, 22)d 21.5 (17, 27)d 40.4 (3, 53)d SIc -41.4 (-52, -32)d -47.7 (-65, -15)d -20.9 (-31, 1)d AI=Antifeedant index, DI=Dissuasion index, SI= Suppression index. Kruskal-Wallis test, a T=27.92, p=9.7x10-5, b T=2.27, p=0.52, c T=7.23, p=0.065. No significance is labeled with the same letter in the same column (Mann-Whitney test, p  0.05). Shown are indexes (quartile 25, quartile 75) calculated with n=10. 14 OAc HO H OR 1 10 3 1 3 AcO O OAc Oi-Bu O 4 12 6 11 15 Table 1: Indexes of extracts from Hymenoxys robusta. Extract n-Hexane (480 ppm) Chloroform (1000 ppm) Methanol (400 ppm) AIa (quartile 25, quartile 75) -18.15 (-30, -1)c 25.0 (-7, 46)cd 54.5 (30, 68)d O OH 13 DIb 9.858.7c NAc 12.958.7c 1 R = iBu 2 R = 2-MeBu SIb -47.821.4c 35.125.5d 89.421.4d 3 14 10 O 1 3 4 H O O AI=Antifeedant index, DI=Dissuasion index, SI= Suppression index. a The alternative hypothesis was considered (Mann-Whitney, p  0.05). No significance is labeled with the same letter in the same column. b Shown is the meanSE. NA = No activity was detected. No significance is labeled with the same letter in the same column (Tukey test p  0.05). Indexes were calculated with n=10. O O O 6 O OH 13 R 4 R = -OH 5 R = -OH 6 Identified compounds from the different extracts were also assayed for their insecticidal activity and the results are summarized in Table 2 (only the best concentration in the assayed range is shown). When the inositol derivative (1) was assayed, an increase in the consumption of treated discs was observed, whereas low antifeedant and deterrent indices were calculated. In the case of the treatment with the mixture of hymenolides (4-5), similar behavior as for inositol was observed. Interestingly, vermeerin (6) showed a moderate antifeedant effect, low deterrent activity, and a suppression rate that indicates that the larvae fed more from the treated discs. Table 3: Post-treatment analyses of extracts and compounds from Hymenoxys robusta. Interestingly, compounds 1-6 were isolated from the CE. This extract shows a better activity regarding the SI, whereas compounds 1, 4, and 5 show negative values of SI, indicating a moderated stimulant activity. In contrast, compound 6 shows only antifeedant activity. Thus, the antifeedant activity of the CE can be explained because compound 6 was the major compound isolated from the extract. Overall, from the analysis of the results, we conclude that (a) compound 1 stimulated feeding in a preference test, (b) the hymenolide mixture (4-5) caused the larvae to choose primarily the treated discs, and (c) in the presence of compound 6, larvae preferred to feed mainly on non- treated discs in a preference test. Compounds 2 and 3 showed no significant results. a On the other hand, the objective of post-treatment analyses was to determine whether consumption of the extracts and pure compounds by the larvae could cause any anomalies in the successive steps of metamorphosis. Post-treatment analyses are summarized in Table 3 and show that 50% mortality was calculated when larvae ingested the CE in a no preference test. The same extract caused a mortality of 50% and 40% in the pupa and adult stages, respectively, when a preference test was performed. Pupae showed malformations and surviving adults showed deformities in their wings. As listed in Table 3, other extracts and compounds also caused a low mortality rate. Interestingly, although the HE caused a low mortality rate, it has high activity in the adult stage by inducing oviposition and the production of infertile eggs. In the case of CE, larvae assessed in the preference test also show that adults laid infertile eggs. It is worth mentioning that feeding of the larvae with ME produced an inhibition in the production of eggs in both tests, even when the previous metamorphosis stages had no mortality. Extract/ Compound No Preference n-Hexane Chloroform Methanol Inositol derivative (1) Hymenolides (4-5) Vermeerin (6) Preference n-Hexane Chloroform Methanol Inositol derivative (1) Hymenolides (4-5) Vermeerin (6) Pupa stagea Adult stagea Oviposition Fertile eggs 20 50 NS NS NS NS NS NS 20 NS NS NS Stimulated Normal Inhibited Normal Decreased Decreased 0% 80% 0% 0% 0% 20 50 30 40 NS 20 NS 40 40 NS NS 20 Normal Decreased Inhibited Normal Normal Decreased 0% 0% 30% 0% 80% Expressed as mortality (%). NS: Not significant Post-treatment analyses of pure compounds showed that larvae treated with compound 1 in the preference test had 40% mortality in the pupa stage. When larvae were treated with compound 6, pupae and moths showed deformation in the wings. In conclusion, the HE has a stimulant effect on food ingestion, and the ME possesses suppressive activity, as well as high antifeedant activity, suggesting that this extract has toxic effects. Pure compounds isolated from the plant show that compound 1 and hymenolides (4-5) stimulate feeding, while compound 6 has a high antifeedant effect. Analyses post treatment indicated that ME inhibited oviposition, whereas consumption of the HE and hymenolides (4-5) produced infertile eggs. Of special interest is the ME, which inhibited oviposition and stopped the insect cycle. Interestingly, compounds 1 and 3 differ only in the presence of an ester group in their molecular structures. It has been reported that a difference in a single functional group is sufficient to change bioactivities substantially. For example, the sesquiterpene lactones cnicin and onopordopicrin, are two compounds that differ only in the presence of a hydroxyl group. Interestingly, this difference is enough to abolish the antibacterial activity of onopordopicrin [6]. Further studies (X-ray crystallography) demonstrated that the lack of this hydroxyl group is sufficient to avoid a proper binding of onopoprdopicrin to the enzyme MurA, responsible for the bacterial cell wall production [7]. Bioactivity of Hymenoxys robusta on Spodoptera exigua Taking in account that H. robusta is a weed, and the extract can be easily prepared, this plant constitutes a cheap, attractive, and alternative source for the development of products for pest control, especially in organic agriculture. Experimental Collection of plant material: The material studied was collected by Dr Antonio M. Fortuna in January 2006 in the province of Jujuy, Humahuaca Department, City of Uquía, Argentina at 2800 m above sea level. Collected plants were classified by Dr Alberto Slanis and a voucher recorded Fortuna s/n (LIL416346) was deposited in the herbarium of the Instituto Miguel Lillo, Tucumán province, Argentina. Extract preparation: Leaves were dried and sequentially extracted with n-hexane, chloroform, and methanol using maceration, as published [8]. Extracts were fractionated by chromatographic methods, including CC, followed by TLC, as published [8]. The HE, CE, and ME at a final concentration from 400 to 1000 ppm and all the identified compounds (1-6) at different concentrations (from 80 to 300 ppm) were tested to determine their antifeedant and insecticidal activities. Spodoptera exigua (Hübner) (Lepidoptera: Noctuidae) breeding: Larvae of S. exigua were collected every 6 months in a beet field in the town of Reyes de Juárez, Puebla, México. It was verified that no insecticides were applied at least one month prior to collection of larvae. The larvae were collected and placed in plastic containers (30x20x6 cm) lined inside with a filter paper bed and beet (Beta vulgaris var. cicla) leaves. The larvae were transported to the laboratory of Biopesticides at the Department of Chemical Biological Sciences of Universidad de las Américas Puebla (UDLAP). The specimens collected were kept in quarantine for three generations on artificial diet, as described below. The larvae of the fourth generation onwards were used for the tests described in this work. To feed the larvae, a beet culture was established in an eco-friendly nursery at the UDLAP. On reaching the adult period, moths were placed in round plastic containers (10 cm diameter x 13 cm height) lined inside with filter paper. Filter paper strips were placed inside the same container where eggs were laid. Adults were fed with a saturated solution of sugar, placed on cotton in small containers inside the round plastic containers. Laid eggs were transferred to plastic containers (30x20x6 cm) lined inside with filter paper. Larvae were kept in a chamber at 25°C, with a 75% relative humidity, and a 12:12 h photoperiod (light: darkness). Insecticidal and antifeedant activity assays: The assays were carried out in a 15 x 90 mm plastic dish with the bottom covered by a layer of 2.5% agar, as published [9]. On the agar base, 6 equidistant holes of 13 mm each were drilled and beet leaf discs of 11 mm in diameter were placed within these holes. These discs were embedded with 12 L of either plant extracts or pure isolated compounds dissolved in either acetone or ethanol. Solvents alone were used as negative controls. After evaporation of the solvent, a fifth instar larva within the first 24 h of age, kept fasted for 6 h, and weighing between 100-120 mg, was placed in each dish. Ten repetitions were performed for each treatment. Dishes were placed randomly in a growth chamber with the same conditions used to grow the larvae for either 6 h or until the larvae consumed 4 out 6 discs when a preference experiment was performed (see below). When a non-preference experiment was performed, the experiment was stopped when 3 out of 6 discs were consumed. At the end of the experiment, larvae were placed on artificial diet prepared as follows: 80 g agar, 250 g beans (Phaseolus vulgaris), 200 g wheat germ, 200 g soy protein, 140 g casein, and 200 g yeast. All the Natural Product Communications Vol. 9 (7) 2014 897 ingredients were mixed in 6 L of water and boiled for 15 min. The mixture was cooled down and when it reached 65oC, 18 g of ascorbic acid, 9 g of sorbic acid, 250 mg of tetracycline, 10 mL of 40% formalin, 15 g of methyl parabenzene, and 30 g of a vitamin complex (Centrum) were added [10]. The mixture was blended and aliquoted before solidification. At the beginning and the end of the assays, disks were weighed and dried at either 60oC for 48 h or until no changes were observed after 3 consecutive weighing. No preference test: This test was based on the ingestion of an average of 75% of the beet leaf discs by larvae in the control experiments. The antifeedant activity was evaluated by recording whether the insect consumes the treated discs or not. In this situation of no preference, an anti-appetitive index (AI) was calculated using the equation [11a]: AI = [(Dc-Dt)/Dc] x 100 where Dc = ingestion of non-treated discs (control), Dt = ingestion of treated discs. Preference test: In this experiment, 3 out the 6 discs placed in the same dish were embedded with 12 L of either extract or isolated compounds solutions, whereas the other 3 discs were embedded only with the solvents. The treated and control discs were distributed alternately in the dish. On each dish, the insect could choose between a beet leaf disc either treated or untreated, allowing the assessment of the degree of perception that a larva has into a compound and its deterrent effect for food. A compound has a deterrent effect when the insect, after feeding on a treated disc, stops feeding on it and continues feeding on control discs. The deterrence index (DI) was calculated using the equation [11b]: DI = [(Dc-Dt)/Dc+Dt] x 100. Suppression rate: With the data obtained from the treated and control plates of the preference test, the suppression index (SI) was calculated as follows [11c]: SI = [(IngTest-IngTreat)/IngTest] x 100 where IngTreat is the ingestion on plates with treated and untreated discs (preference test), while IngTest is the ingestion of external control dishes. This index assesses the effect of inhibition of feeding on untreated beet leaf discs as a result of ingesting treated discs, measuring a potential for acute toxicity. After completion of all of the tests, larvae were placed back on the artificial diet in plastic containers, separated by treatment for posttrial follow-up, and kept in the environmental chamber, separated from breeding. This assay aims to observe the effect of ingested substances in the development and behavior of the insect to the next generation following the trial in order to evaluate anomalies and/or insecticidal activity. Statistical analyses: A statistical analysis of variance was carried out according to a completely randomized design, after verification of homogeneity of variances (Barttlet test), followed by multiple comparisons of treatment means using the Tukey test. All comparisons were made considering a significance level of 5%. In the case of heterogeneous variances, with p<0.05, data were analyzed using a non-parametric Kruskal-Wallis. The medians were compared by the method of overlapping the interquartile range of the box and whisker graph. Activities were defined as low when the index value ranged between 30-45 %, medium when the index was in the range of 45 to 70%, and high when the index was > 70% 898 Natural Product Communications Vol. 9 (7) 2014 [11a-d]. Negative values of indexes indicate an opposite activity. For instance, a negative antifeedant index indicates phagostimulant activity. Juarez et al. Acknowledgments – We thank Alberto Slanis for plant classification. We also thank Jeffrey Helm for helpful discussions. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] Balandrin MF, Klocke JA, Wartele ES, Bollinger WH. (1985) Natural plant chemicals: sources of industrial and medicinal materials. Science, 228, 1154-1160. (a) Ivie GW, Witzel DA, Herz W, Kannan R, Norman JO, Rushing DD, Johnson JH, Rowe LD, Veech JA. (1975) Hymenovin. Major toxic constituent of western bitterweed (Hymenoxys odorata DC). Journal of Agricultural and Food Chemistry, 23, 841-845; (b) López TA, Campero CM, Chayer R, Calderón G, Quiróz J, Cura J. (1997) Toxicidad experimental aguda de Hymenoxys anthemoides (Botón de oro) en ovinos. Revista Veterinaria Argentina, 14, 525-529; (c) Pfister JA, Provenza FD, Panter KE, Stegelmeier BL, Launchbaugh KL. (2002) Risk management to reduce livestock losses from toxic plants. Journal of Range Management Archives, 55, 291-300. Hill DW, Kim, HL, Martin CL, Camp BJ. (1977) Identification of hymenoxone in Baileya multiradiata and Helenium hoopsii. Journal of Agricultural and Food Chemistry, 25, 1304-1307. (a) La Iniciativa de Comunicación. Mejoramiento de Pasturas y Lucha contra el Kellu–kellu, Planta Tóxica del Altiplano Boliviano – Bolivia 2013); (b) Economic Commission for Latin America. http://www.comminit.com/la/node/45085 (Aug. 7th, http://www.eclac.org/dds/Innovacionsocial/i/doc s/Summary.Luchacontrakellukellu.Bolivia.ENG.Final.pdf (Aug. 7th, 2013). SAGARPA. Norma oficial mexicana NOM-081-FITO-2001, Manejo y eliminación de focos de infestación de plagas, mediante el establecimiento o reordenamiento de fechas de siembra, cosecha y destrucción de residuos. In Diario Oficial de la Federación, 18th September, 2002, pp. 47-50. Bach SM, Fortuna MA, Attarian R, de Trimarco JT, Catalan CAN, Av-Gay Y, Bach H. (2011) Antibacterial and cytotoxic activities of the sesquiterpene lactones cnicin and onopordopicrin. Natural Product Communications, 6, 163-166. Steinbach A, Scheidig AJ, Klein CD. (2008) The unusual binding mode of cnicin to the antibacterial target enzyme MurA revealed by X-ray crystallography. Journal of Medical Chemistry, 51, 5143-5147. Fortuna AM, Juarez ZN, Bach, H, Nematallah A, Av-Gay Y, Sanchez-Arreola E, Catalan CA, Turbay S, Henandez LR. (2011) Antimicrobial activities of sesquiterpene lactones and inositol derivative from Hymenoxys robusta. Phytochemistry, 72, 2413-2418. Escoubas P, Lajide L, Mitzutani J. (1993) An improved leaf-disk antifeedant bioassay and its application for the screening of Hokkaido plants. Entomologia Experimentalis et Applicata, 66, 99-107. Cruz Valdez EA. (2002) Estimación de la DL50 y DL90 del virus Poliedrosis nuclear autographa californica en Spodoptera exigua y Helicoverpa zea. Thesis of Science and Farming Production. Zamorano, Honduras. (a) Bentley MD, Leonard DE, Stoddard WF, Zalkow LH. (1984) Pyrrolizidine alkaloids as larval feeding deterrents for spruce budworm, Choristoneura fumiferana (Lepidoptera: Tortricidae). Annals of Entomological Society of America, 77, 393-397; (b) Blaney WM, Simmonds MSJ, Ley SV, Kats B. (1987) An electrophysiological and antifeedant properties of natural and synthetic drimane-related compounds. Physiological Entomology, 12, 281-291; (c) Raffa KF, Frazier JL. (1988) A generalized model for quantifying behavioral desensitization to antifeedants. Entomologia Experimentalis et Applicata, 46, 93-100; (d) Morimoto M, Tanimoto K, Sakatani A, Komai K. (2002) Antifeedant activity of an anthraquinone aldehyde in Galium aparine L. against Spodoptera litura F. Phytochemistry, 60, 163-166.