|USFLDC Home | Tropical Ecology Collection [Monteverde Institute]||| RSS|
This item is only available as the following downloads:
The role of extrafloral nectaries and saponins in plant herbivore interactions of Inga sierrae (Fabaceae: Mimosaceae) Kevin Davis Department of Botany, University of Wisconsin Madison ABSTRACT Trees of the Neotropical genus Inga are noted for the use o f extrafloral nectaries (EFNs) and secondary compounds to protect against herbivores. For Inga sierrae, the production of Extra EFNs (EEFNs) has been suggested to confer fitness on fledgling leaves, while the role of saponins on herbivory remains unclear. This study investigates the potential trade off between EEFN and saponin production and its effect on herbivory. I recorded percent herbivory on fledgling leaves during a three week period and measured saponin contents for I. sierrae trees at two different study sites at 1525m in Monteverde, Costa Rica. Encounter surveys were conducted to asses s the associated herbivore populations. EEFN production, saponin content, percent herbivory, and herbivore abundance varied between study sites. There was a weak neg ative correlation between EEFNs and percent herbivory. Differences in proximity to continuous forest between study sites may be causing observed differences in herbivore abundance, and in turn, percent herbivory. Future studies should focus on the effect o f saponin content on herbivore preferences and fungal and lichen infestation. Resumen El genero neotropical Inga usa los nectarios extraflorales (EFNs) y qumicos secundarios para protegerse contra los herbvoros. Para Inga sierrae, la produccin de EFNs extras (EEFNs) sugeiere una ventaja par a las hojas inmaduras, pero la funcin de los saponinos contra la herbivora no es claro. Este estudio investiga la posibilidad de un trmino medio entre los EEFNs y los saponinos. Determine el porcentaje de h erbivora para las hojas inmaduras durante tres semanas y calcul el contenido de los saponinos por arboles de I. sierrae en dos sitios de 1525m en Monteverde, Costa, Rica. Hice un estudio para tasar las poblaciones de los herbvoros asociados. La producci n de los EEFNs, contenido de los saponinos, porcentaje de herbivora, y la abundancia de los herbvoros fueron diferentes entre los sitios de estudio. Hubo una correlacin dbil entre EEFNs y porcentaje de herbivora. Es posible que las diferencias en pr oximidad al bosque continuo entre los sitios de estudio causen las diferencias observados en la abundancia de herbvoros, y en medio, porciento de herbivora. Estudios en el futuro deben concentrar se en el efecto del contenido de los saponinos para las pr eferencias de los herbvoros y la infestacin de los hongos y los lquenes.
INTRODUCTION Plants and herbivores constitute a majority of the visible biodiversity in tropical forests (Kursar et al. 2009). The co evolution of plants and the herbivores a nd pathogens that affect them represents an important part of any forested ecosystem, particularly in tropical forests, where there are higher rates of herbivory and pathogenic infestation (Coley & Barone 1996). As a result, tropical plants have evolved mo re numerous and varied defenses than their temperate counterparts, apparent primarily in young leaves, which incur the highest levels of herbivory. Extrafloral nectaries (EFNs) and secondary compounds are two well studied tropical examples of defenses ag ainst herbivory (Bently 1977). Using limited resources, plants attempt to allot the least energy into nectaries and secondary compounds while ensuring that leaves are sufficiently protected from herbivores and pathogens so as to return the investment in th e form of photosynthesis (Coley et al 1985). A plant may invest more heavily in secondary compounds to slow the growth of herbivores or to make leaves unpalatable to generalist species (Janzen 1983). EFNs confer protection on young leaves, which are up to ten fold more likely to be foraged on (Coley & Barone 1996). EFNs generally secrete nectar from when a leaf first unfolds until it is fully mature. (Koptur 2000). EFNs may be marginally preferable when they will attract mutualist insects to defend leaves at a lower energy cost than secondary compounds (Bently 1977). Finding the right balance between various herbivore defense s implies a great deal of uncertainty and can be influenced by a range of evolutionary factors, including microclimate nutrient limita tions, stability of mutualist and herbivore populations, the proportion of generalist to specialist herbivores, and the abundance of pathogens (Coley et al. 1985, Janzen 1983). The Neotropical genus Inga, which has radiated since the late Miocene into mo re than 350 species, is a model example of varying approaches to herbivore and pathogen defense (Kursar et al. 2009). Although many species employ abundant EFNs, each species has its own unique complex of secondary metabolites, with closely related species often having fewer compounds in common than distant relations. This diversity of secondary compounds provides a partial explanation for the rapid diversification of Inga and offers opportunities to study the trade offs between EFN and phenol and saponin c hemical defenses against herbivory. Saponins are a group of secondary metabolites commonly found in alfalfas and certain species of Inga including I. sierrae commonly found in the Monteverde area (Nozzolillo 1997; Alvarez et al 1998). Recent studies h ave shown mixed effects of saponins on herbivory, with some studies suggesting that insect herbivory increased in alfalfa plants with higher levels of saponins (Pearson et al 2008). In Inga no relationship has been shown between increased saponins and low er instance of herbivory, leading to the hypothesis that these compounds may serve as protection against pathogens like fungus and lichens, which may be more common in the cool, humid climates of the cloud forest (Koptur 1985a). At high elevations, plants may produce more EFNs to attract small numbers of mutualists that reside in the uplands. A study conducted on a small group of planted I. sierrae trees at 1520 in Monteverde supported this hypothesis, reporting that the presence of extra EFNs (EEFNs) redu ces percent herbivory (Gough 2003). The low instance of ants in Monteverde resulted in the hypothesis that EEFNs could attract parasitoid and predatory wasps, which in turn prey upon herbivores. However, the exact selection factors that lead to this form o f third trophic level defense are uncertain (Pennington 2005). The goal of this study is to investigate the function of EEFNs and the possible trade offs between EEFN and saponin chemical defense in I. sierrae trees at Monteverde. I hypothesize
that I. si errae trees producing greater numbers of EFNs on young leaves will have higher average rates of mutualistic ant and wasp associations. I predict this will result in lower rates of herbivory, and fewer saponins, since trees with fewer EFNs will compensate w ith chemical defenses. METHODS Study site I surveyed two sites around the property of the Estacin Biolgica Monteverde. Population A consisted of 40 I. sierrae trees on the edge of a secondary forest, adjacent to continuous primary forest, at roughly 1 525m. Population B consisted of 10 I. sierrae trees planted near pasture on the property of Alan Masters at a similar elevation. Trees were marked and given alphabetical labels using orange tape and leaves observed were marked with a small rubber band near the petiole. Selected leaves were in their first week of expansion. For each leaf observed, the and the number of EFNs was recorded. Herbivory and mutu alism observations The next phase of the study involved documenting herbivores and arachnids found on I. sierrae trees and monitoring tagged leaves from October 29 to November 21, 2009 for ant and wasp associations. Herbivores were collected during five se parate 1.5 hour surveys between 9pm and 12am. Each herbivore was then preserved in ethanol and classified. Ant associations were estimated by checking for the presence of ants that actively defended I. sierrae leaves from herbivores. Wasp associations wer e estimated by placing 10 red cups filled with water at study sites A and B and counting parasitoid and predatory wasps that were captured. Saponins During the observation period, immature and maturing leaves were collected from nine trees at each site and measured for saponin content and variation. The relative age of a leaf is distinguishable due to the increased pubescence on immature leaves and the shiny glow and darker coloration of maturing leaves. For each leaf type, 1g of dry leaves were boiled f or 4:10 min in 25mL distilled water. 8mL of the extract was put in a 50mL graduated cylinder and shaken 30x. Saponin content was measured as the mm of foam sustained above the 8ml line 10 minutes after shaking (Massad, unpublished data). A t test was used to look for differences in saponin contents between immature and maturing leaves within each study site. Two additional t tests checked for differences between the immature and maturing leaves between study sites. A Chi squared test for Population A was pr eformed to look at the ratio of average saponin content between immature and maturing leaves. Because collecting leaf samples to measure saponins often required complete defoliation of fledgling leaves, percent herbivory was not calculated for many of the trees for which saponin content was calculated.
Percent herbivory At the end of the three week observation period, all banded leaves were collected and measured for percent herbivory. Percent herbivory was determined by laying each leaf flat on graph pap er and calculating the total number of grid boxes with herbivory damage divided by the total number of grid boxes occupied by the leaf. A t test was used to determine if herbivory levels differ between study sites. A correlation was used to determine if th e number of EFNs is related to percent herbivory. RESULTS There was a difference in the average number of EFNs per leaf between study sites, with Population A averaging 2.92 EFNs and Population B averaging 5.53 EFNs (t test, t = 15.9, df = 153, p< 0.0 001; Figure 1). I. sierrae trees from Population B incurred less herbivory than Population A trees during the three week observation period (t test, t = 15.9, df = 152, p< 0.05; Figure 2). Additionally, the number of EFNs on a given leaf was negatively co rrelated with percent herbivory (Spearman rank correlation: rho = 0.25, p = 0.0039, df = 153; Figure 3). Immature leaves from Population A had higher per leaf average saponin content than Population B (t test, t = 3.2, df = 16, p < 0.05; figure 4a). No d ifference in per leaf average saponin content was found between Populations A and B for maturing leaves (t test, t = 0.46, df = 16, p >0.05; figure 4b). For population A, it was found that saponin content is fixed between immature and maturing leaves, and is found in a 3:2 ratio (t test, t = 5.7, df = 16, p <0.05; figure 4c)(chi square; x 2 = 11.93, df = 8, p=.15; see Table 1). In contrast, no difference in average saponin content per leaf was found between immature and maturing leaves for Population B (t te st, t = 0.89, df =16, p > 0.05; figure 4d). Experiments yielded no evidence of strong ant or wasp mutualisms with I. sierrae During six 24 hour periods setting out 10 red cups with water at each site no predatory or parasitoid wasps were caught. During t he three week observation period only one individual of one species of ant was encountered defending a leaf at study site B. Orthopteran crickets were the herbivore morphospecies most commonly encountered, while members of Coleoptera were encountered with the second greatest frequency (see Table 2). Site A had a higher encounter rate for Orthoptera, Coleoptera, and Arachnida (t tests, t = 5.6, df = 8, p < 0.05; t = 3.9, df = 8, p < 0.05; t = 2.8, df = 4, p<0.05). Other herbivores encountered included Thysan ura and occasional Lepdioptera larvae. Sites A and B had species richness of 16 and 7, respectively.
FIGURE 1. One way ANOVA of average EFNs on observed leaves by Populations A and B. EFNs were counted on all observed leaves and Population B was found to produce more EFNs per average leaf than Population A. This suggests that Population A may be subject to greater resource or phylogenetic constraints. FIGURE 2. One way ANOVA of average percent herbivory for observed leaves by site. Population A in curred more herbivory during a three week period of observation. The increased average percent herbivory in Population A may be the result of increased herbivore abundance at site A.
FIGURE 3. Relationship between EFNs and percent herbivory on o bserved leaves. EFN number was weakly correlated with percent herbivory, suggesting that EFNs account for a very small portion of the observed differences in herbivory between Populations A and B.
(A) ( B) (C) (D) FIGURE 4: One way ANOVA of per leaf average saponin content by: a) Immature leaves in Population A and B. b) Maturing leaves in Populations A and B. c) Immature and maturing leaves in Population A. d) Immature and maturing leaves in Population B. Saponin content was determined by boiling 1g of dry leaves for 4:10 and shaking 8ml of the extract in a 50ml graduated cylinder 30x. Saponin content w as recorded as the mm of foam sustained after 10 minutes. Immature leaves in Population A had more average saponin content per leaf than Population B, suggesting a differential use of resources for fledgling leaves. The observed differences in average sapo nin contents within one leaf suggest that saponin production may be advantageous for only a short time during leaf expansion.
TABLE 1. Chi 2 results for Population A average saponin content per leaf. Population A produces saponins in a fixed 3:2 ratio be tween immature and maturing leaves. No similar pattern was found in Population B, suggesting that all Population A trees use a similar resource strategy and are probably more similar than population B trees. Tree Immature Maturing 2 df Crit p 19 160 129 1.24031 173.4 115.6 1.035525 1.55329 2.588812 1 3.84 0.1076 22 132 67 1.970149 119.4 79.6 1.329648 1.99447 3.324121 1 3.84 0.0683 29 136 104 1.307692 144 96 0.444444 0.66667 1.111111 1 3.84 0.2918 30 153 113 1.353982 159.6 106.4 0.272932 0.4094 0.682331 1 3.84 0.4088 31 122 97 1.257732 131.4 87.6 0.672451 1.00868 1.681126 1 3.84 0.1948 40 143 91 1.571429 140.4 93.6 0.048148 0.07222 0.12037 1 3.84 0.7286 41 148 81 1.82716 137.4 91.6 0.817758 1.22664 2.044396 1 3.84 0.1528 42 1 55 118 1.313559 163.8 109.2 0.472772 0.70916 1.181929 1 3.84 0.277 43 138 100 1.38 142.8 95.2 0.161345 0.24202 0.403361 1 3.84 0.5254 13.13756 9 1287 900 1312.2 874.8 0.483951 0.72593 1.209877 1 11.92768 8 15.51 0.1545 TABLE 2 Herbivore and Arachnida encounter data from sites A and B. Encounter data were taken on five 1.5 hr surveys. Site A had a higher encounter rate for Orthoptera, Coleoptera, and Arachnida, suggesting that a higher abundance of insects leads to a higher rat e of herbivory than at site B. 29 Oct 2 Nov 11 Nov 20 Nov 21 Nov Totals Richness p value Site A, B Othoptera 14,2 8,2 10,0 18,3 10,1 60,8 4,1 0.0032 Coleoptera 8,2 10,4 4,0 9,5 8,2 39,13 2,2 0.0047 Lepidoptera 1,0 0,1 0,0 1,0 1,0 3,1 3,1 Oth er 4,4 6,4 5,1 3,2 5,3 23,14 7,4 Arachnida 42,26 62,40 55,32 159,98 0.006 .0 DISCUSSION Inga is well studied for its use of extrafloral nectaries to attract ants, which in turn provide defense against herbivores (Bentley 1976; Koptur 1984). At high er elevations EFNs have been shown to be less effective due to the absence of beneficial ant species, resulting in higher rates of Lepidopteran herbivory than at lower elevations (Koptur 1985b). The deficit of ants at higher elevations has been used to sug gest a balancing approach for herbivore defense, including
increased rates of leaf expansion and higher levels of secondary compounds (Koptur 1985a). An alternative hypothesis is that fewer mutualists result in increased competition, so trees will produce a greater number of EFNs (Pennington 2005). The differences between EFN and saponin production between study sites suggest a differential use of resources in I. sierrae trees. Population A appears to maximize saponin content when leaves are first produced and then gradually reduce the level of secondary compounds as leaves mature. In contrast, Population B appears to invest heavily in EFN production during leaf fledging. As leaves matured, four Population B trees showed similar or increased levels of sapon ins, while five trees showed decreased saponin levels. The greater variation in when saponins are produced in Population B suggests that Population A may be subject to greater phylogenetic or resource constraints, as reduced phenological variation suggests greater similarity among individuals in Population A. Moreover, I observed I. sierrae trees to be common when located near mature conspecifics, but saplings were exceedingly rare a as solitary saplings in the continuous forest. This suggests to me that mo st of the population A trees studied are the offspring of a few large individuals found in the nearby continuous forest, and that long range dispersal of I. sierrae by mammals is uncommon in the forest near the Estacin Biolgica Monteverde. While this may explain a small amount of the variation between sites A and B, it does not seem a likely reason for the dramatically reduced percent herbivory incurred at site B. The observed differences in EFNs present another explanation. However, for Population B tree s that produced EEFNs on fledgling leaves, it is unclear whether increased EFN production translates into higher nectar production or higher frequency of EFN visitation. In this study, all EEFNs observed were smaller than primary EFNs, and insect visitatio n was only observed at primary EFNs. Furthermore, the abundance of mutualist insects at both sites was minute, suggesting that the interaction strength between I. sierrae and potential mutualists is very weak. Combined with the weakly negative correlation between EFNs and percent herbivory, I conclude that EEFNs confer little, if any, increase in protection at high elevations where mutualists are rare. In addition to differences in population structures and EFN production, sites A and B also differed in abu ndance of herbivores and arachnids. I posit that the higher encounter rate for spiders at site A is a result of increased insect herbivore visitation to I. sierrae which would translate to an increase in prey abundance for spiders. For herbivores that were encountered during the observation period, site A had the greatest abundance and richness. This was especially pronounced for Orthoptera and Coleoptera, which were the most commonly encountered herbivore guilds. All potential herbivores collected were siz es that could have been potential prey items for spiders around Inga trees. The observed difference in insect fauna seems to provide the greatest explanatory power for why site A incurs more herbivory: Being closer to the adjacent continuous forest provide s a greater source population of herbivores, and the resulting increase in diversity and abundance translates to greater herbivory. Although there were greater levels of saponins in immature leaves at site A, the changes in saponin levels within one leaf during the observation period meant there was insufficient data to determine the effect of saponins on herbivores. The consistency within site A suggests that saponins do in fact play an important role in protecting leaves during expansion, but the exact b enefits are unclear. The observed decrease in saponin content as Population A leaves matured suggests that production of saponins is expensive for an individual plant, and is only cost effective for a short time during leaf expansion (Coley et al. 1995). P ossible functions of
saponins in I. sierrae include decreasing herbivore metabolism, making leaves less palatable to generalist herbivores, and protection from fungus and lichens (Koptur 1985a). Clearly, variation in saponin content within one leaf and dif ferences in herbivore abundance between sites make direct comparisons difficult in nature, so future studies should focus on the effect of saponins on herbivore preference and fungal and lichen infestation under laboratory conditions. Other studies could l ook at seasonal differences in herbivory to see if there are corresponding differences in saponin content. ACKNOWLEDGEMENTS I would like to thank Alan and Karen Masters providing me access to their property. Thanks to the Estacin Biolgica Monteverde f or the use of their trees. To Jos Carlos Caldern for answering numerous questions and requests. To Pablo Allen and Yimen Araya for statistical advice and general helpfulness. Finally, special thanks to Anjali Kumar for providing thoughtful advice and a c onstant stream of encouragement and new ideas to improve my project. LITERATURE CITED ALVAREZ, C.J., R. SERRANO, L.F. OSPINA, AND L.A.T. TORRES 1998. Biological activity of saponins from the bark of Inga marginata Willd. Revista Colombiana de Cienci as Qumicas y Farmaceuticas 27: 9 17. BENTLY, B.L 1976. Plants bearing extrafloral nectaries and the associated ant communities: interhabitat differences in the reduction of herbivore damage. Ecology 65: 27 38. 1977. Extrafloral nectaries a nd protection by pugnacious bodyguards. Annu, Rev. Ecol.. and Syst. 8: 407 427. COLEY, P.D. AND J.A. BARON 1996. Herbivory and plant defenses in tropical forests. Annu. Rev. Ecol. and Syst. 27: 305 335. COLEY, P.D., J. P. BRYANT, AND F.S. CHAPIN III 19 85. Resource availability and plant antiherbivore defense. Science 230: 895 899 GOUGH, A 2003. Distribution and anti herbivore role of extra extrafloral nectaries and leaflet pair number on Inga sierrae (Fabaceae: Mimosaceae). In CIEE Spring 2003 Tropi cal Ecology and Conservation: 45 61. JANZEN, D.H. Food Webs: Who eats what, why, how, and with what effects in a tropical forest? In Ecosytems of the world. 1983. Ed. by Bourliere, F. Elseiver publishing. Amsterdam KOPTUR, SUZZANNE 1984. Experimental evi dence for defense of Inga (Mimosaceae) saplings by ants. Ecology 65: 1787 1793. 1985a. Alternative defenses against herbivores in Inga (Fabaceae: Mimosaceae) over an elevational gradient. Ecology 66: 1639 1650. 1985b. Inga In Costa Rican Natural History. Ed by D.H. Janzen. 1983. The University of Chicago Press. Chicago and London. 2000. Interactions among Inga herbivores, ants, and insect visitors to foliar nectaries. In Monteverde: Ecology and conservatio n of a tropical cloud forest. 2000. Oxford University Press. New York and Oxford. KURSAR, T.A., K.G. DEXTER, J. LOKVAM, R.T. PENNINGTON, J.E. RICHARDSON, M.G. WEBER, E.T. MURAKAMI, C. DRAKE, R. M C GREGOR, AND P.D. COLEY. 2009. The evolution of antiherbiv ore defenses and their contribution to species coexistence in the tropical tree genus Inga. Proc. Nat. Acad. Sci. Early Edition 42: 10.1073/pnas.0904786106. MASSAD, T.J Personal communication of unpublished data. NOZZOLILLO, C.J., T. J. ARANSON, F. CA MPOS, N. DONSKOV, AND M. JURZYSTA 1997. Alfalfa leaf saponins and insect resistance. J. Chem. Ecol. 23: 995 1002. PEARSON, C.V, T. J. MASSAD, AND L. A. DYER 2008. Diversity cascades in alfalfa fields: from plant quality
to agroecosystem diversity. Envi ron. Entomol. 37:00 00. PENNINGTON, C 2005. Extra extrafloral nectaries and anti herbivory protection in Inga sierrae (Fabaceae: Mimosaceae). In CIEE Fall 2005 Tropical Ecology and Conservation: 12 22.
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
leader 00000nas 2200000Ka 4500
controlfield tag 008 000000c19749999pautr p s 0 0eng d
datafield ind1 8 ind2 024
subfield code a M39-00324
El papel de los nectarios extraflorales y saponinas en las interacciones herbvoras de Inga Sierrae (Fabaceae: Mimosaceae)
The role of extrafloral nectaries and saponins in plantherbivore interactions of Inga sierrae (Fabaceae: Mimosaceae)
Trees of the Neotropical genus Inga are noted for the use of extrafloral nectaries (EFNs) and secondary compounds to protect against herbivores. For Inga sierrae, the production of Extra-EFNs (EEFNs) has been suggested to confer fitness on fledgling leaves, while the role of saponins on herbivory remains unclear. This study investigates the potential trade-off between EEFN and saponin production and its effect on herbivory. I recorded percent herbivory on fledgling leaves during a three week period and measured saponin contents for I. sierrae trees at two different study sites at 1525m in Monteverde, Costa Rica. Encounter surveys were conducted to assess the associated
herbivore populations. EEFN production, saponin content, percent herbivory, and herbivore abundance varied between study sites. There was a weak negative correlation between EEFNs and percent herbivory. Differences in proximity to continuous forest between study sites may be causing observed differences in herbivore abundance, and
in turn, percent herbivory. Future studies should focus on the effect of saponin content on herbivore preferences and fungal and lichen infestation.
El gnero neotropical Inga usa los nectarios extraflorales (EFNs) y qumicos secundarios para protegerse contra los herbvoros. Para Inga sierrae, la produccin de EFNs extras (EEFNs) sugiere una ventaja para las hojas inmaduras, pero la funcin de lass saponinas contra la herbivora no es claro. Este estudio investiga la posibilidad de un trmino medio entre los EEFNs y las saponinas. Determine el porcentaje de herbivora para las hojas inmaduras durante tres semanas y calcul el contenido de los saponinas en los arboles de I. sierrae en dos sitios de 1525m en Monteverde, Costa, Rica. Hice un estudio para tasar las poblaciones de los herbvoros asociados. La produccin de los EEFNs, contenido de las saponinas, el porcentaje de herbivora, y la abundancia de los herbvoros fueron diferentes entre los sitios de estudio. Hubo una correlacin dbil entre EEFNs y el porcentaje de herbivora. Es posible que las diferencias en la proximidad al bosque continuo entre los sitios de estudio causen las diferencias observados en la abundancia de herbvoros, y en medio, porciento de herbivora. Los estudios en el futuro deben concentrarse en el efecto del contenido de las saponinas para las preferencias de los herbvoros y la infestacin de los hongos y los lquenes.
Text in English.
Monteverde Biological Station (Costa Rica)
Estacin Biolgica de Monteverde (Costa Rica)
Tropical Ecology Fall 2009
Ecologa Tropical Otoo 2009
t Monteverde Institute : Tropical Ecology