1 Effects of Passiflora cyanide, toughness and leaf age on Heliconius oviposition Lauren Barlow Department of Biological Aspects of Conservation , University of Wisconsin Madison ABSTRACT Passi floraceae and Heliconiine butterflies have coevolved around a cyanide defense against herbivory. This study focuses on the oviposition preference of Heliconius charithonia and Heliconius hecale on Passiflora biflora and Passiflora vitifolia . C hanges in c yanide cont ent and leaf toughness were studied as a function of leaf age of P. vitifolia and P. biflora in the Ranario Butterfly Garden in Monteverde, Costa Rica. In addition, P. biflora plants were compared with vines out side of the garden. For P. vitifolia , leaf to ughness significa ntly increased with age while cyanide content significantly decreased. P. biflora sampled inside and outside did not show a significant relationship between leaf age and either cyanide or toughness. Cyanide was shown to significantly decre ase as leaf toughness increased in P. biflora , however . P. biflora sampled from inside and outside showed very similar patterns of leaf toughness and CN. H. charithonia females preferred to oviposit on the youngest P. biflora leaves, while H. hecale showed no ovipositing preference based on leaf age. These results suggest that oviposition preferences varied, with H. hecale egg placement a result of interspecific insect interactions and H. charithonia preference based on leaf toughness. RESUMEN Passifloraceae y mariposas de la familia Heliconiinae han coevolucionado alrededor de una defense de cianuro contra herbivoria. Este estudio se enfoca en la preferencia por sitios de oviposiciÃ³n de Heliconius charithonnia y Heliconius hecale en Passiflora biflora y Passiflora vitifolia . Cambios en el contenido de cianuro y la dureza de las hojas fue estudiado en funciÃ³n de la edad de la hoja de P. vitifolia y P biflora en el Mariposario del Ranario de Monteverde, Costa Rica. En adiciÃ³n, plantas de P. bifl ora se compararon con plantas afuera del jardÃn. Para P. vitifolia la dureza de las hojas aumenta significativamente con la edad mientras que el contenido de cianuro decrece significativamente. Las plantas muestreadas dentro y fuera no mostraron diferenc ia alguna en cuanto a la edad de las hojas y el contenido de cianuro y la dureza. El contenido de cianuro decrece significativamente al aumentar la dureza de las hojas en P. biflora , sin embargo, P. biflora muestreadas dentro y fuera mestran patrones simi lares en cuanto a dureza y contenido de cianuro. Las hembras de H. charithonia prefieren ovipositar en hojas jÃ³venes de P. biflora , mientras que H. hecale no muestra ninguna preferencia basada en la edad de la hoja. Estos resultados sugieren que la pref erencia por sitios de oviposiciÃ³n varÃan,
2 con H. hecale el lugar es resultado de interacciones intraespecÃficas con insectos y H. charithonia se basa en la dureza de las hojas. INTRODUCTION Many plants have evolved physical and chemical defenses to deter herbivor e s , such as passion vines ( Passifloraceae ) which ha ve developed cyanogenic glycosides in their tissues (Benson et al. 1975). These secondary compounds have been successful in deterring most insects, though b utterflies in the subfamily Heliconiinae (Nymphalidae) have developed a resistance to these toxins (Benson et al. 1975) . Heliconi ines almost exclusively use Passiflora as their host plants and this association is an example of coevolution between insects and plants (Gilbert 1975). After these butterflies over came Passiflora cyanide (CN) defenses , Passiflora evolved other strategies to det er heliconiines, such as variable leaf shape, egg mimicry , extrafloral nect ari es (EFN) and hairs ( DeVries 1987 ; Benson et al . 1975). Consequently, most heliconiines have become specialized for a subset of Passiflora s pp in an ongoing coevolutionary arms race (Benson et al. 1975 ; DeVries 198 7). Although Passiflora use multiple defense strategies, it is often too energetically costly to invest in them at the same time, resulting in defense tradeoffs. For example, younger leaves tend to have higher cyanide levels while older leaves are generally tougher (Bellush 2010) . Leaf toughness is a more effective defense, but takes longer to develop (Sagers & Coley 1995) . In addition, defense levels and types can vary based on resource availability. For example, P. dioscoreifolia plants in sunny habitats used more expensive defensive compounds, such as lignin and cellulose, than plants in shade habitats. Leaf toughness was also found to be higher in sunny conditions (Bellush 2010). Heliconiines have evolved adaptations to overcome these defenses, such as acute eyesight to distinguish between eggs and egg mimics (Be nson et al. 1975; DeVries 1987). In addition, o vipositing female s, use sensitive chemoreceptors to accurately identify plants based on secondary compounds and despite changing leaf shape (Benson et al. 1975). This may explain why some heliconiine butterflies have been shown to prefer younger Passiflora biflora leaves, that are less tough but have higher CN levels (Burkholder 2008). In this study, I will quantify defense strategies, including cyan ide content and leaf toughness, of Passiflora vitif olia and P. biflora at a Monteverde butterfly garden. I will also compare o viposition preference of Heliconius hecale and Heliconius charithonia on both of these plants. Since the garden is watered more frequently than it rains outside (test conducted in April of dry season) I will compare defenses between P. biflora in side and outside to determine effects of water limitation. MATERIALS AND METHODS
3 Study Organisms This study focused on the oviposition preference of H. hecale and H. charithoni a, two Monteverde natives that use passion vines as host plants. There were about 20 H. hecale and 45 H. charithonia adults present in the Ranario Butterfly Garden, where this s tudy took place (see below) . H. hecale tend to lay eggs singly on P. vitifolia , while H. charithonia were observed laying eggs in clusters of between one and 45 eggs on P. biflora . Oviposition was recorded for both species of passion vine, including the number of eggs, the location (leaf, tendril, stem, etc.), age of the leaf (one being the bud, two being the first full leaf, etc.), and presence of caterpillars , other insects or ab normalities. S ome hatched eggs were observed, but they were excluded from the data analyses because it was impossible to determine leaf age at time of oviposition. Likewise, eggs observed on recently trimmed branches were excluded because it was impossible to determine leaf age . Study Site This study was conducted at the butterfly garden at the Monteverde Frog Pond, and in the surrounding woods. This site is located at 13 00 m, in premontane we t forest, where both Heliconius species and P. biflora are native . There were four trellises (labeled B, C, F and I) supporting P. vitifolia and two trellises (labeled A and E) supporting P. biflora in the garden . In total, P. vitifolia had about 1600 leaves while P. biflora had 220 . M ultiple P. biflora plants w ere also found in the small forest outside the garden. After recording oviposition , ten leaves and tendrils were collected from each of five different age classes (leaves two, four, six, eight and ten) for each species, resulting in 50 leaves each. Fifty leav es of different ages were also collected from the P. biflora plants in the forest. Fresh leaves were refrigerated until cyanide content and leaf toughness could be determined in the lab . Leaf Toughness and Cyanide Tests Leaf toughness was determined using a leaf penetrometer. A leaf was placed between two metal plates, each with a small hole in the middle. A platform with a pin on one side was gently placed on top of the plates, with the pin in the hole. A small con tainer was placed on top and slowly filled with water until the leaf broke under the weight of the water. The combined weight of the water a nd container was recorded as relative leaf toughness. The second test determined the c yanide content in freshly picked leaves and tendrils (t ho ugh it should be noted that P. biflora tendrils were too dry to be tested ) . A sodium picrate solution was used, which contained 2.5 grams of sodium carbonate and 1 gram of moist .5% wt. vol. picric acid. Water was added for a total volume of 100mL (Bellush 2010). Filter paper was cut into 9 cm by 0 .5 cm strips that were dipped in the sodium picrate solution. When possible, 0.1 grams of fresh leaf was cut and placed in a glass vial. The entire leaf was used when it weighed less than 0.1 grams. Ten drops of t oluene were added to each vial and the leaf was macerated for 40 seconds. A moist picrate strip was placed in the vial so that it hung just above
4 the leaf tissue. The vials were capped and incubated for 15 minutes in a dry box. The yellow strips turned ora nge or brown based on the amount of cyanide released. After incubation, the picrate strips were removed and dipped ten times into 5 mL of distilled water to leach the color into the water. This was then poured into 5 mL cuvets and measured in a spectrophot ometer at 510 nm (Bradbury et al . 1999, Egan et al. 1998). The percent transmission was recorded and converted into absorbance using equation 1. Equation 2 was used to determine the total cyanide concentration (Bradbury et al. 1999, Egan et al . 1998). 1. A =2 log(T) ; A=Absorba nce, T=Transmission 2. Total cyanide ( g/g)=396(A)(100/z); z=we ight of fresh tissue used (mg) Additional Observations As I recorded egg placement, I noticed that the presence of other insects varied significantly from plant to plant. On both P. biflora plants, no other insects were found. On the other hand, all P. vitifolia plants hosted other insects. Trellises B and C hosted ant colonies and what was probably white fly larvae. Trellises F and I had few ants, but many other arthropods , including various types of spiders. Trellis F also hosted a significant colony of hemipterans. Most of the H. hecale eggs found were on Trellises B (15) and C (11) , and only a few were observed on Trellises F (2) and I (3). Also , P. biflora plants were s mall, and I was only able to find a few complete branches that ranged fro m bud to leaf ten. RESULTS Egg P lacement Twenty four unhatched H. hecale eggs were observed on P. vitifolia , while 196 H. charithonia eggs were observed on P. biflora. There was no significant linear trend between the age of P. vitifolia tissues (including leaves, tendrils, stems and flowers ; Fig. 1 ) and egg placement by H. hecale (linear regression, p= 0.1986, RÂ²= 0.1974 , N= 10 ) , though this could be caused by the small number of eggs present . On the other hand , the data suggest that there is a very s ignificant trend between P. biflora tissue age and egg placement by H. charithonia (p= 0.0537, RÂ²= 0.5578, N=7 ; Fig. 2 ). There is also a significant relationship between P. biflora tendril age and egg placement (p= 0.0335, RÂ²= 0.5569, N=8 ; Fig. 3 ).
5 Figure 1 . Oviposition by H. hecale on P. vitifolia as a function of tissue age, including leaves, tendrils, stems, and flowers(Linear regression, p= 0.1986, RÂ²= 0.1974, N= 10 ) . Figure 2 . As age of P. biflora leaves, tendrils, etc. increase, there is a nearly significant decrease in egg placement by H. charithonia ( R egression test, p< 0.01, N= 10) . y = 0.1014x + 3.19 0 1 2 3 4 5 6 7 0 5 10 15 20 25 30 # Eggs Age (1=youngest) y = 95.32x 1.783 RÂ² = 0.7659 0 20 40 60 80 100 120 140 0 2 4 6 8 10 12 # Eggs Age (1=youngest)
6 Figure 3 . As age of P. biflora tendrils increases there is a significant decrease in egg placement by H. charithonia (p= 0.0355, N= 8). Leaf Toughness P. vitifolia show ed a significant increase in leaf toughness with age ( linear regression, p< 0 .0001 , RÂ²= 0.3707 , N= 49; Fig. 4 ) though P. biflora did not follow this trend neither inside ( linear regression, p= 0.6043, RÂ²= 0.0059) nor outside ( linear regression, p= 0.3767, RÂ²= 0.0170). Both spec ies within the garden showed a negative correlation between CN content a nd leaf toughness ( P. vitifolia : p= 0.0241, RÂ²= 0.1037; Fig. 5 ; P. biflora : p= 0.0125, RÂ²= 0.1282 ; Fig. 6 ). Leaf toughness was also compared between P. biflora found in the garden and in the woods. Inside, leaves were slightly tougher with mean of 103.332 g (SE= 5.7149439 ) compared to outside with 86.355g (SE= 5.7149439 ) , though not s ignificantly so (ANCOVA, p= 0.7567 ). Figure 4 . Leaf toughness was shown to increase with age for P. vitifolia (p< 0.0001, N= 49) . y = 0.2656x + 4.8128 0 2 4 6 8 10 0 2 4 6 8 10 12 # Eggs Tendril Age (1=youngest) y = 0.008x + 0.0362 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0 2 4 6 8 10 12 Toughness (g) Leaf Age (1=youngest)
7 Figure 5 . Leaf toughness was negatively correlated with CN concentrations in P. vitifolia ( P=.0241 , N= 49). Figure 6 . Leaf toughness was negatively correlated with CN concentrations in P. biflora leaves sampled in the garden (p= 0.0125, N= 48). C yanide Content Cyanide concentrations decline with age in P. vitifolia leaves ( linear regression, p< 0.0001, RÂ²= 0.370 7 ; Fig. 7 ) as well as tendrils ( linear regression, p< 0.0001, RÂ²= 0.5058 ; Fig. 8 ). P. biflora , on the other hand, did not show a significant trend in leaf CN for samples gathered inside ( linear regression, p= 0.2388, RÂ²= 0.0300) nor outside ( linear regression, p= 0.3043, RÂ²= 0.0234). However, CN did decrease as leaf toughness increased in P. biflora sampled inside (linear regression, p= 0.0125, RÂ²= 0.12 82 ; Fig. 6 ). y = 0.1093x + 56.95 20 0 20 40 60 80 100 120 140 160 0 100 200 300 400 500 600 Toughness (g) Leaf [CN] (ug/g) y = 0.2286x + 119.25 0 50 100 150 200 250 50 0 50 100 150 200 250 300 350 Toughness (g) Leaf [CN] (ug/g)
8 Figure 7 . Leaf CN concentrations were shown to decrease as leaf age increased in P. vitifolia (p< 0.0001, N=49 ). Figure 8 . Tendril CN concentrations were also found to be highest in the youngest P. vitifolia tendrils ( P<.0001 , N=46 ). DISCUSSION This study suggests that Heliconius butterflies may select oviposition sites based on characteristics like leaf toughness, and also environmental conditions like presence of other insects. Although we cannot conclusively say that butterfl ies select ovipositing sites based on cyanide content , it may reveal some interesting trends related to Passiflora Heliconius interactions. It is surprising that oviposition did not relate to cyanide content, but there were other patterns that are in fact consistent with past research (Burkholder 2008) . For example, H. charithonia prefer to oviposit on younger P. biflora leaves and P. vitifolia reduce cyanide production, and increase toughness as leaves age (Benson et al. 1975; Bellush 2010; Lowman & Box 1983) . Also, cyanide content of both Passiflora species was shown to decrease as leaf toughness increased. This suggests there is a tradeoff and that once leaves are toughened CN is withdrawn. Toughness is a better overall strategy, even protecting t he leaves from heliconiine larvae, but it takes time to accumulate. Therefore, CN protects younge r, more tender leaves ( Sagers & Coley 1995 ). y = 19.463x + 264.85 0 100 200 300 400 500 600 0 2 4 6 8 10 12 Leaf [CN] (ug/g) Leaf Age (1= youngest) y = 14.649x + 130.22 50 0 50 100 150 200 250 0 2 4 6 8 10 12 Tendril [CN] (ug/g) Leaf Age (1=youngest)
9 In the garden, it seemed that oviposition was related to the most limiting resource for each plant. For example, H. hecale most likely chose oviposition sites based on the relative abundance and threat of other arthropods. Although P. vitifolia cyanide content and toughness were both correlated with leaf age, H. hecale did not choose to lay their eggs on the youngest leaves. Since leaf toughness was relatively low for all leaves, with an average of 41.15 g compared to 103.32 for P. biflora , it is possible that P. vitifolia leaves are never too tough for larvae to chew. With this in mind, we might consider other influences on oviposition, such as competition or predation by other arthropods . From my observations, it seemed apparent that H. hecale avoided laying eggs on Trellises F and I, which were home to a colony of hemipterans and various species of spiders. It is also possible that eggs were laid on these plants, but were subsequently eaten or removed before I could count them. There were many more eggs on Trellises B and C, which contained many ants and white flies larvae. Th ough, overall there was still no pattern between oviposition and leaf age. Though studies have shown that some ants remove eggs (Benson et al . 1975), both species seem to be coexisting in this case. Perhaps ants do remove eggs, but the ones I found were hi dden too well for detection. This would suggest that ovi positing females try to oviposit in more discrete places to avoid detection and removal by ants. Likewise, it is possible that hemipterans are more efficient egg eliminators and therefore fewer were f ound on those plants. If plants and insects were observed more closely it might be possible to determine the exact relationship between these organisms. It would also be interesting to compare the efficiency of egg removal by different insects on different species of Passiflora. In this study, the data supported a relationship between P. biflora leaf age and oviposition by H. charithoni a even though there was no significant trend betwee n leaf age and cyanide content or toughness. Nonetheless, the most probabl e explanation for preference for younger leaves is still related to leaf toughness. The majority (63%) of eggs were actually laid on the youngest leaf, tendril, or bud of P. biflora . Relative toughness for leaves two, four, six, eight and ten was tested and found to be much higher than P. vitifolia. Leaf toughness was not tested for the first leaf because it was found to be too small to accurately measure. Therefore, it is possible that toughness is significantly lower than older leaves, or at least the soft tissue of the first bud is easiest for larvae to chew. It is also possible that the small plant size effected patterns of oviposition, and it would be b eneficial to test these trends on multiple larger individuals. Other explanations, such as interspecific interactions, can be ruled out because there were no other insects observed on these plants. The comparison between P. biflora found inside and outsi de showed no difference between these groups, suggesting that there was no significant difference between resources that might affect defense levels. It has been demonstrated that resources like light can influence cyanide levels and leaf toughness (Bellus h 2010), and one would expect similar trends in relation to water availability. Since leaves within the garden probably received much more water I would have expected more defense production inside than outside. Unfortunately, I was unable to
10 quantify wate r availability in both areas so it is difficult to confidently make conclusions without further res earch. Overall, this study suggests that there are important differences between Passiflora defenses and Heliconius oviposition preference between species. Our results support the theory of defense tradeoffs due to energy limitations, though cyanide and leaf toughness may not always follow a linear trend. In this case, leaf toughness, extrafloral nectaries and the insects they att ract probably had a stronger influence on Heliconius oviposition than chemical defenses. This suggests that although Passiflora employ many defenses, they may have varying effectiveness and energy investment. ACKNOWLEDGEMENTS I would like to thank my proj ect advisor , Alan Masters, for all his guidance during my study. I would also like to thank the staff at the Mariposario for their assistance, especially Evelyn Casares and Dennis Corrales. LITERATURE CITED Bellush, James . 2010. Leaf Toughness and Cyanide Defense of Passiflora dioscoreifolia in Varied Light Habitats. CIEE Tropical Ecology and Conservation. Summer 2010. 37 44. Benson, W.W., K.S. Brown, and L.E. Gilbert. 1975. Coevolution of Plants and Herbivores: Passion Fl ower B utterflies. Evolution. 29(4): 659 680. Bradbury, M.G., S.V. Eg an, and J.H. Bradbury. 1999. Picrate paper kits for determination of total cyanogenesis in cassava roots and all forms of cyanogens in cassava products. Journal of the Science of Food and A gricultur e 79:593 601. Burkholder, P. 2008. Preferential oviposition by Heliconiinae (Nymphalidae) butterflies on Passiflora biflora (Passifloraceae) leaves with higher cyanide concentrations. CIEE Tropical Ecology and Conservation. Spring 2008. 188 199. DeVries, P. J. 1987. The Butterflies of Costa Rica and their Natural History, Volume 1: Papilionidae, Pieridae, Nymphalidae, pp. 186 196. Princeton University Press, Princeton, New Jersey. Egan, S.V., H.H. Yeoh, and J.H. Bradbury. 1998. Simple picrate pap er kit for determination of cyanogenic potential of cassava flour. Journal of the Science of Food and Agriculture 76:39 48. Gilbert, L.E. and M.C. Singer. 1975. Butterfly Ecology. Annual Review of Ecology and Systematics. 6: 365 397. Janzen, D.H. 1968. Rep roductive behavior in the Passifloraceae and some of its pollinators in Central America. B B ehaviour 32: 33 48. Lowman, M.D., and J. D. Box. 1983. Variation in leaf toughness and phenolic content among five species of A ustralian rain forest tress. Australi an Journal of Ecology 8:17 25. Sagers, C.L. and Coley, P.D. 1995. Benefits and costs of defense in a neotropical shrub. Ecology 76: 1835 1843.
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 M39-00236
Efectos del cianuro, la dureza y la edad de la hoja de Passiflora en la ovoposicin de Heliconius
Effects of Passiflora cyanide, toughness and leaf age on Heliconius oviposition
Passifloraceae and Heliconiine butterflies have coevolved around a cyanide defense against herbivory. This study focuses on the oviposition preference of Heliconius charithonia and Heliconius hecale on Passiflora biflora and Passiflora vitifolia. Changes in cyanide content and leaf toughness were studied as a function of leaf age of P. vitifolia and P. biflora in the Ranario Butterfly Garden in Monteverde, Costa Rica. In addition, P. biflora plants were compared with vines outside of the garden. For P. vitifolia, leaf toughness significantly increased with age while
cyanide content significantly decreased. P. biflora sampled inside and outside did not show a
significant relationship between leaf age and either cyanide or toughness. Cyanide was shown to
significantly decrease as leaf toughness increased in P. biflora, however. P. biflora sampled from
inside and outside showed very similar patterns of leaf toughness and CN. H. charithonia
females preferred to oviposit on the youngest P. biflora leaves, while H. hecale showed no
ovipositing preference based on leaf age. These results suggest that oviposition preferences
varied, with H. hecale egg placement a result of interspecific insect interactions and H.
charithonia preference based on leaf toughness.
Las mariposas de la familia Passifloraceae y Heliconiinae han evolucionado en torno a la defensa contra los herbvoros de cianuro. Este estudio se centra en la preferencia por los sitios de ovoposicin de Heliconius charithonia y Heliconius hecale en Passiflora biflora y Passiflora vitifolia. Los cambios en el contenido de cianuro y dureza de las hojas fueron estudiados en funcin de la edad de la hoja de P. vitifolia y P. biflora en el Jardin de Mariposas del Ranario en Monteverde, Costa Rica. Adems, las plantas de P. biflora se compararon con los bejucos afuera del jardn. Para P. vitifolia la dureza de las hojas aument significativamente con la edad mientras que el contenido de cianuro disminuy significativamente. Las plantas muestreadas dentro y fuera no mostraron diferencia alguna en cuanto a la edad de las hojas y el contenido de cianuro y la dureza. El contenido de cianuro decrece significativamente al aumentar la dureza de las hojas en P. biflora, sin embargo, P. biflora muestreadas dentro y fuera muestran patrones similares en cuanto a dureza y contenido de cianuro. Las hembras de H. charithonia prefieren ovipositar en las hojas jvenes de P. biflora, mientras que H. hecale no muestra ninguna preferencia basada en la edad de la hoja. Estos resultados sugieren que la preferencia por los sitios de oviposicin vara, con H. hecale el lugar es el resultado de las interacciones intraespecficas con los insectos y H. charithonia se basa en la dureza de las hojas.
Text in English.
Costa Rica--Puntarenas--Monteverde Zone--Santa Elena
Costa Rica--Puntarenas--Zona de Monteverde--Santa Elena
Tropical Ecology Spring 2011
Ranario Butterfly Garden
Ecologa Tropical Primavera 2011
Jardin de Mariposas del Ranario
t Monteverde Institute : Tropical Ecology