Effect of Altitude on Cyanide Concentration, Leaf Toughness, and Herbivory in Passiflora biflora Passifloraceae Matthew S. Martin Department of Molecular Biology, University of Wisconsin, Madison, WI 53705 ______________________________________________ _______________________________________ Abstract Insect herbivore diversity and abundance often decreases with increasing elevation due to greater extremes in temperature and climate Janzen 1983; Hanson 2000. As a result, herbivore pressure on plants fo und at high altitude is lower. Thus, one might expect a reduction in the levels of antiherbivore defenses such as secondary metabolites or high fiber at higher altitudes. To test this, leaves from the tropical vine Passiflora biflora Passifloraceae were sampled in the Monteverde area of Costa Rica, from 800 1600m. Leaves were tested for cyanide levels, toughness, and percent herbivory. Cyanide in the leaves of P. biflora was found to increase with increasing altitude P < 0. 0001, while leaf toughness dec reased with altitude P = 0.0407. Increasing leaf toughness was associated with decreasing cyanide levels P = 0.0015. Percent herbivory was constant across altitude, and overall was quite low 1 3% on average; P = 0.2381. These results may be due to the different climatic environments at high and low altitude: high altitude being colder, wetter, and less sunny. Such conditions inhibit photosynthesis, which may increase costs to toughen leaves. As a result, high altitude plants may be compensating by i nvesting more heavily in cyanide. These trends may also indicate that a greater percentage of total herbivory on P. biflora at lower elevations is by the highly coevolved Heliconius larvae. Resumen Muchas veces la diversidad y abundancia de los insectos h erbvoros disminuyen con el aumento de elevacin. Este es porque el ambiente es extrema en altos elevaciones Janzen 1983; Hanson 2000. A resulto, la presin de los herbvoros en los plantas es menos at altos elevaciones. Entonces, es posible que se esper a una reducin en los defenses anti herbivoros en los plantas en altos elevaciones. Para examinar este, probaron las hojas de Passiflora biflora Passifloraceae en la regin de Monteverde, Costa Rica, de 800 1600 metros. Las hojas probaron por los niveles de cianuro, resistencia, y porciento herbvoro. Econtro que cianuro en las hojas de P. biflora aumento con ms elevacin P < 0.0001, mientras la resistencia de las hojas disminuyo con elevacin P = 0.0407. Creciente resistencia estaba asociado con el decreciente de cianuro P=0.0015. Porciento herbivorio estaba constante a travs altura y global era muy abajo 1 3% en medio; P=0.2381. Es posible que estas resultadas son impuesto a los ambientes diferente entre alto y abajo altura: alto altura es ma s frio, mojado y con menos sol. Tal condiciones inhiben fotosntesis, que pueden aumentar los cuestos para endurecer las hojas. A resulto, las plantas en alto altura pueden invertir man en cianuro. Estas
tendencias pueden indicar que un mas por ciento d el herbvoro total en P. biflora en las altura bajas es de las larvas de Heliconius. Introduction The interaction between plants and herbivores has resulted in a wide array of plant defense mechanisms. Coley and Barone 1996 categorized these as nutritional, chemical or phenological. Nutritional defenses are those that reduce feeding by increasing the difficulty of ingesting leaf material. Low nitrogen and water content in leaves is often associated with reduced herbivory. Fiber is another example of a nutrit ional defense as it poses digestive problems for herbivores. Therefore, high fiber content and corresponding leaf toughness are both negatively correlated with herbivory. In fact, leaf toughness or high fiber content has been shown to be the most effective ant i herbivore defense Coley and Aide l991. The presence of secondary metabolites in plant tissues are examples of chemical defenses. These compounds are of particular importance in the defense of young leaves as they have not had time to develop high le vels of fiber and toughness Coley and Barone 1996. Examples of secondary metabolites include alkaloids, cyanide, anthocyanins, and tannins. Phenological defense mechanisms include shifting leaf production to peak when herbivore levels are lowest and flus hing leaves synchronously to saturate herbivores Coley and Barone 1996. However, plants cannot invest in all types of antiherbivore defense due to their limited resources. Consequently a plant makes trade offs. For example, plants with very tough leaves have to sacrifice the amount of chemicals in their leaves, and leaves that contain high amounts of chemical have reduced toughness Hay 1992. Most herbivore species, especially insects, decrease in both richness and abundance with increasing altitude Han son 2000. Hanson 2000 states that leaf beetles Chrysomelidae are considerably more diverse in lowland areas as compared to the cloud forests and higher altitudes. Similarly, many taxa of Hymenoptera decrease in abundance as altitude increases due to s evere temperature fluctuations and frequent cloud cover, which reduce foraging times Hanson 2000. Janzen 1983 found the complete absence of ants, termites, wood boring beetles, and dung beetles at the high altitude site of Cerro de la Muerte in Costa R ica 3500 meters. He attributed this to the generally low average temperature of the area. Thus, it might be expected that the reduced number of herbivorous insects at high elevations would lead to lower herbivore pressure and, hence, lower levels of anti herbivore defenses in plants. The genus Passiflora Passifloraceae is ideal for studying the effects of altitude on herbivore pressure and plant defenses. The method by which this genus defends itself from herbivores is very well studied and understood. The leaf vacuoles of these plants contain cyanogenic glycosides, which are capable of releasing hydrogen cyanide HCN upon damage to the leaves Siegler 1991. The cyanogenic glycosides are highly utilized in young leaves and decrease as the leaves age an d toughen Newman 2000. Leaves of plants in the genus Passiflora have extremely distinct shapes and are easy to identify in a large assemblage of other plants. In fact, leaf shape in Passiflora is among the most varied of any plant group Benson 1975. In the area of Costa Rica species of Passiflora are found in a wide range of altitudes from 700m to 1800m Haber 2001. One impor tant consumer of these plants are the larvae of the Heliconius butterflies Gilbert 1991. Heliconius butterflies only deposit th eir eggs on plants of the genus
Passiflora preferring to do so on fresh shoots Gilbert 1982. To deal with the cyanogenic glycosides in the leaves of Passiflora, Heliconius l arvae have developed their own glucosidase, which can prevent cyanide release b y competing with the plant glycoside and preventing its action Gilbert 1991. In Monteverde, Haber 2001 found that the species H. clysonymus is most common and is the only species of Heliconius confined to mountainous areas. In the nearby village of San Luis, which is at a lower elevation than Monteverde, H. cydno H. hecali H. charitonius and H. erato are the species commonly, found. Overall San Luis has a greater abundance and diversity of Heliconius butterflies Haber 2001. In this study I examine h erbivory, toughness and cyanide concentration in Passiflora biflora, a vine that ranges from 700 to 1600 meters in the Monteverde area Haber 2001. This vine grows primarily along road sides and forest edges. Its primary specialist herbivores are H. clyso ny mus, H. cydno, H. hecali, H. cha ritonius and H. erato. As insect herbivores generally decline with altitude, as does Heliconius, I suspect high altitude individuals will exhibit lower levels of ant herbivore defenses. Methods Leaf Collection This study was conducted in the Monteverde area of Costa Rica. Leaves of Passiflora biflora were opportunistically sampled from vines beginning at 800m in San Luis and ending at approximately 1600m near the biological station in Monteverde. Plants were sampled along the road running from Monteverde to San Luis and along trail edges near the biological station. The first five leaves, beginning with the youngest fully open leaf were collected on each vine and the altitude was recorded using an altimeter. Young leaves we re collected because levels of cyanogenic glycosides decrease as the leaf ages and toughens Newman 2000. The leaves were sealed in plastic bags pending measurement of weight, toughness, percent herbivory, and cyanide content later on the same day. Herbiv ory Percent herbivory was calculated by placing a gridded transparency, with 0.5 cm squares over the leaves and counting the number of squares that covered the leaf. Then, the number of squares that would have been pr esent if the leaf were whole were count ed and these numbers were used to calculate percent herbivory on each leaf. Toughness Toughness was measured with a leaf penetrometer. The leaf penetrometer consists of two metal rectangles with a small hole in each. A metal stick with a plastic base 30 grams is placed in one of the holes in the rectangle and allowed to rest on a leaf placed between the two rectangles. A container of known weight was placed on the plastic base and water was added until the metal rod penetrated the leaf. The weight of the water, container, and plastic base with metal pin is equivalent to the leaf toughness. Some leaves could not support the plastic base and were reported as having a toughness of 30 grams.
Cyanide Cyanide content was analyzed using the Sodium Picrate Test Seigler 1991. This test was performed by soaking small pieces one cm. by six cm. of Watman No. 1 filter paper in sodium picrate. Each leaf was placed in a glass vial with five drops of toluene and crushed well. A piece of the filter paper saturated wit h sodium picrate was placed in the vial and held in place with a rubber stopper. The vials were left to sit for 1.5 hours at 32C after which the strips were soaked in five ml of water for 20 seconds. The absorbance values of the water were then read on a spectrophotometer at 540 nm to quantify the amount of color change in the sodium picrate. Leaf cyanide concentration values were calculated from a standard curve plotting transmittance versus cyanide concentration Magee 1995. Results Herbivory Percen t herbivory did not change with altitude, and overall was quite low 1 3% on average; Simple regression, P=0.2381, R ^ 2=0.008, n=168, Figure 1. Although there was no statistical difference in herbivory across altitudes, the majority 78% of high herbivo ry values >10% were from altitudes below 1200m. Toughness Leaves ranged in toughness from 30 g to 186.2g with a mean of 68.2 +/ 37.3 s.d.. Leaf age had a statistically significant impact on leaf toughness 1 way ANOVA, P = 0.0018, F=4.506, DF=4, n=168 Youngest leaves had a mean leaf toughness of 51.8g+/ 30.4 s.d. while older leaves followed an increasing series: leaf 2=60. lg +/ 35.2, leaf 3=68.0g+/ 38.4, leaf 4=77. lg+/ 38.0, and leaf 5, the oldest leaf age examined, had a mean toughness of 85.1g +/ 36.6 Figure 2. When leaves with toughness levels of 30 g were eliminated from the analysis, leaf age ceased to have a significant effect on leaf toughness P=0.0676, F=2.246, DF=4, n=134. This was done because the exact toughness values for these le aves were unknown because they could not support the plastic base 30 grams. Toughness was negatively associated with altitude Simple regression, P=0.0407, R ^ 2=0.025, n= 168, Figure 3. Exclusion of leaves with toughness of 30g did not change the sign ificance of this result P=0.0034, R ^ 2=0.063, n=168. Cyanide Altitude had a significant effect on leaf cyanide levels with increasing elevation associated with increasing cyanide levels Simple regression, P<0.0001, R ^ 2=0.246, n=168, Figure 4. Cyanide va lues ranged from a high of 263446.4 micrograms/ gram of leaf at 1575m to 66.9 micrograms/ gram of leaf at 930m with a mean of 16395.5+/ 31063.6 s.d.. Numerical values for cyanide concentration should be thought of as relative and not absolute. This i s because the standard curve used to calculate leaf cyanide levels Magee 1995 only went to a percent transmittance of 40%, so values below this were extrapolated. Relationship between Herbivory, Toughness, and Cyanide Toughness did not significantly affe ct percent herbivory. Tougher leaves showed similar
amounts of herbivore damage as less tough leaves Simple regression, P=0.5406, R ^ 2=0.002, n=168, Figure 5. However, of the leaves that suffered >10% herbivory, 88% had a leaf toughness of under 70g, whic h approximates the toughness values for leaf ages 1 and 2. Increasing toughness was associated with decreasing cyanide levels Simple regression, P=0.0015, R ^ 2=0.059, n=168, Figure 6. The toughest leaf 186.2g had a cyanide concentration of 16.895 micro grams/ gram of leaf while the least tough leaves 30g had cyanide concentrations between 41.0 and 50609.0 micrograms/ gram of leaf. Herbivory did not show a statistically significant relationship with cyanide levels Simple regression, P=0.2279, R ^ 2=0. 009, n=168, Figure 7. However, graphical analysis shows an all or none relationship where only leaves with very low levels of cyanide suffered herbivory, and leaves with just slightly higher cyanide levels suffered little or no herbivory. Discussion Tough ness and cyanide both confer ant i herbivore defense to leaves. In this study the majority of leaves that suffered high herbivory >10% were young, less tough leaves. This corresponds with the findings of Coley and Barone 1996 who found that daily damage rates of young leaves are 5 25 times higher than on mature leaves. Further evidence for the protective effects of toughness and cyanide is shown by the fact that leaves with low cyanide levels suffered high levels of herbivory, while leaves with just sligh tly higher cyanide levels suffered little or no herbivory. There is an altitudinal trade off between tough ness and cyanide concentration. Increasing altitude was associated with increasing leaf cyanide levels and decreased leaf toughness, while cyanide lev els were negatively associated with leaf toughness. The fact that increasing leaf toughness is associated with decreased cyanide levels corresponds with the idea that plants have limited resources and must invest more heavily in one type of ant i herbivore d efense whether it be nutritional, chemical, or phenological Coley and Barone 1996. This trend of decreasing toughness and increasing cyanide levels in P. biflora at high altitude may reflect the abiotic conditions in which each plant is found. Plants in the San Luis valley are exposed to a hotter, sunnier climate than those found in the forest near the biological station. These conditions are more favorable for the toughening of leaves than the lower temperatures and reduced light conditions of the forest Rundell and Gibson 1996 report that tropical leaves that are exposed to high amounts of light, like canopy leaves are thicker and denser than understory leaves, which receive less light High light and temperature lead to greater photosynthetic capacit y. Since toughness correlates with the amount of fiber in a leaf, it is faster, easier, and cheaper for a plant in high photosynthetic conditions to toughen its leaves Rundell and Gibson 1996. Thus, it may be that P. biflora in the San Luis area is relyi ng more on leaf toughness for ant i herbivore defense because its leaves can rapidly toughen in the environmental conditions present, while the plants in the high elevation forest are compensating for the slow and costly process of leaf toughening by investi ng more highly in chemical defense. The result of the trade off between toughness and cyanide concentration with altitude is that percent herbivory is nearly constant 2% with altitude. Since most serious herbivory >10% occurred at low and intermediate elevations <1200m and on leaves with toughness >70g, my results agree with Coley and Barone 1996 that toughness is important for ant i herbivore defense at lower altitudes as young leaves are most tender.
Toughness is a broad spectrum ant i herbivore defen se that inhibits generalist and specialist herbivores alike. The increased leaf toughness at lower elevations may also reflect greater herbivory by Heliconius larvae who are unaffected by cyanogenic glycosides. William Haber 2001 has observed that Helico nius diversity and abundance is greater in San Luis and I have observed Heliconius eggs and caterpillars on P. biflora leaves in San Luis, but not on leaves in the forest in Monteverde. Therefore, increased Heliconius herbivory at lower elevations may be s electing for plants that have invested in an ant i herbivore defense that the Heliconius caterpillars have not evolved protection against. In contrast the high cyanide levels in plants found at higher altitude may be due to a more general herbivore pressure and not the highly coevolved Heliconius herbivory. Acknowledgements I would like to thank my advisor and the center of my universe Alan Masters for the plethora of help and advice he gave me. To Will Weider and Andrew Rodstrom a hearty thank you as well. I am very grateful for the biological station and the road from Monteverde to San Luis as they provided habitat for Pasi flora biflora. Finally, I would like to thank my host family, Rafael, Lilliam, and Randy for giving me shelter, food and an insight into their lives during my research. Literature Cited Benson, W. W., K. S. Brown, and L. E. Gilbert. 1975. Coevolution of plants and h erbivores: passion flower butterflies. Evolution. 29: 659 680. Col ey, P. D., and T. M. Aide. 1991. Comparison of herbivory and plant defenses in temperate and tropical broad leaved forests. In P. W. Price, T. M. Lewinsohn, G. W. Fernandes and W. W. Benson Eds.. Plant Animal Interactions: Evolutionary Ecology in T ropical and Temperate Regions. Pp. 25 49. John Wiley and Sons, Inc. New York, New York. Coley, P.D., and J. A. Barone. 1996. Herbivory and plant defenses in tropical forests. Annual Review of Ecological Systems. 27: 305 335. Gilbert, L. E. 1991. Biodiversity of a Central American Heliconius Community: Pattern, Process, a nd Problems. In P. W. Price, T. M. Lewinsohn, G. W. Fernandes, and W. W. Benson Eds.. Plant Animal Interactions: Evolutionary Ecology in Tropical and Temperate Regions. Pp. 403 430. John Wiley and Sons, Inc. New York, New York Gilbert, L.E. 1982. The co evolution of a butterfly and a vine. Scientific American. 247:2 110 121. Haber, W. H. 2001. Personal Communication Monteverde Costa Rica. Hanson, P. 2000. Insects and Spiders. In N. M. Nadkarni and N. T. Wheelwright Eds.. Monteverde: Ecology and Conserv ation of a Tropical Cloud Forest. Pp. 95 147. Oxford University Press, New York, New York. Hay, M. E, and P. D. Steinberg. 1992. The chemical ecology of plant herbivore i nteractions in marine versus terrestrial communities. In G. A. Rosenthal and M. R. Be renbaum Eds.. Herbivores: Their interaction with secondary plant metabolites. Pp. 372 408. Academic press Inc., San Diego, California. Janzen, D. H. 1983. Insects. In D. H. Janzen, Ed.. Costa Rican Natural History. Pp. 619 645. The University of Chica go Press, Chicago, IL. Karban, R., and I. T. Baldwin. 1997. Induced Responses to Herbivory. University of
Chicago Press, Chicago, Illinois. 319pp. Magee, A. 1995. The effects of microhabitat and leaf age on qualitative defenses in Passiflora In UCEAP Spr ing 1995. Monteverde, Costa Rica. Newman, T. W. 2000. Anti herbivore defense mechanisms of Passiflora. In Tropical Ecology and Conservation CIEE spring 2000. Monteverde, Costa Rica. Rundell, P. W., and A. C. Gibson. 1996. Adaptive strategies of growth form and Physiological ecology in Neotropical lowland rain forest plants. In A. C. Gibson Ed.. Neotropical Biodiversity and Conservation. Pp. 33 71. Mildred E. Mathias Botanical Garden, Los ngeles, California. Siegler, D.S. 1991. Cyanide and cyanogenic glycosides. In G. A. Rosenthal and M. R. Berenbaum Eds.. Herbivores: Their Interaction With Secondary Plant Metabolites. Pp. 35 70. Academic Press Inc., San Diego, California.