Intraspecific variation in defense against a generalist lepidopteran herbivore in populations of Eruca sativa (Mill.)

Abstract

Populations of Eruca sativa (Brassicaceae) from desert and Mediterranean (Med) habitats in Israel differ in their defense against larvae of the generalist Spodoptera littoralis but not the specialist Pieris brassicae. Larvae of the generalist insect feeding on plants of the Med population gained significantly less weight than those feeding on the desert plants, and exogenous application of methyl jasmonate (MJ) on leaves of the Med plants significantly reduced the level of damage created by the generalist larvae. However, MJ treatment significantly induced resistance in plants of the desert population, whereas the generalist larvae caused similar damage to MJ‐induced and noninduced plants. Analyses of glucosinolates and expression of genes in their synthesis pathway indicated that defense in plants of the Med population against the generalist insect is governed by the accumulation of glucosinolates. In plants of the desert population, trypsin proteinase inhibitor activity was highly induced in response to herbivory by S. littoralis. Analysis of genes in the defense‐regulating signaling pathways suggested that in response to herbivory, differences between populations in the induced levels of jasmonic acid, ethylene, and salicylic acid mediate the differential defenses against the insect. In addition, expression analysis of myrosinase‐associated protein NSP2 suggested that in plants of the desert population, glucosinolates breakdown products were primarily directed to nitrile production. We suggest that proteinase inhibitors provide an effective defense in the desert plants, in which glucosinolate production is directed to the less toxic nitriles. The ecological role of nitrile production in preventing infestation by specialists is discussed.

Evolution of phenotypic plasticity: Genetic differentiation and additive genetic variation for induced plant defence in wild arugula Eruca sativa

“…To study defence evolution, we used the two populations of E. sativa from Ogran et al (): one from the east coast of the Sea of Galilee (32°46ʹ39ʺN, 35°39ʹ29ʺE) and the second from the southern Jordan Valley (32°04ʹ49ʺN, 35°29ʹ46ʺE). The former is characterized as a Mediterranean environment (average annual rainfall of 400–430 mm), whereas the latter site is a desert environment (average annual rainfall of ≤200 mm) (Goldreich, ).…”

“…The extraction and analysis of glucosinolates and trypsin‐PI activity were performed as previously described (Ogran et al, ). After sulfatase treatment, glucosinolates were separated and measured with an HPLC ProStar 240 high‐performance liquid chromatography (Varian) equipped with an Acclaim reverse‐phase C18 column (2.1 × 250 mm, 5 µm) (Dionex).…”

“…Total glucosinolates were quantified using benzyl glucosinolate as an internal standard. Trypsin‐PI activity measures binding potential to digestive proteases in an herbivore (thereby making nutrients unavailable to the herbivore) and was calculated as the percentage of trypsin activity relative to controls (Ogran et al, ; Thaler, Stout, Karban, & Duffey, ).…”

“…Two of these populations, Susita in the northern Mediterranean climate and Sartaba in the southern desert, were chosen for further study; these show strong genome‐wide differentiation, with an F ST estimate of 0.15 from the AFLP markers (Barazani, unpub.). These populations differed in their induced resistance to a generalist herbivore (Ogran et al, ), perhaps due to differential induction of glucosinolates (Textor & Gershenzon, ). Here, we combine quantitative and ecological genetic approaches to estimate within‐ and between‐population variation in induced defensive traits, including two chemical traits, toxic glucosinolates and anti‐nutritive protease inhibitors (Ogran et al, ), and a physical defence (trichomes).…”

“…These populations differed in their induced resistance to a generalist herbivore (Ogran et al, ), perhaps due to differential induction of glucosinolates (Textor & Gershenzon, ). Here, we combine quantitative and ecological genetic approaches to estimate within‐ and between‐population variation in induced defensive traits, including two chemical traits, toxic glucosinolates and anti‐nutritive protease inhibitors (Ogran et al, ), and a physical defence (trichomes). We also monitored herbivory in the natural habitats to relate population differentiation in phenotypes to differences herbivory regimes.…”