The inducible expression of polyphenol oxidase (PPO), a presumed antiherbivore enzyme, was examined in hybrid poplar (Populus trichocarpa ϫ Populus deltoides). Following mechanical wounding simulating insect damage, PPO activity increased dramatically in wounded and unwounded leaves on wounded plants beginning at 24 and 48 h, respectively. A hybrid poplar PPO cDNA was isolated and its nucleotide sequence determined. On northern blots, PPO transcripts were detected within 8 h of wounding, and reached peak levels at 16 and 24 h in wounded and unwounded leaves, respectively. Methyl jasmonate spray and feeding by forest tent caterpillar also induced PPO expression. The induction of PPO was strongest in the youngest four leaves, which were generally avoided by caterpillars in free feeding experiments. This wound-and herbivore-induced expression of PPO in hybrid poplar supports the defensive role of this protein against insect pests.

Peroxidases (PODs) are believed to act as induced and constitutive defenses in plants against leaf-feeding insects. However, little work has examined the mode of action of PODs against insects. Putative mechanisms include the production of potentially antinutritive and/or toxic semiquinone free radicals and quinones (from the oxidation of phenolics), as well as increased leaf toughness. In this study, transgenic hybrid poplar saplings (Populus tremula × Populus alba) overexpressing horseradish peroxidase (HRP) were produced to examine the impact of elevated HRP levels on the performance and gut biochemistry of Lymantria dispar caterpillars. HRP-overexpressing poplars were more resistant to L. dispar than wild-type (WT) poplars when the level of a phenolic substrate of HRP (chlorogenic acid) was increased, but only when leaves had prior feeding damage. Damaged (induced) leaves produced increased amounts of hydrogen peroxide, which was used by HRP to increase the production of semiquinone radicals in the midguts of larvae. The decreased growth rates of larvae that fed on induced HRP-overexpressing poplars resulted from post-ingestive mechanisms, consistent with the action of HRP in their midguts. The toughness of HRP-overexpressing leaves was not significantly greater than that of WT leaves, whether or not they were induced. When leaves were coated with ellagitannins, induced HRP leaves also produced elevated levels of semiquinone radicals in the midgut. Decreased larval performance on induced HRP leaves in this case was due to post-ingestive mechanisms as well as decreased consumption. The results of this study provide the first demonstration that a POD is able to oxidize phenolics within an insect herbivore's gut, and further clarifies the chemical conditions that must be present for PODs to function as antiherbivore defenses.

Transgenic hybrid aspen (Populus tremula x tremuloides) overexpressing the MYB134 tannin regulatory gene show dramatically enhanced condensed tannin (proanthocyanidin) levels, as well as shifts in other phenolic metabolites. A series of insect bioassays with forest tent caterpillars (Malacosoma disstria) and gypsy moth (Lymantria dispar) caterpillars was carried out to determine how this metabolic shift affects food preference and performance of generalist tree-feeding lepidopterans. Both species showed a distinct preference for the high-tannin MYB134 overexpressor plants, and L. dispar performance was enhanced relative to controls. L. dispar reached greater pupal weight and showed reduced time to pupation when reared on the MYB134 overexpressing poplar. These results were unexpected since enhanced condensed tannin levels were predicted to act as feeding deterrents. However, the data may be explained by the observed decrease in the salicinoids (phenolic glycosides) salicortin and tremulacin that accompanied the upregulation of the condensed tannins in the transgenics. We conclude that for these two lepidopteran species, condensed tannin levels are unlikely to be a major determinant of caterpillar food preference or performance. However, our experiments show that overexpression of a single regulatory gene in transgenic aspen can have a significant impact on herbivorous insects.

Prosystemin is the precursor protein of the 18 amino acid wound signal systemin which activates systemic defense in tomato leaves against insect herbivores (McGurl B, Pearce G, Orozco-Cardenas M, Ryan CA, Science 255 (1992) 1570-1573). Here, we report the isolation of cDNA sequences encoding prosystemin from potato (Solanum tuberosum), black nightshade (S. nigrum), and bell pepper (Capsicum annuum), all members of the Solanaceae family, using reverse-transcription polymerase chain reaction (RT-PCR). Pairwise comparisons of the predicted prosystemin proteins from the three species with tomato prosystemin and among each other indicated sequence identities ranging from 73% to 88%. The deduced systemin polypeptides were synthesized and tested for their capacities to induce the synthesis of the defensive proteinase inhibitors in tomato leaves. Potato and pepper systemins were approximately as active as tomato systemin, whereas nightshade systemin was ten-fold less active. The accumulation of proteinase inhibitor mRNA transcripts could be induced in each of these plants by treatment with the homologous systemin. As in the tomato, in potato, black nightshade, and bell pepper plants, prosystemin homologs appear to function as precursors of systemic wound signals.

Polyphenol oxidase (PPO) is commonly believed to function as an effective antiherbivore defense in plants. PPO is induced in plants following herbivory, and insect performance is often negatively correlated with PPO levels. However, induced defenses create numerous changes in plants, and very little work has been done to test the direct effects of PPO on insect herbivores separately from other changes. This study examined the impacts of high levels of PPO on the performance of two species of tree-feeding caterpillars (Lymantria dispar and Orgyia leucostigma) on poplar. Transgenic PPO-overexpressing poplar (Populus tremula x Populus alba) was used as a source of elevated-PPO leaves, thereby controlling for the multiple effects of induction. In addition, the impacts of treating poplar foliage with high levels of purified mushroom PPO were examined on the two caterpillar species. Contrary to expectation, in several cases increased PPO levels had no significant effect on insect consumption or growth rates. Although one of the mechanisms by which PPO is believed to impact herbivores is via increased oxidative stress, the ingestion of large amounts of PPO had little or no effect on semiquinone radical and oxidized protein levels in the gut contents of lymantriid caterpillars. PPO activity in caterpillars is likely limited by the low oxygen and high ascorbate levels commonly found in their gut contents. This study questions whether induced PPO functions as an effective post-ingestive defense against tree-feeding caterpillars, and indicates that controlled, mechanistic studies are needed in other plant-herbivore systems to test for a direct effect of PPO on insect performance.