The quality of available food may affect insect herbivores directly (via growth and survivorship) and/or indirectly (by modifying insect vulnerability to parasitoids and pathogens). We examined the relationship between different phenolic compounds, belonging to various phenolic groups, in Betula pubescens spp. czerepanovii (mountain birch) foliage and the larval performance of the geometrid Epirrita autumnata (autumnal moth). Direct effects on insect performance were described by pupal weight, developmental rate, and survivorship; indirect effects were described by the encapsulation rate of an implant inserted into the insect hemocoel, a commonly used way to describe insect immune defense. We found profound differences in the effects of different phenolic categories: several individual hydrolyzable tannins were associated positively with larval performance but negatively with level of immune defense, whereas flavonoid glycosides were inversely related to larval survival but showed no association with the larvae immune defense.

Plants have been suggested to have an immunological memory comparable to animals. The evidence for this, however, is scarce. In our study with the mountain birchEpirrita autumnata system, we demonstrated that birches exposed as long as 5 yr to feeding of E. autumnata larvae (delayed induced resistance, DIR), responded more strongly to a new challenge than trees without an herbivory history. Pupal weights remained lower, and the duration of the larval period was prolonged in the DIR trees, although immunity, measured as an encapsulation rate, was not affected. We further demonstrated that the effects of birch phenolics on performance of E. autumnata were different in the exposed (DIR) trees from naive control trees, although we found only one significant change in chemistry. The quercetin:kaemferol ratio was increased in DIR trees, suggesting that herbivory caused oxidative stress in birches. In DIR trees, phenolics, especially hydrolyzable tannins (HTs), affected pupal weights negatively, whereas in control trees, the effects were either nonsignificant or positive. HTs also prolonged the duration of the larval period of females, whereas peroxidase (POD) activity prolonged that of males. We suggest that the causal explanation for the induced resistance was an enhanced oxidation of phenolic compounds from the DIR trees in the larval digestive tract. Phenolic oxidation produces semiquinones, quinones, free radicals, and ROS, which may have toxic, antinutritive, and/or repellent properties against herbivores.

1. Population density of Epirrita autumnata (Lepidoptera: Geometridae) reaches outbreak densities regularly in northernmost Scandinavia. During these outbreak years, the most abundant host species, the mountain birch (Betula pubescens ssp. czerepanovii), is regularly exhausted, although larvae may rescue themselves from starvation by using alternative host species. 2. In this paper, the effects of the shift of host species on the immune defence and other life‐history traits of E. autumnata were investigated, and possible consequences for population dynamics were briefly discussed. Moth larvae were reared on the leaves of the main host, mountain birch, until larvae reached their third instar. After this, larvae were allocated randomly to five treatments: larvae were either allowed to finish larval stage on the mountain birch or were shifted onto four alternative host species that are typical species for the area. 3. As expected, the host species had a major effect on fitness traits: body weight, development, and survival rate of the moths. The pupal weight was lower and development rates slower on the three alternative host species, Salix myrsinifolia Salisb., Vaccinium uliginosum L., and Betula nana L., than on the main host, mountain birch. 4. The immunity was, however, the same or better on the alternative hosts than on the main host. The immunity and pupal weights were negatively related, suggesting a trade‐off between body size and immunocompetence. 5. The decreased body size and fecundity of E. autumnata during outbreak years may be partly due to the shift to alternative host species whereas the host‐plant species probably does not affect markedly the rate of parasitism.

Haviola, S., Saloniemi, I., Ossipov, V. and Haukioja, E. 2006. Additive genetic variation of secondary and primary metabolites in mountain birch. Á/ Oikos 112: 382 Á/ 391.Evolutionary adaptations require genetic variation in the traits concerned. Mountain birch populations suffer from regular autumnal moth defoliations that affect tree survival and growth. Earlier studies have found that birch individuals show clear differences in their leaf chemistry, which may be relevant to herbivore performance. A large part of this variation is assumed to be genetic, since quantitative differences between trees remain the same in different years. We estimated the genetic parameters of several herbivory-related chemical compounds. The material consisted of 30 mature wild trees and their progenies. We treated the data both as half-sib and full-sib families, and calculated parent Á/offspring comparisons as well. The compounds included 12 individual phenolics, 5 carbohydrates and 16 protein-bound amino acids. For comparison, we also analysed the weight and germination percent of seeds and the height of seedlings. Two of the most abundant phenylpropanoids (chlorogenic acid and kaempferol-3-O-a-L-rhamnopyranoside) showed significant genetic variation, but this was not true of all individual phenolic compounds. Amino acids had less genetic variation than phenolic compounds. The high genetic variation of the phenolic compounds indicates that the mountain birch is able to evolve its phenol-based resistance.

Temperature directly affects the growth, survival, and development rates of poikilothermic insect herbivores; it may also have an important indirect impact, via the activities of plant defensive enzymes. The effects of wounding birch leaves and temperature on the growth and development rates of a Lepidopteran moth, Epirrita autumnata, were studied. We also examined the activities of a mountain birch (Betula pubescesns spp. czerepanovii) defensive enzymes, specifically the polyphenoloxidases (PPOs), in relation to temperature and wounding. The optimal temperature for early instars in terms of survival and developmental rates was between +15 and 20 degrees C. Wounding treatment had different effects on birch PPO activity depending on the temperature: at +12 degrees C, wounding decreased the activity, suggesting induced amelioration at that temperature, whereas at +25 degrees C, wounding increased the activity, suggesting induced resistance. However, larval growth was retarded slightly, but significantly, on the leaves of wounded twigs at both temperatures. Both PPO activity and larval growth rates were affected within 12 h, indicating the existence of a transcription- and translation-independent defense system in birch leaves. We suggest that underlying the increase in PPO activity and the decrease in larval growth rate may be H2O2, which has been shown to accumulate in response to wounding. Our results also provide a possible biological mechanism for the hypothesis that low temperatures promote the success of E. autumnata and other Lepidopteran larvae via decreased defensive enzyme activities of host plants at lower temperatures.

Previous studies of mountain birch (Betula pubescens spp. czerepanovii) repeatedly have found differences between individual trees in herbivory-related traits, but rarely have yielded estimates of the additive genetic variation of these traits or of their relationship to habitat. We used thirtyyear-old birch half-sibs in a northern common garden to estimate the effect of genetics and local microhabitat on resistance-related traits. Genetic estimates of foliar chemistry have been studied only rarely with trees as old as these. Moth performance (Epirrita autumnata), rust (Melampsoridium betulinum) incidence levels, and the general level of natural herbivory damage to individual trees were used as direct measures of birch resistance. Chemical resistance-related traits in plant chemistry included 15 individual phenolics, 16 amino acids, and phenoloxidase activities in the foliage. We also followed birch phenology and growth. Our results show that the genotype of the birch was the most important determinant of phenolic composition and phenoloxidase activity, but that amino acid levels were best explained by the microhabitat of the birch. We also found that the phenology of the birch had a high heritability, although its variation was low. Our results reveal rich genetic variation in birch chemistry.

Mountain birches are deciduous trees consisting of several clonal and partly autonomous ramets. Autumnal moth, Epirrita autumnata, is an extremely harmful pest of mountain birches that during outbreak years may cause the death of trees over large areas. During the larval development of E. autumnata, leaf characters and nutritional suitability vary considerably both temporally and spatially among and within mountain birches. Compared to variation between trees, however, the extent of within-tree variation is known for only a limited number of potentially defensive traits. Plant oxidases, polyphenoloxidases (PPOs) and peroxidases (PODs) may play an important part in dictating the suitability of a tree for plant-eating animals, including E. autumnata. In this study, we observed changes in the activities of oxidative and antioxidative enzymes along leaf development during the larval period of E. autumnata. We also estimated the relative amounts of spatial variation among and within trees. Our results show that POD activities were remarkably high during the rapid elongation growth phase of leaves but decreased with leaf maturation. Thus PODs are suggested to take part in leaf elongation growth. Time and within tree variation accounted for the most of variation in POD activities whereas there was no among tree variation. In contrast, the activities of antioxidative CATs, which act as antagonists to PODs, were low in young leaves and increased towards leaf maturation, reflecting an increase in the oxidative status of source leaves. Within tree variation accounted for the most of the variation in CAT activities. The absolute PPO activities decreased along leaf growth due to the dilution effect whereas the specific activity of PPOs, which has been shown to be defensive against E. autumnata larvae, did not vary temporally that might reflect the importance of these enzymes in the defence of birches. Acidic PPOs showed marked withinand among-tree variation, which may impact herbivores performance on certain genotypes and increase larval mobility within the tree canopy.

Several studies suggest that ethylene is involved in the responses and resistance of plants against herbivores and pathogens, but the role of ethylene seems to vary depending on the system studied. Here we extend the studies of introduced Arabidopsis thaliana (L.) Heynh. ethylene receptor gene (etr1-1) to a new type of interaction: a woody plant, Betula pendula Roth (silver birch) and its herbivore, Epirrita autumnata (Borkhausen) (autumnal moth). We studied constitutive insect resistance of two wild-type birch genotypes, V and J, which were very different hosts for E. autumnata. On the J background, genetically modified ethylene insensitivity decreased the performance of E. autumnata, measured as leaf damage, larval mass, or pupal mass. However, hardly any changes were detectable in the V background, which is an inferior wild-type genotype for autumnal moth performance, except increased leaf damage caused by E. autumnata in ethylene-insensitive trees. Ethylene insensitivity caused clear side effects on birch phenology and morphology, especially in the J background. In this background, ethylene modification mainly acted via accelerating phenology, but this was hardly detectable in the V genotype. Also, the number of long shoots increased only in the modified J background. Taken together, it seems that the effects of ethylene on several birch traits relevant for E. autumnata and birch morphology depend on the birch genetic background.