Secondary attraction to aggregation pheromones plays a central role in the host colonization behavior of the European spruce bark beetle Ips typographus. However, it is largely unknown how the beetles pioneering an attack locate suitable host trees, and eventually accept or reject them. To find possible biomarkers for host choice by I. typographus, we analyzed the chemistry of 58 Norway spruce (Picea abies) trees that were subsequently either (1) successfully attacked and killed, (2) unsuccessfully attacked, or (3) left unattacked. The trees were sampled before the main beetle flight in a natural Norway spruce-dominated forest. No pheromones were used to attract beetles to the experimental trees. To test the trees' defense potential, each tree was treated in a local area with the defense hormone methyl jasmonate (MeJ), and treated and untreated bark were analyzed for 66 different compounds, including terpenes, phenolics and alkaloids. The chemistry of MeJ-treated bark correlated strongly with the success of I. typographus attack, revealing major chemical differences between killed trees and unsuccessfully attacked trees. Surviving trees produced significantly higher amounts of most of the 39 analyzed mono-, sesqui-, and diterpenes and of 4 of 20 phenolics. Alkaloids showed no clear pattern. Differences in untreated bark were less pronounced, where only 1,8-cineole and (-)-limonene were significantly higher in unsuccessfully attacked trees. Our results show that the potential of individual P. abies trees for inducing defense compounds upon I. typographus attack may partly determine tree resistance to this bark beetle by inhibiting its mass attack.

Bark beetles kill apparently vigorous conifers during epidemics by means of pheromone-mediated aggregation. During non-endemic conditions the beetles are limited to use trees with poor defense, like wind-thrown. To find olfactory cues that help beetles to distinguish between trees with strong or weak defense, we collected volatiles from the bark surface of healthy felled or standing Picea abies trees. Furthermore, living trees were treated with methyl jasmonate in order to induce defense responses. Volatiles were analyzed by combined gas chromatography and electroantennographic detection (GC-EAD) on Ips typographus antennae. Compounds eliciting antennal responses were characterized by single sensillum recording for identification of specific olfactory sensory neurons (OSN). Release of monoterpene hydrocarbons decreased, while oxygenated compounds increased, from spring to early summer in felled trees. In both beetle sexes particular strong EAD activity was elicited by trace amounts of terpene alcohols and ketones. 4-Thujanol gave a very strong response and the absolute configuration of the tested natural product was assigned to be (+)- trans -(1 R, 4 S, 5 S )-thujanol by stereoselective synthesis and enantioselective gas chromatography. One type of OSN responded to all ketones and five other OSN were characterized by the type of compounds that elicited responses. Three new OSN classes were found. Of the eight EAD-active compounds found in methyl jasmonate-treated bark, the known anti-attractant 1,8-cineole was the one most strongly induced. Our data support the hypothesis that highly active oxygenated host volatiles could serve as positive or negative cues for host selection in I. typographus and in other bark beetles.

The semiochemical diversity hypothesis (SDH) states that interference with host‐selection from non‐host volatiles (NHV) is an important mechanism for associational resistance. Inhibition of bark beetle attraction to point sources by non‐host volatiles (NHV) is well established and might be a signal serving in host‐selection also at the habitat scale. In forests dominated by Norway spruce in middle and northern Europe (N Slovakia 2006 & 2007, SE Sweden 2007), we applied a blend of NHV and verbenone, released from dispensers fixed at 2 and 6 m height at forest edges with high Ips typographus populations. In Slovakia, three different doses (0.2–0.7 dispensers/m forest edge) were tested in 20‐tree zones of spruce stand edges. The Swedish experiments used only the middle dose. In Slovakia, there was high tree mortality but dispensers with the anti‐attractants reduced killed trees in a dose‐dependent manner. The reduction in tree killing ranged from 35 to 76% compared to untreated zones. Regression analysis of relative tree kill on log dispenser density was highly significant ( = 0.34, corresponding effect size d≈ 0.98). In Sweden, with lower beetle populations, most attacks (99%) were found outside the experimental areas, with high attack rates (15 trees/ha) in a range of 15–30 m from treated groups, indicating an active inhibitory radius exceeding the previous estimates. The SDH as a functional aspect of biodiversity was tested by converting spruce monocultures into an artificial semiochemically mixed forests. The use of NHV provides the only non‐insecticidal method of direct protection of conifer forests. The demonstrated principle of protection is still too expensive for area‐wide use, but viable for high‐value areas (nature reserves). Further development of push–pull strategies or area‐wide applications may prove more cost efficient. In the long‐term, the only sustainable approach is a forest landscape of mixed habitats.

Convergent evolution of semiochemical use in organisms from different Kingdoms is a rarely described phenomenon. Tree-killing bark beetles vector numerous symbiotic blue-stain fungi that help the beetles colonize healthy trees. Here we show for the first time that some of these fungi are able to biosynthesize bicyclic ketals that are pheromones and other semiochemicals of bark beetles. Volatile emissions of five common bark beetle symbionts were investigated by gas chromatography-mass spectrometry. When grown on fresh Norway spruce bark the fungi emitted three well-known bark beetle aggregation pheromones and semiochemicals (exo-brevicomin, endo-brevicomin and trans-conophthorin) and two structurally related semiochemical candidates (exo-1,3-dimethyl-2,9-dioxabicyclo[3.3.1]nonane and endo-1,3-dimethyl-2,9-dioxabicyclo[3.3.1]nonane) that elicited electroantennogram responses in the spruce bark beetle Ips typographus. When grown on malt agar with 13C d-Glucose, the fungus Grosmannia europhioides incorporated 13C into exo-brevicomin and trans-conophthorin. The enantiomeric compositions of the fungus-produced ketals closely matched those previously reported from bark beetles. The production of structurally complex bark beetle pheromones by symbiotic fungi indicates cross-kingdom convergent evolution of signal use in this system. This signaling is susceptible to disruption, providing potential new targets for pest control in conifer forests and plantations.

Conifer feeding bark beetles (Coleoptera, Curculionidae, Scolytinae) pose a serious economic threat to forest production. Volatiles released by non-host angiosperm plants (so called non-host volatiles, NHV) have been shown to reduce the risk of attack by many bark beetle species, including the European spruce bark beetle, Ips typographus. However, the most active blend for I. typographus, containing three green leaf volatiles (GLVs) in addition to the key compounds trans-conophthorin (tC) and verbenone, has been considered too expensive for use in large-scale management. To lower the cost and improve the applicability of NHV, we aim to simplify the blend without compromising its anti-attractant potency. Since the key compound tC is expensive in pure form, we also tested a crude version: technical grade trans-conophthorin (T-tC). In another attempt to find a more cost effective substitute for tC, we evaluated a more readily synthesized analog: dehydro-conophthorin (DHC). Our results showed that 1-hexanol alone could replace the three-component GLV blend containing 1-hexanol, (3Z)-hexen-1-ol, and (2E)-hexen-1-ol. Furthermore, the release rate of tC could be reduced from 5 mg/day to 0.5 mg/day in a blend with 1-hexanol and (–)-verbenone without compromising the anti-attractant activity. We further show that T-tC was comparable with tC, whereas DHC was a less effective anti-attractant. DHC also elicited weaker physiological responses in the tC-responding olfactory receptor neuron class, providing a likely mechanistic explanation for its weaker anti-attractive effect. Our results suggest a blend consisting of (–)-verbenone, 1-hexanol and technical trans-conophthorin as a cost-efficient anti-attractant for forest protection against I. typographus.