Byers, J.A. 1995a. Host tree chemistry affecting colonization in bark beetles, in R.T. Cardé and W.J. Bell (eds.). Chemical Ecology of Insects 2. Chapman and Hall, New York, pp. 154-213.

INTRODUCTION-- Bark beetles (order Coleoptera: family Scolytidae) comprise a taxonomic group of species that look similar although they differ widely in their ecology and biochemical adaptations to host trees. This diversity of bark beetle biology, in which each species is adapted to only one or a few host tree species, has probably resulted from natural selection due to the great variety of trees and their biochemicals. It also is likely that each species of tree has coevolved various chemicals to defend against the herbivorous selection pressures of bark beetles and other insects (Erlich and Raven, 1965; Feeny, 1975; Cates, 1981; Berryman et al., 1985). Host plant chemicals can be attractive, repellent, toxic, or nutritious to bark beetles and have affects on: (1) finding and accepting the host tree (host selection and suitability), (2) feeding stimulation and deterrence, (3) host resistance, (4) pheromone/allomone biosynthesis and communication, and (5) attraction of predators, parasites and competitors of bark beetles.

Bark and ambrosia beetles contain at least 6000 species from 181 genera worldwide (S.L. Wood, 1982). In the United States there are 477 species, and in North and Central America a total of 1430 species occur from 97 genera. Bark beetles may have originated as early as the Triassic (over 200 million years ago) on conifer hosts (S.L. Wood, 1982). Baltic amber dating from the Oligocene (25-30 million years ago) sometimes contains entrapped insects that look identical to bark beetles from species of present-day genera such as Tomicus (S.L. Wood, 1982).

Since 1970 there have been over 3800 research papers on bark and ambrosia beetles (BIOSIS Previews computer database, Philadelphia, PA, USA). The genus Dendroctonus has been the most studied with over 1196 papers published, primarily on four pest species of North America, D. frontalis, D. ponderosae, D. brevicomis, and D. pseudotsugae. Other genera in order of studies were: Ips, Scolytus, Xyleborus, Trypodendron, Tomicus=Blastophagus, Pityogenes, Hypothenemus, Pityophthorus, Hylastes, and Gnathotrichus. The most studied Ips species (852 papers) also were pests, I. typographus of Europe, and the three North American species, I. paraconfusus = confusus, I. pini and I. grandicollis. Scolytus multistriatus, vector of the Dutch elm disease, made up the majority of papers from this genus. Thus, it is clear that most biological knowledge on bark beetles derives from studies on relatively few pest species (obligate and facultative parasites comprise about 10% of scolytid species in US and Canada, Raffa et al., 1993). This focus on pests is appropriate, however, since only commonly occurring bark beetles that kill living trees or their parts would be expected to have a significant influence on evolution of the host tree and its chemistry.

CONCLUSIONS-- Of the 6000 bark beetle species worldwide, in a particular geographic area there are usually from only a few to some tens of species that colonize any given species of tree, and then only one or a few species that can attack and kill the tree. Each host-tree species has a large variety of chemicals, some of which affect the success of the bark beetle in finding and colonizing its host tree. Bark beetles probably orient to attractive semiochemical sources during flight using odor-modulated anemotaxis, as in moths, but little is known about this process. It is better understood how beetles orient upwind during walking to attractive sources. Most studies have observed chemotaxis in arenas for the purpose of isolating pheromones rather than from the standpoint of basic behavior.

Bark beetles find suitable host trees by orienting to host odors, especially ethanol and monoterpenes, as well as to aggregation pheromones. However, very few studies could be characterized as complete or rigorous because the attractive compounds were discovered usually by screening of semiochemicals previously known for other species. Other studies have selected compounds for testing based on their presence in the insect or related species, but it is likely that many bioactive compounds have been overlooked. Also, blends of compounds have rarely been tested for synergism. Tree volatiles that attract predators and parasites of bark beetles are often the same as those that attract their host, i.e., most of these chemicals have been discovered by chance when testing compounds on bark beetles. Feeding stimulants and deterrents of conifer bark beetles have been isolated in various solvent fractions but not identified. Several compounds that elicit or deter feeding in deciduous bark beetles have been identified, but undoubtedly many behaviorally active compounds remain to be discovered. It is likely that behavioral responses of bark beetles within a species to semiochemicals may vary geographically as well as the semiochemicals produced by the bark beetles and the host trees.

Fractionation of a biological extract by chromatographic methods (usually GC) and then recombination of certain fractions with an additive method has been used to test for synergism among semiochemicals in a behavioral bioassay. This method was used to isolate some of the first multi-component pheromones of bark beetles (Silverstein et al., 1966, 1967; 1968; Pearce et al., 1975). However, due to the substantial work involved with these classical isolation methods, most studies have discovered semiochemicals by screening or comparative methods which are inherently less rigorous and are dependent on chromatographic resolution. The subtractive method, where each of the fractions is subtracted from the blend and tested such that blends with lowered activity indicate subtracted synergists, should aid in isolation of synergistic semiochemicals that otherwise have been hard to detect (Byers et al., 1990a; Byers, 1992b).

Future studies should be careful to report the release rates of test volatiles, and in many cases these should be adjusted to coincide with natural rates (Byers, 1988). This means that measurements of volatile release of semiochemicals must be done in many bark beetle systems for both host- and beetle-released semiochemicals (Birgersson and Bergström, 1989). When testing semiochemicals in the field, the spatial and temporal variation of responding insect populations with respect to trap placement may lead to erroneous conclusions. To counter this potential problem, relatively numerous trap replications have been previously employed; however, the mechanical slow rotation of a pair of traps (1-2 rph at 6 m separation) can even this catch variation (Byers et al., 1990b).

Resistance of trees has been studied for many years and monoterpenes, such as limonene, are implicated in their resistance to bark beetles and their symbiotic microorganisms (mostly fungi). However, there has been little recent toxicological work and the relative importance of the purported toxins remains to be established. Also, the monoterpenes have been tested at much higher vapor concentrations than that in nature, and they have not been evaluated in diets. Synergism or interactions of various "toxins" have not been investigated. Also, geographic variation in toxicity of host compounds has been little studied. A correlation between high-limonene trees and historical "predation" by bark beetles has been suggested as an example of host chemical evolution. However, more studies are needed in stands with ongoing outbreaks of bark beetle to determine if natural selection can alter the genetic frequencies in the population of trees, and at the same time if populations of bark beetle change their tolerance to particular monoterpenes that were initially most toxic.

Host tree monoterpenes alpha-pinene and myrcene can be converted by a simple hydroxylation to ipsenol, ipsdienol, cis-verbenol and trans-verbenol, pheromone components of bark beetles. However, I. paraconfusus is able to make the same amounts of ipsenol and ipsdienol regardless of the myrcene titer in the host tree, suggesting the major pathway is de novo. Recent studies in Ips species also suggest that pheromonal analogues of myrcene may not be derived primarily from myrcene but by synthesis from mevalonate. Although ipsenol/ipsdienol and E-myrcenol biosynthesis in some species of bark beetle are probably not coevolving with myrcene in the tree, it is possible that cis- and trans-verbenol biosynthesis may coevolve with alpha-pinene levels in hosts. Both cis-verbenol and trans-verbenol appear to be directly produced in bark beetles by conversion of alpha-pinene enantiomers from the host tree. However, verbenone, an inhibitor of aggregation in many bark beetle species, may not be directly converted from alpha-pinene. Other bark beetle pheromone components are probably biosynthesized from small molecules into the more complex structures in several or more different biosynthetic pathways.

Evolution of tree chemistry in response to predation by bark beetles is best supported in studies of host compounds that are toxic to bark beetles or that deter feeding. The bark beetle has also coevolved detoxification mechanisms for the toxic monoterpenes, some of which have been secondarily utilized as pheromone components. Volatile host attractants can be termed kairomones, and there is little evidence that trees evolve these compounds to repel herbivores not adapted to this potential host, since the same compounds attract their herbivores. The compounds probably are beneficial in some way to the tree and can not be dispensed with even though bark beetles (and some of their predators and parasites) have evolved to utilize the compounds as kairomones. Host tree chemistry affects most aspects of bark beetle biology, moreover, bark beetles probably differentially affect survival of host trees and alter genotypic frequencies and host chemistry both at the micro-evolutionary scale (cycling of endemic and epidemic insect populations) and at the macro-evolutionary level (host tree selection in response to new species of bark beetle).

Chemical Ecology