5.6 SECONDARY PLANT COMPOUNDS AND BARK BEETLE-TREE COEVOLUTION

5.6 SECONDARY PLANT COMPOUNDS AND BARK BEETLE-TREE COEVOLUTION

Secondary plant compounds are those chemicals not believed to be necessary for the basic metabolism of plant cells (Whittaker, 1970). Since the compounds occur in some plant families, often of very different lineage, and not in other families, they appear to have evolved independently for the defense against a common enemy such as fungi or insects (Feeny, 1975). Secondary plant compounds include many exotic structures: phenolics, cinnamic acids, coumarins, lactones, quinones, phenolic alcohols, glycosides of any of these, furan and pyranes, anthraquinones, flavonoids, anthocyanin, polymeric phenolics (tannins), monoterpenoids, sesquiterpenoids, and di- and triterpenoids, resin acids, steroids, cardiac glycosides, ecdysone analogues, and alkaloids.

Secondary plant compounds important for host-finding from long-range would necessarily be volatile. Some of the most common volatiles in conifers are the monoterpenes and thus it is not surprising that bark beetles are often attracted to these chemicals even though in high concentrations they are toxic. It could be expected that volatile monoterpenes shown to be important for a species in one geographic region might not be attractive to populations of this species in another region, if the trees in the second region have different proportions of the monoterpenes. The monoterpenes do vary in ponderosa pine throughout the western United States (Smith, 1964, 1966, 1967, 1968, 1969). This could explain why different populations of D. brevicomis may have responded differently to myrcene or 3-carene as synergists of pheromone components (Bedard et al., 1969, 1970; Pitman, 1969).

Quantities of volatile compounds such as ethanol and verbenone that are released from killed trees later during colonization by beetles and microorganisms are probably not regulated by the tree's genetics. Thus, trees are not under selection pressure from bark beetles because the tree has no control over how the insects use these compounds (as pheromones, allomones or kairomones) to indicate potential competition or decaying, unsuitable host areas. Kairomones, in contrast to pheromones and allomones, are not used in chemical communication, in which both parties, sender and receiver, must gain a benefit (Burghardt, 1970).

Host plant specificity by insects may be the result of plant- insect coevolution of chemical defenses and interspecific competition. Those insect species that do not adapt to a genetic change in host plant chemistry during evolution are not able to eat the plant as effectively and thus are at a competitive disadvantage to the species that adapt first. The consequence is that one or a few species now become prevalent (major pests) because of reduced interspecific competition.

Hughes (1973, 1974) hypothesized that aggregating pheromones in Dendroctonus and Ips are "waste products from the metabolism of terpenes that have secondarily been utilized as chemical messengers." D. ponderosae have an enzyme system that indiscriminately hydroxylates monoterpenes on allylic methyl groups that are E to a methylene or vinyl group as a way of detoxifying monoterpenes (Pierce et al., 1987). The products may also be more easily excreted due to the increased solubility in aqueous fluids. Some of the detoxification products of myrcene have been used as pheromones, but there appears to have been a selection for beetles that can synthesize these pheromone components from mevalonate (Byers and Birgersson, 1990; Ivarsson et al., 1993). In I. paraconfusus the production of ipsenol and ipsdienol is not primarily for detoxification of myrcene since the conversion is sex-specific, inhibited by streptomycin and mating with females, and induced by JH. On the other hand, the cis-verbenol synthesis is primarily a detoxification process since the compound is not sex- specific (although males produce more cis-verbenol) and JH, streptomycin or mating have no effect on its production (Hughes and Renwick, 1977; Byers et al., 1979; Byers 1981a, b, 1983b; Byers and Wood, 1981b; Hunt and Borden, 1989). Other than cis-verbenol and possibly trans-verbenol, no other bark beetle pheromone components (Fig. 5) are thought to be primarily a detoxification of monoterpenes, but rather are synthesized de novo from acetate or mevalonate (Vanderwel and Oehlschlager, 1987; Lanne et al., 1989; Ivarsson et al., 1993).

A case where coevolution may still occur is in I. paraconfusus and I. typographus that rely on host tree (-)-alpha-pinene as the precursor of cis-verbenol. However, I. typographus is relatively insensitive to large variations in the proportion of cis-verbenol to methyl butenol in the pheromone blend (Schlyter et al. 1987c), indicating that most Norway spruce are adequate hosts regardless of their alpha-pinene content. Thus, a tree genotype that produced less (- )-alpha-pinene might not gain an advantage since beetles could still produce adequate pheromone to cooperate in killing the tree. Certainly, the large number of beetle generations relative to the tree would allow the genes of beetles to track any changes in the host and compensate in production and/or response. Presumably the benefits of alpha-pinene for the tree, such as its toxicity to beetles and microorganisms discussed earlier, also would counter any tendency to select trees with a lower titer of alpha-pinene.

Several authors have suggested that some plants have evolved an indirect tri-trophic mechanism of resistance in which they release compounds (synomones) after being fed upon that are attractive to predators or parasites of their herbivores (Turlings et al., 1990; Whitman and Eller, 1990). The plant could gain benefits if their herbivores are not attracted by these same host volatiles. However, the same host compounds that are attractive to the predators and parasites of bark beetles are often attractive to bark beetles as well. For example, Enoclerus lecontei (Cleridae) and Temnochila virescens (Ostomidae) prey on several bark beetles such as D. brevicomis and I. paraconfusus in California. The predators are attracted to several monoterpenes or n-heptane (Jeffrey pine, Mirov, 1961) from host trees of bark beetles (Rice, 1969; Pitman and Vité, 1971). A clerid predator in Europe, Thanasimus formicarius preys on I. typographus of Norway spruce and T. piniperda of Scots pine and is also attracted to alpha-pinene and other monoterpenes from these conifers (Schroeder, 1988; Schroeder and Lindelöw, 1989). The dipteran parasites Medetera aldrichii of D. pseudotsugae and M. bistriata of D. frontalis are attracted to the host tree monoterpene, alpha-pinene, presumably aiding in their location of attacked trees (Williamson, 1971; Fitzgerald and Nagel, 1972). These plant monoterpenes are attractive to many bark beetles, or are synergists of attractive pheromone components (see 5.2.3). Thus, it seems doubtful that the tree would benefit by producing chemicals that attract bark beetle enemies as well as predators and parasites of these beetles. In this case the host compounds are better thought of as kairomones, i.e., chemicals that are used by receiving individuals to gain advantages. Kairomones are not dispensed with by the plant over evolutionary time because they confer advantages, such as insect resistance, that outweigh any disadvantages of them attracting additional herbivores.

Other parasites and predators of bark beetles are most strongly attracted to volatiles produced by their host bark beetle, although some of the attractive compounds may be derived from monoterpene precursors in the tree. Temnochila virescens feeds on D. brevicomis and the predator is almost as sensitive to exo- brevicomin as is the bark beetle to this pheromone component (Pitman and Vité, 1971; Byers, 1988). However, in Texas T. virescens is more attracted to a mixture of Ips components (ipsdienol, ipsenol and cis-verbenol) than to exo-brevicomin (Billings and Cameron, 1984). Clerid beetles such as Thanasimus formicarius are "generalist" predators of bark beetles, feeding on several species and are attracted to (or perceive) compounds of Ips (ipsenol, ipsdienol, cis-verbenol), Trypodendron (lineatin), and Dendroctonus/Dryocoetes (exo-brevicomin)(Bakke and Kvamme, 1981; Hansen, 1983; Kohnle and Vité, 1984; Lanne et al., 1987; Tommerås, 1988).

Byers, J.A. 1995. 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.