5.2.3 Host plant volatiles attractive to bark beetles

Host volatiles are attractive to a number of forest scolytids including species in the genera Scolytus, Dendroctonus, Hylurgops, Trypodendron and Tomicus (Goeden and Norris, 1964; Rudinsky, 1966; Meyer and Norris, 1967a; Moeck, 1970; Byers et al., 1985; Lanne et al., 1987; Volz, 1988; Lindelöw et al., 1992; Hobson et al., 1993). Species of bark beetle that regularly attack and kill living trees (termed "aggressive") have been shown invariably to possess an aggregation pheromone, usually of two or more components, but are weakly, if at all, attracted by host volatiles alone (Vité and Pitman, 1969; Byers, 1989a). However, so-called "secondary" bark beetle species (those that arrive later after the tree has already been killed by the aggressive bark beetles or that feed as saprophytes in decaying trees) may not use an aggregation pheromone, but generally are strongly attracted to either host monoterpenes, ethanol or a combination (Kohnle, 1985; Klimetzek et al., 1986; Schroeder, 1988; Schroeder and Lindelöw, 1989).

Ethanol, probably released by microorganisms in decaying woody tissue (Graham, 1968; Moeck, 1970; Cade et al., 1970) and stressed plants (Kimmerer and Kozlowski, 1982), is attractive to a wide variety of species of bark beetles (Moeck, 1970, 1981; Magema et al., 1982; Montgomery and Wargo, 1983; Kohnle, 1985; Klimetzek et al., 1986; Schroeder 1987, 1988; Schroeder and Eidmann, 1987; Phillips et al., 1988; Volz, 1988; Chénier and Philogène, 1989; Schroeder and Lindelöw, 1989; Byers, 1992a). Primary alcohols other than ethanol have not been reported as being attractive to scolytids. However, only a few studies have tested methanol (Moeck, 1970; Montgomery and Wargo, 1983; Byers, 1992a); longer chain alcohols up to hexanol did not attract Scolytids in Sweden when they were known to be flying (Byers, 1992a). Electroantennogram (EAG) responses of T. piniperda to a series of straight-chain alcohols indicated that the antennae respond increasingly with longer chain length up to a maximum between pentanol and heptanol and then decrease (Lanne et al., 1987). The response spectrum could be due in part to differences in volatility. Thus, although ethanol plays a role in host selection (discussed subsequently), the EAG response for ethanol is lower than for longer-chain alcohols (which probably are not involved in behavior). Ethanol and CO2 are the usual end products of sugar fermentation by microorganisms whereas methanol is not, which probably explains the evolution of the use of ethanol by forest insects. Moeck (1970) found methanol to be a minor constituent and ethanol a major constituent of extracts from Douglas-fir sapwood attractive to T. lineatum.

Various tree monoterpenes (e.g. alpha-pinene, myrcene, terpinolene, B-pinene, Fig. 4) and turpentine are also attractive to a large number of bark beetle species (Byers et al., 1985, 1992a; Phillips et al., 1988; Schroeder, 1988; Chénier and Philogène, 1989; Schroeder and Lindelöw, 1989; Miller and Borden, 1990; Phillips, 1990; Hobson et al., 1993). Fig. 4 - Host tree monoterpenes
Fig. 4. Major monoterpenes of conifers. Note that the enantiomers of alpha-pinene are identical except that they are non-superimposable (mirror images). Camphene, B-pinene, 3-carene, B-phellandrene, and limonene also have two enantiomers, although only (-)-B-pinene and (+)-3-carene are found in trees (Mirov, 1961). Myrcene and terpinolene are achiral.

Synergism between ethanol and various monoterpenes (or turpentine) is also of widespread occurrence (Nijholt and Schönherr, 1976; Kohnle, 1985; Vité et al., 1986; Phillips et al., 1988; Volz, 1988; Schroeder, 1988; Chénier and Philogène, 1989; Schroeder and Lindelöw, 1989; Phillips, 1990). These compounds are not only important for primary attraction to plants, but also may play a role in enhancing the bark beetles' response to aggregation pheromone (Bedard et al., 1969, 1970; Pitman et al., 1975; McLean and Borden, 1977; Borden et al., 1980, 1981; Paiva and Kiesel, 1985; Byers et al., 1988; Miller and Borden, 1990). Host-tree compounds, ethanol and monoterpenes, elicited increased entering rates of bark beetles T. lineatum and P. chalcographus, respectively, into pipe traps baited with aggregation pheromone (Vité and Bakke, 1979; Bakke, 1983; Byers et al., 1988). B-Phellandrene (Fig. 4) is slightly attractive alone to I. pini and enhances response to pheromone (Miller and Borden, 1990), and so the monoterpene might induce entering of holes.

In most of the above studies, the discovery of host compounds attractive to bark beetles has been by the comparative approach (similar species are known to be attracted) or by surmising that identified chemicals in the host would be attractive. Thus, most studies are incomplete because of the possibility that there are still undiscovered chemicals important for attraction to the host. Byers et al. (1985) used the subtractive-combination bioassay and fractionation method (Byers, 1992b) to rigorously identify the host volatiles responsible for aggregation of T. piniperda. A combination of (-)-(S)-alpha-pinene, (+)-(R)-alpha-pinene, (+)-3-carene, and terpinolene, or each alone, was effective in attracting both sexes (Fig. 4). During the isolation study, designed for detection of synergistic pheromone components, no evidence was found for beetle-produced compounds being attractive, in contrast to most bark beetles that aggregate en masse on hosts (Byers, 1989a). Byers et al. (1985) quantified the release rates of alpha-pinene, terpinolene and 3-carene from a freshly cut log of Scots pine (28 cm x 13 cm diam.) and found them each to be about 15 mg/day. Release of comparable amounts in the field competed favorably with a host log in attracting T. piniperda. They theorized that the attraction to monoterpenes functioned in the beetle's selection of both host- species (other common tree species have less monoterpenes) and recognition of the host's susceptibility (storm-damaged trees are unable to resist attack and have resinous wounds that release monoterpenes).

In the isolation of host volatiles attractive to T. piniperda, a gas-chromatographic adsorbent (Porapak Q), widely used for trapping insect pheromones, was used to collect headspace air from the infested pine logs. Unfortunately, Porapak Q will not retain ethanol molecules due to their small size. Thus ethanol could be a constituent of the attractive host odor. Vité et al. (1986) presented evidence that ethanol enhanced the attraction of T. piniperda to alpha-pinene and terpinolene (identified above) by about eight-fold, but these results are difficult to confirm since the chemical release rates were not given. They proposed that ethanol would be released from diseased trees and thus indicate their suitability to T. piniperda. Ethanol is attractive when released on healthy trees since T. piniperda were caught in ethanol-baited traps on trees, and these beetles also attacked trees baited with ethanol (Schroeder and Eidmann, 1987; Byers, 1992a). However, the attraction to ethanol in traps away from trees is weak or negligible, while monoterpenes in these traps are attractive (Schroeder, 1988; Schroeder and Lindelöw, 1989; Byers, 1992a).

Ethanol, in fact, sometimes reduces response to attractive baits. Klimetzek et al. (1986) tested different release rates of ethanol (24 to 125 mg/day) with an unreported release rate of alpha- pinene and terpinolene and found that the higher releases of ethanol inhibited attraction of T. piniperda. However, a control with ethanol alone, or terpenes alone, was not reported. Schroeder (1988) increased the release of ethanol in five dosages over an even wider range from 0 to 50 g/day in combination with a 240 mg/day release of alpha-pinene. In this case, the attraction of T. piniperda declined linearly with the logarithm of ethanol release, in conflict with the theory of Vité et al. (1986) that ethanol was synergistic with monoterpenes.

Schroeder and Lindelöw (1989) provided the first evidence that could integrate the disparate results. They found that a high release of alpha-pinene was most attractive to beetles and that ethanol releases alone from 0 to 3 g/day were barely attractive. At a low release rate of alpha-pinene (2.4 or 22 mg/day) and thus low attraction, the lower releases of ethanol from 0 to 3 g/day had a synergistic effect with alpha-pinene in attracting beetles (Schroeder and Lindelöw, 1989). Their results are supported by Byers (1992a); i.e., a weak enhancement of attraction by ethanol at low releases of the blend of three host monoterpenes, but no observable effect of ethanol on the greater attraction to higher releases of monoterpenes.

Ethanol released at even higher rates, 120 mg/day (Klimetzek et al., 1986) or 50 g/day (Schroeder, 1988), inhibits the response of T. piniperda to monoterpenes. Therefore, the beetle could find diseased, but physically uninjured, trees by a weak response to a synergism between low monoterpene release rates and moderate ethanol rates - the hypothesis of Vité et al. (1986). These trees would be tested occasionally by beetles, and if low in resistance this would permit beetles to continue feeding. Resinosis and monoterpene release would elicit increased numbers joining in a mass-

attack. Injured trees with wound oleoresin, and trees under attack with "pitch tubes", would have a higher monoterpene release and attract the greatest numbers of beetles, according to Byers et al. (1985). Trees with high ethanol release rates would indicate a tree in advanced decay and unsuitable for reproduction, and thus to be avoided, as theorized by Klimetzek et al. (1986). High monoterpene releases from trees (freshly wounded and not dead) would not naturally coincide with high ethanol release rates (presumably during decay after death). In addition, other compounds such as verbenone from decaying hosts would inhibit response to monoterpenes from unsuitable hosts (discussed in the next part). These studies emphasize the need for releasing semiochemicals at precise rates during tests in the field. In addition, measurements of the natural release rates of ethanol and monoterpenes from various host and nonhost substrates are necessary for further understanding of bark beetle chemical ecology.
Next Page

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.