tree but can vary widely between males from different spruce trees. Very good correlations were found between some host tree monoterpene hydrocarbons and.
HOLARCTIC ECOLOGV 12: 451-456. Copenhagen 1989
Host tree resistance influencing pheromone production in Ips typographus (Coleoptera: Scolytidae) (iiirun Birgersson Birgersson. G. 1989. Host tree resistance influencing pheromone production in Ips typographus (Coleoptera: Scolytidae). - Holarct. Ecol. 12: 451-456. Analyses of volatiles in hindguts of Ips lypgraphus males from different spruce trees and attack phases are reviewed. The cotnposition of monoterpenes. and the chiratity of a-pinene, have been determined in phloem samples. Relationships between compounds emanating from spruce trees and bark beetles, respectively, have been studied. Male beetles depend on their host tree for the production of pinene alcohols. The ratio between the pinene alcohols is almost constant in males boring in the same tree but can vary widely between males from different spruce trees. Very good correlations were found between some host tree monoterpene hydrocarbons and bark beetle produced pinene alcohols. The production of the essential pheromone component 2-methyl-3-buten-2-ol, was not correlated with the monoterpene content in the host trees, while the production of the other essential pheromone component, ri5-verbenol. depends on the amount of the precursor, (-)-a-pinene in the phloem. Male beetles boring in a resistant spruce tree will continue to produce the pinene alcohols, including cK-verbenol, as long as the tree defends itself with resin. G. Birgersson, Dept of Chemical Ecology, Goteborg University, Box 33031, S-40033 Goteborg, Sweden. Present address: Dept of Entomology, Univ. of Georgia, Athens, GA 30602, USA.
Introduiiiun When pioneering bark beetles attack and bore into a host tree, they first meet the primary resin, always present in the inner bark and phloem. This damage to the tree initiates the next part of the defensive mechanism: prodtiction of more resin. The resin contains compounds toxic to attacking bark beetles, such as phenols and terpenes (Cater and Alexander 1982). In addition, it is sticky, and these qualities together hinder the beetles and may even kill them. As a response to the attack, the amounts of monoterpene hydrocarbons increase and are tenfold larger in the phloem surrounding the galleries than in unattacked areas (Leufven and Birgersson 1987). However, the production of resin requires a lot of energy. The amount of stored starch and sugars gives a measure of a tree's vigour, and its ability to produce resin for its defence (Wright et al. 1979). A resistant tree has the capacity to produce more resin than a susceptible tree, and is also able to produce resin for a longer time. © 25'
There are large variations between individual Norway spruce Picea abies trees with regard to the composition of terpenes in their resin (Esteban et al. 1976). Not only do the ratios between the various mono- and sesquiterpenes vary between different trees, but also the enantiomeric composition of single compounds (Lindstrom et al. 1989). Bark beetles boring in different spruce trees will be exposed to different ratios of terpenes. especially between a- and (^-pinene, as they are the quantitatively dominant monoterpenes (Heeman and Francke 1977). Depending on whieh spruce tree a single bark beetle is boring into, it will be exposed to a terpene mixture specific to its host tree. In order to survive, the beetle fighting the resin has to detoxify the terpenes. This process proceeds through hydroxylation (White et al. 1980). It has been shown in laboratory experiments that male Ips typographus are capable of stereospecific hydroxylation of u-pinene (Klimetzek and Francke 1980). Exposed to different ratios of (+)- and (-)-isomers of a-pinene. the males produce trans- and cis-verbenol in corresponding quotients.
HOLARCTIC ECOLOGY HOLARCTIC ECOLOGY 12:4 (1989)
451
rel. to cV*tV 80
Tree n=
A 12:154
B 6:72
D
F 12:210
G 3:33 tOOn
50-
0,99 0,98 1,02 1.03
1.03 l.Ot l.Oi
1.02 0.98 0.99 l.Oi
Tree
I
II
IV
n=
69
A7
T20
x/median
099 0.98 1.09 0.99
i
1.01 1.01 0.99 1.06
VI 61
50
Fig. 1. Differences in relative amounts of pinene alcohols between male Ips typographus collected from different spruce trees. The amounts of m-verbenol (open bars), (/-am-verbenol (hatched bars), myrtenol (cross-hatched bars) and ;ru/i.v-myrtanol (filled bars) are related to the sum of m- and rran^-verbenol. Males from trees A. B. D. FandG were analysed batchwise. males from trees I. II. IV, V and VI. individually, n = numbers of batches: individuals, and individuals, respeetively, analysed from each tree. Ratios of means to medians were elose to 1.0 indicating unskewed distributions. (Data reealculated from Birgersson et al 1984 and Birgersson et al. 19S8.)
Fourteen volatiles. including eleven oxygenated monoterpenes, have been identified in hindguts from male /. typographus boring in spruce phloem (Birgersson el al. 1984). Of these compounds. 2-methyl-3-buten-2-ol (MB) and cts-verbenol (cV) attract both males and females of this species (Bakke ct ai. 1977), and are the only essential compounds in the aggregation pheromone (Schlyter et al. 1987a). These two compounds have also been found to release different behavioural steps in the attraction (Schlyter et al. 1987b). The questions addressed in this paper are: To what degree are the beetles innately "programmed" in their pheromone production, and to what degree does the environment to which the beetles are exposed during the attack in the tree contribute to variation in production? To answer these questions, data from beetles collected both early and late in their attacks and from several spruce trees, will be reviewed.
452
Materials and methods The beetles used for the two studies reviewed in this paper were collected during the main swarming period, in late May and early June 1981 and 1982, in the province of Varmland. one of the largest outbreak areas in Sweden (Loyttyniemi et al. 1979). The population was declining, after having reached a maximum a few years earlier, resulting in a low number of spontaneously attacked trees in the area. As we wanted to study beetles attacking and boring in standing trees, some spruce trees were baited with the commercial pheromone attractant for /. typographus, IPSLURE®, to initiate and concentrate the attacks. Bark beetles boring in Norway spruce were cut out and the attack phase scored, as described by Birgersson et al. (1984). All beetles were kept in liquid nitrogen until dissection and analyses. Two series of analyses were performed. In the first, beetles from five different standing trees and two logs, and from different attack phases, were analysed in groups of, on average, 15 beetles per extract (Birgersson et al. 1984). Male beeHOLARCTIC ECOLOGY
Tab. I. Correlations between amounts of monoterpene hydrocarbons and phcromone components as quantified in 18 hindnut batch samples of male Ips typographus (data from Birgersson et at. 1984). Monoterpene hydrtKiarbons
Pheromone components
ii-pinene
cis-verbenol /rans-verbenol myrtanol trans-myrtanol 2-methyl-3-buten-2-ol ctr-verbenol rran.v-myrtanol
p-pinene
r 0.97 0.98 0.73 0.70 0.22 0.75 0.99
lies in two different attack phases, i.e. with nuptial chambers or with females, which had 2-4 cm-long egg galleries, from four standing and two windthrown trees, were analysed individually in a second series (Birgersson et al. 1988). Phloem samples from unattacked areas on the trees in the second series were collected and stored in liquid nitrogen until extraction and analyses. Hindguts from male beetles were extracted and analysed by GC-MS, and the amount of volatiles were quantified; for conditions, see Birgersson et al. (1984, 1988). The phloem samples were cut in mm-sized pieces and extracted in pentane for analyses of monoterpene composition and the chirality of a-pinene, as described in detail by Lindstrom et al. (1989). Statistical analyses were made using the programs SPSS (Nie et al. 1975) and MINITAB (®Penn. State Univ.).
Results Monoterpene alcohol profiles Males of Ips typographus boring in a specific spruce tree contain the same relative amounts of pinene alcohols in their hindguts. These ratios are independent of the attack phase of the beetles and the absolute amounts of the compounds. When comparing males attacking different spruce trees, the ratios are more or less different. In Fig. I. the relative amounts of cV, /ram-verbenol (tV), myrtenol (Mt), and trans-myrtartol (tM), relative to the sum of cV plus tV, from batchwise or individually analysed male beetles collected from ten different trees, arc shown. On the basis of the ratios between the pinene alcohols, it is possible to identify the host tree from the analysis of a single male beetle. This can be done manually or using principal component analysis (Birgersson et al. 1988). The relative amounts of pinene alcohols, in males from the same host tree, arc almost normally distributed, with a low skewness, indicated by the fact, that the ratios between means and medians are close to 1 (0.98-1.09). HOLARCTIC ECOLOGY 12:4 (198'»)
Correlations between host monoterpenes and pheromone components Batch samples To investigate the correlation between host- and beetleproduced terpenic compounds, the amounts of a- and p-pinene and their oxygenated derivatives were studied in beetles analysed batchwise. The monoterpene hydrocarbons were only present in quantifiable amounts in the hindguts of males collected in the earlier attack phases, i.e. males boring in the phloem and males with females, before egglaying. The amounts of both cV and tV were strongly correlated with the amounts of apinene, but not with p-pinene (Tab. 1) while the amount of tM was correlated with P-pinene. More than 90% of the variation in the amounts of these compounds was explained by the amounts of hosts tree compounds, with the exception of Mt (54% of the variance). The variation in the other pheromone component, MB, was not at all explained by the variation in monoterpene hydrocarbons. Individuals In the second series, when males were analysed individually, the amounts of spruce monoterpenes in the hindguts were mostly too low for reliable quantification. Instead, phloem samples from unattacked areas of the host trees were analysed for their content of monoterpene hydrocarbons. The enantiomeric composition of a-pinene, in each tree, was also determined. Regressions were made between the relative amounts of the host monoterpenes, and their relative amounts of the pinene alcohols, in order to indicate biosyntbetical Tab. 2. Correlations between the relative amounts of monoterpene hydrocarbons in Picea abies pbloem and pinene alcohols in male [ps lypographus hindguts. Host terpene in phloem
Pinene alcohol in Ips typographus
(-)aP (-)aP, pp (-)aP, pp, 3C •1- M, pPh (+)aP ( + )aP! pp 3C -+M, pPh (-)aP ( + )aP
pp
(-)aP pp
{-)aP - a-pjnene {-l-)aP = (-H)a-pinene pp = p-pinene 3C = 3-carene M = myrcene pPh = p-phellandrene All monoterpene hydrocarbons In amounts relative to the sum of ( + )- and (-)apinene
cV eV
cV tv tV tv Mt Mt Mt tM tM
cV tV Mt tM
r
0.93 0.93 1.00 0.93 0.93 1.00 -0.16 0.16 -0.16 0.50 0.91 = cis-verbenol = /ra/ts-verbenol = myrtenol = f ram-myrtanol
All pinene alcohols in relative amounts to sum of cis- and fram-verbcnol 453
ng/mal*
beetles. In the second study (Birgersson et al. 1988), male beetles in one early and one late attack phase were Ms
