Specificity and redundancy in the olfactory system of the bark beetle Ips typographus: Single-cell responses to ecologically relevant odors

Journal of Insect Physiology 55 (2009) 556–567

Contents lists available at ScienceDirect

Journal of Insect Physiology journal homepage: www.elsevier.com/locate/jinsphys

Specificity and redundancy in the olfactory system of the bark beetle Ips typographus: Single-cell responses to ecologically relevant odors Martin N. Andersson *, Mattias C. Larsson, Fredrik Schlyter Chemical Ecology, Department of Plant Protection Biology, Swedish University of Agricultural Sciences, SE-230 53 Alnarp, Sweden

A R T I C L E I N F O

A B S T R A C T

Article history: Received 9 December 2008 Received in revised form 29 January 2009 Accepted 30 January 2009

We screened 150 olfactory sensilla in single-sensillum recordings to unravel the mechanisms underlying host selection in the spruce bark beetle, Ips typographus (Coleoptera: Curculionidae: Scolytinae). Odor stimuli comprised of pheromone (various bark beetle spp.), host, and non-host compounds elicited strong and selective responses from 106 olfactory receptor neurons (ORNs), 45 of which were tuned to pheromone compounds, 37 to host compounds, and 24 to non-host volatiles (NHV). In addition, 26 ORNs responded only weakly to any odor stimulus. Strongly responding ORNs were classified into 17 classes. Seven classes responded primarily to the Ips pheromone compounds: cis-verbenol, ipsenol, ipsdienol (two classes), 2-methyl-3-buten-2-ol, amitinol, or verbenone, respectively. Six classes responded to the host compounds: a-pinene, myrcene, p-cymene, myrcene and p-cymene, 1,8-cineole, or D3-carene, respectively. Four classes responded to NHV: 3-octanol, 1-octen-3-ol, trans-conophthorin, or indiscriminately to the repellent green leaf volatiles (GLVs) 1-hexanol, Z3-hexen-1-ol and E2-hexen1-ol, respectively. Indiscriminate responses from GLV neurons might explain a behavioral redundancy among these GLVs. This is the first description of individual bark beetle ORNs dedicated to NHV perception. These comprise almost 25% of the strongly responding neurons, demonstrating that a large proportion of the olfactory system is devoted to signals from plants that the insect avoids. ß 2009 Elsevier Ltd. All rights reserved.

Keywords: Single-sensillum recordings Odor coding Non-host volatiles Olfactory receptor neuron Host selection

1. Introduction The European spruce bark beetle, Ips typographus (Coleoptera: Curculionidae: Scolytinae) attacks and kills Norway spruce, Picea abies, in Europe. This has stimulated a large amount of research concerning its olfactory-mediated host selection behavior. Attraction to host trees is governed by secondary attraction of both sexes to the male-produced aggregation pheromone that consists of (4S)-()-cis-verbenol and 2-methyl-3-buten-2-ol (Schlyter et al., 1987a). Whether the pioneering males locate host trees at random or if primary attraction (in the absence of pheromone) to host odors (kairomone) also occurs has not been clearly demonstrated, but spruce volatiles, particularly in large amounts, seem to improve the attraction to the pheromone bait (Jakus and Blazenec, 2003; Hulcr et al., 2006; Erbilgin et al., 2007). The Ips-associated compounds, verbenone, E-myrcenol, ipsenol and high doses of ipsdienol, as well as the pheromone components of the sympatric

* Corresponding author at: Chemical Ecology, Department of Plant Protection Biology, Swedish University of Agricultural Sciences, PO Box 102, SE-230 53 Alnarp, Sweden. Tel.: +46 40415268; fax: +46 40461991. E-mail address: [email protected] (M.N. Andersson). 0022-1910/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jinsphys.2009.01.018

Pityogenes chalcographus, (1S,5R)-chalcogran and methyl (E,Z)-2,4decadienoate, inhibit the attraction of I. typographus (Schlyter et al., 1989, 1992; Byers, 1993) and likely mediate avoidance of intraand interspecific competition. Behavioral responses are also modulated by non-host volatiles (NHV) that originate from angiosperm trees, and anti-attractant effects of various NHV have been demonstrated in more than 20 bark beetle species (reviewed by Zhang and Schlyter, 2004). For the spruce bark beetle, the green leaf volatiles (GLVs) 1-hexanol, E2-hexen-1-ol and Z3-hexen-1-ol have inhibitory properties (Zhang et al., 1999). In addition, 3-octanol, 1-octen-3-ol, and the spiroacetal (5S,7S)-trans-conophthorin (Zhang et al., 2002), isolated from bark of angiosperm trees (Betula and Populus spp.), inhibit the attraction to the pheromone (Zhang et al., 2000). The chemical ecology of I. typographus is complex, with behavioral responses to binary and ternary blends of inhibitory chemicals involving both redundancy and synergism (Zhang and Schlyter, 2003). Redundancy occurs among and between the GLV alcohols and the C8-alcohols, whereas trans-conophthorin and verbenone synergize the effect of the alcohols. Previous studies identified olfactory receptor neurons (ORNs) specialized to several bark beetle pheromone compounds and a few host monoterpenes (Mustaparta et al., 1984; Tømmera˚s et al.,

Table 1 Classifications, origins, and behavioral effects of the chemicals used in the study. Abbreviation

Chemical source

Purity %

Chemical class

Biological class

Behavioral effect

Ecological source

Refs

()-cis-Verbenol ()-trans-Verbenol ()-Verbenone ()-Ipsenol ()-Ipsdienol Amitinol E-Myrcenol 2-Methyl-3-buten-2-ol ()-Chalcogran ()-exo-Brevicomin

cV tV Vn Ie Id Am EM MB Cg xB

Borregaard SciTech Ltd., Prague Fluka Bedoukian Bedoukian W.F.a SciTech Ltd., Prague Acros Celamerck W.F.

95 91 >99 97 94 96 94 >99 90 99

MT–OH MT–OH MT O MT–OH MT–OH MT–OH MT–OH HT–OH C9–spiroacetal C9–diacetal

Ph Ph Ph Ph Ph Ph Ph Ph Ph Ph

++ ? – – –+ ? – ++ – +

I. typographus aggr. Ph Tomicus spp. Ph I. typographus anti-aggr. Ph I. typographus anti-aggr. Ph Ips spp. Ph Ips spp. Ph Ips spp. Ph I. typographus aggr. Ph P. chalcographus Ph Dendroctonus spp. Ph

Methyl (E,Z)-2,4-decadienoate (+)-a-Pinene ()-a-Pinene ()-b-Pinene Myrcene p-Cymenea (+)-Limonene ()-Limonene ()-D3-Carene ()-1,8-Cineole Terpinolene ()-Bornyl acetate 1-Hexanol E2-Hexen-1-ol Z3-Hexen-1-ol Hexanal E2-Hexenal Z3-Hexenyl acetate ()-3-Octanol (+)-3-Octanol

MD (+)aP ()aP bP My pC (+)L ()L D3 Ci Te BoAc C6 E2C6 Z3C6 C6Ald E2C6Ald Z3C6Ac C8an (+)C8an

Bedoukian Janssens chimica Fluka Fluka Fluka Acros Fluka Fluka Aldrich Aldrich Fluka Acros Fluka Acros Acros Aldrich Aldrich Sigma Acros Aldrich

92 98 >99 92 95 >99 >99 98 93 >99 97 >99 >99 93 98 98 95 99 >99 98

C10–O–Me MT MT MT MT MT MT MT MT MT–acetal MT MT–OAc C6–OH C6–OH C6–OH C6 O C6 O C6–OAc C8–OH C8–OH

Ph Host Host Host Host Host Host Host Host Host Host Host NHV NHV NHV NHV NHV NHV NHV NHV

– ? + ? ?

()-1-Octen-3-ol ()-1-Octen-3-ol

C8en ()C8en

Acros Acros

98 98 (99)

C8–OH C8–OH

NHV NHV

– ?

5S,7S-trans-Conophthorin

SStC

W.F.

92 (97)

C9–spiroacetal

NHV

–

5R,7R-trans-Conophthorin ()-Linalool ()-Linalool

RRtC Lol ()Lol

W.F. Fluka Fluka

74 (91) 92 (97) 94 (>98)

C9–spiroacetal MT–OH MT–OH

NHV NHV NHV

? 0 ?

P. chalcographus Ph Spruce Spruce. Attractive with Ph Spruce Spruce Spruce Spruce Spruce Pine > spruce Spruce Pine > spruce Spruce GLV GLV GLV GLV GLV GLV Non-host bark Enantiomeric composition in bark unknown Non-host bark Enantiomeric composition in bark unknown Non-host bark, predominant enantiomer Non-host bark Many plants Many plants

Schlyter and Birgersson (1999) Schlyter and Birgersson (1999) Schlyter and Birgersson (1999), Schlyter et al. (1989) Schlyter and Birgersson (1999), Schlyter et al. (1989) Schlyter and Birgersson (1999), Schlyter et al. (1989) Zhang et al. (2007) Schlyter and Birgersson (1999), Schlyter et al. (1992) Schlyter and Birgersson (1999) Schlyter and Birgersson (1999), Byers (1993) Schlyter and Birgersson (1999), Tømmera˚s and Mustaparta (1984) Schlyter and Birgersson (1999), Byers (1993) Persson et al. (1996) Persson et al. (1996), Erbilgin et al. (2007) Persson et al. (1996) Persson et al. (1996) Persson et al. (1996), Schlyter (unpublished data) Persson et al. (1996) Persson et al. (1996) Persson et al. (1996), Wibe et al. (1998) Schlyter (unpublished data), Wajs et al. (2006) Persson et al. (1996) Wajs et al. (2006) Zhang and Schlyter (2004) Zhang and Schlyter (2004) Zhang and Schlyter (2004) Zhang and Schlyter (2004) Zhang and Schlyter (2004) Zhang and Schlyter (2004) Zhang and Schlyter (2004)

(95)b (95) (95) (97) (93)

(>99)

(>99) (95)

(97) (96)

(98)

(97)

? ? ? – ? ? – – – 0 0 0 – ?

Note: Compounds classified as ‘host’ can typically also be found in plants other than spruce (mainly conifers). MT: monoterpene (10C), HT: hemiterpene (5C), –OH: alcohol, ester, Ph: pheromone (i.e. bark beetle-produced), NHV: non-host volatile, GLV: green leaf volatile. a Wittko Francke, University of Hamburg, Hamburg, Germany. b All compound purities were analysed with GC–MS. Parenthesis indicates purity specified on vial if different from our analysis.

Zhang and Schlyter (2004)

M.N. Andersson et al. / Journal of Insect Physiology 55 (2009) 556–567

Chemical

Zhang and Schlyter (2004), Zhang et al. (2002) Zhang and Schlyter (2004), Zhang et al. (2002) Zhang and Schlyter (2004) Zhang and Schlyter (2004) O: aldehyde and ketone, –Me: methyl, –OAc: acetate

557

558

M.N. Andersson et al. / Journal of Insect Physiology 55 (2009) 556–567

1984; Tømmera˚s, 1985). Responses to spruce and birch volatiles have also been obtained without further identification of the active compounds (Tømmera˚s and Mustaparta, 1987). Similar studies were performed on other Scolytids, e.g. I. pini and I. paraconfusus (Mustaparta et al., 1977, 1979, 1980), Dendroctonus pseudotsugae (Dickens et al., 1984, 1985), D. frontalis (Payne et al., 1982), D. micans (Tømmera˚s et al., 1984), Trypodendron lineatum (Tømmera˚s and Mustaparta, 1989), Scolytus scolytus (Wadhams et al., 1982), and Tomicus destruens (Guerrero et al., 1997). Since these studies, our knowledge about bark beetle chemical ecology has increased, especially with regards to repulsive NHV. Therefore, we conducted an extensive SSR study on I. typographus to investigate the peripheral coding of bark beetle pheromones and odors associated with host (conifer) and non-host (angiosperm) plants. We primarily included compounds with known behavioral effects (Table 1), but since only a few host odors have been conclusively shown to influence the behavior of I. typographus, we also included major spruce monoterpenes (Persson et al., 1996; Wajs et al., 2006). We investigated the relative proportions of ORNs that respond to pheromone (i.e. con- and heterospecific bark beetle-produced compounds), host (conifer), and non-host (angiosperm) compounds respectively, using comparable numbers of compounds from these three semiochemical classes. We also investigated if the redundancy and synergism in behavioral responses among the behaviorally inhibitory odors are reflected in the response specificity of the ORNs and tested the hypothesis that redundant compounds might be perceived by the same ORNs, while synergists are likely to excite different sets of receptor neurons (Zhang and Schlyter, 2003). We identified 17 ORN classes based on strong responses. In addition to pheromone responding ORNs, a large number of ORNs responded to spruce-associated compounds, suggesting that host odors are important in the host selection process. We also found that almost one-fourth of the strongly responding neurons responded to NHV, which strengthens the notion that perception of angiosperm volatiles is important for coniferfeeding insects. Finally, we show that most ORNs are very selective in their responses and that response specificity of individual neuron classes can explain much of the behavioral redundancy and synergism observed among repellent non-host volatiles.

2. Materials and methods 2.1. Insects Insects were from a laboratory-reared colony, established in 2005 from individuals collected in their natural environment outside Asa and A¨lmhult, southern Sweden. Beetles were reared on spruce logs in an environmental chamber, simulating Swedish summer conditions. Emerged adults were kept in a state of low activity in a refrigerator (3  2 8C). Individuals of the fourth to sixth generations were used, and they were approximately 2 weeks old when tested. Sex separation by external morphology (Schlyter and Cederholm, 1981) was confirmed by genitalia dissection after recordings. Both sexes were used in recordings. 2.2. Chemicals and stimuli Synthetic compounds were obtained from various commercial and non-commercial sources. Chemical purities were confirmed by means of coupled gas chromatography–mass spectrometry (GC– MS) and deviations from labeled values are noted (Table 1). All 36 compounds were diluted in paraffin oil (product # 1.07162.1000, MERCK, Darmstadt, Germany). The cis-verbenol did not dissolve directly in paraffin and was thus first diluted in tetrahydrofuran (THF) (>99%, Aldrich) and subsequently in paraffin, leaving 1% THF in the paraffin solution. Neurons that responded to cis-verbenol never responded to THF controls. Odors were applied on filter papers (1.5 cm  0.5 cm) inside Pasteur pipettes that were capped with plastic pipette tips (1000 ml). Stimulus pipettes were stored at 18 8C between experiments. Screening pipettes were used during maximum two consecutive days, whereas dose–response stimuli were prepared daily. During screening, the amount of applied compound was 10 mg (i.e. 10 ml solution with 1 mg/ml concentration), while during dose–response trials amounts between 100 pg and 10 mg were tested in decadic steps. 2.3. Single-sensillum recordings Insect preparations: a live I. typographus adult was fixed with dental wax inside a cut section (ca. 1 cm) of a plastic pipette tip (200 ml), leaving the head and part of the pronotum protruding. The insect-containing pipette tip was secured in dental wax on a

Fig. 1. (A) The ventral side of the terminal segment of the Ips typographus antenna. Olfactory sensilla are present in two undulating bands (A and B) and in a distal area (C) (Hallberg, 1982b). Single-walled sensilla type 1 predominate in A and B, and single-walled sensilla type 2 are mainly found in the distal area C. (B) Approximate positions of four ORN classes, representative of the two general distribution patterns. Large red circles: cis-verbenol (cV) sensitive ORNs, small blue circles: 1,8-cineole neurons colocalized with cV neurons, yellow triangles: myrcene cells (one cell found in area C), and green squares: GLV-OH cells. Scale bar: 50 mm. Micrograph provided by Hallberg.

M.N. Andersson et al. / Journal of Insect Physiology 55 (2009) 556–567

microscope slide (76 mm  26 mm). An antenna was fixed in a thin wax layer (ca. 0.5 mm) on a coverslip (18 mm  18 mm), allowing light from below to penetrate the antenna during the recordings that were performed under a Nikon light microscope (FN-S2N) with 500 magnification. Single-sensillum recordings with tungsten microelectrodes were performed using standard equipment and established experimental procedures (Stensmyr et al., 2003; Ghaninia et al., 2007). The indifferent electrode was inserted through a pre-made hole in the pronotum and the recording electrode into the base of a sensillum. The antenna was continuously exposed to a charcoalfiltered and humidified airflow at 1.2 l/min. The flow was controlled by a stimulus controller (CS-01, Syntech, Kirchzarten, Germany) and passed through a glass tube (7 mm i.d.) that terminated approximately 15 mm from the antenna. During odor stimulation, a 0.5 s air-pulse (0.2 l/min) was passed through the stimulus pipette that was inserted into the continuous airflow through a hole in the glass tube. During odor stimulation, the continuous airflow was decreased by the same amount as the airpulse, resulting in a constant total flow over the antenna. The sensilla of I. typographus are almost exclusively found on the ventral side of the terminal flattened antennal club segment, and are mainly positioned in two undulating bands as well as in a distal area (Fig. 1A) (Hallberg, 1982b). Two morphological types of olfactory single-walled sensilla, and one double-walled sensillumtype thought to be involved in chemo- or thermoreception have been described (Hallberg, 1982b). The single-walled sensilla are termed single-walled sensillum type 1 (SW 1) and type 2 (SW 2), which by the use of different nomenclature correspond to sensilla basiconica and trichodea, respectively (Borden and Wood, 1966). The SW 1 sensillum is located in all areas but is predominantly found in the two undulating bands. It normally contains two ORNs but occasionally one or three. SW 2, on the other hand, is mainly found in the distal area and is innervated by one or two olfactory neurons (Hallberg, 1982b). A total of 150 randomly chosen sensilla from 22 males and 16 females were screened for ORN responses to the odor panel. The positions of the sensilla were mapped in order to investigate the spatial distribution of ORNs. Recordings were taken from sensilla distributed all over the terminal segment in order to cover as many locations as possible. Odors were presented to the insect in random order. Based upon response spectra, ORNs were divided into classes. Characterization of response specificity was undertaken without adjusting for differences in compound volatility. Ten classes were subjected to dose–response trials, which were done separately from the screening. In these, odors were presented in increasing doses, with weakly exciting odors tested before odors that elicited stronger responses. 2.4. Data analyses Net responses were calculated using Autospike 3.0 (Syntech) by counting the number of spikes (action potentials) during the first 0.5 s of the response and subtracting the spontaneous spike activity during a 0.5-s pre-stimulation period. The value was multiplied by 2 in order to obtain a response in spikes/s (Hz). If the paraffin oil blank elicited response in a neuron, the odor responses were adjusted by subtracting the net blank response. At 10 mg dose, responses below 20 Hz were regarded as ‘no response’. Excitatory responses were classified as: (+) 20–39 Hz, (++) 40– 69 Hz, (+++) 70–99 Hz, (++++) 100–129 Hz, and (+++++) 130 Hz. Because of overall low (typically 2–8 Hz), and often slightly irregular spontaneous activities, inhibitory responses could not be reliably quantified, and were thus not utilized for ORN characterization. The frequencies of receptor neurons that primarily responded to any of the host, non-host, or pheromone compounds, were analyzed with x2 tests, using SPSS software (v. 11.0). To

559

calculate standardized effect sizes, Crame´r’s phi (wc) (Zar, 1999) was used according to sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

’c ¼

x2

Nðk  1Þ

where N is the total number of observations and k is the smaller of the number of rows or columns. wc ranges from 0 (no association) to 1 (perfect association), and is independent of sample size. 3. Results 3.1. General response characteristics Within a sensillum, we typically observed two ORNs that could be distinguished based on spike amplitudes (Fig. 2A), but in some sensilla one or three amplitudes were discerned. A specific odor typically elicited a response in only one of the neurons, normally the cell with the largest amplitude (from hereon referred to as the A neuron; the second largest amplitude corresponding to the B neuron). Most responding ORNs responded with increased action potential firing to odors in our test panel. The ORNs responded in a typical phasic–tonic manner, exhibiting the highest firing activity during the initial phase of the response, and then declining firing until resting activity was restored. However, some responses were more tonic (Fig. 2A), with spike activity well above resting level for up to several minutes after odor stimulation, whereas others were more phasic (Fig. 2B), lasting for only a few seconds. The variation was observed both between cells and between different compounds tested on the same cell. Generally, more extreme tonic responses were observed at high doses. A given ORN typically responded to only a few structurally similar odors even at the highest dose (10 mg) (Table 2). High selectivity was found, not only among pheromone responding neurons, but also among those that were tuned to host or non-host compounds. The odor that evoked the strongest response at the high dose was normally also the one for which the ORN had the lowest response threshold. The strongest response in a neuron typically ranged between 120 and 160 Hz, although some responses above 200 Hz were recorded. 3.2. Frequencies of responding ORNs From the randomized screening of 150 sensilla (<: 78, ,: 72), we obtained strong responses from 106 ORNs (<: 53, ,: 53) in 99 sensilla (<: 50, ,: 49) (Table 2). The powerful responses of these ORNs suggest that the key ligands for the receptors likely were included in the odor panel. In addition, 26 ORNs (<: 13, ,: 13), housed in 22 sensilla (<: 11, ,: 11), were classified based on weak or intermediate responses. Twenty-three sensilla (<: 13, ,: 10) contained neurons that did not respond by excitation to our odor panel. Three of these housed neurons that exhibited long-lasting inhibitions (Fig. 2C). These neurons varied in their target inhibitors (data not shown); however spike activity of all the three was completely suppressed by verbenone, both verbenols, and chalcogran for 1 to above 3 s. Otherwise, clear inhibitions sometimes occurred in A cells when the co-localized B cells responded (see below) (Fig. 2D). Receptor neurons in as few as 6 sensilla (<: 4, ,: 2) could not be classified due to suboptimal signal quality. Among the 106 strongly responding neurons, we identified 17 ORN classes based on response spectra (Table 2). When comparing the number of receptor neurons that responded to pheromone, host, or non-host compounds and considering only the strongly responding ORNs, we found that the three groups are of similar sizes (Fig. 3). However, a marginally significant difference was present (x2 = 6.4, d.f. = 2, P = 0.04), but the effect size was low (wc = 0.17). There was no difference between males

560

M.N. Andersson et al. / Journal of Insect Physiology 55 (2009) 556–567

Fig. 2. Response characteristics exemplified by four sensillum types stimulated with 10 mg. Excitatory responses ranged from highly tonic to highly phasic. (A) cis-Verbenol elicited a tonic response in the cis-verbenol cell. (B) a-Pinene elicited a phasic response in the a-pinene cell. (C) Three ORNs were clearly inhibited by some odorants and never responded by excitation. This neuron was inhibited for 3 s upon stimulation with cis-verbenol. (D) Inhibitions also occurred in A neurons when co-compartmentalized B neurons responded by excitation, exemplified by a sensillum housing the methyl-butenol B neuron. Horizontal bars indicate the 0.5-s stimulation period.

and females in the number of cells that responded to pheromone, host, or non-host compounds (x2 = 2.0, d.f. = 2, P > 0.05; wc = 0.14). The 26 weakly responding neurons found during screening were divided into 11 ORN classes, and a 12th class was identified during dose–response tests. 3.3. ORNs responding to pheromone compounds

Fig. 3. Frequencies of ORNs tuned to compounds of different ecological origins, i.e. pheromones of various bark beetle species, compounds produced by host (conifer), or non-host (angiosperm) plants. Male and female frequencies do not differ and the different origins differ marginally (see text).

During screening, 45 (<: 25, ,: 20) receptor neurons were found that were tuned to bark beetle pheromone compounds. Among these, 22 ORNs responded primarily to ()-cis-verbenol (Table 2, Fig. 4A). Secondary responses in this neuron type were recorded to ()-trans-verbenol and ()-verbenone, but only at high doses. The response threshold for cis-verbenol, on the other hand, was at approximately 1 ng. Seven of the cis-verbenol neurons were co-compartmentalized with B neurons that responded primarily to 1,8-cineole (described below). This was the only case of co-occurring, strongly responding A and B neurons characterized in our study. When the B cell responded to high doses (1–10 mg) of 1,8-cineole, the spike activity of the cisverbenol cell was reduced. Three ORNs responded primarily to 2-methyl-3-buten-2-ol, the other major component of the aggregation pheromone of I. typographus. In this case it was the B neuron that was excited,

M.N. Andersson et al. / Journal of Insect Physiology 55 (2009) 556–567

561

Table 2 Response spectra of the 17 strongly responding ORN classes to the 10 mg stimulus dose. Compound

ORN class cV

Ie

IdA

IdB

MB

Am

Vn

aP

My

My/pCa

pC

Ci

D3

C8an

C8en

GLV–OH

tC

()-cis-Verbenol ()-trans-Verbenol ()-Verbenone ()-Ipsenol ()-Ipsdienol amitinol E-Myrcenol 2-Methyl-3-buten-2-ol ()-Chalcogran ()-exo-Brevicomin Methyl 2,4-decadienoate (+)-a-Pinene ()-a-Pinene ()-b-Pinene myrcene p-Cymene (+)-Limonene ()-Limonene ()-D3-Carene ()-1,8-Cineole Terpinolene ()-Bornyl acetate 1-Hexanol E2-Hexen-1-ol Z3-Hexen-1-ol Hexanal E2-Hexenal Z3-Hexenyl acetate ()-3-Octanol (+)-3-Octanol ()-1-Octen-3-ol ()-1-Octen-3-ol S,S-trans-Conophthorin R,R-trans-Conophthorin ()-Linalool ()-Linalool

+++++ + ++ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 +++++ ++ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 +++ +++++ +++ + 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 + +++++ + 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 + 0 ++++ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 ++++ +++++ + 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 ++ ++++ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

+++ ++ 0 0 0 0 0 0 0 ++ 0 +++++ +++++ ++++ 0 0 0 0 + + 0 0 0 0 0 0 0 0 0 0 0 0 ++ + 0 0

0 0 0 0 + 0 ++ 0 0 0 0 0 0 0 +++++ 0 ++ + ++ 0 +++ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 ++ 0 + 0 0 0 0 + 0 0 +++++ +++++ ++ ++ ++ + +++ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 + 0 0 0 +++++ ++ ++ ++ + + 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 ++ + 0 0 0 0 0 0 0 0 0 ++++ 0 0 0 0 0 0 0 0 0 0 0 0 + ++ 0 0

0 0 0 0 0 0 0 0 + 0 0 + 0 0 0 0 0 0 +++++ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 + 0 0

0 0 0 0 0 0 0 0 + 0 0 0 0 0 0 0 0 0 0 0 0 0 +++ ++ ++++ 0 0 0 +++++ ++++ +++++ +++ 0 0 + +

0 0 0 0 0 0 0 0 + 0 0 0 0 0 0 0 0 0 0 0 0 0 + 0 + 0 0 0 +++++ +++++ +++++ +++++ 0 0 0 0

0 0 0 0 0 0 0 0 ++++ ++ 0 0 0 0 0 0 0 0 0 0 0 0 +++++ +++++ +++++ ++++ +++ + + ++ +++ + 0 0 0 0

0 0 0 0 0 0 0 0 ++++ ++++ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ++++ +++ 0 0

Neuron

A

A

A

A

B

A

A

A

A

A

A

B

A

A

A

A

A

Antennal distribution

C (B)

AB

AB

B

BC

AB

AB

C (B)

A B (C)

B

AB

C (B)

AB

AB

AB

AB

AB

0 2 2

2 0 2

1 2 3

3 1 4

2 0 2

4 7 11

5 6 11

1 0 1

0 3 3

3 4 7

2 2 4

6 4 10

1 0 1

4 3 7

2 4 6

No. of cells found during screening Male 11 6 Female 11 4 Total 22 10

Note: For ORN classes that were subjected to dose–response tests, average responses are calculated from these cells due to overall higher sample sizes (for n-values, see Fig. 4). Responses in other cell types derive from cells found during screening. (+) 20–39 Hz, (++) 40–69 Hz, (+++) 70–99 Hz, (++++) 100–129 Hz, (+++++) 130 Hz. Abbreviations of ORN classes according to their primary response: cV: cis-verbenol, Ie: ipsenol, IdA and IdB: ipsdienol classes A and B, MB: 2-methyl-3-buten-2-ol, Am: amitinol, Vn: verbenone, aP: a-pinene, My: myrcene, My/pC: myrcene and p-cymene, pC: p-cymene, Ci: 1,8-cineole, D3: D3-carene, GLV–OH: green leaf volatile alcohols, C8an: 3-octanol, C8en: 1-octen-3-ol, and tC: trans-conophthorin. a Responses calculated from one dose–response cell and one screening cell.

while at the same time, the A neuron was inhibited. This A cell never responded by excitation to our odor panel. Ten cells responded primarily to ipsenol and secondarily to ipsdienol. Four cells responded the strongest to ipsdienol. Differences in the magnitude of the secondary responses to ipsenol, amitinol and E-myrcenol, suggested that the four ipsdienol cells should be divided into two ORN classes, Id A and Id B (Table 2). Four other ORNs were strongly excited by amitinol. In these, ipsdienol also elicited a fairly powerful response. The remaining two pheromone-responding ORNs responded primarily to the old-host anti-aggregation compound, ()-verbenone. 3.4. ORNs responding to host odors 37 ORNs (<: 15, ,: 22) responded strongly to odors associated with the host. Eleven cells responded primarily to a-pinene followed by b-pinene and the pheromone component cis-verbenol (a few compounds elicited even weaker responses) (Fig. 4B). The

(+)-enantiomer of a-pinene evoked a somewhat stronger response than the ()-enantiomer at all doses, but both compounds elicited responses classified as (+++++) at 10 mg. The response threshold for both enantiomers was between 1 and 10 ng. Eleven cells were found that primarily responded to myrcene (Fig. 4C). Secondary responses in this neuron class were recorded to other conifer monoterpenes, but also to the bark beetle produced compounds ipsdienol and E-myrcenol. The response threshold to myrcene appeared at approximately 1 ng. Strangely, the response to E-myrcenol increased with the age of the stimulus pipette. The reason for this phenomenon remains unknown, but could perhaps be explained by breakdown of the compound on the filter paper during oxygen exposure. For consistency, the results presented for this compound are derived only from responses to fresh pipettes. Four cells responded strongly to D3-carene (found primarily in Pinus spp.). This ORN type exhibited few and weak secondary responses. Another specific ORN was the one that responded to p-

562

M.N. Andersson et al. / Journal of Insect Physiology 55 (2009) 556–567

Fig. 4. Dose–response curves from ORN classes with primary response(s) to (A) cis-verbenol (n = 7), (B) a-pinene (n = 7), (C) myrcene (n = 7), (D) myrcene/p-cymene (n = 1), (E) p-cymene (n = 7), (F) 1,8-cineole (n = 7), (G) 3-octanol (n = 7), (H) 1-octen-3-ol (n = 7), (I) GLV alcohols (n = 8), and (J) trans-conophthorin (n = 9). cV: ()-cis-verbenol, Vn: ()-verbenone, tV: ()-trans-verbenol, (+) and ()aP: (+)- and ()-a-pinene, bP: ()-b-pinene, SStC and RRtC: S,S- and R,R-trans-conophthorin, xB: exo-brevicomin, D3: D3carene, Ci: 1,8-cineole, My: myrcene, Te: terpinolene, pC: p-cymene, (+)- and ()L: (+)- and ()-limonene, EM: E-myrcenol, Id: ipsdienol, Cg: chalcogran, C8an and (+)C8an: ()- and (+)-3-octanol, C8en and ()C8en: ()- and ()-1-octen-3-ol, C6: 1-hexanol, Z3C6: Z3-hexen-1-ol, E2C6: E2-hexen-1-ol, Lol and ()Lol: ()- and ()-linalool, C6Ald: hexanal, E2C6Ald: E2-hexenal, and Z3C6Ac: Z3-hexenyl acetate.

cymene (Fig. 4E). Three neurons were found and the response threshold was around 1 ng compound. Weaker responses in this neuron were elicited by a few other conifer-associated odorants. As previously mentioned, seven cells responded to 1,8-cineole, with a response threshold around 1 ng (Fig. 4F). These ORNs always occurred together with a cis-verbenol responding ORN. The neuron also responded weakly to chalcogran, exo-brevicomin and both enantiomers of trans-conophthorin.

During screening we found one cell that responded as the myrcene cell, except for an additional strong response to p-cymene (Fig. 4D). Dose–response curves were later obtained from another cell of this type and the curves were remarkably similar to the ones of the myrcene cell (Fig. 4C). To rule out that a second ORN, with spike characteristics indistinguishable from the myrcene cell, was responsible for the p-cymene response, we performed a crossadaptation test. Reciprocal sensory adaptation indicated that the

M.N. Andersson et al. / Journal of Insect Physiology 55 (2009) 556–567

563

same cell responded to both compounds. Subsequent comparisons of response spectra revealed that the response spectrum of the rare cell type was a combination of responses from the myrcene- and pcymene cells (Table 2), suggesting that the olfactory receptor proteins tuned to myrcene and p-cymene may occasionally be coexpressed in the same ORN (see also Section 4).

activated by Z3-hexen-1-ol (+++), 1-hexanol (++), and E2-hexen-1ol (+). Three cells responded to S,S-trans-conophthorin (++) and ()-linalool (+). These cells were co-compartmentalized with B neurons that also responded weakly (+) to S,S-trans-conophthorin. One of these sensilla was subjected to a dose–response test, showing no response below 1 mg dose (data not shown).

3.5. ORNs responding to non-host volatiles

3.7. Spatial distributions of ORNs

The remaining 24 (<: 13, ,: 11) of the strongly responding ORNs, responded primarily to NHV (Table 2). 3-Octanol elicited the strongest response in 10 cells (Fig. 4G). However, 1-octen-3-ol and Z3-hexen-1-ol elicited almost as powerful responses as 3-octanol and still other odors, mainly in the NHV group, evoked responses in this neuron. The racemic mixtures of 3-octanol and 1-octen-3-ol elicited stronger responses than their corresponding (+) and ()enantiomers, respectively. The ORN did not respond to 3-octanol below 100 ng. One cell was found that was most sensitive to 1-octen-3-ol. The naturally occurring ()-enantiomer elicited stronger responses than the racemic blend at all doses (Fig. 4H). 3-Octanol elicited almost as strong response as 1-octen-3-ol at the highest dose but the responses were clearly differentiated at lower doses. A few other odors elicited weak responses. In contrast to the 3-octanol cell, the response threshold was below 1 ng. Seven ORNs were most sensitive to the three behaviorally redundant GLV alcohols: 1-hexanol, E2-hexen-1-ol and Z3-hexen1-ol. Even at the lowest dose (100 pg), this ORN was activated to the same extent by the three compounds (Fig. 4I). At high doses, the cell responded relatively strongly also to two GLV aldehydes and to chalcogran. A few other compounds elicited weaker responses in this neuron. Six receptor neurons responded primarily to S,S-trans-conophthorin, with a threshold for response between 100 pg and 1 ng (Fig. 4J). Relatively strong responses were also elicited by chalcogran and exo-brevicomin. The R,R-enantiomer of transconophthorin elicited responses that were clearly weaker, with a threshold more than two decadic steps higher than for the S,Senantiomer.

Neurons of a particular ORN class were in general distributed in one out of two major patterns that largely correspond to the major distributions of morphological single-walled sensillum types. Either the neurons were located in areas A and/or B (where SW 1 is dominating), or they were predominately found in area C (where most SW 2 are found) (Table 2, Fig. 1B). The two sensillum types (Hallberg, 1982b) could, however, not easily be discerned under the microscope, but the density and arrangement of sensilla in the A and B bands were distinct from the ones in area C (Fig. 1A). The two types of sensilla are not absolutely confined within the specific areas (Hallberg, 1982b), but it is likely that most recordings from areas A and B were from SW 1, whereas recordings from SW 2 have been obtained in area C. The cis-verbenol neurons, and thus also the 1,8-cineole neurons, were mainly located in area C, but a few were found in area B (Fig. 1B). Also, the a-pinene cells were found in C, except one that was found in B. Two methyl-butenol cells were found at the borderline between B and C, and the third one in area B. ORNs belonging to all other strongly responding classes were found in the A and/or B areas, except one myrcene cell that was found in area C (close to the B/C border). Most of the 26 weakly responding ORNs were located in areas A and/or B. However, the two that responded to the pinenes and cis-verbenol were found in area C, and the two that responded weakly to methyl-butenol were located at the B/C border. The three sensilla that contained neurons that solely responded by inhibition were located in areas A and B.

3.6. Weakly responding ORNs

Our results suggest that the olfactory receptor neurons in I. typographus are, with few exceptions, remarkably selective in their response specificity. Typically, only one compound was able to elicit a strong response in a given neuron while a small number of other, often structurally related, compounds elicited responses that were clearly weaker. High specificity was not only characteristic for ORNs responding to pheromone compounds, but was also found among those responding to host and non-host compounds. In some insects, including Drosophila, many ORNs appeared broadly tuned when odors were delivered at high concentrations (Hallem and Carlson, 2006). However, apparent response specificity is dependent on the number of test compounds, their chemical structures, and the stimulus dose. The latter is evident from our dose–response curves that demonstrate higher ORN specificity at lower doses. Increased response specificity at low doses was also shown in Drosophila (Hallem and Carlson, 2006) and in other insects (Hansson et al., 1999; Larsson et al., 2001). In insects, ORNs exhibit stereotypical co-localization within sensillum types that in turn have particular distributions on the antenna (Hansson et al., 1986; de Bruyne et al., 2001; Ghaninia et al., 2007). Pheromone-related ORNs are typically situated in specific morphological types of sensilla that are often spatially segregated from other ORN types (Hansson et al., 1986, 1999; Larsson et al., 2001; but see Renou et al., 1998; Said et al., 2003). In I. typographus, we found no strict spatial segregation between ORNs detecting compounds involved in pheromone communica-

During screening, 26 ORNs (<: 13, ,: 13) could be classified into 11 classes, exclusively based on weak or intermediate responses. One additional ORN class (one cell) was found during dose– response tests (thus not included in the 26 ORNs) together with the neuron that responded strongly to the GLV alcohols. This B neuron responded exclusively to amitinol (+++). Since it is very likely that the key odorants for the weakly responding neurons were lacking, their classification is somewhat preliminary. Two cells were moderately excited (++) by ()-verbenone and ()-trans-verbenol. Two other cells responded (+) to 2-methyl-3buten-2-ol. Like the ORNs that responded strongly to methylbutenol, this neuron was also a B neuron and the corresponding A neuron was inhibited during the response. One cell that responded (++) to ()-verbenone was found together with the strongly responding a-pinene cell described above. Eight cells responded (++) to (+)-a-pinene (but not to the ()enantiomer) and ()-b-pinene. Two cells were activated by both enantiomers of a-pinene (++), ()-b-pinene (+), and ()-cisverbenol (+). Another cell responded (+) to the esters bornyl acetate and Z3-hexenyl acetate. We found two cells that responded to 3-octanol (++), 1-octen-3ol (+) (with weaker responses to the (+)- and ()-enantiomers, respectively) and Z3-hexen-1-ol (+). One cell responded (++) only to 1-octen-3-ol (pure enantiomers not tested). Another cell was

4. Discussion 4.1. Organization of the peripheral olfactory system

564

M.N. Andersson et al. / Journal of Insect Physiology 55 (2009) 556–567

tion and plant odors, respectively. ORNs for cis-verbenol were predominantly found in the C area, and for 2-methyl-3-buten-2-ol in the B/C areas, while ORNs for most plant odors were found in areas A and B. These patterns may correspond to the spatial arrangements of the two types of single-walled olfactory sensilla. However, ORNs for the host volatiles 1,8-cineole and a-pinene were also predominantly found in area C, and the former type was always co-localized with cis-verbenol neurons. Receptor neurons for all the other con- and heterospecific bark beetle-related compounds were all found in areas A and B. This fuzzy segregation between ORNs for pheromones and plant odors in I. typographus may indicate that host finding by a combination of pheromones/ plant odors in bark beetles is an integrated system that does not functionally segregate between different classes of semiochemicals. Alternatively, the distal distribution of the two pheromone ORNs may represent their current roles in the aggregation behavior of I. typographus, as only cis-verbenol and 2-methyl-3-buten-2-ol are essential aggregation pheromone components in this species (Schlyter et al., 1987a). Testing this hypothesis would require comparative studies of the peripheral olfactory systems of related species of bark beetles utilizing different pheromone compounds. Interestingly, behavioral experiments involving selective coating of sensilla and partial antennectomy on Ips confusus, indicated that the sex pheromone receptors are located in the distal area of the antennae and probably are confined within sensilla trichodea (Borden and Wood, 1966).

as narrowly tuned. This indicates that host-produced volatiles may serve a function during host location, possibly acting as habitatscale primary attractants (Saint-Germain et al., 2008) or as pheromone synergists. Specific responses to host-plant odors have been reported also in other insects (Wibe and Mustaparta, 1996; Hansson et al., 1999; Røstelien et al., 2000, 2005; Larsson et al., 2001; Stensmyr et al., 2001; Barata et al., 2002; Bicha˜o et al., 2003, 2005). Among the neurons that responded to host odors, we found one type that exhibited all the responses of the more common myrcene and p-cymene cells. The responses of this neuron could have been explained by the presence of the myrcene cell and a p-cymene responding cell in the same sensillum. However, the spike form and amplitude during the responses to p-cymene and myrcene were indistinguishable and a cross-adaptation test showed reciprocal sensory adaptation, demonstrating that these responses were most likely from the same ORN. This ORN might then express a novel OR with ligand specificity as in Table 2. But since insects occasionally coexpress two ORs in the same ORN (Fishilevich and Vosshall, 2005; Goldman et al., 2005; Ray et al., 2007), and since responses of the neuron were so similar to the other cells, we postulate that the responses might result from coexpression of the myrcene- and the p-cymene ORs in the same ORN. Of course, we are aware that molecular data is needed to support such a hypothesis. 4.3. Redundancy and synergism among non-host volatiles

4.2. ORNs responding to pheromones and host volatiles In this study we found a large number of receptor neurons that responded to compounds that are ecologically relevant to I. typographus. In fact, we obtained responses to all compounds with known behavioral effects except to methyl-2,4-decadienoate, a pheromone component of Pityogenes chalcographus (Byers, 1993). We found seven ORN classes that specifically responded to bark beetle pheromone substances, similar to the results of an earlier study (Tømmera˚s, 1985). However, in that study, a large proportion of the neurons responded to ipsdienol (enantiospecific), whereas only two cis-verbenol cells and no a-pinene cells were found. This might simply be due to the fact that recordings from the distal area of the antenna were under-represented in that study. In agreement with Tømmera˚s (1985), we unexpectedly found only three cells that responded to the aggregation pheromone component 2-methyl-3-buten-2-ol. A low number of olfactory cells might explain the high dose that is needed to elicit EAG response (Dickens, 1981) and behavioral attraction (Schlyter et al., 1987b), but individual cells have also been reported to be quite insensitive to this compound (Tømmera˚s, 1985). However, we found no indications that the methyl-butenol cells would be less sensitive than other ORNs, based on the response to our screening dose (Fig. 2D). We found four neurons that responded primarily to ipsdienol, which we grouped into two classes, based on their secondary responses to ipsenol, amitinol and E-myrcenol. These two classes likely correspond to the two previously described enantiospecific ipsdienol cells, which also varied in their responses to these secondary odors (Mustaparta et al., 1984). The pheromone-responding neurons in the present study were highly selective, even though most pheromone compounds are structurally similar to each other and also to the host monoterpenes that we tested (Table 1). Given that kairomones have not been clearly demonstrated to be important for host selection in I. typographus, we had not expected to find a large proportion of ORNs that responded with great discriminatory power to host compounds. All host (conifer) odors tested were structurally related monoterpenes. Thus, most ORNs that responded to this group of compounds can be described

The potential of non-host volatiles to modulate host selection behavior has been shown in many bark beetle species, and different species respond to different but often overlapping sets of compounds (Schlyter et al., 2000; Zhang and Schlyter, 2004). The fact that almost one fourth of the strongly responding neurons responded to behaviorally inhibitory NHV, highlights their ecological and evolutionary relevance. In I. typographus the GLV alcohols 1-hexanol, E2-hexen-1-ol and Z3-hexen-1-ol as well as the angiosperm bark-associated compounds 3-octanol, 1-octen-3ol and trans-conophthorin, reduce the attraction to the pheromone. Combining the individual GLVs with each other or with the C8-alcohols does not produce a significantly stronger antiattractant effect (defined as redundancy). In contrast, transconophthorin and the anti-aggregation pheromone verbenone, in combination with the alcohols, dramatically increase the activity of the blend (strict sense synergism) (Zhang and Schlyter, 2003). We asked whether these behavioral response patterns could be explained by the specificities of the receptor neurons, or if the integration of negative olfactory signals occurs in the CNS. Interestingly, our results indicate that part of the redundancy between the GLV alcohols is likely to be explained solely by the responses of one ORN class. The ORN that strongly responded to GLV alcohols responded indiscriminately to the three active compounds, even at 100-pg loading (Fig. 4I). This is in sharp contrast to the specific GLV neurons found in scarab beetles (Hansson et al., 1999; Larsson et al., 2001). Narrowly, as well as broadly tuned GLV neurons have been found also in the Colorado potato beetle (Ma and Visser, 1978; De Jong and Visser, 1988). These species feed on angiosperms, and the Colorado beetle is attracted to the specific GLV blend of potato plants (Visser and Ave´, 1978). In the Colorado beetle, specific GLV receptors would be expected to be required for efficient host localization, but for conifer-feeding insects that cannot utilize plants with large amounts of GLV (i.e. angiosperms) there is no need for specialized receptors for angiosperm compounds that signal the same kind of information (i.e. ‘non-host plant’). Instead the insects would probably benefit from having some redundancy in the olfactory system to be able to increase the likelihood of perceiving a non-

M.N. Andersson et al. / Journal of Insect Physiology 55 (2009) 556–567

host signal. Similar phenomena have been found in the pheromone systems of other beetles and moths, where the same neuron detects several behavioral antagonists (Cosse´ et al., 1998; Wojtasek et al., 1998; Larsson et al., 2002). The behavioral redundancy among the two C8-alcohols and between them and the GLV alcohols cannot, on the other hand, be explained exclusively by ORN response specificity. Judging from the dose–response curves of the two C8-alcohol sensitive ORNs (Fig. 4G and H), it is obvious that the behaviorally redundant compounds to different degrees activate both ORNs, and that the ORNs show partly overlapping response spectra. However, the neuron responding primarily to ()-1-octen-3-ol is not nearly as excited by 3-octanol at low doses. Thus, the response specificities are not sufficient to explain the behavioral redundancy between these odors. As expected, the synergizing compounds, verbenone and trans-conophthorin, are perceived by other receptor neurons, indicating that synergism is integrated by the CNS. Recently it was shown that a synergistic behavioral response of Cydia molesta (Lepidoptera: Tortricidae) to host-plant odor mixtures was reflected in the odor-evoked calcium activity in their antennal ˜ ero et al., 2008). Synergistic responses to compound lobes (Pin blends have also been observed at the peripheral level (Ochieng et al., 2002). However, since separate input channels for the active compounds exist in our system, we believe that synergistic interaction at the ORN level is an unlikely explanation. The neuron that responded best to trans-conophthorin was also strongly excited by chalcogran (Fig. 4J), a structurally related compound included in the aggregation pheromone of the competing P. chalcographus. Interestingly, chalcogran is used as a negative cue by I. typographus (Byers, 1993), and based on our results it is possible that its inhibitory behavioral effect mainly results from excitation of the trans-conophthorin neuron. From an evolutionary perspective, it is plausible that the trans-conophthorin receptor in I. typographus and a tentatively existing chalcogran receptor in P. chalcographus have diverged from the same ancestral receptor protein. It would therefore be of great interest to compare the specificities of the two cells in the different species. exo-Brevicomin, a pheromone of Dendroctonus bark beetles, also elicited a relatively strong response in the transconophthorin neuron. A previous study reported receptor neurons in I. typographus that responded to birch bark vapor as well as to exo-brevicomin (Tømmera˚s and Mustaparta, 1987), and in other studies where birch volatiles were not tested, specific exobrevicomin cells were reported (Tømmera˚s et al., 1984; Tømmera˚s, 1985). These cells may conform to the trans-conophthorin cell found in the present study. In contrast to chalcogran, exobrevicomin seems to enhance the attraction to the pheromone, particularly in males (Tømmera˚s and Mustaparta, 1984; Tømmera˚s et al., 1984). Apparently, the reasoning concerning the repellent action of chalcogran cannot be applied to exo-brevicomin, highlighting the complexity of the olfactory system. The ubiquitous plant constituent linalool did not evoke strong responses in any class of receptor neuron. Linalool is behaviorally or electrophysiologically active for many insects utilizing angiosperm plants (Stensmyr et al., 2001; Larsson et al., 2003; Bicha˜o et al., 2005). The absence of linalool-specific neurons in I. typographus agrees with a previously recorded weak EAG response and a seemingly absent behavioral activity of this compound (Zhang et al., 1999). This suggests that, although present also in conifers, linalool may have little informational value for bark beetles. 4.4. Enantiomeric discrimination The trans-conophthorin cell proved to be highly enantioselective, favoring the enantiomer that is predominately found in nature

565

(Zhang et al., 2002). Curiously, its enantioselectivity was higher than its selectivity towards other structural analogs, being almost a thousand times more sensitive to the S,S-enantiomer than to the R,R-enantiomer of trans-conophthorin (Fig. 4J). This is higher than has been demonstrated even for chiral pheromones in bark beetles and many other insects (Mustaparta et al., 1980; Tømmera˚s, 1985; Larsson and Hansson, 1998; Larsson et al., 2002). It is, however, not quite as striking as the highest degrees of enantioselectivity demonstrated in insects (Hansen, 1984; Wojtasek et al., 1998; Larsson et al., 1999; Svensson and Larsson, 2008). Such high selectivity was not observed among the other ORNs for which we had both enantiomers available. The a-pinene cells found in this study all responded slightly stronger to the (+)- than to the ()enantiomer (Fig. 4B), similar to what was found in the pine weevil (Hylobius abietis) (Wibe et al., 1998). This contrasts the results of a previous EAG study that showed that the antenna of I. typographus is more sensitive to the ()-enantiomer (Dickens, 1981). Apart from the fact that the beetles had different geographical origins (Sweden and Germany), one can only speculate about the reason for this discrepancy and stress that EAGs and single-sensillum recordings can provide conflicting results (Wibe, 2004). However, enantioselectivity among neurons was not the main focus of this study, but should definitely be investigated in the future with a complete set of enantiomer pairs. 5. Conclusions In this study we found that I. typographus, besides having ORNs that respond to pheromone compounds of its own and other bark beetles, also have a large number of ORNs that respond to host odors. However, we were more surprised to find that almost 25% of the strongly responding neurons were tuned to compounds typical of angiosperm non-host plants. We also show that part of the behavioral redundancy among NHV probably is integrated exclusively at the peripheral level, whereas the synergism appears to be integrated in the CNS. We identified 17 ORN classes based upon strong excitatory responses and, additionally, 12 classes were preliminary characterized based on weak or intermediate responses giving a total of 29 classes. The antennal lobes of I. typographus have been estimated to contain 60–70 glomeruli (J. Schachtner, pers. comm.). Assuming a one-to-one relationship between the number of glomeruli and the number of ORN types, we have characterized almost half of the existing ORNs. To be able to characterize the remaining neurons and also find the best ligands for the weakly responding ORNs, a more diverse array of test compounds, such as aromatics, esters, sesquiterpenes, and stress-associated plant compounds, should be tested. Sesquiterpenes are of particular interest since EAG responses to at least two unidentified compounds have been found (Q.-H. Zhang, unpublished data). By running odor collections from host and non-host plants through GC-coupled SSR, the identification of novel ligands would be facilitated. Recordings should also be done on the maxillary palps (Hallberg, 1982a). To be able to better relate the electrophysiological responses to the behavior of the insects, recordings should also be performed in the natural environment to naturally occurring doses of pheromones, host, and non-host odors. Acknowledgements We thank Elisabeth Marling and Muhammad Binyameen for rearing assistance. We also thank Eric Hallberg (Dept. Cell and Organism Biology, Lund University, Lund, Sweden) for providing micrographs and Qing-He Zhang (Sterling Rescue, Spokane, WA, USA) for useful comments on a previous version of the manuscript.

566

M.N. Andersson et al. / Journal of Insect Physiology 55 (2009) 556–567

This study was funded by FORMAS, project # 230-2005-1778, ‘‘Semiochemical diversity and insect dynamics’’, and Linnaeusprogram ‘‘Insect Chemical Ecology, Ethology and Evolution’’ (ICE3). References Barata, E.N., Mustaparta, H., Pickett, J.A., Wadhams, L.J., Arau´jo, J., 2002. Encoding of host and non-host plant odours by receptor neurones in the eucalyptus woodborer, Phoracantha semipunctata (Coleoptera: Cerambycidae). Journal of Comparative Physiology A 188, 121–133. Bicha˜o, H., Borg-Karlson, A.-K., Arau´jo, J., Mustaparta, H., 2003. Identification of plant odours activating receptor neurones in the weevil Pissodes notatus F. (Coleoptera, Curculionidae). Journal of Comparative Physiology A 189, 203–212. Bicha˜o, H., Borg-Karlson, A.-K., Wibe, A., Arau´jo, J., Mustaparta, H., 2005. Molecular receptive ranges of olfactory receptor neurones responding selectively to terpenoids, aliphatic green leaf volatiles and aromatic compounds, in the strawberry blossom weevil Anthonomus rubi. Chemoecology 15, 211–226. Borden, J.H., Wood, D.L., 1966. The antennal receptors and olfactory response of Ips confusus (Coleoptera: Scolytidae) to male sex attractant in the laboratory. Annals of the Entomological Society of America 59, 253–261. Byers, J.A., 1993. Avoidance of competition by spruce bark beetles Ips typographus and Pityogenes chalcographus. Experientia 49, 272–275. Cosse´, A.A., Todd, J.L., Baker, T.C., 1998. Neurons discovered in male Helicoverpa zea antennae that correlate with pheromone-mediated attraction and interspecific antagonism. Journal of Comparative Physiology A 182, 585–594. de Bruyne, M., Foster, K., Carlson, J.R., 2001. Odor coding in the Drosophila antenna. Neuron 30, 537–552. De Jong, R., Visser, J.H., 1988. Integration of olfactory information in the Colorado potato beetle brain. Brain Research 447, 10–17. Dickens, J.C., 1981. Behavioural and electrophysiological responses of the bark beetle, Ips typographus, to potential pheromone components. Physiological Entomology 6, 251–261. Dickens, J.C., Payne, T.L., Ryker, L.C., Rudinsky, J.A., 1984. Single cell responses of the Douglas-fir beetle Dendroctonus pseudotsugae Hopkins (Coleoptera: Scolytidae) to pheromones and host odors. Journal of Chemical Ecology 10, 583–600. Dickens, J.C., Payne, T.L., Ryker, L.C., Rudinsky, J.A., 1985. Multiple acceptors for pheromonal enantiomers on single olfactory cells in the Douglas-fir beetle Dendroctonus pseudotsugae Hopk. (Coleoptera: Scolytidae). Journal of Chemical Ecology 11, 1359–1370. Erbilgin, N., Krokene, P., Kvamme, T., Christiansen, E., 2007. A host monoterpene influences Ips typographus (Coleoptera: Curculionidae Scolytinae) responses to its aggregation pheromone. Agricultural and Forest Entomology 9, 135–140. Fishilevich, E., Vosshall, L.B., 2005. Genetic and functional subdivision of the Drosophila antennal lobe. Current Biology 15, 1548–1553. Ghaninia, M., Ignell, R., Hansson, B.S., 2007. Functional classification and central nervous projections of olfactory receptor neurons housed in antennal trichoid sensilla of female yellow fever mosquitoes, Aedes aegypti. European Journal of Neuroscience 26, 1611–1623. Goldman, A.L., van der Goes van Naters, W., Lessing, D., Warr, C.G., Carlson, J.R., 2005. Coexpression of two functional odor receptors in one neuron. Neuron 45, 661–666. Guerrero, A., Feixas, J., Pajares, J., Wadhams, L.J., Pickett, J.A., Woodcock, C.M., 1997. Semiochemically induced inhibition of behaviour of Tomicus destruens (Woll) (Coleoptera: Scolytidae). Naturwissenschaften 84, 155–157. Hallberg, E., 1982a. Sensory organs in Ips typographus (Insecta: Coleoptera). Fine structure of the sensilla of the maxillary and labial palps. Acta Zoologica 63, 191–198. Hallberg, E., 1982b. Sensory organs in Ips typographus (Insecta: Coleoptera)—fine structure of antennal sensilla. Protoplasma 111, 206–214. Hallem, E.A., Carlson, J.R., 2006. Coding of odors by a receptor repertoire. Cell 125, 143–160. Hansen, K., 1984. Discrimination and production of disparlure enantiomers by the gypsy moth and the nun moth. Physiological Entomology 9, 9–18. Hansson, B.S., Lo¨fstedt, C., Lo¨fqvist, J., Hallberg, E., 1986. Spatial arrangement of different types of pheromone-sensitive sensilla in a male moth. Naturwissenschaften 73, 269–270. Hansson, B.S., Larsson, M.C., Leal, W.S., 1999. Green leaf volatile-detecting olfactory receptor neurones display very high sensitivity and specificity in a scarab beetle. Physiological Entomology 24, 121–126. Hulcr, J., Ubik, K., Vrkoc, J., 2006. The role of semiochemicals in tritrophic interactions between the spruce bark beetle Ips typographus, its predators and infested spruce. Journal of Applied Entomology 130, 275–283. Jakus, R., Blazenec, M., 2003. Influence of the proportion of ()-a-pinene in pheromone bait on Ips typographus (Col., Scolytidae) catch in pheromone trap barriers and in single traps. Journal of Applied Entomology 127, 91–95. Larsson, M.C., Hansson, B.S., 1998. Receptor neuron responses to potential sex pheromone components in the caddisfly Rhyacophila nubila (Trichoptera: Rhyacophilidae). Journal of Insect Physiology 44, 189–196. Larsson, M.C., Leal, W.S., Hansson, B.S., 1999. Olfactory receptor neurons specific to chiral sex pheromone components in male and female Anomala cuprea beetles (Coleoptera: Scarabaeidae). Journal of Comparative Physiology A 184, 353–359. Larsson, M.C., Leal, W.S., Hansson, B.S., 2001. Olfactory receptor neurons detecting plant odours and male volatiles in Anomala cuprea beetles (Coleoptera: Scarabaeidae). Journal of Insect Physiology 47, 1065–1076.

Larsson, M.C., Hallberg, E., Kozlov, M.V., Francke, W., Hansson, B.S., Lo¨fstedt, C., 2002. Specialized olfactory receptor neurons mediating intra- and interspecific chemical communication in leafminer moths Eriocrania spp. (Lepidoptera: Eriocraniidae). Journal of Experimental Biology 205, 989–998. Larsson, M.C., Stensmyr, M.C., Bice, S.B., Hansson, B.S., 2003. Attractiveness of fruit and flower odorants detected by olfactory receptor neurons in the fruit chafer Pachnoda marginata. Journal of Chemical Ecology 29, 1253–1268. Ma, W.-C., Visser, J.H., 1978. Single unit analysis of odour quality coding by the olfactory antennal receptor system of the Colorado beetle. Entomologia Experimentalis et Applicata 24, 320–333. Mustaparta, H., Angst, M.E., Lanier, G.N., 1977. Responses of single receptor cells in the pine engraver beetle, Ips pini (Say) (Coleoptera: Scolytidae) to its aggregation pheromone, ipsdienol, and the aggregation inhibitor, ipsenol. Journal of Comparative Physiology A 121, 343–347. Mustaparta, H., Angst, M.E., Lanier, G.N., 1979. Specialization of olfactory cells to insect and host produced volatiles in the bark beetle Ips pini (Say). Journal of Chemical Ecology 5, 109–123. Mustaparta, H., Angst, M.E., Lanier, G.N., 1980. Receptor discrimination of enantiomers of the aggregation pheromone ipsdienol in two species of Ips. Journal of Chemical Ecology 6, 689–702. Mustaparta, H., Tømmera˚s, B.A˚., Baeckstro¨m, P., Bakke, J.M., Ohloff, G., 1984. Ipsdienol-specific receptor cells in bark beetles: structure–activity relationships of various analogues and of deuterium-labelled ipsdienol. Journal of Comparative Physiology A 154, 591–596. Ochieng, S.A., Park, K.C., Baker, T.C., 2002. Host plant volatiles synergize responses of sex pheromone-specific olfactory receptor neurons in male Helicoverpa zea. Journal of Comparative Physiology A 188, 325–333. Payne, T.L., Richerson, J.V., Dickens, J.C., West, J.R., Mori, K., Berisford, C.W., Hedden, R.L., Vite´, J.P., Blum, M.S., 1982. Southern pine beetle: olfactory receptor and behavior discrimination of enantiomers of the attractant pheromone frontalin. Journal of Chemical Ecology 8, 873–882. Persson, M., Sjo¨din, K., Borg-Karlson, A.-K., Norin, T., Ekberg, I., 1996. Relative amounts and enantiomeric compositions of monoterpene hydrocarbons in xylem and needles of Picea abies. Phytochemistry 42, 1289–1297. ˜ ero, J.C., Galizia, C.G., Dorn, S., 2008. Synergistic behavioral responses of female Pin oriental fruit moths (Lepidoptera: Tortricidae) to synthetic host plant-derived mixtures are mirrored by odor-evoked calcium activity in their antennal lobes. Journal of Insect Physiology 54, 333–343. Ray, A., van der Goes van Naters, W., Shiraiwa, T., Carlson, J.R., 2007. Mechanisms of odor receptor gene choice in Drosophila. Neuron 53, 353–369. Renou, M., Tauban, D., Morin, J.-P., 1998. Structure and function of antennal pore plate sensilla of Oryctes rhinoceros (L.) (Coleoptera: Dynastidae). International Journal of Insect Morphology and Embryology 27, 227–233. Røstelien, T., Borg-Karlson, A.-K., Mustaparta, H., 2000. Selective receptor neurone responses to E-b-ocimene, b-myrcene, E,E-a-farnesene and homo-farnesene in the moth Heliothis virescens, identified by gas chromatography linked to electrophysiology. Journal of Comparative Physiology A 186, 833–847. Røstelien, T., Stranden, M., Borg-Karlson, A.-K., Mustaparta, H., 2005. Olfactory receptor neurons in two Heliothine moth species responding selectively to aliphatic green leaf volatiles, aromatic compounds, monoterpenes and sesquiterpenes of plant origin. Chemical Senses 30, 443–461. Said, I., Tauban, D., Renou, M., Mori, K., Rochat, D., 2003. Structure and function of the antennal sensilla of the palm weevil Rhynchophorus palmarum (Coleoptera, Curculionidae). Journal of Insect Physiology 49, 857–872. Saint-Germain, M., Buddle, C.M., Drapeau, P., 2008. Primary attraction and random landing in host-selection by wood-feeding insects: a matter of scale? Agricultural and Forest Entomology 9, 227–235. Schlyter, F., Cederholm, I., 1981. Separation of the sexes of living spruce bark beetles Ips typographus (L) (Coleoptera Scolytidae). Zeitschrift fuer Angewandte Entomologie 92, 42–47. Schlyter, F., Birgersson, G.A., 1999. Forest beetles. In: Hardie, J., Minks, A.K. (Eds.), Pheromones of Non-Lepidopteran Insects Associated with Agricultural Plants. CAB International, Oxford, pp. 113–148. Schlyter, F., Birgersson, G., Byers, J.A., Lo¨fqvist, J., Bergstro¨m, G., 1987a. Field response of spruce bark beetle, Ips typographus, to aggregation pheromone candidates. Journal of Chemical Ecology 13, 701–716. Schlyter, F., Byers, J.A., Lo¨fqvist, J., 1987b. Attraction to pheromone sources of different quantity, quality, and spacing: density-regulation mechanisms in bark beetle Ips typographus. Journal of Chemical Ecology 13, 1503–1523. Schlyter, F., Birgersson, G., Leufve´n, A., 1989. Inhibition of attraction to aggregation pheromone by verbenone and ipsenol. Density regulation mechanisms in bark beetle Ips typographus. Journal of Chemical Ecology 15, 2263–2277. Schlyter, F., Birgersson, G., Byers, J.A., Bakke, A., 1992. The aggregation pheromone of Ips duplicatus and its role in competitive interactions with I. typographus (Coleoptera: Scolytidae). Chemoecology 3, 103–112. Schlyter, F., Zhang, Q.-H., Anderson, P., Byers, J.A., Wadhams, L.J., Lo¨fqvist, J., Birgersson, G., 2000. Electrophysiological and behavioural responses of Tomicus piniperda and Tomicus minor (Coleoptera: Scolytidae) to non-host leaf and bark volatiles. Canadian Entomologist 132, 965–981. Stensmyr, M.C., Larsson, M.C., Bice, S., Hansson, B.S., 2001. Detection of fruit- and flower-emitted volatiles by olfactory receptor neurons in the polyphagous fruit chafer Pachnoda marginata (Coleoptera: Cetoniinae). Journal of Comparative Physiology A 187, 509–519. Stensmyr, M.C., Dekker, T., Hansson, B.S., 2003. Evolution of the olfactory code in the Drosophila melanogaster subgroup. In: Proc. R. Soc. Lond. Ser. B, vol. 270. pp. 2333–2340.

M.N. Andersson et al. / Journal of Insect Physiology 55 (2009) 556–567 Svensson, G.P., Larsson, M.C., 2008. Enantiomeric specificity in a pheromone– kairomone system of two threatened saproxylic beetles, Osmoderma eremita and Elater ferrugineus. Journal of Chemical Ecology 34, 189–197. Tømmera˚s, B.A˚., 1985. Specialization of the olfactory receptor cells in the bark beetle Ips typographus and its predator Thanasimus formicarius to bark beetle pheromones and host tree volatiles. Journal of Comparative Physiology A 157, 335–342. Tømmera˚s, B.A˚., Mustaparta, H., 1984. Enhanced attraction of Ips typographus by adding exo-brevicomin to pheromone traps. Naturwissenschaften 71, 375–377. Tømmera˚s, B.A˚., Mustaparta, H., 1987. Chemoreception of host volatiles in the bark beetle Ips typographus. Journal of Comparative Physiology A 161, 705–710. Tømmera˚s, B.A˚., Mustaparta, H., 1989. Single cell responses to pheromones, host and non-host volatiles in the ambrosia beetle Trypodendron lineatum. Entomologia Experimentalis et Applicata 52, 141–148. Tømmera˚s, B.A˚., Mustaparta, H., Gregoire, J.-C., 1984. Receptor cells in Ips typographus and Dendroctonus micans specific to pheromones of the reciprocal genus. Journal of Chemical Ecology 10, 759–770. Wadhams, L.J., Angst, M.E., Blight, M.M., 1982. Responses of the olfactory receptors of Scolytus scolytus (F.) (Coleoptera: Scolytidae) to the stereoisomers of 4methyl-3-heptanol. Journal of Chemical Ecology 8, 477–492. Wajs, A., Pranovich, A., Reunanen, M., Willfo¨r, S., Holmbom, B., 2006. Characterisation of volatile organic compounds in stemwood using solid-phase microextraction. Phytochemical Analysis 17, 91–101. Wibe, A., 2004. How the choice of method influence on the results in electrophysiological studies of insect olfaction. Journal of Insect Physiology 50, 497–503. Wibe, A., Mustaparta, H., 1996. Encoding of plant odours by receptor neurons in the pine weevil (Hylobius abietis) studied by linked gas chromatography–electrophysiology. Journal of Comparative Physiology A 179, 331–344. Wibe, A., Borg-Karlson, A.-K., Persson, M., Norin, T., Mustaparta, H., 1998. Enantiomeric composition of monoterpene hydrocarbons in some conifers and receptor

567

neuron discrimination of a-pinene and limonene enantiomers in the pine weevil, Hylobius abietis. Journal of Chemical Ecology 24, 273–287. Visser, J.H., Ave´, D.A., 1978. General green leaf volatiles in the olfactory orientation of the Colorado beetle, Leptinotarsa decemlineata. Entomologia Experimentalis et Applicata 24, 538–549. Wojtasek, H., Hansson, B.S., Leal, W.S., 1998. Attracted or repelled? A matter of two neurons, one pheromone binding protein, and a chiral center. Biochemical and Biophysical Research Communications 250, 217–222. Zar, J.H., 1999. Biostatistical Analysis, 4th ed. Prentice Hall, New Jersey. Zhang, Q.-H., Schlyter, F., 2003. Redundancy, synergism, and active inhibitory range of non-host volatiles in reducing pheromone attraction in European spruce bark beetle Ips typographus. Oikos 101, 299–310. Zhang, Q.-H., Schlyter, F., 2004. Olfactory recognition and behavioural avoidance of angiosperm nonhost volatiles by conifer-inhabiting bark beetles. Agricultural and Forest Entomology 6, 1–19. Zhang, Q.-H., Schlyter, F., Anderson, P., 1999. Green leaf volatiles interrupt pheromone response of spruce bark beetle, Ips typographus. Journal of Chemical Ecology 25, 2847–2861. Zhang, Q.-H., Schlyter, F., Birgersson, G., 2000. Bark volatiles from nonhost angiosperm trees of spruce bark beetle, Ips typographus (L.) (Coleoptera: Scolytidae): chemical and electrophysiological analysis. Chemoecology 10, 69–80. Zhang, Q.-H., Tolasch, T., Schlyter, F., Francke, W., 2002. Enantiospecific antennal response of bark beetles to spiroacetal (E)-conophthorin. Journal of Chemical Ecology 28, 1839–1852. Zhang, Q.-H., Schlyter, F., Liu, G.-T., Sheng, M.-L., Birgersson, G., 2007. Electrophysiological and behavioral responses of Ips duplicatus to aggregation pheromone in inner Mongolia, China: amitinol as a potential pheromone component. Journal of Chemical Ecology 33, 1303–1315.