European Symposium for

Insect Taste and Olfaction



Welcome to the


July 2-7, 2003 - Harstad, Norway


Main Page Venue Travel and Costs Main Program Social Program Abstracts of Talks Abstracts of Posters Registration Abstract Submission Deadlines ESITO Sponsors ESITO Homepage


Abstracts of Talks




Mamiko Ozaki and Atsushi Seto

Kyoto Institute of Technology, Department of Applied Biology, Faculty of Textile Science, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan
[email protected]

Effects of floral scents and their memory on the feeding preference or appetite in insect: In case of mushroom body-missing fly.

Flowers use scented nectar to attract and reward pollinator insects. However, Flowers of different plant species have different scents having different chemical compositions. Hence, it is not likely that every floral scent will uniformly affect the feeding preferences or appetites of insects. Here we show that the floral scents of 50 plant species brought various effects on the feeding threshold of the blowfly, Phormia regina. Moreover, memory of feeding experience in the presence of the floral scents variously influenced feeding threshold.
As the next step of experiment, we chosen several pure chemicals, which induce the same effects on the feeding preference of the fly as the representative floral scents do. Using these chemicals, we repeated the similar behavioural tests in the flies whose antennae and/or maxillary palps were cut and the flies whose mushroom body were missing. The results indicated that some olfactory memories were formed through antennae and others were through maxillary palps. The mushroom body-missing fly could change feeding preference depending on the existing olfactory cues but not by the memory of any feeding experience with odours, even if they have intact antennae and maxillary palps.
We expect that this study contributs to understandings of underlying mechanism of constructing a variety of olfactory-taste associative memories in insects.
Supported by a grant of ProBRAIN to MO.



Bente G Berg1,2, Tor J Almaas2, Jan G Bjaalie3, Hanna Mustaparta2

1 Dept of Psychology, Norwegian University of Science and Technology, Trondheim, Norway
2 Dept of Biology, Norwegian University of Science and Technology, Trondheim, Norway
3 Dept of Anatomy, University of Oslo, Oslo, Norway
[email protected]

Encoding of olfactory information underlying reproductive behaviour: Similarities and differences between two species of tobacco budworm moths, the Oriental Helicoverpa assulta and the American Heliothis virescens

The encoding of olfactory information in reproductive behaviours has been thoroughly investigated in several moth species. Among the heliothine species studied so far, the Oriental tobacco budworm moth, Helicoverpa assulta, is unique as concerns the composition of the pheromone blend. Whereas most of the species, as for instance the American tobacco budworm moth, Heliothis virescens, produces cis?11?hexadecenal (Z11?16:AL) as the major pheromone component, H. assulta uses this substance as the second component, ¾ and cis?9?hexadecenal (Z9?16:AL) as the major component. The pheromone information is detected by numerous sensory neurons, housed in the long sensilla trichodea type 1 on the antennae, and conveyed directly to the macroglomerular complex (MGC) of the antennal lobe. In addition to the pheromone neurons, it is well known that heliothine males possess particular neurons for detecting interspecific signals. Based on results from electrophysiological recordings combined with stainings, we have compared functional and morphological properties of the male specific sensory neurons in Helicoverpa assulta and Heliothis virescens. Digital atlases of the antennal lobe glomeruli in the two species (Berg et al. 2002) have been used to determine the target areas of the axon terminals. The data show similarities and differences, and it is interesting to uncover how nature has created general principles as well as fine variations in the structure and function of the neural system dealing with information about the insect produced substances in the two species.


Berg BG, Galizia CG, Brandt R, Mustaparta H 2002 Digital Atlases of the Antennal Lobe in Two Species of Tobacco Budworm Moths, the Oriental Helicoverpa assulta (Male) and the American Heliothis virescens (Male and Female). J Comp Neurol 446: 123-134



Rickard Ignell and John R. Carlson

Department of MCDB, Yale University, New Haven, CT 06520, USA
[email protected]

Physiological characterization and topographic mapping of coeloconic sensilla in Drosophila melanogaster

Drosophila melanogaster has emerged as an important model species to understand cellular coding mechanisms of peripheral olfactory systems. The main olfactory organ of Drosophila, the funiculus of the antenna, has been shown to be covered by three morphologically distinct sensillum types, sensilla basiconica, s. trichodea and s. coeloconica (grooved pegs sensu stricto). These sensory units are part of the peripheral interface where ultimately olfactory receptor neurons (ORNs) interact with the odor environment. It is the specificity, sensitivity and temporal dynamics of these ORNs that ultimately underlie the olfactory code. Several fundamental principles of this code have already been elucidated through analysis of basiconic ORNs (de Bruyne et al., 1999, 2001). Here, however, we report a number of novel features of the olfactory code based on in-depth functional analysis of the response characteristics of ORNs housed in coeloconic sensilla.

Based on the response spectra of individual ORNs to a panel of 104 odors we characterized five functional types of antennal coeloconic (ac) sensilla. While many of the ORNs observe a pairing rule, in some cases the rule appears to be violated: for example ac1B and ac2B appear identical, whereas ac1A and ac2A display different response spectra. This sharing is also observed in coeloconic sensillum types containing three neurons types. In all, coeloconic ORNs displayed overlapping but in some cases unique response spectra compared to basiconic ORNs.

Do coeloconic sensilla make a distinct contribution to odor coding? Data from this study indicate that ORNs in basiconic and coeloconic sensilla relay different temporal dynamic information to the primary olfactory center. Based on the response to a set of 32 odors we identified three temporal response profiles of basiconic and coeloconic ORNs; phasic, short-lasting and long-lasting phasic tonic responses. Interestingly, basiconic ORNs only displayed phasic and short-lasting phasic tonic responses, whereas coeloconic ORNs, in general, displayed long-lasting phasic tonic responses.

Topographic mapping of coeloconic sensilla revealed that most ORN classes are restricted to spatial domains on the antennal surface as also found for basiconic sensilla (de Bruyne et al. 2001).

De Bruyne M, Clyne PJ and JR Carlson (1999) Odor coding in model olfactory organ: The Drosophila maxillary palp. J Neurosci 19(11):4520-4532
De Bruyne M, Foster K and JR Carlson (2001) Odor coding in the Drosophila antenna. Neuron 30:537-552



Marien de Bruyne 1 , Sabine Schwarz 1 , Manja Wendt 1 , Barbara Regnery 1 , C. Giovanni Galizia 1 , Andre Fiala 2 , Sören Diegelmann 2, Erich Buchner 2 , John R. Carlson 3

1 Freie Universitaet Berlin, Berlin, Germany
2 Theodor-Boveri-Institut, Wuerzburg, Germany
3 Yale University, New Haven, CT, USA
[email protected]

Carbon dioxide perception in Drosophila: receptor expression, physiology, and behavior

Olfaction in Drosophila employs ca. 1300 olfactory receptor neurons (ORNs), which can be divided into different classes depending on their response spectra. Each class is thought to express a different member of a family of olfactory receptors (OR). In addition a few gustatory receptors (GR) are also expressed in antenna, one of which is called Gr21a (Scott et al., 2001). Recognition and discrimination of odorants most likely involves combinatorial coding. Alternatively, certain odors may have a special signaling function and be encoded via ‘labelled lines’ with very specific response properties.
In antennal sensilla basiconica, 22 classes of ORNs with different response spectra have been described (de Bruyne et al., 2001). We recorded from ab1C neurons that respond exclusively to CO2 with thresholds around ambient concentrations (0,03%). We then used the GAL4-UAS system to drive several reporter genes under control of the Gr21a promoter. The expression pattern of GFP suggests that Gr21a is expressed in ab1 sensilla. Gr21a-driven expression of the cell-death gene rpr deletes most GFP positive cells. Electroantennograms of such flies show a marked reduction in responses to CO2 but not to other odorants. We also expressed calcium sensitive cameleon in ab1C neurons. In optical recordings on transgenic flies we measured a dose-dependent activation by CO2 but not by other odorants. In addition, the V glomerulus was labeled and showed similar responses. We conclude that ab1C neurons express the gustatory receptor GR21a.
It is noteworthy that the Anopheles gambiae Gr22 receptor shows high homology with Drosophila Gr21a. Mosquitoes are known to use CO2 as a cue in host finding. In a simple orientation assay, Drosophila avoids a wide range of CO2 concentrations. The response reaches saturation at 0.3%, a dose well below anesthetic levels which generates 50 spikes/s from ab1C neurons. At this dose behavioral responses of GR21a-rpr flies are reduced but not absent. A negative response to elevated CO2 concentrations may serve to avoid noxious levels in fermenting fruits.

Supported by the DFG, Sonderforschungsbereich 515
- de Bruyne, M., Foster, K. & Carlson, J.R. (2001) Neuron 30, 537-552.
- Hill, C.A., Fox, A.N., Pitts, R.J., Kent, L.B., Tan, P.L., Chrystal, M.A., Cravchik, A., Collins, F.H., Robertson, H.M. & Zwiebel, L.J. (2002) Science 298, 176-178.
- Scott, K., Brady, R.Jr., Cravchik, A., Morozov, P., Rzhetsky, A., Zuker, C. & Axel, R. (2001) Cell 104, 661-673.



I. Masante-Roca

Institut National de la Recherche Agronomique, UMR Santé Végétale, Centre de Rercherches de Bordeaux, 33883 Villenave d’Ornon cedex, France
[email protected]

Plant odour processing in the antennal lobe of male and female grapevine moths, Lobesia botrana (Lepidoptera: Tortricidae)

Moths of Lobesia botrana (Lepidoptera: Tortricidae) are confronted with different volatiles emitted from the host plant during the different seasons. To test the hypothesis of a plasticity of central plant odour processing in moths of the different generations in the future, we first investigated the responses of antennal lobe interneurons of laboratory-reared virgin and mated males and females. We used intracellular recording and staining techniques while stimulating the antenna with a range of host and non-host plant odours. The antennal lobe structure of L. botrana is similar to that found in other Lepidoptera species studied. The most frequent responses for all types of moths were obtained with (E)-2-hexenal, and with thujyl alcohol and b-thujone, components of tansy, a behaviourally attractive non-host plant. Some broadly responding neurons were capable to distinguish between different compounds through different response patterns (excitation/inhibition) and/or different dose-response characteristics. Whereas these response patterns did not correlate with specific odours or the sex or mating status of the moths, neurons in unmated males and in mated females responded more frequently than in mated males and unmated females to the tested odours. This indicates a possible effect of mating on plant odour responsiveness of antennal lobe neurons in both sexes in correlation with behavioural changes.



Sandrine Gouinguené and Erich Städler

Eidg. Forschunganstalt für Obst-, Wein-, und Gartenbau, 8820 Wädenswil, Switzerland.
[email protected]

Comparison of the perception of plants surface compounds by three closely related Delia flies.

The majority of insect herbivores are specialised on a limited number of plant species, genus, or families. However, the reason for such specialisation, that narrows the food available in the environment, is still a matter of debate. Bernays (2001) proposed that generalist insects are, compared to specialist, less efficient in their host-plant choice. Thus they are more exposed to predators and parasite, which can reduce their fitness. The more efficient host-plant choice of specialists is supposed to be linked to their higher selectivity and sensitivity to host specific chemicals. The key element of this hypothesis is the specialised sensory physiology and the associated information processing. How can it be tested? We choose to investigate the “behavioural facilitation hypothesis”, proposed long ago by Dethier, that postulates that during evolutionary time changes in the behaviour (choice) of herbivore insects lead to the first step in new colonisation of novel hosts. Since chemoreception is crucial, either the sensory organs or the central nervous system or both must be involved. Evolutionary history and /or present host-plant is expected to have affected the behavioural and chemosensory responses to plant compounds. We are comparing the chemoreception to host plant compounds of related Delia flies, which are specialised on different host plants. In the cabbage root fly (Delia radicum), the egg laying behaviour is dependent on the presence of some specific compounds (glucosinolates and CIF (Fluorene compound)) on the leaves surface of cabbages. The oviposition behaviour of the onion fly (Delia antiqua) is activated by the odour of onion (chopped or attacked my bacteria or/and mould). While the stimuli activating the egg deposition of the bean seed fly (Delia platura) have been less studied, so far in the last two flies, the olfaction has been considered to be the major sensory modality mediating the choice of the ovipositional sites. Contact with the plant surface has not been considered in this respect. We are comparing the sensitivity level of these three closely related flies to both non-volatile and volatile chemicals present on the leaf surface of the host plants. This comparative approach will hopefully shed more light on the evolution of chemoreception in the flies tested.



Wolfgang Rössler

University of Würzburg, Biocenter, Zoology II, Am Hubland, 97074 Würzburg, Germany
[email protected]

Functional organization and plasticity of olfactory glomeruli

Glomeruli are a hallmark of primary olfactory centers in vertebrates (olfactory bulb) as well as in insects (antennal lobe). Various studies suggest that olfactory glomeruli serve as functional units in odor processing. However, only little is known about mechanisms and rules underlying the establishment and plasticity of glomeruli and about the consequences of plastic changes of glomerular organization for olfactory behavior. I will present examples of our recent investigations on insect and vertebrate olfactory systems that address these questions.
Role of sensory axons and glia in organizing glomeruli: In the moth Manduca sexta, axons of olfactory receptor neurons (ORNs) cross a glia-rich axon-sorting zone just before they enter the antennal lobe. Interactions with glial cells in this zone during a particular time window are crucial for correct sorting and targeting of ORN axons and for the establishment of a chemotopic organization of glomeruli.
Potential role of F-Actin in plasticity of glomeruli: In a comparative study including various species of insects and vertebrates we found that, in all cases, filamentous (F-)-actin is strongly aggregated in olfactory glomeruli indicating a potential role of the actin-based cytoskeleton in synaptic and structural plasticity within glomeruli. Interestingly, the distribution of F-actin in axonal and dendritic neuronal compartments appears different between insects and vertebrates.
Phenotypic plasticity of glomeruli: In our recent studies on olfactory communication in two closely related species of leafcutter ants we discovered macroglomeruli in the antennal lobes of worker ants that are most likely involved in processing of information about trail pheromones. 3D-reconstructions revealed striking similarities as well as very distinct differences in the arrangement of macroglomeruli among the two species. Workers of both species express an extraordinary size polymorphism which is solely caused by environmental factors. The difference in body size is correlated with differences in the response to trail pheromones. Our results show that this phenotypic plasticity in olfactory behavior is reflected in distinct differences in the organization of macroglomeruli.
Taken together, the results indicate that glomeruli in primary olfactory centers express a remarkable degree of developmental plasticity which can be shaped by multiple environmental and endogenous factors. The plastic changes in glomerular organization can lead to significant changes in odor-guided behavior.



Martin Schwaerzel1, Maria Monastirioti2, Henrike Scholz3, Florence Friggi-Grelin4, Serge Birman4 and Martin Heisenberg3

1 Division of Biological Sciences, 219 Lefevre Hall, University of Missouri-Columbia, Columbia, MO 65211-6190, USA.
2 Institute of Molecular Biology & Biotechnology, Foundation for Research and Technology, Heraklion, Greece.
3 Lehrstuhl für Genetik und Neurobiologie, Biozentrum, Am Hubland, D-97074 Würzburg, Germany.
4 Institute of Developmental Biology Institute of Marseille, Campus de Luminy - Case 907, F-13288 Marseille Cedex 9, France.
[email protected]

Dopamine and octopamine establish separate memory traces in the same set of neurons in Drosophila

Behavior is organized as a dynamic balance of motivational states such as activity and rest, perseverance and flexibility, attraction and repulsion. The catecholamines play a major role in this regulation. Here we investigate, in the fly Drosophila melanogaster, how appetitive and aversive memory traces of the same odor can invoke different behaviors.

We localize the memory traces to the same set of approximately 700 Kenyon cells in the mushroom bodies of the brain and find that the two memories are distinguished by the requirement for different catecholamines, dopamine for aversive and octopamine for appetitive conditioning. The two memory traces may be located in the same Kenyon cells, but in separate subcellular domains, one modulated by dopamine, the other by octopamine. The study suggests that memories in associative conditioning are linked to the motivational system by their specific modulatory pathways.



Marit Stranden 1, Tonette Rostelien 1,2, Ilme Liblikas 3,4, Wilfried A. König 5, Anna-Karin Borg-Karlson 1,3 and Hanna Mustaparta 1

1 Norwegian University of Science and Technology, Department of Biology, Neuroscience Unit, NO-7489 Trondheim, Norway
2 Gjøvik University College, Faculty of Health Studies, NO-2802 Gjøvik, Norway
3 The Royal Institute of Technology, Department of Chemistry, Ecochemistry, SE-100 44 Stockholm, Sweden
4 Estonian Agricultural University, Laboratory of Ecochemistry, Tartu EE-51005, Estonia
5 University of Hamburg, Institute of Organic Chemistry, DE-20146 Hamburg, Germany
[email protected]

Similar molecular receptive range of five plant odour receptor neurone types in three heliothine moth species.

One major question in olfaction is which volatile chemicals the receptor neurones (RNs) detect. The complexity of the blends and instability of the compounds in the chemical environment of an organism, make the identification of important odorants difficult. In our laboratory we have used gas chromatography linked to single cell recordings (GC-SCR) and to mass spectrometry (GC-MS) to identify odorants in naturally produced plant volatiles that are detected by the receptor neurones in herbivorous insects.

In this study we have examined the RNs in the three heliothine moth species, the polyphagous Heliothis virescens (America) and Helicoverpa armigera (south Europe, Asia, Australia, Africa, eastern Pacific) and the oligophagous Helicoverpa assulta (Asia). Olfactory RNs responding to plant odorants have been classified in 20-30 types, according to the compounds eliciting responses. For five RN types presented here, the odorants have been chemically identified by GC-MS and retested with authentic material. Interestingly, the structure-activity relationships of all five RN types were similar in the three species. On of them, a major type in all three species, is very sensitive for the sesquiterpene (-)-germacrene D, present in leaves and flowers of host and non-host plants (Røstelien et al. Chem Senses 2000, Stranden et al. Chem Senses 2002, J Comp Physiol 2003, submitted). The ranking of the secondary responses to seven other related sesquiterpenes, including (+)-germacrene D with ten times lower effect, was similar in all three species. The other four RN types were co-located in the same sensilla, of which three types responded strongest to the inducible compounds E-b-ocimene, E,E-a-farnesene and E,E-TMTT (4,8,12-trimethyl-1,3,7,11-tridecatetraene), respectively (Røstelien et al. 2000, J Comp Physiol A, Stranden et al. 2003 Chemoecology, submitted). The fourth type responded strongest to geraniol, which is a common floral volatile. All five RN types were narrowly tuned, showing responses exclusively to a few structurally similar compounds out of the hundreds of volatiles present in all plant materials tested. No overlap of the molecular receptive range appeared. These results indicate the presence of similar types of plant odour RNs in the three heliothine moths, and suggest conservation or reappearance of functional similar olfactory receptor proteins, independent of the evolution of polyphagy and oligophagy in these species.



Thomas A. Christensen

ARL Division of Neurobiology, University of Arizona, Tucson, AZ 85721 USA
[email protected]

Olfactory Conditioning in the Moth: Central Representations and Behaviors Evoked by Odor Are Modulated by Stimulus Context

Surprisingly, most contemporary studies of olfactory coding fail to consider the influence of stimulus context on an animal`s perception of ambient odors. In the moth, we have shown that the neural representation of a given odor stimulus observed at the earliest stage of processing in the antennal lobe is not fixed, but strongly dependent on the design of the stimulus protocol. Even though the chemistry of the odor stimulus does not change, different stimulation protocols produce different responses across ensembles of antennal lobe neurons as well as distinct flight behaviors. In this talk I will discuss the need to design a stimulation protocol that matches an animal`s behavioral requirement for olfactory input, and I will also discuss how both the glomerular representations and flight behaviors triggered by even a single odorant are drastically transformed by simple olfactory conditioning. Our findings have important implications for understanding hypotheses that relate patterns of neural synchrony and information coding in the brain. Supported by grants from the NIH.



Menzel, R, Galizia, CG, Müller, D, Sachse, S, Syzszka, P, and Weidert, M,

Neurobiologie, Freie Universität Berlin, Königin-Luise-Str. 28-30, 14195 Berlin, Germany

Odor perception and odor learning in honeybees

Honeybees discriminate a large range of odors, and learn odors as signals for food. Because of these capacities bees are important model-animals for the analysis of the mechanisms of olfactory coding and of memory formation. We have used optical imaging and intracellular recording techniques to examine the neural substrates of odor coding and memory formation in the primary and secondary neuropil of the honeybee olfactory system, the antennal lobe (AL) and the mushroom bodies (MB), respectively. Each AL consists of 160 glomeruli, 40 of which can be imaged for their Ca2+ activity during olfactory stimulation and learning. Local interneurons and projection neurons which link the ALs to the MBs were recorded either optically or electrophysiologically, or both. Behavioral experiments were performed to relate the physiological results to odour perception and learning.

Odors elicit combinatorial patterns of activity in the glomeruli of the AL. These patterns can be visualized using either bath-applied calcium-sensitive dyes or dyes backfilled via particular nerve bundles (projection neurons, Kenyon cells). The individual glomeruli can be morphologically identified according to a 3D atlas. The responses of the postsynaptic elements within the glomeruli allow the reconstruction of network properties of the AL. Some glomeruli respond with a calcium increase during odor presentation (‘on’-glomeruli), others with a calcium increase at stimulus offset (‘off’-glomeruli), and other glomeruli again with a calcium decrease during odor presentation (‘inhibited’ glomeruli). A comparison of odor-evoked glomerular activity patterns in bath-stained preparations and those recorded selectively in projection neurons reveal that the antennal lobe sharpens the spatial activity patterns. Two inhibitory networks can be separated pharmacologically: one globally controls overall activity in the antennal lobe, with each neuron branching off equally into many glomeruli (10-50), while neurons of the other connect single glomeruli to a subset of other glomeruli, thus performing a more glomerulus-specific odor-tuning task.

A systematic analysis of odor discrimination among many odors reveals a close relationship between discriminability and similarity of odor-induced glomerular activity patterns. Odors that are less well-discriminated induce more similar glomerular activity patterns indicating that the spatial combinatorial activity pattern in the AL is a substantial component of the neural basis of odor discrimination.

Electrophysiological recordings from olfactory projection neurons provide evidence for complex patterns of action potential rate increases or decreases at stimulation time. These are indicative of particular odors, and are fully compatible with the calcium responses observed in the imaging experiments. There are two populations of uniglomerular projection neurons. One leaves the AL via a medially located tract (mACT neurons, median Antenno Cerebralis Tract), while the other passes laterally (lACT nerons). lACT neurons arborize in different glomeruli than those of the mACT, and these two populations of glomeruli are also innervated by distinct tracts coming from the antenna. Responses to odors also differ between the two pathways: projection neurons in the lACT have low spontaneous activity, broad response profiles, and response latencies which reflect the stimulus onset, while neurons in the mACT tract have higher spontaneous activity, narrower response profiles, and latencies that are odor-dependent. Since neurons of the two pathways respond to general odors, we conclude that the two parallel pathways code for the same odors but with different properties.

The effect of odor learning can be studied under conditions in which glomerular activity patterns are imaged or neurons are recorded intracellularly. After differential conditioning, the response to the positively reinforced odor is enhanced, while the response to the negatively trained odor is unchanged. The glomeruli involved in both response patterns are not affected, but the correlation between the two patterns decreases, indicating that learning may improve the discriminatory power of the olfactory system with respect to the trained odor.



Erika Plettner , Taraneh Lajevardi , Nicolette Honson , James Inkster

Simon Fraser University, Department of Chemistry, Burnaby B. C. V5A 1S6, Canada
[email protected]

Recovery and gain control in an insect pheromone olfactory system

The pheromone olfactory system of male moths is highly sensitive and responds to a wide range of concentrations. The pheromone detection threshold has been estimated with the silk moth to be ~ 200 molecules/s (Kaissling and Priesner 1970, Naturwissenschaften, 57, 23). The range of detection has been estimated at 8-10 orders of magnitude in concentration, from the threshold to the highest concentration that does not cause complete adaptation (Kaissling 1977, In: Chemical Control of Insect Behavior, pp. 45). In the laboratory, a highly sensitive detector (such as a mass spectrometer) can detect 3-4 orders of magnitude in concentration in a single run. To extend the range, a sample has to be either diluted or the gain control on the detector needs to be adjusted. In order to understand pheromone-mediated behaviour and to adapt pheromones and related compounds for pest control, it is important to understand how the moth is able to respond over such a wide range of concentrations, yet at the same time have such a low detection threshold.

Pheromone olfaction occurs in three stages. 1) The pheromone adsorbs on the antennae and diffuses into hollow sensory hairs via pores in the cuticle. The pores are filled with an aqueous lymph, through which the hydrophobic pheromone is transported by the pheromone-binding protein (PBP). 2) An olfactory receptor on the sensory neuron is stimulated. Whether the pheromone dissociates from PBP and then binds to the receptor or the PBP-pheromone complex stimulates the receptor is not known. 3) Pheromone is deactivated by specific enzymes. We have focussed on the first and third stages to gain insights into the molecular basis for the high sensitivity and wide range of detection in pheromone olfactory systems.

We have studied the interaction between various ligands and PBPs. During competition assays we observed that the addition of certain ligands significantly increases the binding affinity of the PBP for the pheromone. Furthermore, when the concentration of ligand becomes high relative to the concentration of PBP, the binding constant increases significantly. These observations can only be explained by invoking ligand-mediated "head-to-tail" multimerization of the PBP, such that a portion of the ligand becomes trapped in the PBP multimers. This mechanism would be equivalent to an "automatic gain control": at low ligand concentration, little multimerization occurs and most PBP-bound ligand has an exit path; at high ligand concentration, extensive multimerization occurs and most PBP-bound ligand is trapped and unavailable to go onto the next stage of olfaction. We are developing a photochemical assay to study this phenomenon further.

We have also studied the degradation of a behavioural antagonist and of the pheromone in the gypsy moth. Our studies have revealed an unexpected degradation pathway for (7Z) 2-methyloctadec-7-ene. The alkene is hydroxylated at the allylic positions, desaturated to the ketones, converted to a set of esters by a Baeyer-Villiger reaction and further desaturated to give a series of ynoates. A portion of this pathway may also be relevant to the degradation of (7R,8S) epoxy 2-methyloctadecane, the pheromone of the gypsy moth.



Alex Bäcker

Sandia National Laboratories and California Institute of Technology. MC 139-74. Pasadena, CA 91125. USA
[email protected]

Gain control in an olfactory system: Invariance to volatility, priming and a homeostatic balance of excitation and inhibition in the antennal lobe of the locust contribute to concentration invariance

Object recognition requires both specificity, to ensure that stimuli with distinct behavioral relevance are distinguished, and invariance, to ensure that different instances of the same stimulus are recognized as the same under varied conditions (intensity, pitch, position, …). Psychophysical studies show that an odor can be perceived as identical over significant ranges of concentrations. Whether concentration invariance results, at least in part, from low-level neural phenomena rather than cognitive grouping is so far unknown.
I explore the contribution of projection neurons (PNs) in the antennal lobe of the locust, the analog of the vertebrate olfactory bulb, to the recognition of odor identity across concentrations.

I show the following:
A computerized odor delivery system capable of delivering binary mixtures in arbitrary ratios and with arbitrary timecourses selected in real-time.The locust can recognize odors, and shows innate olfactory preferences.

PNs solve the task of encoding both odorant concentration and odorant identity, independently of concentration, in at least three ways. First, by multiplexing information in different response dimensions using a code that involves neuronal identity, spike timing and synchronization across a neuronal assembly. Second, via a phenomenon
of experience-dependent plasticity that contributes to PNs’ invariance to concentration and sensitizes PNs after exposure to an odor at high concentration, contrary to the adaptation exhibited by receptors. Third, a phenomenon of gain control, whereby excitatory and inhibitory responses balance out massive changes in receptor activity as a function of odorant concentration, maintains the average output of PN assemblies within a small dynamic range.

Volatility has hitherto been believed to play a fundamental role in an odorant’s effectiveness. A further mechanism of gain control contributing to keep the activity of olfactory circuits relatively constant across the wide dynamic range of odorant concentrations in the air is the physical chemistry of odorant reception. I hypothesize it confers the olfactory system with invariance to odorant volatility, and advance supporting evidence.

Response patterns sometimes exhibit stable representations over large composition ranges and then abrupt transitions as a function of concentration and mixture composition, suggesting the difference between “same” and “different” odors may be delineated by sharp boundaries in odor space.



Blanka Pophof, Gunde Ziegelberger, Rosario Maida, and Karl-Ernst Kaissling

Max-Planck Instut für Verhaltensphysiologie, D-82319 Seewiesen, Germany
[email protected]

Functional investigation of moth pheromone binding proteins

The sensilla trichodea of the silkmoth Antheraea polyphemus are innervated by three types of receptor cells, each responding specifically to one of three pheromone components. The sensillum lymph surrounding the sensory dendrites in these sensilla contains three different types of pheromone binding proteins (PBPs). Binding and competition assays show that each PBP preferentially binds one of the pheromone components1. The sensilla trichodea of the silkmoth Bombyx mori are supplied with two receptor cells each tuned specifically to one of the two pheromone components bombykol and bombykal, but only one type of PBP was found. Recombinant PBPs of both silkmoth species in various combinations with recombinant pheromone components were applied to the receptor cells via tip-opened sensilla during electrophysiological recordings.

Over a broad range of pheromone concentrations the responses of receptor cells of Antheraea polyphemus depended on both, the pheromone component and the PBP2. For example, an artificial combination of a pheromone component A with PBP B preferentially binding pheromone B, elicited nerve impulses not only in cell A tuned to pheromone A, but also in cell B tuned to pheromone B. Since cell B is insensitive to pheromone A and pheromone B was absent, the response of cell B must have been mediated by PBP B. The response required the presence of pheromone A since PBP B alone was ineffective. This observation suggests a contribution of the PBP to the excitation of the receptor cell and its specificity. The latter was impaired by the artificial pheromone-PBP combination.

In Bombyx mori, bombykol combined with the recombinant PBP of B. mori (rBmorPBP) activated only the bombykol receptor cell. In combination with PBPs of A. polyphemus the same dose of bombykol elicited weaker responses. However, bombykal in combination with the expressed PBP of B. mori failed to activate the bombykal receptor cell, but did so if combined with PBP1 of A. polyphemus. In this combination it even activated the bombykol cell. The failing activation of the bombykal cell by bombykal + rBmorPBP might indicate a lack of binding also observed in electrospray mass spectrometry. It remains to be shown whether an unknown protein solubilizing bombykal exists.

The possible functions of the PBP as carrier and scavenger of the pheromone will be discussed in the light of structural analysis by X-ray3 and NMR4 and modeling of receptor cell responses5. The modeling revealed that the measured binding affinity of pheromone and PBP (60 nM) is much higher than the calculated one of the pheromone and the receptor molecule (40 µM). An estimation of the number of receptor molecules per receptor cell reveals that the plasma membrane of the outer dendritic segment is densely covered by receptor molecules, with a minimum value of 6,000 units per µm2.

1) Maida, R., Ziegelberger, G., Kaissling, K.E. (subm.). 2) Pophof, B. Naturwissenschaften (2002) 89, 515-518. 3) Sandler, B.H., Nikonova, L., Leal, W.S. and Clardy, J. (2000) Chemistry & Biology, 7, 143-151. 4) Horst, R., Damberger, F., Luginbühl, P., Güntert, P., Peng, G., Nikonova, L., Leal, W., Wüthrich, K. (2001) Proc. Natl. Acad. Sci. USA 98, 14374-14379. 5) Kaissling, K.E. (2001) Chem. Senses, 26:125-150.



J.J.A. van Loon1, Lin-er Luo2 and L.M. Schoonhoven1

1 Laboratory of Entomology, Wageningen University, P.O. Box 8031, 6700 EH Wageningen, The Netherlands
2 Laboratory of Biophysics, Department of Biology, Beijing University, Beijing 100871, P.R. China
[email protected]

Chemosensory discrimination between deterrents in Pieris brassicae butterflies

Hostplant recognition by butterflies is mediated by tarsal taste receptors. Females of the large white butterfly, Pieris brassicae L. possess ca. 500 taste sensilla on their legs, each innervated by four contact chemosensory neurons. Electrophysiological studies of these taste neurons in different Pieris species have previously demonstrated that they respond to different classes of plant compounds: sugars, host-plant specific oviposition stimulants (glucosinolates); conspecific oviposition deterrents (contained in egg washes); deterrent compounds occurring in the host-plant family (cardenolides). We document here the response to non-hostplant deterrents (triterpenoids) of sensilla in a lateral row on the fifth tarsomere. Two neurons respond to a pure triterpenoid compound or extracts rich in these compounds whereas they are insensitive to cardenolides. In contrast, sensilla in a medial cluster of sensilla are unresponsive to the triterpenoids but very sensitive to cardenolides, which excite responses from one neuron. We have previously found two types of deterrent chemoreceptor neurones in P. brassicae caterpillars, which enabled these to discriminate deterrent compounds occurring in their host-plant family from deterrents isolated from non-hostplants. The two types of deterrent neurons in the adult female seem to correspond with the two larval types.



Sèverine Jansen, Guillaume Jacquemin, Lina Siauciaunaite, Yohann Gaubard, Christer Löfstedt and Jean-François Picimbon

University of Lund, Department of Ecology, Pheromone Group, SE-223 62 Sweden
[email protected]

Genetics of Odorant Reception and Chemosensation

In insects, the detection of environmental chemical stimuli is mediated by specific olfactory neurons located in specialized cuticular hair structures from the antennae, the sensilla. A cross-section of one antennal sensillum reveals a typical feature, that is dendrites of olfactory neurons expending into a large lymphatic cavity and immerging in an aqueous fluid, the antennal lymph. This lymph may protect the dendritic structures, but separates the receptor neurons from the cuticular pores through which the chemical molecules enter the sensilla. Hydrophobic molecules such as odorants will have to be solubilized in the lymph to access the olfactory receptors. This task may be attributed to a particular class of proteins, the Odorant Binding Proteins, which sieve the lymphatic space. “Proteinaceous fluid”, “hydrophobic chemicals”, “dendritic receptors” and “sensillar pores” are not “keywords”restricted to the olfactory systems. A similar structure is found in generalist contact chemosensory sensillae, suggesting that olfaction and contact chemosensation in the sensillae are mediated by similar mechanisms. In chemosensory sensillae, another class of proteins, the Chemosensory Proteins (CSPs), may transport and deliver stimuli molecules to chemosensory receptors (Picimbon, 2003 and citations therein).
In the course of a study on Evolution and Biochemistry of insect olfactory and chemosensory proteins, we have identified the structures of the CSP and OBP genes in different noctuid species of moths. Before all, we have demonstrated that the genes encoding CSPs and the OBP genes are different in structure. The OBP genes display a conserved motif exon1-intron1-exon2-intron2-exon3, with differences found mainly in intron2. The CSP genes are much shorter and exhibit only one intron sequence. Observed variations in intron sizes and nucleotide compositions in both CSP and OBP genes across species denote the influence of evolution not only on expressed sequences but also on untranslated regions of olfactory and chemosensory genes. In addition, our results suggest that the genes encoding olfactory OBPs and the genes encoding general chemosensory CSPs may have different ancestral genes and may have undergone very specific evolutionary routes. Further work should address Evolution of CSPs and OBPs across the living kingdom of insects that have developed senses to volatile or contact chemicals.

1) Picimbon J.F. Biochemistry and Evolution of OSD and OBP proteins. Chapter 14 In: PHEROMONE BIOCHEMISTRY AND MOLECULAR BIOLOGY new edition. Ed. GJ. Blomquist and RG. Vogt, NY Academic Press, USA, in press.
2) Abraham D. , Löfstedt C. and Picimbon J.F. . Molecular evolution and characterization of GRP1 and GRP2 pheromone binding protein genes in moths. submitted.
3) Vogt R.G. , Rogers M.E. , Franco M.D. and Sun M. . A comparative study of odorant binding protein genes: differential expression of the PBP1-GOBP2 gene cluster in Manduca sexta (Lepidoptera) and the organization of OBP genes in Drosophila melanogaster (Diptera). J. Exp. Biol. 205: 719-744.



Joseph C. Dickens1, Benedict Hollister1 and Bryan Vinyard2

U. S. Department of Agriculture, Henry A. Wallace Beltsville Agricultural Research Center, Plant Sciences Institute, Chemicals Affecting Insect Behavior Laboratory, and Biometrical Counsulting Service Beltsville, MD 20705

Patterning of Gustatory Neural Inputs for Host Selection and Conspecific Recognition in the Colorado Potato Beetle

Spike patterns of neural complements within sensilla on the antennae, galeae, and legs of the Colorado potato beetle were investigated for responses to feeding stimulants, deterrents, and sexual recognition factors associated with elytra. Rather than measuring activity of individual spikes relative to salt (electrolyte), neural responses to all stimuli including salt were characterized by calculating spike distribution patterns in different sensilla for each combination of gender and stimulus. Thus the percentage of activity of each of three chemosensory neurons within a sensillum was calculated from several replicates. Cluster analysis provided r2 values indicative of the level of similarity of neural response patterns for each stimulus for each sensillum type. Significant differences among neural response patterns by gender were determined using Fisher’s exact test. A feeding deterrent was detected by sensilla on all appendages. The highest levels of responsiveness to feeding stimulants occurred in sensilla on mouthparts, and antennae, perhaps prescient to a meal. Sexual recognition (represented by extracts of male and female elytra) occurred through sensilla on the legs of males and females, and antennae of males. These results provide a functional view of the localization and specialization of gustatory sensilla, and a better understanding of input channels important for selection and utilization of resources by the Colorado potato beetle and possibly other insects.



Frederic Mery

University of Fribourg, Unit of Ecology and Evolution, CH1700 Fribourg, Switzerland
[email protected]

Experimental evolution of learning ability in Drosophila melanogaster

The presence of genetic variation for learning ability in animals opens the way for experiments asking how and under what ecological circumstances improved learning ability should evolve. Here we report experimental evolution of learning ability in Drosophila melanogaster. We exposed experimental populations for 51 generations to conditions that we expected to favor associative learning with regard to oviposition substrate choice. Flies that learned to associate a chemical cue (quinine) with a particular substrate, and still avoided this substrate several hours after the cue had been removed, were expected to contribute more alleles to the next generation. From about generation 15 on the experimental populations showed marked ability to avoid oviposition substrates which several hours earlier had contained the chemical cue. The improved response to conditioning was also expressed when the flies were faced with a choice of novel media. We demonstrate that these behavioural changes are due to the evolution of both a higher learning rate and a better memory.
This experiment open the way to study potential fitness cost associated to increased learning ability.



Teun Dekker1,2 and Ring T. Cardé1

1 Department of Entomology, University of California Riverside, CA 92521, USA
2 Division of Chemical Ecology, Department of Crop Science, Swedish Agricultural University, Alnarp, Sweden

Odor-mediated flight in mosquitoes: implications for the filament-response model.

Our understanding of how flying insects use odors to locate the source is based primarily on decades of experiments with male moths orienting to female pheromone. In the filament-response model, single filament encounters form the ‘building blocks’ of the behavioral response, which includes brief upwind surges, followed by expression of counterturning and ultimately casting if no new filament is encountered. Reiterative filament encounters can, at the ‘optimal’ filament-encounter rate, induce almost straight upwind flights in at least some moth species orienting to pheromone. It remains unclear, however, how applicable the filament-response model is for other odors and insects. In the current study we tested the filament-response model for female Aedes aegypti mosquitoes to skin odor and CO2 by exposing the mosquitoes to ribbon, broad turbulent and broad homogeneous plumes of CO2 and skin odor in a laminar flow wind tunnel. Flight tracks were reconstructed in 3-D. Our detailed analysis of the flight tracks reveals that the flight behavior of mosquitoes in response to CO2 is consistent with the filament-response model. In contrast, the flight behavior of mosquitoes in response to skin odor is not modulated from moment-to-moment or from filament-to-filament. Possible explanations for the difference between CO2 and skin odor are discussed.



Daniela Pelz1, Christina C. Roeske1, and C. Giovanni Galizia2

1 Institut für Neurobiologie, FU Berlin, Königin-Luise-Str. 28/30, 14195 Berlin, Germany
[email protected] , [email protected]

2 Department of Entomology, University of California, Riverside, CA 92521, USA
[email protected]

Functional response spectrum of genetically identified olfactory sensory neurons in the fruit fly Drosophila melanogaster

Olfactory sensory neurons (OSNs) provide an animal with information about odors in its environment. In mammals and insects, individual OSNs are likely to express only a single olfactory receptor (OR). OSNs expressing the same OR gene converge onto one or a few spatially invariant glomeruli in the olfactory bulb (mammals) or antennal lobe (AL, insects), the first olfactory processing center in the brain. About 60 OR genes have been identified in D. melanogaster to date. However, only little is known about which odors activate a particular receptor.
In order to fill this gap, we are currently characterizing the response spectrum of OSN populations expressing the same OR gene in great detail. Using the Drosophila UAS-GAL4 system, we create flies expressing Cameleon, a calcium sensitive fluorescent dye, under the control of a particular OR gene promotor. This allows us to monitor the odor-evoked responses of the OSNs expressing this OR gene in vivo by using optical imaging. We have tested a panel of 93 odors from a variety of chemical groups. We are measuring odor evoked responses from the primary dendrites and somata of the OSNs on the antenna, and from their terminals in the glomeruli of the AL. The receptor Or22a had a very broad response spectrum: it responded to a third of the odors tested, a large number of which, like ethyl butanoate or isoamyl acetate, can be found in banana extract. Comparing the response spectrum measured in the AL to that measured on the antenna will allow us to see whether the olfactory response profile in the AL has been modified with respect to the primary response in the OSN dendrites.
Knowledge about the response spectrum of the OSNs is crucial for understanding the mechanisms of olfactory coding.



Helge Ro1 , Dirk Müller2 and Hanna Mustaparta1

1 Norwegian University of Science and Technology, Departement of Biology, Neuroscience Unit, NO-7489 Trondheim, Norway
2 Freie Universität Berlin, Department of Biology-Neurobiology, 14195-Berlin, Germany
[email protected]

Antennal lobe interneurones of heliothine moths involved in processing of plant odour information.

Great emphasis has been given to the processing of olfactory information in the sex pheromone specific macroglomerular complex of the antennal lobes in male moths. The other important olfactory system dealing with plant odour information is far less understood. This is partly due to the scarce knowledge about biologically relevant plant odourants defining functional types of receptor neurones, as well as the larger number of relevant odourants indicated by the number of ordinary glomeruli in the antennal lobe. In H. virescens 64 female and 66 male glomerular units have been mapped in a 3-D atlas (Berg et al. 2002). In this study, we are examining by intracellular recordings the response properties of antennal lobe interneurones in heliothine moths to stimulation with single compounds and mixtures, including key odourants for the receptor neurones (Røstelien et al., 2000a, b; Stranden et al., 2002). Subsequent staining with fluorescent dye have revealed the dendrite arborizations and the projections of the neurones in the calyces of the mushroom bodies and in the lateral protocerebrum, by the use of confocal microscopy. The Amira 3.0 software has been used to create 3-D polygonal surface models from the optical sections. Superimposed onto the map of glomeruli in the 3-D atlas, the arborization pattern of the neurones is intended to give a functional characterisation of identified glomeruli. The aim is to determine whether one glomerulus primarily mediates information about a specific odourant, possibly a set of chemically related odourants. The responses of the interneurones were recorded as an increased firing rate, sometimes preceded by a short hyperpolarization. In some cases the firing outlasted the stimulus period, whereas in others inhibition occurred after the end of stimulation. Some neurones responded by plain inhibition while others showed excitation as well as inhibition, e.g. by an initial short burst of spikes followed by a silent period. Repeated bursts of spikes also occurred during a single stimulus period. Ability to follow pulsed stimulation was found in some units for certain pulse frequencies. Selective responses to odours were obtained for projection as well as some local interneurones, examplified by a local interneurone responding strongly to geraniol and weaker to citral and citronelool, but not to other tested compounds. Interneurones could readily be distinguished from projection neurones based on their spike characteristics. This was verified by confocal microscopy. Stainings of the neurones revealed local interneurones innervating all or nearly all glomeruli, and projection neurones with multi- and uniglomerular projections. Multi- and uniglomerular projection neurones were found in separate antenno-cerebral tracts.

Berg B.G., Galizia C.G., Brandt R., Mustaparta H. (2002)Digital atlases of the antennal lobe in two species of tobacco budworm moths, the Oriental Helicoverpa assulta (male) and the American Heliothis virescens (male and female. J comp Neurol 446:123-134
Røstelien T., Borg-Karlson A. -K., Mustaparta, H. (2000) Selective receptor neuron 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: Sensory, Neural, and Behavioral Physiology 186(9): 833-847.
Røstelien T., Borg-Karlson A. –K., Fäldt J., Jacobsson U., Mustaparta H. (2000) The Plant Sesquiterpene Germacrene D Specifically Activates a Major Type of Antennal Receptor Neuron of the Tobacco Budworm Moth Heliothis virescen. sChem. Senses 25: 141-148.
Stranden M., Borg-Karlson A. –K., Mustaparta H (2002) Receptor Neuron Discrimination of the Germacrene D Enantiomers in the Moth Helicoverpa armigera. Chem. Senses 27: 143-152.



Jürgen Krieger1, Klaus Raming2, Youssef M.E. Dewer1, Oliver Klink1, Sidonie Conzelmann1 and Heinz Breer1

1 University of Hohenheim, Institute of Physiology (230), 70599 Stuttgart, Germany
2 Bayer CropScience AG, Target Research, 40789 Monheim, Germany
[email protected]

Olfactory receptors of the moth Heliothis virescens

Sensing of odor in moths is accomplished by sensory neurons housed in sensillar hair structures of the antennae. These highly specialized cells supposedly recognize and discriminate pheromones and odorants by means of seven transmembrane domain receptor proteins located in the dendritic membrane. Due to their key role in the initial step of odor perception and their proposed potential in insect control, great efforts have been made over the past decade to identify the olfactory receptors in moth, however, their molecular identity remained elusive. We have assessed a genome database of the moth Heliothis virescens for sequences which may represent exons encoding heptahelical proteins related to candidate olfactory receptors. Putative exons sequences were employed as probes to screen an antennal cDNA library of Heliothis. Analysis of isolated cDNA-clones led to the discovery of a divergent gene family encoding putative seven transmembrane domain proteins. Using receptor-specific primers in RT-PCR experiments with different tissues, several subtypes were found to be specifically expressed in the antennae, supporting the notion that they may encode candidate olfactory receptors of the moth. Moreover, in situ hybridization experiments revealed that they are indeed expressed in antennal sensory neurons and demonstrated that each receptor subtype appears to be expressed in a distinct population of sensory cells. RT-PCR analyses of developmental stages revealed an onset of receptor expression as early as five days before eclosion.

Comparing their amino acid sequences, candidate olfactory receptors of Heliothis virescens are extreme diverse from those of dipteran species. This general rule has one exception; the moth receptor type HR2 shares a substantial degree of sequence identity (about 70%) with the olfactory receptor types Dor83b from Drosophila and AgGPRor7 from Anopheles. Moreover, this unique receptor type was found to be expressed in a large number of antennal neurons, in contrast to all other receptors. We have identified HR2 homologues in two additional lepidopteran species, the moths Antheraea pernyi and Bombyx mori. Moreover, RT-PCR-based experiments led to the discovery of HR2 homologues also in antennal cDNA of the honey bee (Apis mellifera; Hymenoptera), the blowfly (Calliphora erythrocephala; Diptera) and the mealworm (Tenebrio molitor; Coleoptera). Comparison of all HR2-related receptors revealed a high degree of sequence conservation across insect orders. In situ hybridization with tissue sections from bee and blowfly antennae confirmed that R2-type receptors are generally expressed in a very large number of antennal cells. Both, a high degree of sequence conservation and an expression in numerous sensory cells suggests a special role of this unique receptor subtype.

This work was supported by the Deutsche Forschungsgemeinschaft and the BayerCrop Science AG.



Kei Ito1,2,3, Takeshi Awasaki1,2,4, Nobuaki Tanaka1,2,5 and Ryuichi Okada1,2,3

1: Institute of Molecular and Cellular Biosciences, Univ. Tokyo, 2: National Institute for Basic Biology, 3: BIRD, JST, 4: PRESTO, JST, 5: Graduate Univ. for Advanced Studies

Anatomy of local and projection neurons in the Drosophila antennal lobe

Insects detect airborne odorants by the olfactory receptor neurons (ORNs) on the antennae. The ORNs send their axons to the first-order olfactory centre of the brain: the antennal lobe (AL). ORNs expressing the same type of olfactory receptor project to the same glomerulus. The distribution of the glomeruli activity in the antennal lobe thus represents the odourtypic map of the currently active types of ORNs. For the processing of a particular odour, the brain must compare and integrate the level of activities across a large number of glomeruli.

To understand the neural basis of this process, we systematically identified the neurons that receive olfactory information in the AL. Through the screening of 3939 GAL4 enhancer-trap strains, we found several lines that label AL local neurons (LNs) and projection neurons (PNs) rather specifically. Single-cell labeling of LNs revealed that there are two types of LNs. 1: Those that send processes to both core and the periphery regions of the glomeruli and 2: those that innervate only the core of glomeruli. Both types of LNs each innervate all or a very large subset of glomeruli. The combination of fluorescent in situ hybridisation (FISH) and GFP-immunostaining revealed that essentially all these LNs are GABAergic and have GABA receptors. It is likely, thus, that the main function of LNs is to spread inhibitory signals across glomeruli.

There are two major pathways of PNs: iACT and mACT. Most PNs running via the iACT contribute to a single glomerulus. They have GABA receptors but are not GABAergic. Almost all the PNs running via the mACT, on the other hand, are GABAergic. They receive information from either one or a large subset of glomeruli. Among the two second-order olfactory centres, the lateral horn receives the combination of both excitatory and inhibitory signals from the AL. The olfactory input to the mushroom bodies, on the other hand, is essentially limited to the excitatory signals.



Jean-Christophe Sandoz

Centre de Recherches sur la Cognition Animale, Université Paul-Sabatier, 118, Route de Narbonne, 31062 Toulouse cedex 04, France.
[email protected]

Calcium responses to queen pheromones, social pheromones and plant odours in the antennal lobe of the honey bee drone Apis mellifera L.

The functional role of honey bee males (drones) in the colony is to mate with queens, a behavior relying heavily on the olfactory detection of queen pheromone. How the brain of drone honey bees processes olfactory stimuli, in particular pheromones, is still largely unknown. Anatomically, olfactory receptor cells located in sensillae on the antennae project to the antennal lobes (AL), the first brain structures to process olfactory information, forming a number of functional units, the glomeruli. In drones, there are about 100 glomeruli and an additional 4 large compartments which form the macroglomerular complex (MGC). Because in moths the MGC is responsible for sexual pheromone detection and processing, such a function was also proposed in drones. One of these compartments (MG1) is particularly prominent at the surface of the AL, together with about 30 usual-size glomeruli, which makes them accessible for optical imaging studies. Using calcium imaging, we measured odor-evoked responses in the glomeruli of the drone AL. Seventeen different odors belonging to three main classes of stimuli were presented: (i) queen pheromonal components, used by drones for the recognition of queens during nuptial flights, (ii) social pheromonal components used for social cohesion in the colony, and (iii) floral odors, which are present in the food stores of the hive and/or brought back by foragers. All three classes of stimuli produced signals in a variety of glomeruli of the antennal lobe. Queen pheromonal compounds, in particular 9ODA and 9HDA induced signals in the MGC, whilst HOB and HVA each triggered activity in only one, but not the same, ordinary glomerulus. Social pheromones and floral odors evoked responses in a combinatorial manner in ordinary glomeruli. This work suggests that the most active queen pheromonal components are processed in the macroglomeruli of the drone AL, and that social pheromones and floral odors, as well as other queen pheromone components are processed in ordinary glomeruli. These results are discussed in relation to the behavior and role of drones in the honey bee society, as well as with regards to the evolution of sexual signals in social insects.



Leslie P. Tolbert, Lynne A. Oland, Mark Higgins, Nicholas J. Gibson, and Eric Tucker

ARL Division of Neurobiology, University of Arizona, Tucson, AZ, USA 85721
[email protected]

Reciprocal interactions between neurons and glial cells are essential in the developing olfactory system of Manduca sexta

The olfactory systems of certain insects serve as excellent model systems for study of the development of neural circuits specialized for processing odor inputs. Recent findings have begun to unravel the complex cellular and molecular underpinnings of key interactions among neurons and glia cells at several important steps in olfactory development.

We use Manduca sexta to probe these interactions. In metamorphosing M. sexta, where we can block mitosis during glial proliferation to produce animals deficient in glial cells but with normal numbers of neurons, we have found that glial cells have essential roles both in the sorting of ORN axons into fascicles destined for particular target glomeruli and in the creation of those glomeruli during development of the adult antennal system. Glial cells are induced by axons to proliferate to populate the base of the antennal nerve, where the ORN axons enter the antennal lobe of the brain. Interaction of subsequently ingrowing axons with these "sorting zone" glial cells is required for the axons to shed their neighbor-neighbor relationships established in the antenna and to sort into new fascicles likely to have olfactory specificity. Once in the antennal lobe, the growing ORN axons form the templates for glomeruli, which antennal-lobe glial cells migrate to surround. These "border" glial cells in turn help to constrain the branching ORN axon terminals, as well as the dendrites of antennal-lobe neurons, to the developing glomeruli, where they form synapses.

We recently developed methods for co-culturing explants of developing antenna with glial cells derived from the antennal nerve, the sorting zone, or the antennal-lobe neuropil. In time-lapse video microscopy of living cultures, we found that after as little as a single filopodial contact with a centrally derived (sorting-zone or antennal-lobe) glial cell, ORN growth cones often flatten, become highly complex, and pause their growth for many hours. In contrast, contact with glia from the antennal nerve typically causes growth cones to turn and grow along the glial cells. Also, antennal-nerve glial cells divide robustly in culture (unlike their brain-derived counterparts) and, with time, line up to form cellular chains, sometimes along growing axons. When compared with the behavior of growing receptor axons and glial cells in situ, these findings suggest to us that the brain-derived glial cells have special properties that are important for axonal sorting and termination, while the glial cells of the antennal nerve more likely serve as simple struts for growth of late-ingrowing axons. To increase the number of growth cones that can be analyzed, we studied growth cone complexity in cultures fixed and labeled for f-actin and microtubules. These studies corroborated the live-cell studies, revealing a statistically significant increase in morphological complexity of growth cones when they encounter sorting-zone or antennal-lobe, but not antennal-nerve, glia. We hypothesize that this is a direct reflection of the change in adhesive properties that ORN axons exhibit as they grow through the sorting zone in situ and as they subsequently become constrained to branching within single glomeruli.

One likely candidate for a signal between ORN axons and glial cells is nitric oxide (NO). ORN axons express NO synthase during appropriate stages of development, and when NO signaling is prevented, peripheral glia do not migrate down the antennal nerve as usual, and central glia do not migrate to surround developing glomeruli. We plan to use the in vitro co-culture system to explore specific biochemical mechanisms by which NO may exert its effect.



Helena Bichão1,2, Atle Wibe3, Albert Steen, Anna-Karin Borg-Karlson4 and Hanna Mustaparta1

1 Norwegian University of Science and Technology, Department of Biology, Neuroscience Unit, NO-7489 Trondheim, Norway
2 Évora University, Centre for Applied Ecology, Apartado 94, 70002-554 Évora, Portugal
3 The Norwegian Centre for Ecological Agriculture, N-6630 Tingvoll, Norway
4 The Royal Institute of Technology, Department of Chemistry, Ecological Chemistry, SE-100 44 Stockholm, Sweden
[email protected]

Classification of olfactory receptor neurones according to response specificity in the strawberry blossom weevil Anthonomus rubi (Coleoptera, Curculionidae)

Identifying the plant produced volatiles that have a biological relevance for insect species is a central question in insect olfaction. In the search for compounds that contribute to the perception of host odour quality in the strawberry blossom weevil (Anthonomus rubi Herbst), we have screened the antennae for sensitivity to naturally produced plant volatiles and synthetic standards. We used gas chromatography linked to electro-antennogram recordings (GC-EAG) as well as single cell recordings (GC-SCR).

The GC-EAG experiments showed sensitivity of the antenna to ten different compounds in the headspace mixture of strawberry (Fragaria ananassa), identified as aliphatic esters, ketones and aldehydes. The GC-SCR increased the number of active compounds to 18, showing responses also to aromatic alcohols and esters, monoterpenes and sesquiterpenes.

We here present results showing that olfactory neurones respond selectively to one or a few of the numerous compounds released by host and non-host plants. All the neurones exhibited high selectivity and were classified into distinct types according to the compound eliciting best responses. Except for two pairs, the neurone types showed no overlap of the molecular receptive range. Responses were found for both major and minor components of the host blend mixture. Chemical identification of the active compounds using gas chromatography linked to mass spectrometry is in progress. The active compounds are present in the host F. ananassa as well as in non-hosts, supporting the idea that plant odour quality is mediated by the ratio of several compounds rather than specific odorants present in certain plant species.



Philip L. Newland

School of Biological Sciences, University of Southampton, Bassett Crescent East, Southampton SO16 7PX, U.K.
[email protected]

Gustatory Processing in the Locust Central Nervous System

Finding the right kinds of food and in sufficient quantity is one of the most basic behaviours an animal must accomplish to survive. The sense of taste, or contact chemoreception, is a vital component in this process, since for all animals chemicals have to be detected, encoded in the central nervous system, processed and acted upon. At higher levels tastes have to be remembered and recognised so we do not eat things that are unpleasant or toxic. Despite many studies on insect taste receptors we still know little of where chemosensory neurones project in the central nervous system, what central neurones process chemosensory signals and how chemical identity is encoded in neuronal circuits.

In the desert locust, Schistocerca gregaria, taste receptors, called basiconic sensilla, are located on the mouthparts and body, and over all the surfaces of the legs. These sensilla are bimodal and on the leg are innervated by four chemosensory and one mechanosensory neurone, that project to their local thoracic ganglion. Surprisingly, both mechanosensory and chemosensory neurones from individual sensilla project to the same region of neuropil, showing no modality or sensitivity specific segregation of neurites. Moreover, the central projections of these receptors are organised somatotopically such that the spatial location of a sensillum on the leg is preserved in the relative location of the arborisations of it’s sensory neurones in the central nervous system.

Chemosensory neurones make convergent monosynaptic connections with the same spiking local interneurones in the thoracic ganglia that receive mechanosensory sensory inputs from the leg. This organization and pattern of connections leads to overlapping chemosensory and mechanosensory receptive fields of spiking local interneurones. We have developed a behavioural assay of chemosensory responsiveness that is mediated by the basiconic sensilla on the legs, in which the probability of a locust withdrawing its leg from a droplet of a chemical solution applied to the hind leg is measured for concentration series of various chemicals. All tested chemicals, even nutrients, can elicit withdrawal responses but critically the concentration at which different stimuli are effective stimuli in eliciting the behaviour varies over several orders of magnitude. The relative size of the response of spiking local interneurones to these same solutions, as well as their outputs onto leg motor neurones, closely correlate with the probability of eliciting a withdrawal response. We suggest that chemosensory processing in the thoracic ganglia directly assesses a specific chemosensory quality, that of aversiveness, rather than individual chemical identities and concentrations.

This work was supported by awards from the BBSRC (UK) and The Royal Society.



Patrizia Muroni, Maria Laura Scorciapino, Isabella Urru, Maria Dolores Setzu, Iole Tomassini Barbarossa and Anna Maria Angioy

Department of Experimental Biology - Section of General Physiology, University of Cagliari, S.P. Monserrato-Sestu Km 0.7, 09042 Monserrato-Cagliari
[email protected]

Distinct phases of memory for habituation to different training patterns of olfactory stimulation in the blowfly.

Cardiac responses to olfactory stimulation sensitively indicate insect reactivity to odours. In blowflies, repetitive stimulation produces response habituation and short-term memory of olfactory information.

In the present investigation we studied the dynamics of acquisition and storage of memory for habituation on intact blowflies, as a function of cardiac response threshold and training pattern.

By recording electrocardiograms on adult, restrained Protophormia terraenovae flies, we monitored cardiac responses to olfactory stimulation with a concentration series of 1-hexanol vapours. Cardiac response threshold was measured in each specimen, and the corresponding odour concentration was adopted for inducing habituation. Several training protocols consisting of various numbers of stimulations and different interstimulus intervals (ISI) were tested for investigating habituation and memory temporal dynamics.

Independently of the training pattern adopted, the highest percentage of habituated flies showed the lowest cardiac response threshold. Habituation was retained for the shortest duration (< 2 hours) after a stimulation training at 20s ISI. By increasing the latter to 1 min, habituation was detected 7 hours after a training of six stimulations. Flies showed a maximum of 3-hour memory when a series of three stimulations was performed at 1 min ISI. On the other hand, habituation was still retained 24 hours after a training of four stimulations at 30 min ISI.

Results show that the greater the fly sensitivity, the more learning is induced. Besides, greater sensitivity was found to be decisive in determining the longest period of memory retention. Finally, training patterns play a major role in inducing distinct memory phases, massed- and spaced training sessions producing short/medium-term and long-term memory retention respectively.

Supported by MIUR (PRIN – COFIN 2001 to A.M. Angioy)



Dagmar Malun 1, Arnim Jenett 2 and Ulrike Schröter 3

1 Neurobiologie, Institut für Biologie, Freie Universität Berlin, 14195 Berlin, Germany
2 Institut für Genetik und Neurobiologie, Biozentrum am Hubland, Universität Würzburg, 97074 Würzburg, Germany
3 Visual Sciences Group, Research School of Biological Sciences, Canberra ACT 2601, Australia
[email protected]

Developmental and adult expression of Octopamine-like-immunoreactivity in the brain of the honeybee, Apis mellifera.

The transmitter and neuromodulator Octopamine (OA) has been shown to play a role in reinforcement processing of olfactory conditioning in honeybees.(Hammer and Menzel, 1998, Learn & Memory 5:146-156). OA-like immunoreactivity has been found to be widely spread in the brain and suboesophageal ganglion of adult honeybees (Kreissl et al., 1994, J Comp Neurol 348:583-595). In order to elucidate the contribution of OA-immunoreactive processes in the developing olfactory pathway, we examined the expression of OA-like immunoreactivity in the antennal lobe, the mushroom body, and the lateral protocerebrum of early to mid pupal stages (prepupal stage to pupal stage 5). Within this time window, compartmentation of the mushroom body calyx neuropil and the antennal lobe neuropil takes place (Schröter and Malun 2000 J Comp Neurol 422:229–245). The pattern of anti-OA-staining of pupal brains was then compared with that of the adult brain. Furthermore, we tried to allocate OA-immunoreactive fibres with dye-labelled process of morphological and physiologically identified neurons described as putatively containing OA. Examination of brains stained for anti-OA immunoreactivity and of brains containing dye-labelled neurons was performed in whole munts and in vibratome sections using a confocal laser scanning microscope. Subsequently, reconstructions of neuronal fibres were carried out with the Amira software (Indeed, Berlin).

OA-immunoreactivity was detected in the neuropils of the olfactory pathway as early as prepupal stage. In the mushroom body calyx neuropil, with proceeding development the pattern of anti OA-staining became distinguishable between the three calyx compartments, the lip, the collar and the basal ring. Within the antennal lobe, OA-immunoreactive processes were first homogenously distributed in a peripheral layer of the antennal lobe neuropil. Processes then became segregated into glomerular arborizations indicating the formation of olfactory glomeruli. The antenno-cerebralis tracts (ACTs) projecting from the antennal lobe to the mushroom body calyces and the lateral protocerebrum contained OA-immunoreactive processes at early pupal stages. Also, from early pupal stage on, OA-immunoreactive somata and their primary neurites were found in the ventral median part of the suboesophageal ganglion. Taken together, these data indicate an early contribution of OA-immunoreactive processes to the olfactory pathway. A comparative approach of OA-immunoreactivity pattern in the adult bee with individually fluorescent marker stained neurons revealed, that among the OA-immunoreactive neurons are most likely the VUMmx1 neuron (ventral unpaired median neuron of the maxillary neuromere) and the VCBN neuron (neuron with a neurite in the ventral cell body neurite tract).



Pål Kvello, Tor Jørgen Almaas and Hanna Mustaparta

Norwegian University of Science and Technology, Department of Biology, Neuroscience Unit, NO-7489 Trondheim, Norway

Taste sensilla on the proboscis and the projection pattern of the associated receptor neurones in the SOG of Heliothis virescens.

Stimulation of the taste sensilla with sucrose elicits extension of the proboscis in nectar feeding insects, including Heliothine moths. This proboscis extension reflex ( PER) is used to study appetite learning, where sucrose stimulation represents the unconditioned stimulus and odours the conditioned stimulus (Menzel & Müller 1996). In trying to determine the neuronal connection between the taste and the olfactory pathways in Heliothine moths, we are studying the projections of the taste receptor neurones in the central nervous system. We here present the morphology of the taste sensilla (s. styloconica) on the proboscis and the projections in the CNS of the associated receptor neurones. The morphology of these sensilla were studied by light and electron microscopy (SEM, TEM). The sensilla are 80mm long and 15mm in diameter. In cross-section they have a hexagonal, star-like shape and the lumen contain 3 or 4 dendrites enveloped by a dendritic sheath made by one of the supporting cells. The tip of the sensillum is protruded and have a single pore allowing the taste chemicals to enter the wall. The receptor neurones were stained with tetramethylrhodamindextran and the projections were viewed in a confocal laser scanning microscope. The stained axons of the receptor neurones were traced to the suboesophageal ganglion (SOG). Here they entered the SOG via the middle of the three nerve pairs and showed an extensive arborizations in the ipsilateral neuropil. A few fibers also showed projections dorsally in the contralateral SOG.



Mark Stopfer1,2, Vivek Jayaraman2, and Gilles Laurent2

1 National Institute of Health, NICHD, Bethesda, MD, USA
2 California Institute of Technology, Pasadena, CA, USA
[email protected]

Spatiotemporal codes for odor identity and concentration in the locust

We examined the encoding and decoding of odor identity and concentration by neurons in the first and second relays of the locust olfactory system. We found that both odor identity and odor concentration are represented as spatiotemporal patterns of activity across the population of projection neurons (PNs) in the first relay, the antennal lobe. These distributed patterns can be classified by similarity as concentration groups contained within larger odor identity clusters. Thus, the potential confounds between the coding of odor identity and concentration that appear within individual PN responses are resolved by examining responses across groups of PNs. When observed as functions of time, the high-dimensional PN activity patterns can be described as families of trajectories lying on low-dimensional manifolds. Each manifold represents odor identity and contains trajectories representing different concentrations. The spatiotemporal activity patterns are then decoded piecewise over very short time periods (single oscillatory cycles) and are compressed into highly informative responses about both identity and concentration by downstream neurons (Kenyon Cells) in the second relay (Mushroom Bodies).



Irena Valterová1, Blanka Kalinová1, Lucie Ceganová1,2, Vladimír Ptácek3

1 Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic, Flemingovo 2, 166 10 Praha, Czech Republic
2 Department of Chemistry of Natural Compounds, Institute of Chemical Technology, Technická 5, 166 28 Praha, Czech Republic
3 Department of Animal Physiology, Faculty of Science, Masaryk University, Kotlárská 2, 637 11 Brno, Czech Republic
[email protected]

Chemical communication in Bombus terrestris bumblebees

The life in a bumblebee society is directed by chemical signals produced by individuals of each cast. All life-important activities such as foraging, taking care of the brood, finding the way to the food source and back to the nest as well as mating depend on pheromones to high extent. Thus, the information between individuals is chemically mediated.

We focused on the male marking pheromones and queen’s semiochemicals. The males' marking pheromones have been studied extensively. It was shown that the pheromone is produced in the male‘s labial gland. It serves primarily as an attractant and arrestant for females and as a short-term aphrodisiac for males themselves (Kullenberg et al., 1973). On the other hand, semiochemicals of queens were much less studied. Van Honk et al. (1978) reported that the mandibular glands of young queens contain a sex pheromone eliciting mating behaviour in conspecific males.

We present detailed chemical analyses of the labial gland extracts of males and of several queens’ exocrine glands (mandibular, labial, and Dufour’s glands). While the composition of labial gland extracts in males remains more or less constant during the lifespan, the composition of the labial gland extract of queens changes with their age. Differences were also found in the labial gland extracts between groups of virgins and egg-laying queens.

The supplemented EAG experiments showed that all casts, e.g. males, queens and workers, perceive male labial gland volatiles equally. The male antennal sensitivity to the male marking pheromone remains unchanged during the lifetime.

Queen labial gland extract gave significantly higher EAG responses in all casts studied than mandibular and Dufour glands. Workers responded to queen labial gland extracts more than males (two different age categories). The EAG active compounds found in queen labial gland were identified as series of methyl- and ethyl esters of fatty acids C10-C18 and the diterpenic alcohol geranylcitronellol.

The financial support of the Grant Agency of the Czech Republic (grant No. 203/02/0158) is gratefully acknowledged.

Kullenberg B.: Zoon Suppl. 1973, 1, 31-42.
van Honk C. G. J., Velthius H. H. W., Röseler P.-F.: Experientia 1978, 34, 838-839.



Wynand van der Goes van Naters, Anna Dobritsa, Coral Warr, Elissa Hallem, Aaron Goldman, Derek Lessing, Anandasankar Ray, Takashi Shiraiwa, Anupama Dahanukar, Jennifer Perry, and John R. Carlson

Yale University, Dept Molecular Cellular and Developmental Biology, PO Box 208103, New Haven CT 06502-8103, USA
[email protected]

Expression and function of chemosensory receptors in Drosophila

Odour response specificities of olfactory receptor neurons (ORNs) may be dictated solely by the receptors they express or, alternatively, by an ensemble of receptor and peri-receptor molecules. A family of at least 60 G-protein coupled receptor (GPCR) genes, discovered in Drosophila melanogaster, was proposed to encode odour receptors. Investigation of one olfactory receptor gene, Or22a, has shown that it is expressed by a subset of ORNs in the antenna and that these ORNs are uniform in their odour response specificity. These ORNs, but not others in the antennae, lack odour sensitivity entirely in a deletion mutant missing Or22a; rescue experiments whereby the receptor is supplied on a transgene fully restore the odour response. Ectopic expression of a different receptor gene, Or47a, or even a gene from a different Drosophila species endows the mutant neuron with the odour response profile of the donor. Taken together these results suggest the possibility that, at least for some ORNs in Drosophila, differences in odour response specificities arise solely from the expression of different receptor proteins. Extending the analysis from the antennae, we have mapped a number of individual receptors to the neurons that express them in the maxillary palps. We are also investigating the molecular mechanisms by which neurons select the receptor proteins they express and have identified a DNA motif, the “MP Dyad”, which lies upstream of all maxillary palp Or genes and which appears necessary for their selection by maxillary palp ORNs. A second family of G-protein coupled receptor genes identified from the genome database was proposed to encode gustatory receptors. This Gr gene family is similar in size to the Or gene family suggesting that the “gustatory world” of flies may rival the breadth of olfactory perception. We have shown that one member of this family, Gr5a, is necessary for the reception of the sugar trehalose and are investigating the specificity of the receptor protein. For both taste and olfaction, we are now trying to identify the protein domains likely to be involved in ligand-binding. Our studies contribute to an understanding of how information enters the two chemosensory systems.



Yelena Fishilevich and Leslie B. Vosshall

The Rockefeller University, Department of Neurogenetics and Behavior, 1230 York Avenue Box 63, New York, NY 10021 USA
[email protected]

Mapping the Molecular Logic of the Maxillary Palp

Drosophila melanogaster possesses two olfactory organs: the third antennal segment and the maxillary palp. The contribution of these two organs to olfactory-guided behavior is currently unknown. We have taken a genetic approach to compare the patterns of connectivity of antennal and maxillary palp neurons as an initial step toward understanding the logic of olfactory coding in these distinct sensory organs. At the periphery, olfactory specificity is likely determined by the distributed expression of Drosophila odorant receptor (DOR) genes in individual olfactory sensory neurons (OSNs). To understand the relationship between DOR gene expression in the periphery and the organization of OSN projections into the antennal lobe (AL), we used a genetic approach to trace axonal connections of antennal and maxillary palp neurons expressing a given DOR gene. Among the 43 glomeruli located in the AL, we have identified those receiving input from the maxillary palp neurons. The results show that two distinct strategies of wiring the different olfactory organs have evolved. Antennal neurons expressing a given DOR gene show segregated projection into the AL. However, maxillary palp neurons show a convergent projection strategy between “functionally” different populations of neurons. These studies represent a starting point to understand the biological significance of these differences within the context of the function of the maxillary palps in odor recognition and discrimination of the fly.



Troy Zars, Martin Schwaerzel

Division of Biological Sciences, University of Missouri, 114 Lefevre Hall, Columbia, MO 65211, USA.
[email protected]

Extinction of olfactory memories in Drosophila.

Everyday experience tells us that some memories are more stable than others. That the more important memories are maintained indicates that selective memory loss is regulated. It is obvious, therefore, that to completely understand what a memory is, one must include mechanistic investigations of memory loss. Memory extinction is one memory decrement phenomenon especially well suited for investigation as it is rapidly induced, has interesting temporal dynamics, and may have implications for treatment of traumatic memories. Little is known about the neural systems or molecular mechanisms of memory extinction.

Experimental extinction can be induced, after an animal has learned the association of a conditioned stimulus with an unconditioned stimulus, by presenting again the conditioned stimulus alone (Bouton, 1993; Bouton et al., 1999). This leads to a decrease in response toward the original conditioned stimulus. For example, if an animal has learned the association of an odorant with electric shock, presentation of that odorant in the absence of the electric shock will decrease that animal’s avoidance behavior of that odorant.

Our recent discovery that extinction of olfactory memories in Drosophila occurs within the same cells in which memory is stored (the Kenyon cells of the mushroom bodies) provides a unique opportunity to address the molecular mechanism of memory extinction within a circuit (Schwaerzel et al., 2002; Zars et al., 2000). We have determined that extinction of a memory is an intracellular process that antagonizes the molecular mechanisms of memory formation. The cAMP / PKA 2nd-messenger cascade is important for synaptic plasticity and memory formation (Byrne and Kandel, 1996; Lechner and Byrne, 1998). We propose that just as memory formation is thought to rely on the increase in synaptic strength between two neurons, extinction involves the weakening of synaptic strength by antagonizing cAMP / PKA signalling (Schwaerzel et al., 2002).

Current work focuses on the fine details of stimulus presentation and timing in extinction of olfactory memories. In addition, the roles of candidate molecules (e.g. calcineurin) in memory extinction are being tested. Recent results will be presented.

Bouton, M. E. (1993). Context, time, and memory retrieval in the interference paradigms of Pavlovian learning. Psychol Bull 114, 80-99.
Bouton, M. E., Nelson, J. B., and Rosas, J. M. (1999). Stimulus generalization, context change, and forgetting. Psychol Bull 125, 171-186.
Byrne, J. H., and Kandel, E. R. (1996). Presynaptic facilitation revisited: state and time dependence. J Neurosci 16, 425-435.
Lechner, H. A., and Byrne, J. H. (1998). New perspectives on classical conditioning: a synthesis of Hebbian and non-Hebbian mechanisms. Neuron 20, 355-358.
Schwaerzel, M., Heisenberg, M., and Zars, T. (2002). Extinction antagonizes olfactory memory at the sub-cellular level. Neuron 35, 951-960.
Zars, T., Fischer, M., Schulz, R., and Heisenberg, M. (2000). Localization of a Short- Term Memory in Drosophila. Science 288, 672-675.



Marcus C. Stensmyr, Teun Dekker and Bill S. Hansson

Division of Chemical Ecology, Department of Crop Science, Swedish University of Agricultural Sciences, P.O. Box 44, SE-23053 Alnarp, Sweden.
[email protected]

Evolution of the olfactory code in the Drosophila melanogaster subgroup

The Drosophila melanogaster subgroup has been the focus of numerous studies regarding evolution. Here we address the question of how the olfactory code has evolved among the nine sister species. By using in vivo electrophysiological measurements, so called single-cell recordings, we have established the ligand affinity of a defined subset of olfactory receptor neurons (ORNs) across all nine species. We show that the olfactory code as relayed by the investigated subset of ORNs is conserved to a striking degree. Distinct shifts in the code have only occurred within the simulans clade. However, these shifts are restricted to an altered tuning profile of the same single ORN type in all three of the simulans siblings and a more drastic change unique to D. sechellia, involving a complete loss of one sensillum type in favor of another. The alterations observed in D. sechellia may represent a novel host-specific adaptation to its sole host, morinda fruit. The overall high degree of similarity of the code within the subgroup is intriguing when considering the great variety in distributions as well as in habitat and host choice of the siblings, factors that could greatly affect the olfactory system.



Störtkuhl, K.F., Kettler, R., Hovemann, B.T.

Molekulare Zellbiochemie, Ruhr Universität Bochum, Gebäude NC 5 /173, 44780 Bochum, Germany

Misexpression of Drosophila olfactory receptor Or43a changes olfactory behavior

We investigated olfactory processing in the antennal lobe by using olfactory receptor Or43a that has been functionally described recently (Störtkuhl & Kettler 2001). In these tests benzaldehyde was identified as ligand for this receptor. Using green fluorescence marker protein we identified DA4 as target glomerulus in the antennal lobe for Or43a expressing olfactory receptor neurons (ORNs)(Störtkuhl et al., submitted).

Benzaldehyde is a volatile substance that provokes a repellent behavior reaction in the T-maze. We wanted to know if expression of Or43a in additional ORNs could change this escape reaction. Since misexpression of Ors has no influence on the projection pattern of ORNs into the antennal lobe, we tested olfactory behavior of GH320/UAS-Or43a flies.

According to our assumption, in line GH320/UAS-Or43a application of benzaldehyde should result in the activation of additional glomeruli that are connected with Or43a misexpressing ORNs. In that case in which glomerulus DA4 is part of a neuronal network dominantly governing the escape reaction to benzaldehde, the response index in the T-maze assay should not be affected and equal the response of wild type flies. Interestingly, GH320/UAS-Or43a flies performed a reduced response index T-maze assay as compared to wild type. The repelling odorant became more attractive through the activation of additional glomeruli in the antennal lobe. We compare this reaction with the application of a mixture of benzaldehyde and the attractive odorant ethyl acetate in the T-maze. Again the response index was reduced while application of benzaldehyde alone generated elevated repelled behavior indexes. We conclude that during information processing DA4 has no dominant function in the antennal lobe.

Störtkuhl, K.F. and Kettler, R. (2001) Functional analysis of an olfactory receptor in Drosophila melanogaster. PNAS 98: 9381-9385
Störtkuhl, K.F., Pargmann, D., Fischer, S., and Hovemann, B.T. (2003) Drosophila olfactory receptor Or43a is expressed in basiconic and coeloconic sensilla and targets glomerulus DA4 (Chem. Senses submitted)



Wilfried A. König1, Detlev H. Hochmuth2, Simla Basar1 and Melanie Junge1

1 Institut für Organische Chemie, Universität Hamburg, Martin-Luther-King-Platz 6, 20146 Hamburg;
2 Thetis-IBN GmbH, Notkestrasse 85, 22607 Hamburg
[email protected]

Strategies for the Identification of known and unknown Plant Constituents

Plants usually produce a great variety of volatile metabolites. In most cases the rationale for the generation of such a complexity of compounds is not understood. However, in many cases it was proven, that some of these volatiles play an important role in interspecies communication. Even minor constituents which are difficult to identify may exhibit high activity in the chemical communication system. We have established a mass spectral data bank with presently almost 1.800 entries for a quick screening of plant volatiles by semi-automatic comparison of mass spectral patterns in conjunction with retention indices on a non-polar stationary phase. The GC-MS data of all reference compounds were acquired under identical GC and MS parameters in order to guarantee an optimum of reproducibility and reliability. Of course this tool can only serve for the identification of known constituents in a complex mixture, but this includes the bulk of all plant volatiles, which occur redundantly in essential oils and plant extracts. Unknown constituents are readily recognized and selected for isolation and NMR investigations by comparing their mass spectra and retention indices with the reference compounds of the data bank. For the isolation of unknowns we apply preparative chromatographic methods including GC, HPLC and TLC. Preparative GC is very efficient when thick-film capillary columns are employed. TLC isolations can be supported by using silver ion impregnation and particularly by running the TLC at very low temperature.

For the investigation of the structure and relative configuration of chiral compounds the use of NMR methods is indispensible. In addition, chemical microreactions such as hydrogenation, dehydration, oxidation or rearrangement reactions can be carried out and unknown structures can be correlated with known compounds. The absolute configuration, a crucial property for the activity of a molecule, is investigated by enantioselective GC with cyclodextrin derivatives as chiral stationary phases. We have recently prepared a new cyclodextrin phase by selective substitution of individual hydroxy groups of the different glucose units. In this way we succeeded in transferring the properties of two different cyclodextrin derivatives onto one. Cyclodextrin derivatives have also proved very useful for preparative GC, particularly in the separation of isomers which tend to co-elute on polysiloxane phases.



Anna-Karin Borg-Karlson1, Marit Stranden2, Hanna Mustaparta2, Raimondas Mozuraitis1,3 and Ilme Liblikas1,4

1 Department of Chemistry, KTH; Sweden
2 Department of Zoology, MTFS, NTNU; Norway
3 Institute of Ecology, Laboratory of Chemical Ecology, Lithuania
4 Laboratory of Ecochemistry, Estonian Agricultural University, Estonia

(-)-Germacrene D: an unstable and flexible host-plant compound.

The (-)-enantiomer of the sesquiterpene germacrene D is produced by a large number of plant species. Also the (+)-enantiomer is frequently occurring in plants. The molecule is flexible and shows a number of conformations which can explain the large number of cooccurring rearrangement products. We have developed a simple technique to isolate (-)-germacrene D and used the single cell responses as a tool to analyse the structure activity of (-)-germacrene D, its enantiomer and rearrangement products.



Martin Heisenberg

Biozentrum, University of Würzburg, Germany

A simple circuit model of odor memory in the fly

Since their discovery the mushroom bodies of the insect brain have been assumed to be involved in cognitive processing. Genetic intervention in the fly Drosophila melanogaster has now provided strong evidence that they are the seat of a memory trace for odors. This localization of the 'engram' to a single layer of synapses allows designing a simple circuit model of odor memory based on the functional anatomy of the olfactory system. In the model, complex odor mixtures are assumed to be represented by neuronal activity in sets of intrinsic mushroom body neurons (Kenyon cells). Conditioning renders an extrinsic mushroom body output neuron (CR neuron) specifically responsive to such a set (and hence the respective odor).

The localization of the memory trace for odors is based on the assumption that associative olfactory learning is mediated by synpatic plasticity. Evidence for the memory trace to be represented by the output synapses of Kenyon cells relies on three findings. First, the mushroom bodies are necessary for olfactory learning. Second, in many animals (molluscs, mammals) cAMP signaling has been shown to be crucial for synaptic plasticity. In the fly, this has been confirmed for the larval neuromuscular junction. Third, in Drosophila several genes involved in cAMP regulation are required for olfactory learning and memory. They are all preferentially expressed in the mushroom bodies, some of them specifically in the mushroom body lobes (output region). One of them, rutabaga, has been shown to be required for olfactory learning exclusively in a set of about 700 Kenyon cells. Another one, amnesiac, reveals that cAMP regulation is required for olfactory learning only during the learning experiment (rather than during development of the mushroom bodies). Remarkably, the output of the Kenyon cell synapses can be blocked while they are modulated, and the memory can still be retrieved later.



Walter S. Leal

Honorary Maeda-Duffey Lab, Department of Entomology, University of California, Davis, CA 95616
[email protected]

Proteins that make sense (and some that do not)

Our studies support the hypothesis that odorant-binding proteins (OBPs) participate in the selective transport of pheromones and other semiochemicals to their olfactory receptors. While engulfed in the binding pocket of an OBP-ligand complex, the semiochemical is protected from odorant-degrading enzymes. The solubilized pheromone is thus transported through the sensillar lymph until it reaches certain negatively-charged sites on the surface of dendrites. The low pH at these sites triggers a rapid conformational change of the OBP-odorant complex that leads to the formation of a new alpha-helix at the C-terminal. This helix ejects the ligand out of the protein so as to activate the olfactory receptors. I propose that the overall specificity of the insect olfactory system is achieved by “layers of filters” with OBPs participating in the first step.

Newly developed “cold” binding assays clearly demonstrate that the pheromone-binding protein from BmPBP binds bombykol at high (pH 7) but not at low (pH 5) pH. Bind at the sensillar lymph pH and the lack of binding at low pH are not affected by presence of KCl even if salt is added at concentration 3x higher than the physiological concentration. If the C-terminal does not penetrate the binding pocket, BmPBP binds at high and low pHs, thus, corroborating that the intramolecular arrangement in BmPBP prevents binding at low pH.

One of the hallmarks of OBPs is the six cysteine (six half cystines) residues, but this criterion alone should not be misleading used to classify a certain protein as olfactory protein. Proteins derived from the “PBP-gene family” do not necessarily have an olfactory function. It seems, however, that members of the PBP-gene family encapsulate hydrophobic ligands and thus have the ability to transport them in an aqueous environment. I propose that proteins of this group should be named “encapsulins” to imply the common role of encapsulating small ligands. The encapsulins family would thus encompass odorant-binding proteins, chemosensory proteins, and non-olfactory proteins like GP-9.

We are now testing the hypothesis that fast inactivation of chemical signals is achieved by an enzymatic process regulated by odorant-degrading enzymes (ODEs). Because ODEs are expressed in such low amounts that hitherto isolation and protein-based gene cloning has not been possible, the molecular basis of signal inactivation is still terra incognita. Using a bioinformatics approach, we were able to clone the first cDNAs from the giant silkmoth Antheraea polyphemus encoding an odorant-degrading enzyme (with a putative catalytic site characterized by the sequence Gly-Glu-Ser-Ala-Gly-Ala) and an “integumental esterase.” In-depth studies on the developmental changes in esterase activity and protein expression led us to the conclusion that the so-called “integumental esterase” is in fact a ubiquitous protein.



Rudolf Alexander Steinbrecht

MPI für Verhaltensphysiologie, D-82319 Seewiesen, Germany
[email protected]

What can we learn from localizing odorant-binding proteins?

Insect antennae are sometimes called compound noses, as their olfactory receptor neurons (ORNs) are compartmentalized into numerous separate organules, the sensilla. Thus, the extracellular sensillum lymph surrounding the sensory processes of these ORNs may have a different composition in the different sensilla, which possibly is the most profound distinction to vertebrate noses, where all ORNs extend their sensory processes into a common mucus layer. Immunocytochemical localization of different odorant-binding proteins (OBPs) has revealed a complex mosaic of distribution patterns of various pheromone-binding proteins, general odorant-binding proteins, chemosensory proteins, in moths as well as in Drosophila. These distribution patterns often concur with morphological subtypes of sensilla and their functional specificity. Such a correlation hints at a contribution of OBPs to the specificity of the responses of ORNs, a notion that has been more directly verified by electrophysiological experiments in Antheraea pernyi (Pophof 2003). Sometimes, however, OBPs have been localized in compartments where they cannot take part in the primary signal transduction processes. These data underline the multiple functions of OBPs which are not restricted to transporting odorants to the receptor sites.

With specific antibodies now available not only against OBPs but also against olfactory receptor proteins (ORs), it is now possible to correlate the distribution of OBPs and of ORs in any specific sensillum. For the electrophysiological study of the function, these specific sensilla can be recognised by their morphological features and more precisely by genetic coupling of specific ORs with fluorescent in vivo markers (Dobritsa et al. 2003). Thus, all tools are available for integrating the molecular and cellular basis of odour coding in insects.

Dobritsa AA, van der Goes van Naters W, Warr CG, Steinbrecht RA, Carlson JR (2003) Neuron 37:827-841
Pophof B (2003) Naturwissenschaften 89:515-518



Zainulabeuddin Syed and Patrick M. Guerin

Institute of Zoology, University of Neuchâtel, Rue Emile Argand 11, 2007 Neuchâtel, Switzerland
[email protected]

Tsetse flies are attracted to the invasive plant Lantana camara

Both tsetse fly sexes feed exclusively on the blood of vertebrates for a few minutes every 2-3 days. Tsetse flies seek cover from high temperatures to conserve energy between blood meals. Plants provide shelter for tsetse in allbiotopes they occupy. Recently, tsetse flies have taken cover in plantations and under the invasive bush Lantana camara that has invaded large areas of the testes fly belt of Africa. Flies from such refugia are implicated in sleeping sickness epidemics. In a wind tunnel we show that both foliage and an extract of volatiles from foliage of L. camara attract three tsetse spp. from different habitats: Glossina fuscipes fuscipes (riverine), G. brevipalpis (sylvatic) and G. pallidipes (savannah).

Gas chromatography analysis of volatiles extracted from leaves and flowers of L. camara coupled to electroantennogramme recordings show that 1-octen-3-ol and ß-caryophyllene are the major chemostimuli for the antennal receptor cells of the three tsetse spp. studied. A binary mixture of these products attracted the three spp. in the wind tunnel. Furthermore, chromatographic analysis of the L. camara extracts using the antennae as detectors for other biologically active plant volatile fractions show that the receptor cells of the three tsetse spp. respond similarly to groups of volatiles derived from the major biosynthetic and catabolic pathways of plants, i.e. to mono- and sesquiterpenes, to lipoxidation products and to aromatics. Mixtures of these plant volatiles also attracted the three tsetse spp. in the wind tunnel. These findings show that these flies have conserved a strong sensitivity to volatile secondary products of plants, underlining the fundamental role of vegetation in tsetse survival.



Yu Tong Qiu and Joop van Loon

Laboratory of Entomology, Wageningen University and Research Centre, Wageningen, The Netherlands
[email protected]

Identification of antennal odour receptor neurones of the malaria mosquito Anopheles gambiae

Anopheles gambiae is one of the major vectors of human malaria in Africa. Female mosquitoes of An. gambiae feed preferentially on human blood. Host location is guided predominantly by olfaction. Odour coding of this mosquito is studied by means of extracellular recordings of antennal olfactory receptor neurones (ORN). Two subtypes of sensilla trichodea as well as sensilla basiconica (grooved peg sensilla) were examined for their response specificity to 35 individual host or oviposition site related odours. From most of the sensilla trichodea spontaneous activity of two ORNs can be recorded. Activity of each of the two neurones can be recognised by differences in action potential amplitude and shape. Most of the ORNs responded to the tested compounds by excitation, whereas inhibition responses were recorded in a few cases. So far three sub-populations of short sharp-tipped sensilla trichodea have been identified according to their response profiles. The first type responded to indole but not to geranyl acetone. This sub-population of sharp-tipped sensilla trichodea was also responsive to C5 and C6 carboxylic acids and 1-hepten-3-ol. The second subpopulation of sensilla responded to geranyl acetone but not to indole. This sub-population of sensilla responded to a broader array of compounds: 2-nonanone, several phenols, 1-hepten-3-ol, 1-octen-3-ol, carboxylic acids (C2, 4MC4, C6, C9, C12, C14 and 7-octenoic acid) and three oxocarboxylic acids. The third sub-population of the short sharp-tipped sensilla trichodea was not sensitive to either indole or geranyl acetone. These sensilla contain ORNs responsive to phenols, 1-hepten-3-ol and hexanoic acid (C6). The sub-population of sensilla trichodea housing indole and geranyl acetone receptors was previously documented by Meijerink et al. (2001). A sub-population of medium sharp-tipped sensilla trichodea houses two ORNs, one of which was sensitive only to geranyl acetone. Spontaneous activities of more than three ORNs can be recorded from the grooved peg sensilla. ORNs innervating grooved peg sensilla responded to lactic acid, ammonia and oxocarboxylic acids by excitation, other compounds tested only elicited an inhibition response if any response at all. Compounds such as 4MC3, C6, 7-octenoic acid showed attractive effect when combined with ammonia and lactic acid in behavioural assays (unpublished results). Oxocarboxylic acids were found to increase landing response of An. gambiae (Healy et al. 2002). A more thorough study of ORN response profiles may increase our understanding of the semiochemistry involved in the host seeking behaviour of this malaria mosquito.

Meijerink, J.; Braks, M.A.H.; Loon, J.J.A. van 2001. Olfactory receptors on the antennae of the malaria mosquito Anopheles gambiae are sensitive to ammonia and other sweat-borne components. Journal of Insect Physiology. 47: 455-464.
Healy, T.P.; Copland, M.J.W.; Cork, A; Przyborowska, A. and Halket, J.M. 2002. Landing responses of Anopheles gambiae elicited by oxocarboxylic acids. Medical and Veterinary Entomology 16: 126-132.



Silke Sachse and Leslie B. Vosshall

The Rockefeller University, Laboratory of Neurogenetics and Behavior, New York, USA.
[email protected]

Odor Experience-Dependent Plasticity in the Antennal Lobe of Drosophila melanogaster

The ability of the brain to adapt structurally and functionally in response to sensory stimuli is a striking property across animal phyla. Previous studies have reported that continuous exposure of adult flies to a single odor for several days results in morphological changes in glomeruli, the morphological and functional units of the antennal lobe (Devaud et al., 2001). However, the cellular and biochemical mechanisms for these stimulus-dependent changes in glomerular size are currently unknown. We have extended this study by investigating olfactory sensory neurons (OSNs) with known odor response profiles innervating an identifiable glomerulus. Using the GAL4/UAS system to visualize OSNs expressing specific odorant receptors, the corresponding glomerulus could be identified and its morphological changes due to odor exposure investigated. In further studies we will investigate the anatomical changes of other components of the antennal lobe neuronal circuit using GAL4 lines labeling local interneurons and projection neurons. Finally, we wish to understand the behavioral effects produced by these stimulus-evoked changes in olfactory circuitry.

Reference: Devaud et al. (2001). J Neurosci. 21 :6274-6282.



Jocelijn Meijerink, Mikael Carlsson and Bill Hansson

Department of Crop Sciences, Chemical Ecology, Swedish Agricultural University
Box 44, Sundsvägen 14, 230 53 Alnarp, Sweden
jocelijn,[email protected]

Spatial representations in the moth antennal lobe

We investigated the spatial representation of odorants by imaging calcium activity within the primary olfactory integration centre, the antennal lobe (AL), in moth. We focussed on two main questions: how is odorant structure spatially represented in the AL of the moth Spodoptera littoralis and how conserved are these spatial representations between different moth species.

Spatial representation of odorant structure. The odorant features chain length and function group were investigated by using chemically ‘simple’ molecules like straight chain aliphatic alcohols and aldehydes. To determine (near)-threshold levels for calcium activity in the AL several low doses were tested. Studies showed that activity patterns of a given odour were most similar to so called ‘neighbouring compounds’, compounds with the same functional group, differing in chain length by only one carbon atom. A chain length dependency was present as the most activated glomerulus in the lobe shifted from a medial to a lateral position with increasing chain length of the molecule. Statistical analysis revealed that in both classes of chemicals the chain length of the molecule was represented in a similar way. No topographically fixed domains were observed for any of the classes. However, activity patterns evoked by lower chain length molecules were spatially more distinct than patterns evoked by higher chain length molecules. The number of activated glomeruli for both classes of chemicals increased with increasing chain length to reach a maximum at eight or nine C atoms followed by a decrease as the chain length further increased.

Evolution of olfactory ‘coding’ in moths.
Odour evoked activity patterns in the AL were compared for the noctuid moth species Spodoptera littoralis, Autographa gamma and Agrotis segetum. Several behaviourally relevant plant and flower odours like for example phenylacetaldehyde, geraniol and linalool were used. Each of these odours elicited very distinct activity patterns in the AL. Results revealed that odour evoked spatial representations are very preserved for the above mentioned species. More distantly related species, like Manduca sexta, were investigated. Preliminary results indicate that odour evoked activity patterns differ from those observed in noctuid species. These studies will gain more insight in the evolution of olfactory ‘coding’ in insects.



Helge Rø1, Hanne Therese Skiri1, Dirk Müller2, C. Giovanni Galizia2,3 and Hanna Mustaparta1

1 Norwegian University of Science and Technology, Departement of Biology, Neuroscience Unit, MTFS, NO-7489 Trondheim, Norway
2 Freie Universität Berlin, Department of Biology-Neurobiology, 14195-Berlin, Germany
3 Department of Entomology, University of California, Riverside, CA 92521, USA
[email protected]

Coding of plant odour information in the antennal lobe of heliothine moths

Great emphasis has been given to the processing of olfactory information in the pheromone specific macroglomerular complex of the antennal lobes in male moths. The other important olfactory system dealing with plant odour information is far less understood. This is partly due to the scarce knowledge about biologically relevant plant odorants defining functional types of receptor neurones, as well as the larger number of odorants indicated by the number of ordinary glomeruli in the antennal lobe. In Heliothis virescens 62 female and 66 male glomerular units have been mapped in a 3-D atlas (Berg et al. 2002).

In this study, we examined the antennal lobe responses in heliothine moths to stimulation with single compounds and mixtures, including key orourants for the receptor neurones (Røstelien et al., 2000a, b; Stranden et al., 2002). Two approaches were used:
1) Optical imaging recordings from the upper part of the antennal lobe showed that stimulation with plant odours at concentrations used in electrophysiology elicited responses in the ordinary glomeruli, but not in the MGC. In general, stimulation with single odorants elicited activity in one or a few ordinary glomeruli and different activity patterns were obtained for the different odorants. It can therefore be concluded that it is a functional organisation of all the glomeruli in the antennal lobe.
2) Intracellular recordings were used to study the response properties of antennal lobe interneurones. Subsequent staining with fluorescent a dye revealed the dendrite arborizations and the projections of the neurones in the calyces of the mushroom bodies and in the lateral protocerebrum, by the use of confocal microscopy. The Amira 3.0 software was used to create 3-D polygonal surface models from the optical sections. Superimposed onto the map of glomeruli in the 3-D atlas, the arborization pattern of the neurones is intended to give a functional characterisation of identified glomeruli. The aim is to determine whether one glomerulus primarily mediates information about a specific odourant, possibly a set of chemically related odourants. The responses of the projection neurons were recorded as plain inhibition or excitations as well as complex pattern including phases of increasing and decreasing firing rates and hyper-or depolarised membrane potentials. Ability to follow pulsed stimulation was found in some units for certain pulse frequencies. Selective responses to odours were obtained for projection as well as some local interneurones, exemplified by a local interneurone responding strongly to geraniol and weaker to citral and citronelool, but not to other tested compounds. Interneurones could readily be distinguished from projection neurones based on their spike characteristics, e.g. broad and varying spike amplitudes and higher background activity. This was verified by confocal microscopy. Staining of the neurones revealed local interneurones innervating several up to nearly all glomeruli. Multi- and uniglomerular projection neurones were found in separate antenno-cerebral tracts.

Berg B.G., Galizia C.G., Brandt R., Mustaparta H. (2002) Digital atlases of the antennal lobe in two species of tobacco budworm moths, the Oriental Helicoverpa assulta (male) and the American Heliothis virescens (male and female. J Comp Neurol 446:123-134
Røstelien T., Borg-Karlson A. -K., Mustaparta, H. (2000) Selective receptor neuron 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. J Comp Physiol, A: Sensory, Neural, and Behavioral Physiology 186(9): 833-847.
Røstelien T., Borg-Karlson A. –K., Fäldt J., Jacobsson U., Mustaparta H. (2000) The plant sesquiterpene germacrene D specifically activates a major type of antennal receptor neuron of the tobacco budworm moth Heliothis virescens. Chem Senses 25: 141-148.
Stranden M., Borg-Karlson A. –K., Mustaparta H (2002) receptor neuron discrimination of the germacrene D enantiomers in the moth Helicoverpa armigera. Chem Sen ses 27: 143-152.



Uli Müller

Free University of Berlin, -Neurobiology-, Koenigin-Luise-Str. 28/30, D-14195 Berlin, Germany
[email protected]

Olfactory memory formation: a complex network of molecular processes

Memory can exist in a variety of temporal domains ranging from short-term memories lasting minutes to long-term memories (LTM) lasting days and, in the limit a lifetime.
Evidence from organisms as diverse as mollusks, insects, birds and mammals shows that different molecular mechanisms contribute to this multiphasic process of memory formation. While short-living covalent modifications of proteins are believed to underlie short-term changes, long-lasting neuronal changes require translation and transcription. However, the interaction between these processes, which occur in very different time-windows during memory formation, is presently unknown.

Associative olfactory learning in honeybees provides a perfect system for studying the highly dynamic modulations of molecular processes spanning time-windows from seconds up to several days. In the honeybee, as in other animals, memory formation depends on the sequence and succession of the stimuli during training: only repeated conditioning trials induce a stable olfactory long-term memory. Our analysis revealed that during the few minutes of multiple conditioning trials, the action of different molecular processes is essential for the formation of long-term memory. Blocking these processes affects neither learning itself nor the formation of mid-term memory. Moreover, these processes are not essential for the memory induced by a single conditioning trial, suggesting a specific contribution of these processes to long-term memory formation.

In addition to these parallel-acting molecular processes, we have identified processes required in very distinct time-windows after the olfactory conditioning. Again these processes are specifically involved in LTM formation. The identification of these parallel- and serially -acting processes is now the basis for studying the interaction between these molecular processes occuring in very different time-windows. Understanding the molecular network required for memory formation provides the opportunity to search for mechanisms in memory formation that are targets of physiological processes like stress and circadian rhythm, which are known to influence memory formation.

Supported by the Deutsche Forschungsgemeinschaft



Préat1, T., Isabel1, G., Pascual1, A., Coulom2, H. and Birman2, S.,

1 DEPSN, CNRS, 91190 Gif-sur-Yvette, France.
2 LGPD, Université de la Méditerranée, 13288 Marseille Cedex 9, France.
[email protected]

Dynamic of olfactory memory phases in drosophila

Animals and humans store information into distinct memory phases that interact dynamically. In Drosophila, two forms of consolidated memories have been described, Anesthesia Resistant Memory (ARM), and Long Term Memory (LTM). As in most species, LTM formation requires repetitive presentation of the same stimuli (1). How does the brain recognize that the same sequence of events is repeated, and how does the newly acquired information interact with previously stored information? Up to now it was thought that ARM and LTM could coexist (2). On the opposite, we show here that LTM training leads to extinction (or blocking) of ARM.

We previously reported that the alpha-lobes-absent mutant lacks either the two vertical lobes of the mushroom body or two of the three median lobes. Long-term memory requires the vertical lobes but not the median lobes, which correspond to a branching of the same neurons (3). We now show that LTM conditioning is still able to erase ARM in those animals that cannot form LTM, resulting in a paradoxical situation: the more these flies learn, the less they remember.

The molecular and cellular basis of those interactions will be discussed. The dynamic of memory phases in drosophila will be compared to that observed in other insects.

(1) Tully T., Préat T., Boynton S. and DelVecchio M. (1994). Cell, 79: 35-47.
(2) Dubnau J. and Tully T. (1998). Annu. Rev. Neurosci. 21:407-44.
(3) Pascual A. and Préat T. (2001). Science, 294:1115-1117.



Maartje Bleeker1, Hans Smid1, Johannes Steidle2, Joop van Loon1 and Louise Vet1

1 Laboratory of Entomology, Wageningen University, Binnenhaven 7, 6709 PD Wageningen, The Netherlands.
2 Universität Hohenheim, Institut für Zoologie 220c, Fg. Tieroekologie Postfach 700562, D-70599 Stuttgart, Germany
[email protected]

Associative learning of odors in two parasitoid wasps: A comparison of memory structure

Cotesia glomerata and C. rubecula (Hymenoptera: Braconidae) are parasitoid wasps of the caterpillars of Pieris butterflies. They find their hosts by use of plant odors, which are induced by the feeding of caterpillars (Steinberg et al., 1993; Geervliet et al., 1994). In host searching, these two species differ in their use of olfactory associative learning (Geervliet et al., 1998). C. glomerata learns to associate new odors with the presence of caterpillars and alters its preference accordingly, whereas C. rubecula has an innate preference (Geervliet et al., 1998).

We studied the memory structure of both wasp species after an associative learning experience. Female wasps are given a single oviposition experience in the presence of an odor. The response of the females to this odor is tested in the windtunnel. We tested the response at different time points after the experience to gain insight in the memory duration. To be able to distinguish different memory phases in the memory development of these two species, we used different chemicals that are known to interfere with specific memory phases in other insect species.

Geervliet, J.B.F., Vet, L.E.M. and Dicke, M., 1994. Volatiles from damaged plants act as major cues in long-range host-searching by the parasitoid Cotesia rubecula. Entomologia Experimentalis et Applicata 73, 289-297.
Geervliet, J.B.F., Vreugdenhil, A.I., Dicke, M. and Vet, L.E.M., 1998. Learning to discriminate between infochemicals from different plant-host complexes by the parasitoids Cotesia glomerata and Cotesia rubecula (Hymenoptera: Braconidae). Entomologia Experimentalis et Applicata 86, 241-252.
Steinberg, S., Dicke, M. and Vet, L.E.M., 1993. Relative importance of infochemicals from first and second trophic level in long-range host location by the larval parasitoid Cotesia glomerata. Journal of Chemical Ecology 19, 47-59.
Vet, L.E.M. and Dicke, M., 1992. Ecology of infochemical use by natural enemies in a tritrophic context. Annual Review of Entomology 37, 141-172.



E. Städler

Sinnesphysiologie, Forschungsanstalt CH-8820 Wädenswil, Switzerland
[email protected]

Evolutionary history of herbivorous insects mirrored in their chemoreceptors?

Different researcher have explored this question earlier. Dethier (1973) compared the sensitivity of contact chemoreceptors of different caterpillar species to plant saps and some compounds. The recent review by Schoonhoven & van Loon (2002) and Chapman (2003) confirmed the earlier results and concluded that even closely related species have rather different contact-chemoreceptor systems. It could be expected that pheromones are different in this respect. Mustaparta (1990) reviewed the pheromone perception in different species and pointed out different peripheral and central processes that have been identified in differently related species. The author concluded, and more recent research is supporting it, that the olfactory receptor neurones of related species differ relatively little in their specificity for individual pheromone components and that species specificity is based to a large extend on central processing neurones of the antennal lobe and the higher centres of the CNS.

The behavioural facilitation hypothesis (Feeny, 1995) postulates that during evolutionary time changes in the host-plant choice was the first step in the colonisation of novel hosts. Since chemoreception is crucial, either the sensory organs or the CNS or both, must be involved. Comparative studies of behaviour and sensory physiology of insect herbivores along the approach taken by Dethier (1973) should allow us to test the different hypotheses. In our lab we concentrate on the plant compounds mediating oviposition in flies of the genus Delia. We try to isolate and identify plant extracts, fractions or compounds of host and non-host plants that influence the behaviour. The starting point of our comparative investigation is the oviposition behaviour and the chemoreceptors of Delia radicum. In this species contact-chemoreception is the key sensory modality, but olfaction needs to be considered too. We expect to obtain a complex, but realistic view of the sensory physiology of host-plant selection in the species studied. But no doubts, we will also experience the limits of such a study in terms of time and resources to obtain the satisfactory results.

Based on our preliminary results (Gouinguené et al., this symposium) and in accordance with other studies, we can confirm that closely related species, strains or hybrid species can differ significantly in the sensitivity to host-plant compounds. Further, we have evidence that the type of compounds (polarity and volatility) perceived and the coding mechanisms can differ between the species of the same genus. In agreement with the hypothesis of Bernays (2001), we have first results indicating that the more polyphagous species studied is more sensitive to non-host compounds than the specialized species. Thus, in an evolutionary context, the sensory organs seem indeed to be rather variable allowing the colonisation of new food plants.

Bernays, E.A., 2001. Neural limitations in phytophagous insects: implications for diet breadth and evolution of host affiliation. Annu.Rev.Entomol., 46, 703-727.
Chapman, RF, 2003. Contact chemoreception infeeding by phytophagous insects. Annu Rev Entomol 48, 455-484.
Dethier, V.G., 1973. Electrophysiological studies of gustation in lepidopterous larvae. II. Taste spectra in relation to food plant discrimination. J.Comp.Physiol.A 80, 103-134.
Feeny, P., 1995. Ecological opportunism and chemical constraints on the associations of swallowtail butterflies. In: Swallowtail Butterflies: Their Ecology and Evolutionary Biology, Scriber, J.M., Tsubaki, Y., Lederhouse, R.C., eds., Scientific Publ.Inc, Gainesville, Florida, 9-15.
Mustaparta, H., 1990. Evolutionary aspects of pheromone perception. ISOT 10, 164-174.
Schoonhoven, L.M., van Loon, J.J.A., 2002. An inventory of taste in caterpillars: each species its own key? Acta Zool Acad Sci Hung, 48, 215-263.



Kari Jørgensen, Tor Jørgen Almaas and Hanna Mustaparta

Norwegian University of Science and Technology, Department of Biology, Neuroscience Unit, NO-7489 Trondheim, Norway

Taste sensilla on the antennae of heliothine moths: Physiological characterisations of the associated receptor neurones and projection patterns in the suboesophageal ganglion.

The heliothine moths can learn odours by pairing odour stimulation (conditioned stimulus) with stimulation of the taste sensilla with sucrose (unconditioned stimulus). With the aim to reveal the neuronal connection between the two sensory pathways, we have studied the taste sensilla (s. chaetica) on the antennae and the projection of the associated receptor neurones in the primary taste centre, the suboesophageal ganglion (SOG), of the CNS. Scanning electron microscopy (SEM) imaging showed that four to six s. chaetica were present on all annuli of the flagellum, except for the most distal having a higher density of s. chaetica.

S. chaetica were characterised by long, solid hairs having a single pore at the tip and a socket at the base of the hairshaft, indicating a combined chemosensory and mechanosensory function. Staining with tetramethylrhodamine dextran by cutting the s. chaetica near the base while the antennae was submerged in the dye resulted in stained axons of the receptor neurones. Imaging in confocal laser scanning microscope (CLSM) showed that the axons by-passed the antennal lobe posterior- laterally, and proceeded towards the SOG where they terminated in a fingerlike pattern. Axons from the same sensillum kept tightly together and projected in the same area, i.e. the antennal mechanosensory and motor centre (AMMC) and the dorsal SOG. Electrophysiological recordings (tip recordings) were performed using glass capillary microelectrodes filled with 0,1 M KCl and various chemicals to be tested. The recordings indicated the presence of three or four taste receptor neurones responding mainly to sucrose, water, leucin and sinigrin. These compounds were tested on PER, showing that stimulation with phagostimulants (sucrose, leucin and water) elicited PER in all individuals, whereas the deterrent sinigrin elicited PER only in half of them. An interesting aspect of these studies was that the individuals clearly were affected by previous exposure to a stimulant, which also served as a reward. Retesting with sinigrin showed a reduced response compared to retesting with the phagostimulants.


Main Page Venue Travel and Costs Main Program Social Program Abstracts of Talks Abstracts of Posters Registration Abstract Submission Deadlines ESITO Sponsors ESITO Homepage


Copyright © ESITO 2003

Last updated: June 21, 2003