1

Schneider Dietrich

Max-Planck-Institut für Verhaltensphysiologie
Seewiesen

Cycads - Palmferns - and their Insects (Herbivores and Pollinators)

This communication provides some background information on the biology and history of palm-ferns (plant order Cycadales), which are neither ferns nor palms, but true gymnosperms. After an overview on cycads, I will discuss chemo-ecological aspects of cycad evolution, which probably was a specific co-evolution with some cycad-feeding and cycad pollinating insects. In this context, I like to report also on our* own descriptive and experimental work with two insect species (a butterfly Eumaeus atala, and cucurlionid weevils) which are ecologically connected to two neotropical cycad species of the genus Zamia.

The cycads evolved from the seed-ferns (pteridosperms) and thus are - with the conifers and the Ginkgo-tree - one of the earliest groups of extant gymnosperm, seed bearing plants. Cycads were first seen in the upper layer of the Paleozoic and had their richest phase from the Trias to the Jurassic ages. But cycads still exist with few changes as "living fossils" with several genera and ca.160 species (most of them underwent speciation within the last 20 millions years), mainly in the tropical zones of the former Gondwana continent. - The evolution of seed producing plant groups from sporophytes was a major step which eventually involved the reduction of the haploid gametophyte-generation of the early plants to the ovum and pollen. Finding of the partner gamete is an overall problem of sexual reproduction. This appears to be a simple, pheromone guided affair on the wet prothallium of a fern. Cycads, however, produce pollen, which deposits haploid male spermatozoids directly at the egg. The flagellated mobile male cell is a remnant of cycad predecessors. In all gymnosperms, male and female organs are more or less spatially separated. Pollen, with little motility on its own, therefore needs a carrier, often wind as in conifers. The dioecious, remnant cycads were earlier also thought to be wind pollinated, yet now found to be also (or even mainly?) pollinated by beetles. Was this always so? Perhaps, since the extinct sistergroup of the cycads, the Benettitales is said to have partially depended on beetle pollination.

The cucurlionid beetles in question are olfactorially and/or visually attracted to the male cycad cones where the beetle eggs are deposited. The larvae feed (only!! why?) on somatic cone tissue and develop and eventually pupate. The adults which emerge from the male cone, are in this process passively dusted with pollen which covers the surface of the cone. Both sexes of the beetles now fly to neighboring male and female specimens of the given cycad species and accidentally (?) lose some of the pollen on the micropyle of the egg chamber of the cycad. Such behavioral "key relation" in two species of the cycad genus Zamia and two beetle species is species-specific. This observations now explains why cycads of which both sexes are cultivated in botanical gardens far from their origin, remain unfertile unless their beetles came with them. Such cycads need to be hand-pollinated by the gardener (Norstog et al. 1986; Norstog 1987; Tang 1987).

The cone tissue, not the pollen as in the honeybee, is the "reward" for the behaviourally unintended, but evolutionary very meaningful happening of pollination. These basic facts lead directly to the interest which our research group* takes in the plant order Cycadales and the insects connected with them. Our curiosity is based on the chemo-ecological view that major evolutionary trends are steered by co-evolution of the green biomass producers and the consumers, the animals. Plants, more than animals, defend their position in nature by the production of chemicals which render them more or less unpalatable for the herbivores which in turn try to overcome, avoid or even exploit this defence (Rosenthal & Janzen 1979). Cycads are reported not to suffer much from herbivory, probably due to their chemical protectants but, without doubt, also by their tough and even prickly older leafs.

All recent cycads appear to share two specific secondary compounds: a) azoxy-glycosides such as cycasin and macrozamin with their aglykon methylazomethanol (MAM) and b) a peculiar, non protein amino-acid: alpha-amino-beta-methylaminoproprionic acid (BMAA). While the complete glycoside (a) is not a cell-toxin, MAM, when liberated by beta-glucosidase, and further degradation products (such as diazomethane or another methyldonor), interfere fatally with genetic structures, clearly in the mammalian liver cells but probably with any living cell. Cycasin, or rather MAM (?), was also found to embitter otherwise palatable food for livestock and for some insects. The role of (b) is less well known. This strange amino-acid is strongly suspected to be a powerful neurotoxin in man and other mammals and the cause of amylotrophic lateral sclerosis and/or parkinsonism. These toxic effects on humans are the result of the ingestion of cycad seeds or starch from the trunks, which are only good for consumption after specific treatment. Sailors of Captain Cook were the first reported people suffering from eating Cycas seeds.

The production of a seed, a lasting embryo, is an adaptation to the life on land. But this evolutionary step requires fertilization. All seed plants use pollen for this, which in Zamia is carried by beetles which eat the plant. We now ask: how do the pollinators and also the larvae of the Lepidopera cope with the chemical weapons of cycads.

Some, if not all of the Lepidoptera which feed on cycads seem to use the acquired cycasin, the glycoside, for their own protection. Unknown is at present, how they overcome the risk of a cycasin hydrolysis and thus the production of the MAM. Yet even the plant has to take care to handle the MAM. In the beetles we are not yet certain of a cycasin storage. The very low amounts of cycasin which we found in a pollinating weevil could be residues from the gut contents: The guts of the beetle might not even have a biochemical carrierprotein to transfer the cycasin into the body, as cycasin is a polar molecule which cannot pass biomembranes by simple diffusion.

The specific "cycad-animal" which we were interested to study in more detail is the Atala Butterfly (Eumaeus atala), the larvae of which feed in the South of Florida on the local cycad (Zamia integrifolia). Pioneering work on the atala-cycad relation was done by Rothschild et al. (1986) and Bowers et al. (1989/90). The butterfly stores the ingested cycasin and is avoided by some of its potential predators. Recently, geometrid moths in South Africa were also found to store cycasin and macrozamin of their local cycad foodplants (Donaldson 1991, Donaldson & Bösenberg 1995). All these cycad feeders are brightly coloured in at least one of their living stages. This fits to the claim that chemically protected animals warn their predators with their "aposematic" coloration (Rothschild 1985). Long known examples are several arctiid moths which store pyrrolizidine alkaloids (PA) and danaine (Monarch) butterflies which store cardenolides and PA (for lit see Boppré 1990, Schneider 1992).

In our recent work we were able to corroborate the essence of the earlier reports on cycasin uptake and sequestration by the atala life-stages and observed puzzling, hitherto unknown quantitative relations which need further studies. Storage of secondary plant substances by a lepidopteran larva is a complex process which does even require molecular carrier processes in the gut tissue (Wink and Schneider 1988, 1990). We also found that the adult butterflies store much of their cycasin in the wings, but also considerable amounts in the eggs and also the spermatophore.

The atala and its neotropical cousins of the Eumaeini subfamily all feed as larvae exclusively on cycads. All, like atala, display striking coloration in the imago, eggs, larva and pupa. The imagos show a relaxed flight behaviour, comparable to other chemically protected butterflies. Eumaeini and some related groups of peculiar lycaenid butterflies also possess abdominal androconial organs, which occur in a variety of Lepidoptera and dissipate male sexual pheromones.

In atala, we found these male organs to be of the corema type, which are flexible tubular structures kept invaginated in the resting position in the abdomen, but evaginated by hemolymph pressure during courtship. Such organs have been described from a number of Lepidoptera, in particular from the danaines, the "Monarchs" (Boppré 1990, Schneider 1992). In atala, the tip area of these tubes is covered by long yellow hairs (scales) which form a very visible spherical hairbrush when fully expanded. This organ dissipates an odour which activates receptor cells on the antennae of both atala sexes. The still unidentified scent can also be sniffed with our nose. During courtship, males expose their hairbrushes in a hovering flight to the females.

The extract of the whole atala corema contained also cycasin. To date, we are unable to localize the cycasin in this tissue. When coremata are prepared for the chemical analysis, the organs are artificially expanded and then cut from the body. Does the cycasin now come from the residual hemolymph, from the epidermal and glandular tissue or even from the many yellow hairs? In the latter case, cycasin would even be a candidate for the pheromone or its precursor as in analogous cases of alkaloids in other Lepidoptera (Schneider 1992). But cycasin neither smells for us nor did it induce a remarkable EAG in the olfactory organ of atala, the antenna.

* The Eumaeus atala research is a collaborative project of D. Schneider with: K. Fiedler & B. Kornmaier, (U. Bayreuth), P. Lounibos, (U. Florida), M. Wink, F. Sporer & A. Tei (U. Heidelberg), J. Ziesmann (MPIV, Seewiesen).

References
Boppré, M. (1990) Lepidoptera and pyrrolizidine alkaloids: Exemplification of complexity in chemical ecology. J. chemical Ecol. 16: 165-185.
Bowers, M.D. & Farley, S. (1990) The behavior of the grey jays, Perisoreus canadensis, towards palatable and unpalatable Lepidoptera. Anim. Behav. 39: 699-705.
Bowers, M.D. & Larin, Z. (1989) Acquired chemical defense in the lycaenid butterfly, Eumaeus atala. J.Chem.Ecol. 15:1133-1146.
Donaldson, J.S. (1991) Adaptation to the host plant and the evolution of host relationship in cycad weevils (Coleoptera: Brentidae). PhD Thesis, U. Cape Town, S.Afr.
Donaldson J.S. & Bösenberg, J.D. (1995) Life history and host range of the leopard magpie moth, Zerenopsis leopardina Felder (Lepidoptera: Geometridae). Afric. Entomol. 3: 103-110.
Norstog, K. (1987) Cycads and the origin of insect pollination. Am. Scientist 75: 270-279.
Norstog, K.J., Stevenson, D.W. & Niklas, K.J. (1986) The role of beetles in the pollination of Zamia furfuracea fil. (Zamiaceae). Biotropica 18: 300-306.
Rosenthal, G. & Janzen, D. (1979) Herbivores: Their interaction with secondary plant metabolites. Academic Press, London New York.
Rothschild, M. (1985) British aposomatic Lepidoptera. In: The Moths and Butterflies of Great Britain & Ireland. 2: 8-62 Colchester: Harley Books.
Rothschild, M., Nash, R.J., & Bell, E.A. (1986) Cycasin in the endangered butterfly Eumaeus atala Florida. Phytochemistry 25: 1853-1854.
Schneider D. (1992) 100 years of pheromone research- an essay on Lepidoptera. Naturwissenschaften 79: 241-250.
Tang, W. (1987) Insect pollination in the cycad Zamia pumila (Zamiaceae). Am. J. Bot. 74: 90-99.
Wink, M. & Schneider, D. (1988) Carrier mediated uptake of pyrrolizidine alkaloids in larvae of the aposematic alkaloid-exploiting moth Creatonotos transiens. Naturwissenschaften 75: 524-525.
Wink, M. & Schneider, D. (1990) Fate of plant derived secondary metabolites in three moth species (Syntomis mogadorensis, Synomeida epilais, Creatonotos transiens). J. Comp. Physiol. B 160: 389-400

2

Vosshall Leslie B., Amrein Hubert, Morozov Pavel S., Rzhetzky Andrey and Axel Richard

Center for Neurobiology and Behavior
Howard Hughes Medical Institute, Columbia University
College of Physicians and Surgeons
New York, NY 10032, U.S.A.
LV25@columbia.edu

The molecular biology of olfaction in Drosophila melanogaster

Large and divergent families of G protein-coupled odorant receptors are utilized by vertebrates and the nematode C. elegans to recognize a vast array of odorants. Despite many functional experiments that predict the existence of similarly large receptor families in insects, such genes have not been isolated from any arthropod. Since homology-based approaches have failed to identify homologues of either the vertebrate or nematode receptor families in Drosophila, we assumed that insect odorant receptors would be extremely divergent. We have used two complementary, non-homology-based approaches to identify candidate odorant receptor genes in the fruit fly, Drosophila melanogaster: differential screening to identify antennal- and maxillary palp-specific genes and computer-based screens of the Drosophila genome for multi-transmembrane proteins. Using these approaches, we have identified a novel family of seven transmembrane domain proteins, encoded by 100 to 200 genes, that is likely to represent the family of Drosophila odorant receptors. Members of this gene family are expressed in topographically defined subpopulations of olfactory sensory neurons in either the antenna or the maxillary palp. Sensory neurons express different complements of receptor genes, suggesting that individual neurons are functionally distinct. The precise expression of individual odorant receptor genes in distinct regions of the antenna is in sharp contrast to the apparently random organization of receptor expression in vertebrate olfactory epithelia. We are interested in the logic of olfactory processing in this insect and how ordered receptor expression at the periphery is reflected in synaptic organization of antennal lobe glomeruli.

3

De Bruyne Marien, Warr Coral, Clyne Peter, Kim Junhyong, Freeman Marc, Lessing Derek, Foster Kara, Carlson John

Yale University
Molecular, Cellular and Developmental Biology
Kline Biology Tower
P.O. box 20 8103
New Haven CT 065020-8103, USA

Odor coding in Drosophila melanogaster

The study of olfaction in higher organisms poses two basic problems. First, we do not understand how the multitude of odor molecules is translated into neural signals at the peripheral interface where olfactory receptor neuron (ORN) meets the environment. How specific is this process and what are the mechanisms of ligand-receptor interaction? Second, it has proven difficult to get a grip on the number of physiologically different ORNs that make up an entire peripheral olfactory system and somehow "covers" this odor multitude for the animal. How many different cells does an animal need to do this adequately and how are they organized? We are characterizing the olfactory system of Drosophila melanogaster to address these issues. This tiny fly has long been a favorite model for geneticists and developmental biologists. The availability of many genetic and molecular techniques to manipulate its biology makes this insect an ideal experimental model. Moreover it will soon be the first insect species with its entire genome sequenced.

Using single-sensillum recordings with tungsten wire electrodes, we have characterized responses from basiconic sensilla on the two olfactory organs of Drosophila. On the maxillary palp, 120 ORNs are distributed pairwise in 60 morphologically identical sensilla (1). The neurons can be clearly classified in 6 classes based on response spectra. On the antenna, basiconic sensilla house at least 14 different ORN classes. Here they occur in two morphologically different sensillum types. Big basiconics are of three functional types; one of them housing four ORNs, the others only two. Small basiconics are of three types, each housing two different neurons.
The response spectra of these 20 cell types include mainly responses to simple esters, alcohols and ketones but also to CO₂, some specific aromatic compounds, E2-hexenal, valerolactone, a sulfur compound and a terpene. Some spectra are partially overlapping but dose-response curves for some odors show a sigmoid relation with sensitivity to low doses of only one tested compound. ORNs not only differ in odor response spectra but also show divergent response characteristics. Stimulation can result in inhibition or excitation with distinctly different temporal aspects. This may provide additional degrees of freedom so that multiple messages can be sent from a single neuron to expand odor coding capacity of a system with a limited number of cells.
The distribution of these cell types across antennal and palpal surfaces is organized. First, ORNs form characteristic sets that are almost always found together in one sensillum. Second, although these sensillum types occur mixed together, they are not randomly distributed. They form distinct zones, probably reflecting developmental processes.

The organization of such a complex system must require a sophisticated developmental program, regulated by many different genes. One gene that is partly responsible for endowing ORNs with specific identities is acj6, which encodes a POU-domain transcription factor. It is turned on early in the development of sensilla in antennae and palps. Mutations in this gene cause certain neurons to lose their responsiveness while others require new response characteristics (2). Another mutation, Scutoid, causes complex changes in the physiology of antennal basiconic sensilla. Some neurons are missing, while others appear to show altered response characteristics. In addition, flies carrying this mutation show disruption in the morphological layout of the antennal big basiconics. This mutation likely affects the development of the antennal sensory field.

What are the genes that are directly responsible for giving an ORN its unique identitiy and response characteristics? The entire genome of Drosophila is presently being sequenced and it is possible to analyze these DNA sequences for the presence of particular kinds of genes, such as receptors. This approach has led to the isolation of a large family of putative odor receptor genes from the database of the Berkeley Drosophila Genome project (3,4). Clyne et al. (3) designed a computer algorithm to search for a minimum of three transmembrane domains (TMs) in the predicted proteins for all open reading frames in this DNA sequence database. It led to the identification of two genes with seven TMs, which is characteristic of G-protein-coupled receptors. With these genes as a template, other genes could be found by virtue of their similarity. A total of 24 genes have been found in 20% of the genome, all having the predicted 7 TMs but otherwise they are extremely divergent in their amino-acid sequences. Many of these genes are specifically expressed in subsets of ORNs in the antennae or palps. This and the fact that similar genes in vertebrates and nematodes have been shown to mediate odor responses makes it likely that the DOR family of genes encodes the olfactory receptor proteins in Drosophila ORNs.

The family may contain on the order of 100 genes with probably ca. 15 expressed in the palp. If this holds true, each functional ORN class on the palp would express an average of 2-3 receptors. This suggestion, the complex response spectra and diverse response characteristics may indicate multiple transduction pathways within one neuron. The number of glomeruli in Drosophila and our characterization of neuron classes also suggests that the total number of receptor cell classes is limited (ca. 40) and we may be halfway in having a complete view of the coding possibilities of this fly's nose.

References:
1) de Bruyne M, Clyne PJ, Carlson JR (1999) Odor coding in a model olfactory organ: the Drosophila maxillary palp. J.Neurosci. 19:4520-4532.
2) Clyne PJ, Certel S, de Bruyne M, Zaslavsky L, Johnson W, Carlson JR (1999) The odor-specificities of a subset of olfactory receptor neurons are governed by acj6, a POU domain transcription factor. Neuron 22:339-347.
3) Clyne PJ, Warr CG, Freeman MR, Lessing D, Kim J, Carlson JR (1999) A novel family of divergent seven-transmembrane proteins: candidate odoant receptors in Drosophila. Neuron 22:327-338.
4) Vosshall LB, Amrein H, Morozov PS, Rzhetsky A, Axel R (1999) A spatial map of the olfactory receptor expression in the Drosophila antenna. Cell 96:725-736.

4

Maida Rosario¹ , Pröbstl Therese¹ , Lange Ute² and Ziegelberger Gunde¹

¹ Max-Planck-Institut für Verhaltensphysiologie
D-82319 Seewiesen, Germany

² Klinikum der Friedrich-Schiller-Universität
D-07740 Jena, Germany

Pheromone-binding proteins in isolated olfactory sensilla of the male silkmoth Antheraea polyphemus

Three different pheromone-binding proteins (PBPs) could be identified in pheromone-sensitive hairs of the male silkmoth Antheraea polyphemus by combining pheromone-binding assays, N-terminal sequencing, mass spectrometry and immunoblotting.
In a first set of experiments, we have used sensillum lymph droplets for MALDI-TOF mass spectrometry and for Westernblots with four antisera raised against different silkmoth odorant-binding proteins. Three peaks with a m/z of 15702, 15752 and 15780 were observed in the mass spectra and all proteins detected seem to belong to the class of pheromone-binding proteins, since immunoreactivity was found only with the anti-PBP antiserum, but not with two antisera against general odorant-binding proteins nor with the antiserum against the antennal binding protein X.
In a second set of experiments, proteins of antennal side branches were separated by free flow isoelectrofocusing. By means of Westernblots using the anti-PBP antiserum, at least three anti-PBP immunoreactive proteins with pI values between 4.4 and 4.7 could be identified. Similar results were obtained using isolated pheromone-sensitive hairs and ion-exchange chromatography on Mono-Q leading again to three anti-PBP immunoreactive proteins. N-terminal sequencing of these three proteins revealed two proteins sharing identical amino acid sequence at the first 49 amino acids, representing appr. 35% of the full sequence. This sequence is identical to the one of Apol PBP of Raming et al. (1989), which we now call Apol PBP1. Our two proteins with identical N-terminal sequence may be called Apol PBP1a and Apol PBP1b. Our third protein represents a new PBP with 57% homology to Apol PBP1, which we call Apol PBP2.
PBP-subtypes bound the 3H-labelled acetate, but to a different extent, whereas no binding of the the 3H-labelled aldehyde was found. PBP multiplicity and the pheromone binding specificity support the idea that PBPs participate in odour discrimination.

5

Vogt Richard

Department of Biological Sciences
University of South Carolina
Columbia, SC 29208
vogt@biol.sc.edu

Ecdysteroid influences on neurogenesis and the expression of neurogenic genes during early antennal development in Manduca sexta.

Olfactory sensory neurogenesis in the developing adult antenna of the tobacco hawkmoth Manduca sexta: DNA replication and ecdysteroid regulation.
DNA replication events were followed in early developing olfactory sensory epithelia during the 6 days following pupation and were assayed by the incorporation of 5-Bromo-2=92deoxy-Uridine (BrdU). BrdU incorporation occurred in two phases separated by antennal apolysis which took place about three days after pupation. Pre-apolysis incorporation was due to the mitotic events of which establish sensilla cells from Sensory Mother Cells, while post-apolysis incorporation was the consequence of endoreplicative events in sensilla support cells. Pre-apolysis incorporation showed a temporal-spatial progression, initiating along the annular borders around one day after pupation and progressing inward to the middle region by around three days after pupation. These mitoses progressed through regions which yield spatially distinct distributions of different classes of olfactory sensilla, and may thus contribute to the patterning of these regions. Post-apolysis incorporation also showed a temporal-spatial progression similar to the pre-apolysis pattern. To test whether the temporal-spatial progression of olfactory antennal cell proliferation might be hormonally regulated, BrdU incorporation was assayed in pre-apolysis tissue cultured +/- 20-hydroxy ecdysone, alpha-ecdysone, or the ecdysteroid agonist RH5992. Ecdysteroids and agonist developmentally advanced the temporal-spatial pattern of BrdU incorporation, suggesting that olfactory sensory neurogenesis is regulated by endogenously rising pupal ecdysteroids during this period.

6

Kennedy¹ Linda M., Foster¹ Kara D., Spielman² Andrew I.

¹ Neuroscience Program
Department of Biology
Clark University
Worcester, MA 01601, USA
lkennedy@clarku.edu

² New York University College of Dentistry
Basic Science Division
New York, NY 10010, USA

Rapid kinetics of receptor cell firing and second messenger formation in response to sucrose

Biochemical events in taste transduction must occur within the time frame of behavioral responses to the stimuli. Phormia regina behavioral responses occur within 100 msec after sucrose contacts the taste sensillum (1), and in this species, the contact time and initial receptor cell firing response can be measured with precision. Here we have studied the behaviorally-significant first 100 msec of receptor cell firing and second messenger formation to sucrose in taste tissues of P. regina.

In physiological studies, receptor cell action potential responses to 50 mM sucrose were tip-recorded from single taste sensilla. First, cyclic nucleotide (cNMP) levels were increased by inhibiting phosphodiesterases: the first 100 msec firing rates were significantly decreased by 3 min after IBMX (p <0.013), and the effects of hodulcin and IBMX appeared to be additive over a 10 min period (2,3). Next, the time course of firing within the first 100 msec was investigated. Firing began within 10 msec after contact of the sensillum with sucrose. The firing rate began to decline at about 50 msec, indicating the onset of adaptation. By 75 msec after contact, the rate was significantly decreased (p= 0.05). and there was a further significant decrease by 100 msec (p = 0.0003) (Repeated Measures ANOVAs) (4,5).

In biochemical studies, real time changes in concentrations of cAMP, cGMP and IP3 in the presence of (a) buffer (basal), (b) sucrose, or (c) sucrose and IBMX were measured in homogenates of taste sensilla from 8,000 flies using a quench flow system (QFM5) (6). Preliminary data show that after 25, 50 and 75 msec, sucrose induced clear decreases in cAMP (to 28-39% of basal levels). There was a marked increase in cAMP to 311% of basal levels after 100 msec, and then cAMP decreased again to 23% of basal levels after 200 msec of stimulation. Levels of cGMP also were decreased after 25, 50, 75, and 100 msec (55-71% basal), and then increased to 110% of basal at 200 msec. The sucrose/IBMX mixture decreased levels of cAMP at all time points and cGMP after 50, 75, and 100 msec. There was no change in IP3 from basal levels in sucrose-stimulated tissue at 25, 50 75 and 100 msec (4,5,7).

The preliminary biochemical results are consistent with the physiological effects of IBMX. If the biochemical results hold with further replications, the combined biochemical and physiological data would support our proposal that transduction involves cyclic nucleotide decreases (2,3), while cNMP increases may mediate adaptation or some other later aspect of responses to sucrose in P. regina (4,5,7). Since biochemical and physiological studies of second messenger roles in vertebrate sweet taste have not included data within the time frame of vertebrate behavioral responses to sweet stimuli, it is not possible to relate these data from P. regina to vertebrate sweet taste mechanisms in vertebrates at the present time.

Supported by NIH grants DC01563 and DC/OD02663 to LMK and NIH grant DC10754 to AIS.

REFERENCES:
1) Dethier, V.G. (1955) Quart. Rev. Biol. 30: 348-371.
2) Foster, K.D. and L.M.Kennedy (1994) Chem. Senses 19: 470.
3) Foster, K.D. and L.M. Kennedy (1995) Chem. Senses 20: 176.
4) Foster, K.D., A.L.Spielman (1998) Chem. Senses 23: 549.
5) Foster, K.D.(1998) Ph.D. Thesis, Clark University.
6) Spielman, A.I., H.Nagai, G.Sunavala, M.Dasso, H.Breer, I.Boeckhoff, T.Huque, G.Whitney and J.G.Brand (1996) Am. J. Physiol. 270 (Cell Physiol. 39), C926-C931.
7) Kennedy, L.M., K.D. Foster, A.L. Spielman and D.E.Kolodny (1998) Japn. J. Taste Smell Res. 5: 249-250.

7

Røstelien¹ Tonette, Stranden¹ Marit, Borg-Karlson² Anna-Karin and Mustaparta¹ Hanna

¹ Department of Zoology
Norwegian University of Science and Technology, Trondheim
tonetter@alfa.itea.ntnu.no

² Organic Chemistry
Royal Institute of Technology, Stockholm

Receptor neuron responses to naturally produced host plant odours in three moth species Heliothis virescens, Helicoverpa armigera and Helicoverpa assulta.

Gas chromatography linked to electrophysiological recordings from single receptor cells (GC-SCR) were emplyed to identify volatile compounds released by plants, that are detected by receptor neurones (RNs) in three species of the subfamily Heliothine, Heliothis virescens, Helicoverpa armigera and Helicoverpa assulta. Volatiles of various host (e.g. sunflower, tobacco and tomato) and non-host materials were collected by a "head-space" procedure and separated in the gas chromatograph (GC). Two GC-columns were installed, each with a split at the end, leading half of the effluent to the respective GC-detectors and the other half out of the GC-oven, over the insect antenna during recordings from single RNs. Thus, each neuron was tested for all compounds of a volatile mixture, separated in a polar and a nonpolar column. The chemical structures of the compounds were identified by coupled gas chromatography-mass spectrometry (GC-MS). The RNs recorded (N=3D80) on the antenna of the H. virescens females were classified into 12 different types, each type with neurons showing identical response spectra. Three frequently recorded RN types in H. virescens were also found on the female antenna of H. assulta and H. armigera. These neurones responded selectively to one or two compounds. One RN type, appearing in 80% of all experiments on the H. virescens females, responded with high sensitivity and selectivity to a particular hydrocarbone sesquiterpene present in host aswell as non-host materials. A second RN type responded selectively to another two sequiterpene hydrocarbons, and a third type, functionally similar in the three species, responded to the monoterpenoids, trans-*-ocimene and *-myrcene. These results suggest that plant odour information in the females of H. virescens, H. assulta and H. armigera, is received by specialized RN types that were conserved during evoulution.

8

Pophof Blanka

Max-Planck-Institut für Verhaltensphysiologie, Seewiesen
pophof@mpi-seewiesen.mpg.de

Modulation of the sensitivity of moth pheromone receptor neurons by octopamine

Octopamine is a biogenic amine which functions in insects as a neurotransmitter, neuromodulator and neurohormone. High amounts of octopamine are secreted into the antennal hearts of insects, pumping haemolymph into the antennae (1). Octopamine receptors were cloned from antennal tissue of the moths Bombyx mori and Heliothis virescens (2). In the wind tunnel octopamine improved orientation towards pheromone sources and blend discrimination in several moth species (3, 4). Therefore, effects of octopamine and its antagonist epinastine (5) on the sensory reception of the pheromone components (E,Z)-6,11-16:acetate and (E,Z)-6,11-16:aldehyde by specialized receptor neurons of Antheraea polyphemus were investigated. Octopamine and epinastine had no effect on the transepithelial potential of the preparation and on the spontaneous nerve impulse frequency in both types of receptor neurons. However, in the presence of a continuous low intensity pheromone stimulation (lower than stimulation with 1 pg pheromone per filter paper), octopamine significantly increased the nerve impulse frequency in the acetate receptor neuron, but not in the aldehyde receptor neuron.
Epinastine and octopamine had no significant effect on the maximum receptor potential amplitudes elicited in both types of receptor neurons by pheromone stimulation. The peak nerve impulse frequency in response to pheromone was in both receptor neurons significantly decreased by epinastine and increased by octopamine. In control experiments, haemolymph ringer increased the peak nerve impulse frequency slightly (not significantly). Possible mechanisms of the modulatory effect are either a direct action of octopamine on the receptor neurons, or a indirect action via effects on the metabolism.

References
1. Pass, G. et al. (1988) J. Exp. Biol. 135: 495-498
2. Nikisch-Rosenegk E. v. et al. (1996) Insect Biochem. & Mol. Biol. 26:817-827
3. Linn, C.E. & Roelofs W.L. (1986) Arch. Insect Biochem. & Physiol. 3:161-172.
4. Linn C.E. et al. (1992). Arch. Insect Biochem. & Physiol. 20: 265-284
5. Roeder, T. et al. (1998) Europ. J. Pharmacol 349:171-177

9

Grosmaitre* Xavier , Marion-Poll Frédéric , Renou Michel

Unité de Phytopharmacie et des Médiateurs chimiques
INRA Station de Phytopharmacie,
route de St-Cyr,
78 026 Versailles Cedex, FRANCE

Octopamine modulate sex pheromone receptors in a moth? An electrophysiological study.

Octopamine (OA) receptors are expressed within sensilla of moths. Using electrophysiological techniques, we evaluated if OA could modulate the olfactory responses of male Mamestra brassicae antennae or sensilla, stimulated with their major sex pheromone component (Z11-16:Ac), at time intervals ranging from a few ms up to a few hours. The electroantennogram (EAG) responses decreased significantly after OA injection into the abdomen, at doses ranging from 0.1-100 µg. EAG depletion reached a plateau within 10-30 min, and lasted up to a few hours. At the level of individual sensilla, OA injected at the base of the antenna rapidly increased the spontaneous firing activity of the neurons. This increase culminated at 30 min after injection, and then slowly decreased. In response to sex pheromone stimulation, the total number of spikes elicited was significantly higher in OA treated insects. However, this increase was primarily due to an elevated background firing activity. We could not detect any difference in the shape of the dose-response curves. These results indicate that OA receptors within sensilla are functional, and suggest that OA can modulate the sensitivity of the olfactory sensilla.

Key words: Octopamine, single sensillum recordings, electroantennogram, modulation, olfactory receptor neurons, insect.

*Corresponding author, tel + 33 (0)1.30.83.31.45, fax + 33(0)1.30.83.31.19

10

Thurm Ulrich

Institut für Neuro- und Verhaltensbiologie
Universität Münster
thurmu@uni-muenster.de

Contact-chemoreception at a naked cilium: stinging cells of Hydrozoa

Nematocytes (stinging cells) of Hydrozoa perform contact-chemoreception in the verbal sense: they detect solid chitin and water-unsoluble lecithin. Outputs of these cells are: excitatory synaptic transmission at the basal pole of the cell, like a secondary sensory cell, and discharge of the stinging capsule (cnidocyst) at the apical pole of the cell (1). In other respects the cells are similar to mono-ciliary sensory cells of invertebrates. Different from the sensory cells of insect sensilla, however, the stiffened cilium (cnidocil) of nematocytes projects into the surrounding sea or fresh water nakedly. The basis of the cnidocil is surrounded by a ring of stereovilli and is covered by a thin and loose cuticle. Cnidocil and stereovilli are interconnected and constitute a concentric hair bundle (2). This assembly is mechanically sensitive, comparable with the hair bundle of vertebrate hair cells. Deflection of the cnidocil induces a usual mechanoreceptor potential (3, 4).

On the other hand, chemical sensitivity is contributed by the naked membrane of the distal cnidociliary shaft: chemically adequate surface contact of e.g. 200 ms duration in the next seconds strongly changes the probability for a mechanoreceptor potential to elicit a postsynaptic response and to induce the discharge of the stinging capsule. In Hydra, chemical stimulation of the ciliary shaft also increases the responsiveness of mechanoreceptor channels. The chemical stimulation itself, i.e. the effective contact at the cnidocil without its deflection, is not followed by a receptor potential. That means, this chemosensory process has a modulatory effect on the probability of response of ion channels and synaptic secretion, comparable to the neural modulation that is known from ganglion cells to be induced by some synaptic inputs and to be performed via second messengers. The chemosensory modulation found in the nematocytes makes nematocytes functioning as bimodal sensory cells which do not simply add information on chemical and mechanical stimuli but detect chemo-mechanical coincidences only. These coincidences characterize prey.

Keeping mechanical stimuli constant (saturating), characteristics of the chemosensitivity were studied in the freshwater polyp Hydra and marine capitate polyps (Corynidae), using mainly the probability of discharge of capsules as the ouput (1). Strong effects were found for cnidocil contact of the water unsoluble phosphatidylcholine (in both systematic groups) and for purified chitin (in Hydra only). Though the water-soluble head segments of phosphatidylcholine were ineffective, the headgroup-analogue carbamylcholine became efficient at 10 pM. Similarly, the monomere of chitin, N-acetylglucosamine was efficient from 100 pM upwards. In Hydra, some but lower sensitivity also exists for aminoacids and purines; the site of reception was not determined for these soluble substances.

The contact-chemosensitivity of the distal cnidociliary shaft is parallelled by a locally high concentration of irregularly distributed intramembrane particles within this membrane area, as revealed by freeze fracturing (2). This indication for the distribution of receptor molecules corresponds to very similar results for the ciliary membranes of olfactory chemoreceptor cells of insects and vertebrates. Cnidocils were collected from mass cultures of Hydra using the finding that nematocytes are easily caused to decnidociliate (by 6% ethylene glycol, e.g.) (5). The cnidocils regenerate within about 3 hours. The sensitivity of the cells is lost with decnidocilation andis fully restituted with the time course of regeneration.

References:
1) Thurm et al., 1998, in "From Structure to Information in Sensory Systems" Ed. Taddei-Ferretti and Musio, World Scientific, 237-253.
2) Golz and Thurm, 1991, Cell Tissue Res. 263: 573-583.
3) Brinkmann et al., 1996, J. comp. Physiol. A 178: 125-138.
4) Thurm et al., 1998, in From Structure to Information in Sensory Systems" Ed. Taddei-Ferretti and Musio, World Scientific, 228-236.
5) Golz and Thurm, 1990, Protoplasma 155: 95-105.

11

Mikus Stefan & Kaib Manfred

Department of Animal Physiology
University of Bayreuth
D-95440 Bayreuth, Germany
mailto:stefan.mikus@uni-bayreuth.de

Evidence for a water and sodium receptor site in the same taste receptor cell in the termite Schedorhinotermes lamanianus (Isoptera: Rhinotermitidae)

The termite Schedorhinotermes lamanianus (Isoptera: Rhinotermitidae) is widely distributed in humid lowland forests throughout Africa south of the Sahara (Harris 1968). In a broader project on food selectivity of termites it has been shown that S. lamanianus is selective for food under undisturbed environmental conditions (Mikus et al. 1997, see also poster: Food choice in a termite: Characterization of a deterrent signal). This termite probes food sources by antennal taste receptors (Reinhard & Kaib, 1995). To understand the sensory input modulating behaviour during food choice we, therefore, investigated antennal sensilla by morphological and electrophysiological techniques. On each antennal segment three types of taste receptors can be morphologically distinguished, one of which - type TP I - contains five chemosensitive receptor cells. Using the tip recording technique, spikes of four of the five cells can be distinguished based on amplitude, width and frequency. In this study we focus on the physiology of cell 1. For this cell we provide strong evidence for two distinct receptor sites: a water site and a sodium site.
Cell 1 responds to water with a mean frequency of 15 Imp./200 ms. As typical for water receptor cells in insect sensilla, this response drops to zero with increasing stimulus concentrations of KCl and CaCl2. The response to pure water is reduced to about half at 0.01 mM CaCl2 and 1 mM KCl, thus making CaCl2 about two orders of magnitude more effective than KCl. In contrast, NaCl does not inhibit the cell at higher concentrations but increases the response with increasing concentrations. This suggests a site specific for sodium in addition to the site specific for water. To investigate this sodium site individually, the water site can be specifically inhibited by low concentrations of CaCl2. Under these experimental conditions the cell shows no activity to NaCl concentrations below 0.1 mM and shows for higher concentrations a dose-response curve typical for salt receptor cells in insects. The unexpected response of cell 1 to increasing NaCl concentrations with a response to water and elevated response levels at stimulation with high NaCl concentrations is due to a simultaneous response of both sites in a ratio depending on the stimulus concentration.
Food and water uptake of termites can be inhibited by increasing concentrations of various salts. However, this behaviour cannot be explained by the response characteristic of cell 1 to this salts.

Harris WV (1968) African termites of the genus Schedorhinotermes (Isoptera: Rhinotermitidae) an associated termitophiles (Lepidoptera: Tineidae). Proc. R. ent. Soc. Lond. 37: 103-113
Mikus S, Brandl R & Kaib M (1997): Tree-use by Schedorhinotermes lamanianus (Isoptera: Rhinotermitidae). Mitt. Dtsch. Ges. allg. angew. Ent. 11: 193-197
Reinhard J & Kaib M (1995) Interaction of pheromones during food exploitation by the termite Schedorhinotermes lamanianus. Physiol. Ent. 20: 266-272

12

Städler Erich, de Jong Ruurd, Baur Robert

Eidg. Forschungsanstalt
Schloss 334
CH-8820 Wädenswil, SWITZERLAND
Erich.Staedler@faw.admin.ch

Contact-chemoreceptor neuron as a detector for the isolation of a novel oviposition stimulant.

Two compounds present on the surface of cabbage (Brassica oleracea L.) leaves have been isolated and identified which stimulate very effectively oviposition in the cabbage root fly, Delia radicum L. and which are perceived by a specific receptor neuron in the tarsal sensillum C5 of the female fly. Activity of extracts and chromatographic fractions were bioassayed, using oviposition experiments and mainly electrophysiological recordings from the C5 tarsal contact chemoreceptor sensillum of female flies. The major compound is 1,2-dihydro-3-thia-4,10,10b-triaza-cyclopenta[.a.]fluorene-1-carboxylic acid (CIF-1). The second compound is the glycine conjugate of the major compound.

13

Ozaki, Mamiko

Department of Applied Biology
Faculty of Textile Science
Kyoto Institute of Technology
Kyoto, Japan
mamiko@ipc.kit.ac.jp

Effects of Olfactory Experience on Taste and Feeding

It has been generally accepted that food preference or appetite is influenced by olfactory memory accompanied by daily meal in childhood. Using the blowflies, Phormia regina, we examined influences of thus the memolized odor being associated with the taste of sugar on the appetite or the feeding sensitivity of the fly. Animals were exposed to odors while being fed with sucrose over the course of six days since emergence, starved for two days, and then tested for response to sucrose.

1. Preference in odors of natural foods
The analysis was initiated with complex odor mixtures, either odor of fish (Group 1), odor of cheese (Group 2) or no odor (Group 3). When flies were then tested for sucrose appetite in the absence of odor, Groups 1 and 2 showed a reduced appetite compared to Group 3. The appetites of Groups 1 and 2 recovered to the same level as Group 3 when presented with the odors to which they had previously been exposed, but not the other odor mixture. These results provide evidence that in Phormia, feeding behavior is influenced by prior olfactory experience, in an odor-specific manner.

2. Acquired or inherent preference in odor
Next, the analysis was applied to the examination with odors of pure chemicals, either odor of coumarin (Group 1), odor of limonene (Group 2) or no odor (Group 3). When flies were then tested for sucrose appetite in the absence of odor, Groups 1 and 2 showed a reduced appetite compared to Group 3. The appetite of Group 1 recovered to the same level as Group 3 when presented with the odor of coumarin, but the appetite of Group 2 did not recover even when presented with the odor of limonene. The flies of group 2 kept thus the poor appetite for weeks. These results suggest that Phormia inherently knows odor of limonene is hatable (Actually, limonene prescripted is toxic for the fly.), while odor of coumarine influences food preference with by prior experience.

3. Recall of a preferable memory of odor
We prepared another Group, where the flies were exposed to coumarine while being fed with sucrose for the first six days, chacked in the appetite, then exposed to limonene for the following 6 days, and chacked in the appetite again. At the first checking between the experiences of odors of coumarine and limonene, they showed a reduced appetite but the appetite recovered when presented with the odor of coumarine. At the checking after the experience of odor of limonene, the appetite was preserved to be reduced and would not recover when presented with the odor of coumarin. However, after several days being fed with no odor sucrose, the flies dramatically restored in their appetite when presented with the odor of coumarin. These results provide evidence that a prior olfactory experience, even once blocked in its memoly by unwelcomed experience of odor, was recalled and restore the appetite.

We have begun to investigate the neural basis of this phenomenon, by ablation experiments. Flies contain two olfactory organs, the antenna and the maxillary palp. Ablation of the maxillary palp, but not the antenna, eliminated the effect of olfactory experience; the maxillary palp appears to provide the sensory input used to construct an experiential memory associated with food.

14

Kleineidam Christoph, Romani Roberto & Isidoro Nunzio

Biozentrum
University of Wuerzburg
Dep. Zoology II
Am Hubland
D-97074 Wuerzburg

Ultrastructure of the sensillum ampullaceum and sensitization of the associated CO₂ receptor cell after long-term stimulation in the leaf cutting ant Atta sexdens

Social insects like the leaf-cutting ants can regulate the microclimate inside their nests. CO₂ is besides temperature and humidity an important parameter to be regulated. Therefore the ants have to measure the absolute concentration of CO₂ which is possible with the non adapting response character of the sensory cells. We investigated this unusual feature of an olfactory receptor cell with continuous CO₂ stimulation of 60 min. The concentration used in this experiment was in the range the ants normally encounter inside their nests (1.2% CO₂). After 10 min of stimulation the activity of the sensory cells was lowest and about 20% less than at the end of stimulation. 10 min after termination of the stimulus (to a concentration of 0.05% CO₂) the cellular activity was the same as before stimulation. Repeated stimulation revealed that the sensitization lasts for less than 10 min. The ecological significance of this response characteristic remains to be shown in behavioral experiments.
CO₂ receptors are associated with antennal sensilla of leaf-cutting ants. The sensilla are typical 'sensilla ampullacea' (sensu Snodgrass) consisting of a peg located in a deep, flask-shaped pit with a narrow opening in the antennomer cuticle. The peg surface has a constant number of about 20 finger-like ridges. These cuticular fingers taper from the base to the tip and are separated by furrows. At the base of the peg the walls of neighboring fingers are fused whereas at the distal part of the peg the sensillar wall is reduced and the profile of the fingers becomes mushroom-like. In the furrows, separating the cuticular fingers, pores are present which enlarge in the cuticular wall and form pore kettles. The cellular components of each sensillum consists of a single sensory neuron and an undefined number of accessory cells, wrapped around each other. The sensory cell somata is noticeably larger than those of other surrounding sensilla. The outer dendritic segment, enclosed in a dendritic sheath, enters the shaft lumen and extends unbranched to the middle of the peg where first branches are visible. At the tip the number of branches increased to more than 100 completely filling the shaft lumen. The dendritic sheath is conspicuously thick at the base of the peg and closely envelopes the outer dendritic segment up to the middle of the peg. From thereon to the tip the thinner dendritic sheath comes into contact with the sensillar wall of the cuticular fingers.

15

Koch Uwe T.

Universität Kaiserslautern
ukoch@rhrk.uni-kl.de

Interaction of Odor Stimuli in EAG Signals

EAG system have been used to measure concentrations of airborne pheromone in the field to provide new information for the improvement of mating disruption as a method of environmentally safe insect control. Recently, experiments were presented which seemed to prove that the EAG response to pheromone is influenced by the presence of plant odors in such a way that an EAG system would not be capable to measure pheromone quantitatively in the presence of plant odors.
We will present evidence that an artifact caused by the experimental arrangement may explain some of the postulated interactions in the receptive channels.
In addition, results of measurements will be presented and discussed in which the successful suppression of the plant odor background can be demonstrated in antennae of Sesamia nonagrioides.
The measurements involving Sesamia were made in collaboration with Dr. B. Férot, INRA.

16

Hölldobler Bert

Biozentrum, Würzburg
bertholl@mailserv.biozentrum.uni-Wuerzburg.de

Multimodal signals in Ant Societies

In the early stage of the study of chemical communication in animals, scientists assumed that insects' behavioral responses are released by single chemical substances, whereas in vertebrates and particularly in mammals, chemical signals are complex blends of substances mediating inter-individual recognition and interactions. However, most insect semiochemicals have proven to consist of several compounds, whereby either different components of a complex pheromone mixture may have different effects on the receiver, or the pattern of the particular mixture of substances comprises the signal. Thus with respect to the sophistication of their chemical communication systems, vertebrates and insects do not differ greatly.
It is also well known, and studied in several vertebrate groups, that communication can work through several sensory channels; that is, a signal can be composed of distinct physical components, transmitted simultaneously or in tightly paced sequence. Such cross-modal perception of composite signals has been investigated particularly well in humans and non-human primates and in birds, and, in fact, already was recognized by Darwin in his book "The expressions of the emotions in man and animals" (1872) where he noted that the power of communication by language is much enhanced by "the expressive movements of the face and body." The paragon example of multimodal communication signals in insects is the recruitment communication in honeybees, where the information about distance, location and quality of food source is transmitted through four or five sensory modalities. Though the dance-communication of the honeybees is clearly unique in its complexity of cross-modal integration, we find multimodal signals in many other forms of communication in social insects and particular in the ants.
We have learned that chemical signals mediating communication in ant societies are usually complex mixtures of substances with considerable variation in molecular composition and in relative proportions of components. Such multicomponent signals can be produced in single exocrine glands, but they can also be composed with secretions from several glands. This variation is often functional, identifying groups or specific actions on a variety of organizational levels. Chemical signals can be further combined with signals from other sensory modalities, such as vibrational or tactile stimuli. These kinds of accessory signals usually serve in modulatory communication, lowering the response threshold in the recipient for the actual releasing stimulus. Comparative studies suggest that modulatory signals evolved through ritualization from actions originally not related to the same behavioral context, and modulatory signals may further evolve to become independent releasing signals.

17

Sachse Silke, Galizia C. Giovanni, Rappert Angelika & Menzel Randolf

Institut für Neurobiologie,
FU Berlin,
Königin-Luise-Str. 28/30,
14195 Berlin

The conserved representation of different chemical structures in the honeybee antennal lobe: steps towards the spatial olfactory code

Odors are represented by specific ensembles of activated glomeruli in a combinatorial manner in the olfactory bulb of vertebrates or in the antennal lobe (AL) of insects. Here, we optically measured glomerular calcium activities in vivo in the honeybee Apis mellifera during olfactory stimulation with 36 pure chemicals differing systematically in carbon chain length (between C5 and C10) and functional group (alkanes, alcohols, aldehydes, ketones and carboxylic acids). We show their glomerular representations in 38 identified glomeruli of the honeybees 160. These results provide evidence that the activity patterns of the AL are conserved between individuals. Furthermore, we can show that the glomerular receptive range depends on carbon chain length. 20 of the 38 glomeruli measured show maximal activity in a specific range of chain length. Most of these responses have their maximum range at an intermediate chain length (50%), the other are either maximally stimulated by a short chain length (35%) or at a longer carbon chain (15%). Glomeruli specifically responding to functional groups are always also broadly tuned to a particular chain length. Decreasing vapor pressure and lack of electronegativity of the odorants reduces the overall AL activity. The position of functional groups of alcohols influences chain length coding. For example, the pattern elicited by 2-heptanol is intermediate to the patterns for 1-heptanol and 1-hexanol. These results allow conclusions concerning the interactions between olfactory receptors and odor molecules.

18

Van Naters van der Goes Wynand M., Inouchi Jun

Max-Planck-Institut für Verhaltensphysiologie, Seewiesen
naters@mpi-seewiesen.mpg.de

The GABA-blocker picrotoxin affects spatial patterns of neural activity in the antennal lobes of Periplaneta americana on stimulation with odours

Optical mapping of fluorescence changes on the primary olfactory center using potential and calcium sensitive probes shows that different odours generate different spatial patterns in the glomerulus population 1,2. Glomeruli are distinct morphological entities in the primary olfactor center of many species and consist of dense, synaptically connected arborizations of olfactory receptor neurons, local neurons, and neurons which project to higher brain centers. The majority of local neurons in the antennal lobes of the cockroach Periplaneta americana are GABAergic and are therefore thought to be inhibitory. In our poster, we present how disruption with picrotoxin of GABA-mediated connections affects the spatial pattern of active glomeruli on stimulation with three alcohols and their binary mixtures. After picrotoxin treatment, spatial patterns on stimulation with the three alcohols became significantly more similar. For the patterns generated by the binary mixtures, additivity of response patterns of component odours increased after pharmacological block of GABAergic connections. We suggest that, in addition to providing a better separation of odour quality for single odors, the inhibitory connections may help determine the across-fibre representation of odour-mixtures in tracts leaving the antennal lobes.

1 Joerges J, Küttner A, Galizia CG and Menzel R (1997) Nature 387, 285-288
2 Friedrich RW and Korsching SI (1997) Neuron 18, 737-752

19

Anton¹ Sylvia and Gadenne² Christophe

¹ Dept of Ecology, Lund University
Ecology Building, S-223 62 Lund, Sweden
Sylvia.Anton@ekol.lu.se

² INRA-Bordeaux,
Unite de Zoologie,
BP 81, 33883 Villenave d'Ornon cedex,
France

Hormonally-regulated sensitivity of olfactory antennal lobe neurons in a moth brain

In males of the noctuid moth, Agrotis ipsilon, behavioural responses to female-produced sex pheromone increase as a function of age and juvenile hormone (JH) status. Young, immature males with low levels of JH do not respond to sex pheromone, although their peripheral olfactory system is fully functional. Only sexually mature males with high JH levels show full mating behaviour in response to the female-produced sex pheromone. To investigate the neuronal basis for the JH-linked modulation of behaviour, we studied the response characteristics of olfactory interneurons in the antennal lobe (AL) of A. ipsilon males at different ages and with manipulated JH levels using intracellular recording techniques.
The sensitivity of AL interneurons increased significantly with age when the antennae were stimulated with different amounts of a behaviourally active sex pheromone blend. The sensitivity changes of AL interneurons concerning single sex pheromone components were, however, not as pronounced as for the blend. AL interneurons of mature A. ipsilon males, surgically deprived of JH by removing the corpora allata (CA), were significantly less sensitive to the pheromone blend and to single pheromone components than neurons of intact mature males. High sensitivity of the AL interneurons to the pheromone blend, but not as pronounced for single pheromone components, could be restored by injecting the hormone into JH-deprived males. Moreover, JH injected into young immature males induced high sensitivity to the sex pheromone blend, but only to a lesser extent to single pheromone components in AL interneurons. The effect of JH in general seems to be strongest for the behaviourally active sex pheromone blend, while sensitivity in response to single components might need a longer time to be affected by JH. Our results indicate that JH is involved in the plasticity of the adult moth brain by modulating the central nervous processing of olfactory information, thus allowing mate recognition and reproduction at the optimal time.

Supported by Swedish and French Research Councils (NFR, SJFR, INRA).

20

Ignell Rickard and Anton Sylvia

Dept. of Ecology, Lund University
Ecology Building
S-223 62 Lund

Effect of Juvenile Hormone on Aggregation Behaviour and Central Nervous Processing of Aggregation Pheromones in the Desert Locust

Juvenile hormone (JH) have been shown to effect the central nervous processing of aggregation pheromones in the antennal lobe (AL) of the desert locust, Schistocerca gregaria resulting in changes in aggregation behaviour.

The aggregation behaviour of desert locusts was studied using animals reared under crowded conditions. Male and female adult locusts were subjected to either of three treatments; allatectomy, sham-operation or left as control. Aggregation behaviour was monitored in individual insects as (a) large-scale, i. e. tendency to aggregate measured as an aggregation index (AI) and (b) small-scale, i. e. changes in behavioural parameters, in a two-choice bio-assay applying the main adult aggregation pheromone component phenylacetonitrile (PAN) in the treatment arena. Individual locusts were monitored at day 1, 8, 15, 22 and 29 after adult emergence in order to evaluate behavioural changes over time.

Control and sham-operated male locusts displayed large- and small-scale behavioural changes coinciding with the natural fluctuating JH-titre. These changes were characterised by an initial period of inactivity in the treatment area at day 1 with a low AI. Days 8 to 15 were characterised by a high AI and an increased behavioural activity, e. g. walking, leg- and antennal movements in the treatment area, which may suggest a searching behaviour. Locusts tested at day 22 showed a negative AI with an increased activity in the treatment area interpreted as an escape reaction, which coincides with the time of adult maturity. At day 29 locusts did not show any preference to or behavioural differences in neither of the two sides indicating a behavioural indifference to PAN. Female locusts displayed similar behaviour with the exception of the escape behaviour at day 22.

Allatectomised male locusts tested at day 8 showed similar behaviour and AI as control and sham-operated locusts. However, locusts tested from day 15 showed an increase in AI and in behavioural activity up to day 29. Allatectomised female locusts displayed a lower AI and behavioural activity than allatectomised males and control locusts early in the testing period. Behavioural activity and AI, however, increased and reached the corresponding level of control locusts at day 15 at the end of the test period.

Intracellular recordings from projection neurons (PNs) in the antennal lobe of 8 and 29 days old control locusts revealed differences in the number of neurons responding suggesting a regulatory role of JH on the activity of PNs. The number of projection neurons responding in 29 days old allatectomised locusts, however, did not differ from PNs of 8 days old control locusts. In order to exclude the possibility of a diminished sensory input to the AL, elektroantennograms (EAG) on control and allatectomised locust antennae were performed. No significant differences in EAG amplitude between the two groups were observed. Whether or not the response spectra of individual PNs change during development is still unclear.

We conclude that the observed changes in aggregation behaviour may be regulated through a direct or an indirect effect of JH on the central nervous processing. We postulate that the observed behavioural indifference to PAN at day 29 arise due to a significant decrease in the central nervous processing of aggregation pheromone components.

21

Kalinová¹ Blanka and Hansson² Bill S.

¹ Institute of Organic Chemistry and Biochemistry
Academy of Sciences of the Czech Republic
Flemingovo nám. 2
166 10 Praha 6

² Department of Ecology
Lund University
S-233 62 Lund, Sweden

Physiological responses and central nervous projections of antennal olfactory receptor neurons in the sphinx moth, Manduca sexta (Lepidoptera, Sphingidae)

Olfactory receptor neurons in two morphologically different sensillum trichodea (type I and type II) on the male M. sexta antenna were investigated electrophysiologically when stimulated with sex pheromone components and some host plant odours. Neurons present in both type I and type II sensilla responded to pheromonal stimuli. The most abundant receptor neurons found on the male antenna were tuned to E10,Z12-hexadecadienal (bombykal) and E10,E12,Z14-hexadecatrienal, the two most prominent pheromone components in the female pheromone blend. These receptor neurons were associated with both sensillum types. Neurons tuned to E10,E12,E14-hexadecatrienal were occasionally found in type I, but were more frequent in type II. Receptor neuron types tuned to minor pheromone components Z11-hexadecenal and E10,E12-hexadecadienal respectively, were found in type II trichoid sensilla. Neurons sensitive to host plant odours were found among shorter sensilla and were never present in a sensillum with neurons responding to pheromonal stimuli.

The receptor neurons in type I sensilla trichodea were selectively stained using cobalt lysine. The selectivity of staining was achieved by stimulation by the key stimulus (e.g. bombykal and/or E10,E12,Z14-hexadecatrienal) during the staining period. Stimulation facilitates cobalt uptake by the activated neuron. The arborization of primary afferents within the antennal lobe were reconstructed and projection areas were determined. Different projection patterns were observed when different stimuli were used. After bombykal stimulation, stained afferent fibres projected into the cumulus part of the macroglomerular complex. When E10,E12,Z14-hexadecatrienal (or its more stable mimic E11,Z13-pentadecadienal) was applied during the staining period, several single neurons projecting to the toroid were detected. More frequently, both neurons within the sensillum were stained revealing one neuron projecting to the cumulus, and a second one to the toroid. These results suggest 1) that input from the two neuronal types are separated and 2) that arborization areas of primary afferents and projection neurons of similar specificity do not correlate. If true, the functional organisation of the macroglomerular complex in Manduca sexta is more complex than previously thought.

22

Berg BG, Almaas TJ, Mustaparta H.

Dept of Zoology Nowegian University of Science and Technology
NTNU, N-7034 Trondheim, Norway
Bente.Berg@chembio.ntnu.no

The Macroglomerular Complex in Two Related Species of Moths: Specified Subdivision According to Input Information.

Projections of functionally different receptor neuron types in the macroglomerular complex (MGC) of the antennal lobe have been studied in two allopatric species of heliothine moths: Heliothis virescens using cis-11-hexadecenal and Helicoverpa assulta using cis-9-hexadecenal as the major pheromone component. Histological investigations including camera-lucidae drawings and 3-D computer reconstructions have revealed that the MGC consists of four compartments in H. virescens and three in H. assulta. In both species one large unit, the cumulus, was located at the entrance of the antennal nerve. A second larger compartment was located dorso-medially of the cumulus, and one or two smaller units ventrally. Electrophysiological recordings from single sensilla (tip-recordings) combined with cobalt application revealed marked axon terminals in the MGC. In both species the cumulus received input from receptor neurons tuned to the major pheromone component. The dorso-medial compartment received input from neurons tuned to cis-9-tetradecenal, which is important as a second pheromone component in H. virescens and as a behavioural antagonist in H. assulta. Whereas the two ventral compartments in H. virescens received input from neurons responding to two behavioural antagonists, the single ventral compartment in H. assulta seemed to receive information about the second pheromone component. Thus, a striking similarity is expressed between these and other heliothine species by the function of the cumulus as a relay for transmitting information about the major pheromone component, whereas the function of the other compartments differs between species.

23

Galizia C.G., Sachse S., Brandt R., ¹Berg B. and ¹Mustaparta H.

Inst. für Neurobiologie, Freie Universität Berlin,
Königin Luise Str. 28-30,
D-14195 Berlin.

¹Dept. of Zoology, Norwegian University of Science and Technology,
N-7034 Trondheim, Norway

Mapping responses in the antennal lobe of the moth Heliothis virescens using calsium imaging.

Receptor neurons responding selectively to insect and plant produced compounds have been identified in the tobacco budworm moth H. virescens. In the males, four sex specific neuron types are classified, each projecting in one compartment of the macroglomerular complex (MGC) of the antennal lobe. Neurons responding selectively to the pheromone components A (cis-11-hexadecenal) and B (cis-9-tetradecenal), projected in the cumulus and dorso-medial compartments, respectively, whereas the neurons responding to interspecific signals projected in the two ventral compartments. In the females, several plant odour receptor neuron types have been identified, each responding to one or two compounds of mono- and sesquiterpenes. In this study, we used in-vivo calsium imaging in order to measure the responses of olfactory glomeruli. We found that in the male moth stimulation with the pheromone component A and B leads to activation in the area of the cumulus and the dorso-medial compartments, whereas the interspecific signals elicit responses in the ventral area. No responses to the insect produced compounds were found in the females. Stimulation with the plant odours did not elicit reponses in the area of the MGC of males, but in the "ordinary" glomeruli. In both sexes, different glomeruli were activated by the various odorants and blends. For instance, head-space blends of the host plants tobacco and sunflower elicited activity in 2-3 restricted areas (glomeruli), with overlap of one. Ocimen and trans-b-myrcene, which activate the same receptor neuron type, elicited responses in one antennal lobe area which overlapped for the two stimuli. A sesqiterpene fraction of Pipera cubeba elicited activity in a clustered area of several glomeruli, including the area activated by one sesuiterpene present in the fraction and known to activate the majority of receptor neurons of the females.
These data confirm and extend results from electrophysiological recordings of pheromone receptors and antennal lobe projection neurons, showing that pheromone information is coded in a combinatorial manner with each glomerulus corresponding to one information channel. The results further suggest that the principle for coding plant odour information in the antennal lobe is also based on a functional organization of the ordinary glomeruli, where one or a few specific glomeruli are activated by a particular compound, similar to what has been shown in other species.

24

Christensen Thomas A.

ARL Div. of Neurobiology
Gould-Simpson University of Arizona
625 Gould-Simpson
PO BOX 210077
Tucson AZ 85721-0077
tc@manduca.neurobio.arizona.edu

Contextual Influences on the Central Processing of Chemical Signals by Ensembles of Olfactory Interneurons in the Moth Antennal Lobe

In order to understand the coding mechanisms that are used by olfactory circuits in the brain to discriminate different odorants, it is important to consider that this task must be performed under a wide range of stimulus conditions. Using multiunit recording methods, we are investigating how changes in the olfactory environment affect activity patterns in ensembles of neurons in the glomeruli of the macroglomerular complex (MGC) of male Manduca sexta. As in previous studies of this neuropil, we observed robust and selective responses to the species-specific sex-pheromone blend. The overall pattern of the ensemble response reflected the time course of the stimulus, but there was considerable variability in a number of different spike-train parameters from trial to trial. These included spike latency, the number of evoked spikes, as well as the temporal pattern of spike trains. With longer duration stimuli, moreover, new units were recruited into the ensemble. Since functional interactions among neurons may enhance the representation of sensory stimuli, we also examined the responses of pairs of MGC neurons in the ensemble for evidence of odor-induced synchronization. Pairwise cross-correlations show that coincident events tend to occur only early in the response to a single pulse, regardless of stimulus duration (from 50-1000 msec). We furthermore found that a pulsatile stimulus evokes a greater degree of coactivity in the ensemble than does a continuous stimulus over the same time period. These data show that the chemical identity of the stimulus is not represented by a unique temporal code in the MGC, but these ensemble patterns are greatly affected by stimulus context. Furthermore, response patterns are not constrained by a coherent network event (like an oscillation), but instead faithfully represent the stimulus dynamics, thus optimizing the amount of information available for higher processing. The implications of these results for regulating mate-seeking behaviors will be discussed.

Supported by a contract from DARPA.

25

Baker Thomas C.

Department of Entomology
Iowa State University
Ames, Iowa 50011
U.S.A.

Neuroethology of olfaction in heliothine moths

We have discovered a type of sensillum on the antennae of male corn earworm moths, Helicoverpa zea, that house both the agonistic and antagonistic types of receptor neurons (RN) that explain both the upwind flight attraction of males to their own species' pheromone blend and arrestment to the quite similar blends emitted by females of three other sympatric North American heliothine species. The first RN type is a large-spiking neuron that is most sensitive to (Z)-9-hexadecenal (Z9-16:Ald), the secondary H. zea pheromone component that along with the major component, (Z)-11-hexadecenal (Z11-16:Ald), causes attraction to the female. This RN is also equally responsive to (Z)-9-tetradecenal (Z9-14:Ald), not a H. zea pheromone component.

The second type of RN in this sensillum is a small-spiking neuron that is again responsive to Z9-14:Ald (which in larger proportions acts as a strong antagonist to upwind flight) but this RN is actually more sensitive to two other strong behavioral antagonists, (Z)-11-hexadecenyl acetate and (Z)-11-hexadecenol (Z11-16:Ac and Z11-16:OH, respectively. Thus, activation of this single broadly-tuned "antagonist" RN could explain why H. zea males will orient only to their conspecific females. These latter two compounds are emitted by females of three other North American species, Heliothis subflexa, Heliothis phloxiphaga, and Z11-14:Ald by Heliothis virescens, as agonists in their blends.

We have found recently by experimentally generating odor filaments that are either always, or never coincident, that male H. zea can behaviorally distinguish filaments of pheromone and the Z11-16:Ac antagonist that are spatially separated by no more than one millimeter and temporally separated by 0.001s. We suggest that this high degree of filament resolution begins at the antennal receptor level, in the co-compartmentalization of the antagonist- and antagonist-tuned receptor neurons in the same sensilla. Only by sampling the air at the same point in space and time can two functionally different receptor neurons, tuned to different compounds and by default, oblivious to the arrival of the compound detected by the other, determine with optimal accuracy whether there is complete spatial an temporal coincidence in the arrival of two different odorants. Due to reduced fitness caused by mating with non-conspecific females, male moths have evolved the ability to detect key compounds from non-conspecific females that will prevent upwind flight, even in response to emissions that are quite similar to, but not identical with, their own females' pheromone. The trade-off for having RNs tuned to such antagonistic compounds and for ceasing upwind flight is that males could miss opportunities to find their own females in plumes from non-conspecifics that are co-mingled in the shifting wind-fields. A male with the ability to respond to the pure strands of pheromone and to continue to fly upwind in the presence of antagonist filaments due to the incomplete admixing of the strands should be at a fitness advantage in being able to arrive at a conspecific female and mate. Conversely, our results suggest that if a male can discern that every portion of every filament hitting his antenna contains the antagonist along with the pheromone components, then the plume that he detects must be being emitted upstream as a single strand from a single gland, one that cannot possibly be from his own female.

Supported by USDA/NRICGP grants 95-37301-1805 and 97-02521 to T.C.Baker

26

Hansson Bill S. & Hong Lei

Department of Ecology, Lund University
S-223 62 LUND, Sweden
bill.hansson@ekol.lu.se

Protocerebral Processing of Olfactory Input in the Male Turnip Moth, Agrotis segetum

The functional morphology and physiology of sex pheromone-responding antennal receptor neurons and antennal lobe neurons have been investigated in detail in several moth species. The fate of the pheromone signal at higher brain levels is, however, less well known. We have performed intracellular recording and subsequent staining among protocerebral neurons in the turnip moth male. The pheromone communication syustem of this moth has previously been studied in detail, as has the antennal receptor system and the antennal lobe neurons.

In studies of the protocerebrum in Manduca sexta, it was shown that the pheromone blend played an important role at these higher levels, and elicited specific, so called long lasting excitations. In A. segetum, we have earlier identified blend-specific neurons in the antennal lobe, and our hypothesis was that we would find higher and higher proportions of these neurons at higher levels in the system. Surprisingly, we found that this is not the case. A large proportion of the neurons investigated responded exclusively to one of the pheromone components. Labelled lines for single pheromone components do thus persist to and beyond protocerebral levels in the turnip moth.

Among the neurons investigated we found several morphological types. These are presently being analysed and will be presented along with the physiological data at the ESITO VI meeting.

27

Strausfeld Nicholas J.

Division of Neurobiology
University of Arizona
Tucson, AZ 85721, USA
flybrain@neurobio.arizona.edu

Distributed olfactory imformation processing in the insect nervous system

Most of our knowledge about olfactory pathways in the insect central nervous system derives from studies at three levels: olfactory receptor terminals in the antennal lobes, projection neurons from the antennal lobes, and descending neurons supplying thoracic motor circuits. Idealized elements from these levels could be constructed into a circuit that would mediate up-wind flight to an odor target. However, comparative studies of neopteran insects find no evidence for a direct pathway between the antennal lobes and premotor channels. Instead, information about odors is routed through many diverse regions of the insect brain. Indeed, amongst all the sensory modalities, that of olfaction appears to be the most widely distributed amongst different brain areas.

Divergence of olfactory pathways

Intracellular and neuroanatomical studies of the cockroach (Strausfeld, 1976; Malun et al., 1993) demonstrate that antennal lobe projection neurons supply a second olfactory neuropil in the superior lateral protocerebrum (Stocker, 1994; Homberg et al., 1989; Li and Strausfeld, 1997). This relationship has been identified in Diptera, Orthoptera, Hymenoptera, Lepidoptera, Dictyoptera, Coleoptera and Blattodea. In all these, the superior lateral protocerebrum receives antennal lobe inputs via three ascending pathways: the inner, medial, and outer antennocerebral tracts (Homberg et al., 1989). The inner antennocerebral tract differs from the others in providing collaterals to the mushroom body calyces. In Periplaneta, for example, different groups of glomeruli provide inputs to different zones of calycal neuropil. Other afferents, most of which are GABAergic and carry acoustic, tactile and visual information, also distinguish calyx zones.

Mushroom body lobes and calyces have different evolutionary histories

The calyces are neopteran innovations, as are the antennal lobes. The primitive calyx-less condition is represented in paleopteran insects such as zygentomans and odonates whose mushroom bodies lack calyces. In these groups afferents invade the mushroom body alpha and beta lobes. This is also seen in neopterans. In cockroaches, for example, afferents to the lobes carry information about visual, tactile, and acoustic events. They are not GABAergic and are likely to contain excitatory transmitter substances. The lobes also give rise to many dozens of multimodal output (efferent) neurons, the axons of which project back to the superior lateral and superior medial protocerebra.

Kenyon cells provide local circuits in the lobes

Kenyon cell axons make diverse synaptic connections. They are postsynaptic to afferent endings, are pre- and postsynaptic to each other, and are presynaptic to the dendrites of efferent neurons. Thus, Kenyon cells essentially provide local circuits between inputs and outputs in the lobes. The function of the lobes appears to be that of determining which sensory stimuli are relevant to context. This is evidenced by the responses of efferent neurons, which are context-dependent: activation of an efferent neuron depends on the specific coincidence of modalities or their sequences.

What does this mean with respect to olfactory processing?

The evolutionary acquisition of dendrites by Kenyon cells has resulted in a specialized region of the mushroom bodies called the calyces. With the acquisition of olfactory and (in hymenopterans) visual inputs to the calyces, local circuits between afferents and efferents are regulated by these modalities. Thus, olfaction (and in some insect groups vision) is the ubiquitous modality against which all other sensory signals are assessed by circuits in the mushroom body lobes. Thus, mushroom bodies are not higher order odor-discriminating neuropils but are odor-modulated neuropils that determine which of the many stimuli challenging the insect are relevant for further processing. Odor discrimination itself occurs in the antennal lobes, which send information to the superior lateral protocerebrum. This region is also supplied by multimodal inputs from the mushroom body lobes.

Pathways from the superior lateral protocerebrum to premotor neurons

Intracellular studies of efferent neurons from the superior lateral protocerebrum suggest that although none respond exclusively to odorants, no responses are independent of odorants. Neurons from the superior lateral protocerebrum target several other protocerebral nuclei that then send connections to the central complex. The central complex is a center that supervises locomotion (Strauss and Heisenberg, 1993). It receives information about gait from the thoracic ganglia and integrates this with information about sensory events that have been analyzed by circuits in the mushroom bodies and in distributed protocerebral nuclei. Outputs from the central complex terminate in the lateral accessory lobes. There they contact descending neurons supplying motor circuits in the thoracic ganglia that are involved in odor-guided orientation (Kanzaki et al., 1994).

Homberg U, Montague RA, Hildebrand JG. 1989. Anatomy of antenno-cerebral pathways in the brain of the sphinx moth Manduca sexta. Cell Tissue Res. 254: 255-281.
Kanzaki R, Ikeda A, Shibuya T. 1994. Morphology and physiology of pheromone-triggered flip-flop descending interneurons of the male silkworm moth, Bombyx mori. J. Comp. Physiol. A. 175:1-14.
Li Y-S, Strausfeld NJ. 1997. Morphology and sensory modality of mushroom body efferent neurons in the brain of the cockroach, Periplaneta americana. J. Comp. Neurol. 387:631-650.
Malun, D, Waldow U, Kraus D, Boeckh J. 1993. Connections between the deutocerebrum and the protocerebrum, and neuroanatomy of several classes of deutocerebral projection neurons in the brain of male Periplaneta americana. J. Comp. Neurol. 329:143-162.
Stocker RFM. 1994. The organization of the chemosensory system in Drosophila melanogaster. Cell Tissue Res. 275:3-26.
Strausfeld NJ. 1976. Atlas of an Insect Brain. Heidelberg: Springer.
Strauss R, Heisenberg M. 1993. A higher control center of locomotor behavior in the Drosophila brain. J. Neurosci. 13:1852-1891.

28

Rössler Wolfgang, Tolbert Leslie P., Hildebrand John G.

ARL Division of Neurobiology,
University of Arizona,
Tucson, AZ 85721, USA

Establishment of an odotopic organization of olfactory glomeruli in the developing antennal lobe of Manduca sexta.

Primary olfactory centers in most vertebrates and invertebrates typically are organized into spheroidal compartments of neuropil called glomeruli. In the glomeruli, axons of olfactory receptor cells (ORCs) and dendrites of local interneurons and output neurons form a dense synaptic neuropil. Recent studies suggest that olfactory glomeruli function as discrete processing units for olfactory information, which is represented spatially across the glomerular neuropil. Only little is known about the establishment of such an odotopic organization in primary olfactory centers. The pheromone-specific macroglomerular complex (MGC) in the antennal lobes (ALs) of male Manduca sexta is one of the best studied examples for an odotopic organization of olfactory glomeruli. To gain a more precise understanding of the formation of individual glomeruli with known odor specificity, we are studying the morphogenesis of the MGC during postembryonic metamorphic larval-adult development.

The ALs of adult M. sexta males contain 64 plus-minus 1 'ordinary' glomeruli and the MGC, a set of 3 anatomically specialized glomeruli that process information about individual components of the female's sex pheromone. Recently, we also identified a pair of sexually dimorphic glomeruli unique to the female AL, the lateral (or large) female glomeruli (LFGs). We explored the cellular events underlying odotopic segregation of the MGC glomeruli during normal development and after specific experimental manipulations using fluorescent dye-labeling techniques and confocal microscopy. During normal development, the formation of sexually dimorphic glomeruli diverges in males and females as soon as the first antennal ORC axons grow into the AL and male- and female-specific ORC axons form protoglomerular branches in the characteristic MGC or LFG pattern (Rössler et al. 1998, J.Comp.Neurol. 396:415). Trans-sexual transplantation experiments revealed that axons of MGC-specific ORCs from grafted male antennae were also able to express their typical branching patterns in the corresponding regions of the female antennal lobe (Rössler et al. 1999, J. Neurobiol. 38:521-541). Even in cases in which the antennal nerve entered the AL via an abnormal route, MGC-specific receptor axons grew to their correct target region within the AL. Within ectopic neuromas formed by misrouted antennal nerves that did not contact the AL, MGC-specific receptor axons did not become organized odotopically but instead became completely intermixed with other ORC axons terminating in glomerulus-like structures. This suggests that guidance and odotopic segregation of ORC axons depends on cues present in the AL tissue. To test this hypothesis further, we surgically removed lateral parts of antennal lobe tissue before ORC-axon ingrowth. In this case, MGC-specific axons expanded their normal territory and MGC subdivisions did not form normally.
What is the influence of ORC axons on the organization of cellular components within the AL? After formation of protoglomerular branches in the MGC or LFG pattern, ORC axons induced the establishment of glial borders. Glial cells in a female AL were able to respond to innervation by MGC-specific ORC axons from a grafted male antenna by forming borders in the typical MGC-pattern. In the opposite case, glial borders arranged in the LFG pattern. Sex-specific ORC axons also had profound influences on the morphology of AL neurons. Following trans-sexual transplantation of antennae, normally LFG-specific uniglomerular projection neurons (uPNs) in female ALs extended branches into the induced MGC, and branches of normally MGC-specific uPNs in male ALs arborized in the induced LFGs. The morphology of a unique serotonin-immunoreactive neuron showed a similar plasticity in response to antennal innervation. To find out if there is a sensitive period for these intercellular interactions, we experimentally manipulated the timing of ORC-axon ingrowth by selectively cooling or warming the developing antenna. The results showed that proper formation of glial borders requires interaction of ORC axons with AL glial cells during a specific time window in early development. After delayed ingrowth of ORC axons, glial cells were no longer able to migrate and form normal glomerular borders. In the MGC, disruption of glial borders resulted in a less discrete segregation of neuronal processes into subdivisions. Precocious ingrowth of ORC axons resulted in an increase in targeting errors and a significant number of axons projected to areas outside the typical MGC position.

Our results suggest an important organizing role of ORC axons in the establishment of an odotopic organization of glomeruli during development of the primary olfactory centers of the moth. This organizing potential of ORC axons, however, depends on signals within the AL and cell-cell communication with AL glial cells which are necessary for odotopic segregation of axon terminals. ORC axons need to interact with AL glial cells during a sensitive period to induce the formation glomerular borders which are important to stabilize developing glomeruli. Interaction of ingrowing ORC axons with AL neurons influences the dendritic morphology of these neurons. We currently are beginning to explore the possible nature of signals involved in ORC-axon targeting and interaction of ORC-axons with AL glial cells and neurons.

29

Meijerink Jocelijn and van Loon Joop J. A.

Laboratory of Entomology, Wageningen Agricultural University
P.O. Box 8031, 6700 EH Wageningen, The Netherlands.

Electrophysiological responses of the malaria mosquito, Anopheles gambiae to sweat-borne components

30

Van den Broek Ingrid V.F. and Den Otter Cornelis J.

Group Sensory Organs and Behaviour, Department of Animal Physiology,
University of Groningen,
P.O. Box 14,
9750 AA Haren,
The Netherlands.
broekivf@biol.rug.nl

Properties of the olfactory systems of Anopheles mosquitoes with different host preferences

Mosquito species of the African Anopheles gambiae complex show different vectorial capacities for human malaria, depending on their anthropophilic biting preferences. We compared the olfactory sensitivity of female mosquitoes of three sibling species with different host preferences: the antropophilic An. gambiae s.s., the opportunistic An. arabiensis and the zoophilic An. quadriannulatus. Responses of sensory cells associated with the most important antennal olfactory hairs, the sensilla trichodea and the grooved pegs, were recorded. Stimuli were vapours of some natural host materials and various synthetic host odours.

Both the grooved pegs and the sensilla trichodea responded to the vapours from natural materials such as cow manure and urine and Limburger cheese. Sensilla trichodea responded to carboxylic acids, 1-octen-3-ol, 3-methylphenol, ammonia and butylamine. Grooved pegs also responded to carboxylic acids, ammonia and butylamine, and occasionally to lactic acid and acetone. The majority of grooved peg sensilla were responsive to water vapour. Inhibitory as well as excitatory responses to synthetic host odours were found from both sensilla-types. Differences in host preferences appeared to be reflected in differences in the number and sensitivities of cells of trichoid sensilla responding to vapours of carboxylic acids and 1-octen-3-ol, of the anthropophilic An. gambiae and the zoophilic An. quadriannulatus. No differences were found between the responses of grooved pegs to synthetic host odours. The three species responded similarly to the natural odour mixtures, however some differences between the response profiles of the two types of olfactory sensilla may exist.

31

Bosch Oliver Jörg, Geier Martin and Boeckh Jürgen

Institute for Zoology,
University of Regensburg,
Universitaetsstr. 31
D-93040 Regensburg
Germany
oliver.bosch@biologie.uni-regensburg.de

Identified volatiles emitted from human skin attract female Aedes aegypti

In a dual choice olfactometer odour components identified on human skin were tested in different concentrations for their attractivness to female Aedes aegypti (L.) (Diptera: Culicidae). None of the odours were attractive when tested alone. However, in combination with L-(+)-lactic acid, a major component on human skin, some of the odours increased the attractivness of the latter stimulus. These compounds had no effect in combination with carbon dioxid, a attractive component of human breath, although carbon dioxid synergize the effect of L-(+)-lactic acid. This indicates (1) the need of L-(+)-lactic acid in the olfactory host finding of Aedes aegypti, and (2) that its essential role as a synergist can not be compensated by carbon dioxid. Furthermore, a step wise increase of attractiveness was found when distinct combinations of odours were mixed to lactic acid.

Key words: mosquito, Aedes aegypti, host odour, olfaction, attractant

32

Steib B. M., Geier M. and Boeckh J.

Institut fuer Zoologie, Universitaet Regensburg
93040 Regensburg, Germany
birgit.steib@biologie.uni-regensburg.de

What makes us attractive to yellow fever mosquitoes - The effect of lactic acid on host selection of Aedes aegypti

Host selection of yellow fever mosquitoes, Aedes aegypti (L.) was observed in a behavioural study. It was investigated (a) whether body odours of human individuals differ in their attractiveness to yellow fever mosquitoes, and (b) whether mosquitoes respond to skin odours of mammals, other than humans. The odour samples were tested alone as well as in combination with lactic acid, and the responses to the odours were studied in a Y-tube wind tunnel. As a measure for attraction the percentage of mosquitoes was taken that were trapped at the two upwind ends of the wind tunnel. Additionally, different aspects of the response of single mosquitoes to host odours were observed in a flight chamber (e.g. flying, landing, probing). By comparing rubbings from hands of different humans in a two-choice test in the wind tunnel it could be shown that certain individuals attract a significantly higher percentage of mosquitoes than others. In the flight chamber mosquitoes were shown to prefer the same odours as they spent more time in the sector with these odours. It could be demonstrated that the addition of lactic acid to less attractive odour samples significantly influences mosquito preference. Such "blended" odours were preferred even to the originally most attractive skin odours. The attractiveness of different animals (cow, goat, cat) was tested by using extracts from skin rubbings. Only one to ten percent of the mosquitoes responded to such odour samples. The attractivenes of animal odours, however, increased significantly when lactic acid was added. In combination with lactic acid odour samples of some mammals were as attractive as extracts from human skin rubbings.

33

Geier M., Bosch O.J. and Boeckh J.

Institut fuer Zoologie
Universitaet Regensburg
Universitaetsstr. 31
martin.geier@biologie.uni-regensburg.de

Effects of plume structure on upwind flights of mosquitoes towards host odours

Both the concentration and fine-scale plume structure of host odours influence the upwind flight of female mosquitoes Aedes aegypti (L.) (Diptera: Culicidae) in a wind tunnel. The attractive effects of carbon dioxide, human skin odour, and L-(+)-lactic acid were tested in (a) homogeneous, (b) turbulent and (c) filamentous odour plumes. With carbon dioxide, the percentage of upwind-flying mosquitoes increased with the increasing fluctuations in concentration that occur in turbulent and filamentous plumes. In homogeneous plumes an initial activation effect was observed, but sustained upwind flights were less frequent than in the other plumes. The opposite was found with plumes of human skin odour: the highest number of mosquitoes flew upwind in the homogeneous plume, whereas in turbulent or filamentous plumes their numbers were significantly lower. Regardless of plume type, the percentage of upwind-flying mosquitoes increased with increasing concentrations of carbon dioxide and of skin odour. With L-(+)-lactic acid, the dose-response characteristics were not consistent, and the relative effects of different plume types upon upwind flights were different within different ranges of concentration. Even maximum reactions to this compound were modest compared with those to carbon dioxide or to skin odour. Our findings demonstrate (1) that mosquitoes are able to orient upwind under continuous odour stimulation and (2) that upwind flight is dependent upon plume structure in different ways for different host odour components.

34

McMahon C., Guerin P.M.

Dep. of Animal Physiology
Institute of Zoology
University of Neuchâtel
Rue Emile Argand 11
CH-2007 Neuchâtel, Switzerland
conor.mcmahon@zool.unine.ch

Behavioural responses of the tropical bont tick, Amblyomma variegatum, to components of its pheromone and to vertebrate odours on a servosphere.

35

Donzé Gérard and Guerin Patrick

Institute of Zoology, University of Neuchâtel,
Emile-Argand 11, BO 2,
2007 Neuchâtel, Switzerland
gerard.donze@zool.unine.ch

Ticks love signals transited the ruminant stomachs

36

Widmer A., Vlimant M., Diehl P.A. and Guerin P.M.

Universite de Neuchâtel
Institute of Zoologie
Emile-Argand 11
CH-2007 Neuchâtel
alexandre.widmer@zool.unine.ch

Cold receptors on the body surface of the tick Amblyomma variegatum

37

Visser J.H. & Piron P.G.M.

Research Institute for Plant Protection (IPO-DLO)
P.O. Box 9060, 6700 GW Wageningen, The Netherlands
J.H.Visser@IPO.DLO.NL

Sensory and behavioural responses of aphids to plant odours

The aphid antennae bear specific sets of olfactory neurones, placoid sensilla in the so-called distal and proximal primary rhinaria on segments six and five, and, in winged aphids, the secondary rhinaria on segments three and four. By means of the electroantennogram technique, olfaction of plant odour components was studied in a number of aphid species, namely the peach-potato aphid Myzus persicae, the black bean aphid Aphis fabae, the cabbage aphid Brevicoryne brassicae and the vetch aphid Megoura viciae. The shapes of electroantennograms revealed the three elements of sensory transduction: (1) the transport of odour molecules towards receptor sites, (2) the interaction of odour molecules with receptor sites, and (3) the deactication after stimulation, measured by the rise, peak and decay of EAGs, respectively. Besides striking similarities in the odour response profiles of species, differences occur between species, clones, winged and wingless aphids, and responses were affected by the food on which aphids have fed. In order to investigate the role of plant odours in the host-plant selection of aphids special olfactometers were constructed for recording their responses towards plant odours and volatile components.

38

Kelling Frits J. and den Otter Cornelis J.

Group Sensory Organs and Behaviour
Department of Animal Physiology
University of Groningen
P.O. Box 14, 9750 AA Haren, the Netherlands
f.j.kelling@biol.rug.nl

Does ambient odour affects olfactory sensitivity of houseflies?

House flies can be vectors of pathogens for man and cattle. In order to control house fly populations in stables, traps with attractant odours may be used. Luring chemicals are only effective when they can be distinguished by the flies from the ambient odours. This research is done to investigate the effect of background odour on the sensitivity of the olfactory system of house flies. The sensitivity of the olfactory system of house flies was measured using electroantennography and single cell recordings. 1-Octen-3-ol yielded clear EAGs and was chosen as a pure-chemical background odour. Chicken manure was used as a natural background odour in this study. A continuous flow of air was led through a clean bottle or a bottle with chicken manure or with a filter paper onto which different amounts of 1-octen-3-ol, diluted in silicon oil, had been pipetted. This airstream was combined with a flow of clean charcoal-filtered air directed towards the antenna of a fly. Stimulus pulses were given by blowing air for 0.2 second through a pasteur pipette containing an odour source, into this airstream.

Dose-response curves of 1-octen-3-ol, R-limonene and 2-pentanone were made prior to, during and after application of background 1-octen-3-ol or manure. At low background odour concentrations (manure or 1 mg or 10 mg 1-octen-3-ol in the bottle), sensitivity did not differ significantly from the sensitivity in clean air for 1-octen-3-ol, R-limonene and 2-pentanone. 0.1 g Background 1-octen-3-ol reduced the EAG-responses to both 1-octen-3-ol and 2-pentanone reversibly; the EAGs returned to higher levels again, when testing the doses after switching back to clean air. No differences in sensitivity were found for R-limonene during 0.1 g background 1-octen-3-ol. With 1 g background 1-octen-3-ol, EAG-responses to all chemicals were reduced and remained low after switching back to clean air.

Single cell tests with 0.1 g background 1-octen-3-ol showed no effect on the sensitivity for 2-pentanone, a small and reversible decline of sensitivity for R-limonene, and a small, irreversible decline of sensitivity for 1-octen-3-ol. When background 1-octen-3-ol was switched on, some cells responded with a sustained elevated firing frequency, while other cells increased their firing rate for a short time only, returning to their initial clean-air firing frequency despite continuous application of the background odour.

We conclude that some cells in the antenna are sensitive to both 1-octen-3-ol and 2-pentanone or R-limonene as the EAG responses to 2-pentanone and R-limonene decreased in background 1-octen-3-ol. When background odour concentration is not too high, the olfactory system is still capable of responding to both the odour used as background, and a different odour. No effect of the natural background odour, chicken manure, on the sensitivity of other odours was found in the dose tested. This offers good perspectives for using attractive odours to lure flies in smelly stables.

This research is supported by the Technology Foundation (STW).

39

Dickens Dr. Joseph C.

U. S. Department of Agriculture, Agricultural Research Service
Beltsville Agricultural Research Center
Plant Sciences Institute, Vegetable Laboratory
Bldg. 010A, Rm. 112
Beltsville, MD 20705
USA

Olfactory Adaptations of the Colorado Potato Beetle and Associated Insects to Chemical Signals Emitted by Host Plants

Olfactory responses of the Colorado potato beetle (CPB), Leptinotarsa decemlineata (Say) (Coleoptera: Chrysomelidae), a generalist predator, Podisus maculiventris (Say) (Hemiptera, Heteroptera: Pentatomidae) (Pm), and a specialist predator, Perillus bioculatus (F.) (Hemiptera, Heteroptera: Pentatomidae) (Pb) were investigated. Volatiles tested included 20 compounds emitted by undamaged potato plants (Solanum tuberosum), plants that had been artificially damaged, or plants damaged by feeding by CPB larvae. Coupled gas chromatography/electroantennogram detector recordings revealed five compounds for which reliable responses were recorded.

Both Pm and Pb responded selectively to the same compounds as the CPB with few exceptions. CPB was at least 100 x more sensitive to a green leaf volatile than were the predators; both predators were more sensitive to each of the other compounds than were CPB. Both CPB and Pm were attracted to blends of host plant volatiles.

These results show that the herbivore (CPB) has olfactory receptors which are more sensitive to constitutive host plant volatiles, e.g. green leaf volatiles, while both generalist (Pm) and specialist (Pb) predators are more sensitive to systemic volatiles produced in response to prey feeding.

Return to program
Go to posters
Go to schedule

Copyright © ESITO 1999

Last updated: Aug 10, 1999