Spatially resolved time‐frequency analysis of odour coding in the insect antennal lobe

Antennal lobes constitute the first neurophils in the insect brain involved in coding and processing of olfactory information. With their stereotyped functional and anatomical organization, they provide an accessible model with which to investigate information processing of an external stimulus in a neural network in vivo. Here, by combining functional calcium imaging with time‐frequency analysis, we have been able to monitor the oscillatory components of neural activity upon olfactory stimulation. The aim of this study is to investigate the presence of stimulus‐induced oscillatory patterns in the honeybee antennal lobe, and to analyse the distribution of those patterns across the antennal lobe glomeruli. Fast two‐photon calcium imaging reveals the presence of low‐frequency oscillations, the intensity of which is perturbed by an incoming stimulus. Moreover, analysis of the spatial arrangement of this activity indicates that it is not homogeneous throughout the antennal lobe. On the contrary, each glomerulus displays an odorant‐specific time‐frequency profile, and acts as a functional unit of the oscillatory activity. The presented approach allows simultaneous recording of complex activity patterns across several nodes of the antennal lobe, providing the means to better understand the network dynamics regulating olfactory coding and leading to perception.     

Antennal lobe interneurons in the desert locust Schistocerca gregaria (Forskal): Processing of aggregation pheromones in adult males and females

Physiological and morphological characteristics of antennal lobe interneurons in male and female Schistocerca gregaria were studied by using intracellular recording and staining techniques. For the first time, the responses of projection neurons to behaviourally active and potential aggregation pheromone compounds and plant odour compounds were recorded in young adult locusts. Excitatory, inhibitory, or combined excitatory/inhibitory responses to the presented odours were detected. The stained neurons had their cell bodies in the frontal cell group, arborized in 10 to 25 glomeruli at equal distances from the central fibre core, and sent their axons to the calyces of the mushroom body and to the lateral protocerebrum. The projection neurons responded to the set of different stimuli with varying specificity. In females, more neurons responded specifically to single plant or aggregation pheromone compounds than in males, where more generalist responses were found. “Blend specialist” neurons, responding only to mixtures of behaviourally active aggregation pheromone compounds, but not to the single compounds, were present in both males and females. Most neurons responded to the behaviourally active aggregation pheromone mixtures and to single compounds present in these mixtures, as well as to plant odours. Fewer neurons responded to the potential aggregation pheromone compounds tested. In several experiments, two spike sizes in the recording were correlated with two stained neurons in the antennal lobe, suggesting electrical coupling of the neurons. No response to any of the stimuli was found in antennal lobe interneurons in old adults. The morphological and physiological features of the projection neurons in S. gregaria are compared with projection neuron characteristics in other insects. © 1996 Wiley-Liss, Inc.     

Early olfactory experience modifies neural activity in the antennal lobe of a social insect at the adult stage

In the antennal lobe (AL), the first olfactory centre of the insect brain, odorants are represented as spatiotemporal patterns of glomerular activity. Whether and how such patterns are modified in the long term after precocious olfactory experiences (i.e. in the first days of adulthood) remains unknown. To address this question, we used in vivo optical imaging of calcium activity in the antennal lobe of 17-day-old honeybees which either experienced an odorant associated with sucrose solution 5–8 days after emergence or were left untreated. In both cases, we imaged neural responses to the learned odor and to three novel odors varying in functional group and carbon-chain length. Two different odor concentrations were used. We also measured behavioral responses of 17-day-old honeybees, treated and untreated, to these stimuli. We show that precocious olfactory experience increased general odor-induced activity and the number of activated glomeruli in the adult AL, but also affected qualitative odor representations, which appeared shifted in the neural space of treated animals relative to control animals. Such effects were not limited to the experienced odor, but were generalized to other perceptually similar odors. A similar trend was found in behavioral experiments, in which increased responses to the learned odor extended to perceptually similar odors in treated bees. Our results show that early olfactory experiences have long-lasting effects, reflected in behavioral responses to odorants and concomitant neural activity in the adult olfactory system.     

Environment‐specific modulation of odorant representations in the honeybee brain

Ca²⁺ imaging techniques were applied to investigate the neuronal behavior of projection neurons in the honeybee antennal lobe (AL) to examine the effects of long‐lasting adaptation on odorant coding. Responses to eight test odorants were measured before, during, and after an odor adaptation phase. Bees were exposed to the adapting odor for 30 min. Test odorant responses were only recorded from a sub‐population of accessible glomeruli on the AL surface. Projection neurons, the output neurons of the antennal lobes, are projecting through the lateral, mediolateral, and medial AL tract to higher centers of the olfactory pathway. Due to our staining techniques, we primarily focused our study on projection neurons going through the lateral and medial tract. Test odorants comprised compounds with different functional groups (alcohol, aldehyde, ketone, and ester) representing floral and/or pheromone odorants. Strength and discriminability between combinatorial activity patterns induced by the test odorants were quantified. In two independent experiments, we investigated one group of animals adapted to a colony odor and another adapted to a synthetic odor. Within the experimental groups, we found test odorant responses either decreased or increased in AL projection neurons. Additionally, the discriminability between test odorant patterns became less distinct in the colony odor experiment and more distinct during adaptation in the synthetic mixture experiment. These results are interpreted as odor dependent adaptation effects, increasing or decreasing response strength and discriminability by altered neural coding mechanisms in the AL neuropile.     

Nonassociative plasticity alters competitive interactions among mixture components in early olfactory processing

Experience‐related plasticity is an essential component of networks involved in early olfactory processing. However, the mechanisms and functions of plasticity in these neural networks are not well understood. We studied nonassociative plasticity by evaluating responses to two pure odors (A and X) and their binary mixture using calcium imaging of odor‐elicited activity in output neurons of the honey bee antennal lobe. Unreinforced exposure to A or X produced no change in the neural response elicited by the pure odors. However, exposure to one odor (e.g. A) caused the response to the mixture to become more similar to that of the other component (X). We also show in behavioral analyses that unreinforced exposure to A caused the mixture to become perceptually more similar to X. These results suggest that nonassociative plasticity modifies neural networks in such a way that it affects local competitive interactions among mixture components. We used a computational model to evaluate the most likely targets for modification. Hebbian modification of synapses from inhibitory local interneurons to projection neurons most reliably produced the observed shift in response to the mixture. These results are consistent with a model in which the antennal lobe acts to filter olfactory information according to its relevance for performing a particular task.     

Neuropeptides in the antennal lobe of the yellow fever mosquito,Aedes aegypti

For many insects, including mosquitoes, olfaction is the dominant modality regulating their behavioral repertoire. Many neurochemicals modulate olfactory information in the central nervous system, including the primary olfactory center of insects, the antennal lobe. The most diverse and versatile neurochemicals in the insect nervous system are found in the neuropeptides. In the present study, we analyzed neuropeptides in the antennal lobe of the yellow fever mosquito, Aedes aegypti, a major vector of arboviral diseases. Direct tissue profiling of the antennal lobe by matrix‐assisted laser desorption ionization time‐of‐flight (MALDI‐TOF) mass spectrometry indicated the presence of 28 mature products from 10 different neuropeptide genes. In addition, immunocytochemical techniques were used to describe the cellular location of the products of up to seven of these genes within the antennal lobe. Allatostatin A, allatotropin, SIFamide, FMRFamide‐related peptides, short neuropeptide F, myoinhibitory peptide, and tachykinin‐related peptides were found to be expressed in local interneurons and extrinsic neurons of the antennal lobe. Building on these results, we discuss the possible role of neuropeptide signaling in the antennal lobe of Ae. aegypti. J. Comp. Neurol. 522:592–608, 2014. © 2013 The Authors. The Journal of Comparative Neurology Published by Wiley Periodicals, Inc.     

Ramification pattern and ultrastructural characteristics of the serotonin-immunoreactive neuron in the antennal lobe of the moth Manduca sexta: A laser scanning confocal and electron microscopic study

The two antennal lobes, the primary olfactory centers of the brain, of the moth Manduca sexta each contain one neuron that displays serotonin immunoreactivity. The neuron projects out of the antennal lobe and sends branches into ipsi- and contralateral protocerebral areas. An axon-like process extends from the contralateral protocerebrum to, and terminates in, the contralateral antennal lobe. In order to begin to investigate the possible role of this unique neuron in olfactory information processing, we have used laser scanning confocal microscopic and electron microscopic immunocytochemical techniques to study the ramification pattern, ultrastructural characteristics, and synaptic connections of the neuron in the antennal lobes of female adult Manduca sexta. The neuron ramifies extensively in the antennal lobe contralateral to the cell body. The ramifications, mainly in the base and center of each glomerulus, do not overlap with those of the sensory axons from the antenna. This finding suggests that the serotonin-immunoreactive neuron may not receive direct input from sensory neurons, and that it may modulate the activity of the neurons of the antennal lobe rather than that of the sensory neurons. In the electron microscope, the neuron exhibits large dense-cored vesicles and small, clear round vesicles. In the antennal lobe ipsilateral to the cell body, the primary neurite of the serotonin-immunoreactive neuron is unbranched and lacks detectable synaptic connections. The ramifications in the contralateral antennal lobe, however, participate in synaptic connections. At very low frequency, contralateral branches form synapses onto unlabeled processes and also receive synapses from unidentified neurons in the glomeruli, indicating that the neuron may participate directly in synaptic processing of olfactory information. The high ratio of output to input synapses made by the serotonin-immunoreactive processes in the contralateral antennal lobe is consistent with the idea that this neuron may receive synaptic input via its bilateral branches in the protocerebrum and then send information to the contralateral antennal lobe where the neuron may exert feedback or modulatory influences on olfactory information processing in the glomeruli. © 1993 Wiley-Liss, Inc.     

Central processing of sex pheromone,host odour,and oviposition deterrent information by interneurons in the antennal lobe of female Spodoptera littoralis (Lepidoptera: Noctuidae)

Physiological and anatomical characteristics of antennal lobe interneurons in female Spodoptera littoralis (Boisd.) were investigated using intracellular recording and staining techniques. Responses of local interneurons and projection neurons to female sex pheromone components, host plant odours, and behaviourally active oviposition deterrents were recorded. We found local interneurons and projection neurons that responded specifically to only one or two of the tested odours, but we also found less specific cells, and neurons that responded to most of the tested odourants. These findings show that there are not only specific olfactory pathways in female moths up to the protocerebral level, but also that integration can begin in the antennal lobe. No correlation was found between the degree of specificity of either local interneurons or projection neurons and their respective morphological characteristics. Specialized and unspecialized local interneurons arborized throughout the antennal lobe. Specialized and unspecialized projection neurons had uniglomerular arborizations in the antennal lobe and sent their axons to the calyces of the mushroom body, and to the lateral horn of the protocerebrum. One specific projection neuron had multiglomerular arborizations and projected only to the lateral horn of the protocerebrum. Projection neurons arborizing in the glomeruli closest to the entrance of the antennal nerve always responded to pheromone components. No other correlations were found between the arborization pattern of projection neurons in the antennal lobe or in the protocerebrum and their response characteristics. The sensitivity of local interneurons and projection neurons was in the same range as that of receptor neurons in olfactory sensilla on the antennae, suggesting a much lower convergence in the central nervous system in females than in the pheromone-processing pathway in males. © 1994 Wiley-Liss, Inc.     

Synaptic organization of the uniglomerular projection neurons of the antennal lobe of the moth Manduca sexta: A laser scanning confocal and electron microscopic study

The detailed branching pattern and synaptic organization of the uniglomerular projection neurons of the antennal lobe, the first processing center of the olfactory pathway, of the moth Manduca sexta were studied with laser scanning confocal microscopy and a technique combining laser scanning confocal microscopy and electron microscopy. Uniglomerular projection neurons, identified electrophysiologically or morphologically, were stained intracellularly with neurobiotin or biocytin. Brains containing the injected neurons were treated with streptavidin-immunogold to label the injected material for electron microscopy and with Cy3-streptavidin to lable the neurons with fluorescence for laser scanning confocal microscopy, and then embedded in Epon. Labeled neurons were imaged and reconstructed with laser scanning confocal microscopy (based on the retained fluorescence of the labeled neuron in the Epon block), and thin sections were cut at selected optical levels for correlation of light microscopic data and electron microscopic detail. Each neuron had a cell body in one of the three cell-body clusters of the antennal lobe, a primary neurite that extended across the coarse neuropil at the center of the antennal lobe and then formed a dense tuft of processes within a single glomerulus, and an axon that emanated from the primary neurite and projected from the antennal lobe via the antenno-cerebral tract to the lateral horn of the ipsilateral protocerebrum and, collaterally, to the calyces of the mushroom body. In the electron microscope, the fine dendritic branches in the apical zones of the glomeruli, where sensory axons terminate, were found to receive many input synapses. In deeper layers across the glomeruli, the processes participated in both input and output synapses, and at the bases of the glomeruli, the most proximal, thickest branches formed output synapses. In both of the protocerebral areas in which axonal branches terminated, those branches formed exclusively output synapses. Our findings indicate that, in addition to conveying olfactory information to the protocerebrum, uniglomerular projection neurons in the antennal lobes of M. sexta participate in local intraglomerular synaptic circuitry. J. Comp. Neurol. 379:2-20, 1997. © 1997 Wiley-Liss, Inc.     

[H]2-Deoxyglucose mapping of odor-induced neuronal activity in the antennal lobes of Drosophila melanogaster

Olfactory stimultion results in an enhanced uptake of [³H]2-deoxyglucose in specific glomeruli in the antennal lobes of Drosophila melanogaster. Unilateral stimulation induces activity in receptor axons and lobe interneurons on the ipsilateral side. Collaterals from the receptor axons to the contralateral lobe are also active but stimulate either weak or no postsynaptic activity. This difference in signal transfer properties could be relevant to odor detection by flies.     

Spatial coding of olfactory information in the antennal lobe of Drosophila melanogaster

[³H]2-deoxyglucose autoradiography has revealed foci in the antennal lobe of Drosophila melanogaster that are active during olfactory stimulation. Labelling in the receptor axons and the lobe interneurons can be observed. Each class of odors stimulates activity in a specific subset of antennal glomeruli. This defines the activity domain, which is recognizable distinct, although overlapping, for different classes of volatile chemicals. Unilateral stimulation with attractants predominantly results in excitation of the ipsilateral glomeruli. Collaterals of the receptor neurons to the contralateral lobe stimulate no visible postsynaptic activity. On the other hand, unilateral stimulation with repellent odors, such as benzaldehyde, results in an equally strong activity in both antennal lobes.     

Astrocyte‐like glial cells physiologically regulate olfactory processing through the modification of ORN‐PN synaptic strength in Drosophila

Astrocyte‐like glial cells are abundant in the central nervous system of adult Drosophila and exhibit morphology similar to astrocytes of mammals. Previous evidence has shown that astrocyte‐like glial cells are strongly associated with synapses in the antennal lobe (AL), the first relay of the olfactory system, where olfactory receptor neurons (ORNs) transmit information into projection neurons (PNs). However, the function of astrocyte‐like glia in the AL remains obscure. In this study, using in vivo calcium imaging, we found that astrocyte‐like glial cells exhibited spontaneous microdomain calcium elevations. Using simultaneous manipulation of glial activity and monitoring of neuronal function, we found that the astrocyte‐like glial activation, but not ensheathing glial activation, could inhibit odor‐evoked responses of PNs. Ensheathing glial cells are another subtype of glia, and are of functional importance in the AL. Electrophysiological experiments indicated that astrocyte‐like glial activation decreased the amplitude and slope of excitatory postsynaptic potentials evoked through electrical stimulation of the antennal nerve. These results suggest that astrocyte‐like glial cells may regulate olfactory processing through negative regulation of ORN–PN synaptic strength. Beyond the antennal lobe we observed astrocyte‐like glial spontaneous calcium activities in the ventromedial protocerebrum, indicating that astrocyte‐like glial spontaneous calcium elevations might be general in the adult fly brain. Overall, our study demonstrates a new function for astrocyte‐like glial cells in the physiological modulation of olfactory information transmission, possibly through regulating ORN–PN synapse strength.     

A Sexually Dimorphic Olfactory Neuron Mediates Fixed Action Transition during Courtship Ritual in Drosophila melanogaster

Animals perform a series of actions in a fixed order during ritualistic innate behaviors. Although command neurons and sensory pathways responding to external stimuli that trigger these behaviors have been identified, how each action is induced in a fixed order in response to multimodal sensory stimuli remains unclear. Here, the sexually dimorphic lateral antennal lobe tract projection neuron 4 (lPN4) in male Drosophila melanogaster mediates the expression of a fixed behavioral action pattern at the beginning of the courtship ritual, in which a male taps a female body and then extends a wing unilaterally to produce a courtship song. We found that blocking the synaptic output of lPN4 caused an increase in the ratio of male flies that extended a wing unilaterally without tapping the female body, whereas excitation of lPN4 suppressed the transition from the tapping phase to the unilateral wing extension phase. Real-time calcium imaging showed that lPN4 is activated by a volatile pheromone, palmitoleic acid, whose responses were inhibited by simultaneous gustatory stimulation with female cuticular hydrocarbons, showing the existence of an “AND-gate” for multimodal sensory inputs during male courtship behaviors. These results suggest that the function of lPN4 is to suppress unilateral wing extension while responding to a female smell, which is released by appropriate contact chemosensory inputs received when tapping a female. As the female smell also promotes male courtship behaviors, the olfactory system is ready to simultaneously promote and suppress the progress of courtship actions while responding to a female smell.SIGNIFICANCE STATEMENT Although it has been 80 years since Konrad Lorenz and Niko Tinbergen introduced how multiple acts comprising separate innate behaviors are released in a fixed order according to external stimuli, the neural circuits responsible for such fixed action patterns remain largely unknown. The male courtship behavior of Drosophila melanogaster is a good model to investigate how such a fixed behavioral sequence is determined in the brain. Here, we show that lateral antennal lobe tract projection neuron 4 (lPN4) in D. melanogaster functions as an “AND-gate” for volatile and contact chemosensory inputs, mediating the expression of tapping behaviors before unilateral wing extension during male courtship rituals.     

Functional MRI using molecular imaging agents

Contrast agents for magnetic resonance imaging (MRI) have recently been used as cellular-level probes of neural function. New in vivo labeling strategies now enable researchers to follow plasticity of brain activation patterns and cellular structure over time. On the horizon is the prospect that molecular imaging agents specifically designed for functional imaging (fMRI) on a relatively fast timescale could offer an alternative to conventional hemodynamics-based approaches. Development of several MRI sensors has defined principles by which imaging agents for "molecular fMRI" can be constructed; application of engineered sensors for cellular-level correlates of neuronal activity would allow researchers to combine the noninvasiveness of MRI with spatial resolution of tens of microns and temporal resolution of 100ms or less. Facilitated by advances in imaging-agent delivery methods and model systems appropriate for high-resolution neuroimaging, novel molecular imaging strategies continue to potentiate MRI as a tool for mechanistic investigation of neural systems.     

Manganese-enhanced 3D MRI of established and disrupted synaptic activity in the developing insect brain in vivo

The antennal lobe of the sphinx moth Manduca sexta serves as a model for the development of the olfactory system. Here, the establishment of the glomerular synaptic network formed by the olfactory receptor axons and antennal lobe neurons at pupal stage P12 was followed by transection of the right antenna and - within 24 h - by injection of MnCl2 into the hemolymph. In vivo 3D MRI at 100 and 60 microm isotropic resolution was then performed at P13 to P17. Whereas the left antennal lobe revealed a pronounced increase of the signal-to-noise ratio (SNR) reflecting normal synaptic activity, the observation of only a small SNR increase within the right antennal lobe indicated the disruption of pertinent activity after antennal transection. The accumulation of manganese in the intact antennal system became observable within 3 h and lasted for at least 2 days after injection. Intra-individual comparisons between the right and left side yielded a statistically significant differential SNR increase in the left antennal lobe. Because such an effect was not observed in younger animals studied at pupal stages P10/P11, the MRI findings confirm the development of functional synapses in the antennal lobe of Manduca sexta by P13.     

A multisensory centrifugal neuron in the olfactory pathway of heliothine moths

We have characterized, by intracellular recording and staining, a unique type of centrifugal neuron in the brain olfactory center of two heliothine moth species; one in Heliothis virescens and one in Helicoverpa armigera. This unilateral neuron, which is not previously described in any moth, has fine processes in the dorsomedial region of the protocerebrum and extensive neuronal branches with blebby terminals in all glomeruli of the antennal lobe. Its soma is located dorsally of the central body close to the brain midline. Mass‐fills of antennal‐lobe connections with protocerebral regions showed that the centrifugal neuron is, in each brain hemisphere, one within a small group of neurons having their somata clustered. In both species the neuron was excited during application of non‐odorant airborne signals, including transient sound pulses of broad bandwidth and air velocity changes. Additional responses to odors were recorded from the neuron in Heliothis virescens. The putative biological significance of the centrifugal antennal‐lobe neuron is discussed with regard to its morphological and physiological properties. In particular, a possible role in multisensory processes underlying the moth's ability to adapt its odor‐guided behaviors according to the sound of an echo‐locating bat is considered. J. Comp. Neurol. 521:152–168, 2013. © 2012 Wiley Periodicals, Inc.     

Magnetic resonance imaging,microscopy, and spectroscopy of the central nervous system in experimental animals

Over the last two decades, microscopic resolutionin vivo magnetic resonance imaging (MRI) techniques have been developed and extensively used in the study of animal models of human diseases. Standard MRI methods are frequently used in clinical studies and in the general clinical practice of human neurological diseases. This generates a need for similar studies in experimental animal research. Because small rodents are the most commonly used species as animal models of neurological diseases, the MRI techniques need to be able to provide microscopic resolution and high signal-to-noise ratio images in relatively short time. Small animal MRI systems use very high field-strength magnets, which results in higher signal to noise ratio; however, the contrast characteristics of live tissue are different at these field strengths. In addition to standard MRI techniques, several new applications have been implemented in experimental animals, including diffusion and perfusion studies, MR angiography, functional MRI studies, MRI tractography, proton and phosphorous spectroscopy, cellular and molecular imaging using novel contrast methods. Here we give an overview of how to establish a small animal imaging facility with the goal of CNS imaging. We describe the basic physical processes leading to MR signal generation, highlighting the differences between standard clinical MRI and small animal MRI. Finally, typical findings in the most common neurological disease categories and novel MRI/magnetic resonance spectroscopy methods used in their study are also described.     

The Relative Contributions of MRI, SPECT, and PET Imaging in Epilepsy

Summary: Functional and structural neuroimaging techniques are increasingly indispensable in the evaluation of epileptic patients for localization of the epileptic area as well as for understanding pathophysiology, propagation, and neurochemical correlates of chronic epilepsy. Although interictal single photon emission computed tomography (SPECT) imaging of cerebral blood flow is only moderately sensitive, ictal SPECT markedly improves yield. Positron emission tomography (PET) imaging of interictal cerebral metabolism is more sensitive than measurement of blood flow in temporal lobe epilepsy. Furthermore, PET has greater spatial resolution and versatility in that multiple tracers can image various aspects of cerebral function. Interpretation of all types of functional imaging studies is difficult and requires knowledge of time of most recent seizure activity and structural correlates. Only magnetic resonance imaging (MRI) can image the structural changes associated with the underlying epileptic process, and quantitative evidence of hippocampal volume loss has been highly correlated with seizure onset in medial temporal structures. Improved resolution and interpretation have made quantitative MRI more sensitive in temporal lobe epilepsy, as judged by pathology. When judged by electroencephalography (EEG), ictal SPECT and interictal PET have the highest sensitivity and specificity for temporal lobe epilepsy; these neuroimaging techniques have lower sensitivity and higher specificity for extratemporal EEG abnormalities. Regardless of the presence of structural abnormalities, functional imaging by PET or SPECT provides complementary information. Ideally these techniques should be used and interpreted together to improve the localization and understanding of epileptic brain.     

Synaptic connection between olfactory receptor cells and uniglomerular projection neurons in the antennal lobe of the American cockroach,Periplaneta americana

Both antennal receptor cell axons and uniglomerular projection neurons of the antennal lobe were specifically labeled, and their synaptic relationship was studied at the fine structural level. The labelings were applied in different combinations: i) Experimentally induced anterograde degeneration of sensory-afferent axons was combined with injection of horseradish peroxidase into uniglomerular projection neurons. ii) Lucifer Yellow was injected into uniglomerular projection neurons, and receptor cell axons were anterogradely labeled with the lipophilic dye DiI. The fluorescent dyes were transformed by immuno- or photochemical treatment into electron-dense markers. In both types of preparations, a considerable number of monosynaptic output synapses from antennal receptor neurons onto processes of uniglomerular projection neurons were identified within the glomeruli of the lobe. In most cases, the receptor axon was connected in a dyadic fashion, firstly to a process of a projection neuron and secondly to a nonlabeled process. The results clearly demonstrate a direct connection between receptor cells and output neurons of the cockroach antennal lobe which exists in parallel to the already proposed and demonstrated polysynaptic connection via inhibitory local interneurons. © 1996 Wiley-Liss, Inc.