In many regions of the visual system, the activity of a neuron is normalized by the activity of other neurons in the same region. Here we show that a similar normalization occurs during olfactory processing in the Drosophila antennal lobe. We exploit the orderly anatomy of this circuit to independently manipulate feedforward and lateral input to second-order projection neurons (PNs). Lateral inhibition increases the level of feedforward input needed to drive PNs to saturation, and this normalization scales with the total activity of the olfactory receptor neuron (ORN) population. Increasing total ORN activity also makes PN responses more transient. Strikingly, a model with just two variables (feedforward and total ORN activity) accurately predicts PN odor responses. Finally, we show that discrimination by a linear decoder is facilitated by two complementary transformations: the saturating transformation intrinsic to each processing channel boosts weak signals, while normalization helps equalize responses to different stimuli.

Here we describe several fundamental principles of olfactory processing in the Drosophila melanogaster antennal lobe (the analog of the vertebrate olfactory bulb), through the systematic analysis of input and output spike trains of seven identified glomeruli. Repeated presentations of the same odor elicit more reproducible responses in second-order projection neurons (PNs) than in their presynaptic olfactory receptor neurons (ORNs). PN responses rise and accommodate rapidly, emphasizing odor onset. Furthermore, weak ORN inputs are amplified in the PN layer but strong inputs are not. This nonlinear transformation broadens PN tuning and produces more uniform distances between odor representations in PN coding space. In addition, portions of the odor response profile of a PN are not systematically related to their direct ORN inputs, which probably indicates the presence of lateral connections between glomeruli. Finally, we show that a linear discriminator classifies odors more accurately using PN spike trains than using an equivalent number of ORN spike trains.

Each odorant receptor gene defines a unique type of olfactory receptor neuron (ORN) and a corresponding type of second-order neuron. Because each odor can activate multiple ORN types, information must ultimately be integrated across these processing channels to form a unified percept. Here, we show that, in Drosophila, integration begins at the level of second-order projection neurons (PNs). We genetically silence all the ORNs that normally express a particular odorant receptor and find that PNs postsynaptic to the silent glomerulus receive substantial lateral excitatory input from other glomeruli. Genetically confining odor-evoked ORN input to just one glomerulus reveals that most PNs postsynaptic to other glomeruli receive indirect excitatory input from the single ORN type that is active. Lateral connections between identified glomeruli vary in strength, and this pattern of connections is stereotyped across flies. Thus, a dense network of lateral connections distributes odor-evoked excitation between channels in the first brain region of the olfactory processing stream.

Neural processing in the brain controls behavior through descending neurons (DNs) - neurons which carry signals from the brain to the spinal cord (or thoracic ganglia in insects). Because DNs arise from multiple circuits in the brain, the numerical simplicity and availability of genetic tools make Drosophila a tractable model for understanding descending motor control. As a first step towards a comprehensive study of descending motor control, here we estimate the number and distribution of DNs in the Drosophila brain. We labeled DNs by backfilling them with dextran dye applied to the neck connective and estimated that there are ~1100 DNs distributed in 6 clusters in Drosophila. To assess the distribution of DNs by neurotransmitters, we labeled DNs in flies in which neurons expressing the major neurotransmitters were also labeled. We found DNs belonging to every neurotransmitter class we tested: acetylcholine, GABA, glutamate, serotonin, dopamine and octopamine. Both the major excitatory neurotransmitter (acetylcholine) and the major inhibitory neurotransmitter (GABA) are employed equally; this stands in contrast to vertebrate DNs which are predominantly excitatory. By comparing the distribution of DNs in Drosophila to those reported previously in other insects, we conclude that the organization of DNs in insects is highly conserved.

Signaling by heterotrimeric GTP-binding proteins (G proteins) drives numerous cellular processes. The number of G protein molecules activated by a single membrane receptor is a determinant of signal amplification, although in most cases this parameter remains unknown. In retinal rod photoreceptors, a long-lived photoisomerized rhodopsin molecule activates many G protein molecules (transducins), yielding substantial amplification and a large elementary (single-photon) response, before rhodopsin activity is terminated. Here we report that the elementary response in olfactory transduction is extremely small. A ligand-bound odorant receptor has a low probability of activating even one G protein molecule because the odorant dwell-time is very brief. Thus, signal amplification in olfactory transduction appears fundamentally different from that of phototransduction.Odorants activate specific receptor proteins (1) on the cilia of olfactory receptor neurons (ORNs) and, by way of a G protein (G olf ), stimulate an adenylyl cyclase (type III) to synthesize adenosine 3′,5′-cyclic monophosphate (cAMP) (2,3). cAMP opens a cyclic-nucleotide-gated (CNG) cation channel to produce a membrane depolarization (2,3). Influx of Ca 2+ through the CNG channel opens a Ca 2+ -activated chloride (Cl) channel, leading to Cl − efflux and further depolarization (2,3). Simultaneously, the Ca 2+ influx decreases cAMP synthesis and the effective affinity of CNG channels for cAMP, both effects producing olfactory adaptation (2, 3).Little is known about signal amplification in olfactory transduction. It has been suggested (4) that, in physiological (Ringer) solution, a single odorant-receptor molecule triggers an elementary (or unitary) olfactory response of ~ 1 pA in membrane current, indicating an amplification similar to that in phototransduction. However, this conclusion has been challenged (5,6). The supralinear relation (i.e., Hill coefficient > 1) reported between odorant concentration and response amplitude (7) is also puzzling because it may suggest a nonlinear summation of the elementary responses. At odorant concentrations low enough to give few odorant-binding events, the overall response should arise from spatially segregated, noninteracting transduction domains on the cilia triggered by individual activated membrane receptor molecules (the "units"). Thus, despite intrinsic transduction nonlinearities [multiple cAMP molecules are required to open a CNG channel (2,3) and multiple Ca 2+ to open a