IMAGING AND CODING IN THE OLFACTORY SYSTEM. John S Kauer and Joel White INTRODUCTION. Key Words

Transcription

1 Annu. Rev. Neurosci : Copyright c 2001 by Annual Reviews. All rights reserved IMAGING AND CODING IN THE OLFACTORY SYSTEM John S Kauer and Joel White Department of Neuroscience, Tufts University School of Medicine, 136 Harrison Avenue, Boston, Massachusetts 02111; john.kauer@tufts.edu, joel.white@tufts.edu Key Words smell, odor coding, physiology, fluorescent dyes, functional imaging Abstract Functional imaging methods permit analysis of neuronal systems in which activity is broadly distributed in time and space. In the olfactory system the dimensions that describe odorant stimuli in odorant space are still poorly defined. One way of trying to characterize the attributes of this space is to examine the ways in which its dimensions are encoded by the neurons and circuits making up the system and to compare these responses with physical-chemical attributes of the stimuli and with the output behavior of the animal. For documenting distributed events as they occur, imaging methods are among the few tools available. We are still in the early stages of this analysis; however, a number of recent studies have contributed new information to our understanding of the odorant coding problem. This paper describes imaging results in the context of other data that have contributed to our understanding of how odors are encoded by the peripheral olfactory pathway. INTRODUCTION Functional imaging has played an important role in developing an understanding of olfactory function because of its ability to reveal events distributed in space and time. The distributed nature of olfactory information processing was first observed 50 years ago by Adrian (1953). Subsequent studies using serial single electrode recordings (Leveteau & MacLeod 1966) and, in some cases, multi-electrode recording (Moulton 1976) provided additional support for the idea that odor information is represented by activity occurring in parallel in different locations within the olfactory pathway. Until the advent of functional imaging methods, it was not possible to observe these events with spatial resolution that gives global views of distributed processes. These early electrophysiological studies stimulated exploration of distributed neuronal events that might not have been further scrutinized had there been more recognizable relationships between odorant structure and the single unit recordings. The present discussion reviews the contributions of functional imaging. We focus on methods that permit observation of spatially distributed events related to X/01/ $

2 964 KAUER WHITE activity generated by odorants. Imaging methods that solely record spatial properties of structure, such as histological or immunohistochemical staining, are reviewed elsewhere. This structure/function dichotomy is somewhat blurred by recently developed vital staining and immunohistochemical methods and genetically generated molecular marking techniques that reveal functionally important topographical or biochemical features (see for example LaMantia et al 1992, Mori et al 1985, Mombaerts et al 1996, Dynes & Ngai 1998, Siegel & Isacoff 1997). While the emphasis here is on methods that permit observation of relatively rapid odor-generated events within the time frames of short sniff odorant applications, data from more static imaging methods that have contributed to ideas about odor coding are also included. In the course of assembling a view of the coding process, we examine several levels of the olfactory pathway extending from the inhalation of odorants, to events in the olfactory sensory neurons (OSNs) in the nose, to the circuits of the olfactory bulb (OB). Most data are from vertebrate species, but invertebrate studies have contributed to a number of seminal observations that have influenced much thinking about basic olfactory system functioning (see Hildebrand & Shepherd 1997, Krieger & Breer 1999 for review and Ai et al 1998, Gervais et al 1996, Joerges et al 1997, Kimura et al 1998 for examples of specific experiments). Although technical details of the various methods are briefly discussed, this is not intended to be a comprehensive analysis of either neuronal imaging methods (e.g. see Ebner & Chen 1995, Lieke et al 1989, Kobal & Kettenmann 2000 for reviews) or olfactory coding (e.g. see Buck 1996, Christensen & White 2000, Hildebrand & Shepherd 1997, Laurent 1999, Mori et al 1999 for reviews). This discussion emphasizes functional imaging methods that resolve the spatial (and, in fortuitous cases, temporal) aspects of many neuronal events occurring simultaneously. Space and time are not immediately apparent as stimulus characteristics related to odor quality, but appear to be important neural dimensions representing aspects of odorant quality that are accessible to analysis by these methods. IMAGING METHODS AND THE OLFACTORY SYSTEM Imaging methods potentially provide advantages for examining a number of important functional attributes of nervous systems. These include the study of (a) spatially distributed events that occur during development, (b) changes in real time activity in response to perturbations by stimulation, (c) plastic responses to longterm external or internal environmental changes, and (d) changes due to injury, senescence, and normal (e.g. apoptotic) and abnormal (disease-associated) degeneration. To characterize each of these processes completely, one would need to obtain information with (a) spatial resolution that ranges from nanometers (molecular size) to meters (organism size); (b) temporal resolution extending from

3 IMAGING AND OLFACTORY SYSTEM CODING 965 microseconds (for molecular events such as enzymatic reactions or isomerization, as in rhodopsin) to years (for developmental, plastic, and degenerative changes); (c) the ability to observe in three, as well as two, dimensions (using tools like confocal or multiple photon microscopy); (d) the ability to access the signals in tissue without invasive surgery; and (e) the ability to assess function without perturbation by the observation process. Methods presently available can approach some of these goals and, as with all analytical methods, one has to chose the appropriate method for the question being asked. Figure 1 is an illustration of the application of voltage-sensitive dye recording to the analysis of potentials distributed through the dendritic architecture of a rat somatosensory pyramidal neuron (Antic et al 1999). These data are not taken from the olfactory system, but are presented here to illustrate several attributes of methods discussed in this paper. For most of these methods one must have optical access to the structure to be observed (Figure 1A). The intrinsic pixelation of imaging methods permits analysis of both a single point (pixel) on the structure over time (Figure 1B,C) as well as analysis of spatially distributed events over time (Figure 1D) by use of time-series pseudo color representations. This figure shows explicitly how a single imaging approach can provide high-resolution spatial characterization of where and when rapid electrical changes occur. Imaging systems, in general, include use of markers for neuronal activity [radiolabels (2-deoxyglucose, c-fos), antibodies (c-fos), fluorescent dyes (voltagesensitive or Ca ++ reporters), or intrinsic optical properties of the tissue], optical detectors with good spatial and temporal resolution such as video cameras or photodiode arrays, and various computational devices for acquiring, digitizing, and storing the data and for manipulating, filtering, and presenting the images. These components, along with unimpeded optical access to the tissue, are required for all the imaging methods discussed here. Imaging methods that have been applied to analysis of olfactory function include (a) observation of inhaled and exhaled air flows by schlieren photography (DA Kester, GS Settles & LJ Dodson-Dreibelbis, manuscript in preparation); (b) examination of neuronal activity by measuring glucose uptake with radiolabeled 2-deoxyglucose (2DG) (Kennedy et al 1975); (c) observation of changes in activity-related gene expression, such as c-fos (Sagar et al 1988); (d) measurement of intracellular calcium concentrations using fluorescent markers such as fura-2 (Grynkiewicz et al 1985) or calcium green (Friedrich & Korsching 1997); (e) measurement of transmembrane voltage using fluorescent voltage-sensitive dyes (Cohen & Lesher 1986, Blasdel & Salama 1986); and ( f ) measurement of activity-dependent changes in light-scattering and/or oxy- or deoxy-hemoglobin absorption (Grinvald et al 1986). Earlier results using some of these approaches are presented by Cinelli & Kauer (1992). Descriptions of other methods not applied to olfactory analysis can be found in some of the references mentioned above as well as in Lieke et al 1989 and Zochowski et al 2000.

4 966 KAUER WHITE AN EMERGING VIEW OF OLFACTORY CODING Analysis of the results of imaging methods in the context of much new electrophysiological, biophysical, biochemical, and molecular biological data on the transduction process (for reviews see Ache & Zhainazarov 1995, Mombaerts 1999, Schild & Restrepo 1998) begins to describe some of the events that participate in the encoding of odorant information. We emphasize the peripheral pathway here because of the paucity of information on odor representation at higher levels (for a review of responses seen with electrical stimulation in animals see Litaudon et al 1997; for results using noninvasive methods with odorant stimulation in humans, see Levy et al 1999, Sobel et al 1998, Yousem et al 1999; reviews in Kobal & Kettenmann 2000, Zald & Pardo 2000). Odorant Stimulus Intake and Distribution in the Nasal Cavity The first step in the detection and discrimination of odors in terrestrial vertebrates is inhalation of the vapor phase substance into the nose and delivery of the stimulus to the OSNs in the olfactory epithelium (OE) lining the nasal cavity, a relatively simple structure in amphibians and reptiles, but highly convoluted in macrosmatic mammals. Distribution of odorant molecules to the OE is accomplished via the intrinsic characteristics of the inhaled air flow as well as via the aerodynamic properties of the cavity. Odor access to OSNs is also likely influenced by chromatographic interactions with the mucus lining (Mozell et al 1987). Although there are a number of studies that have calculated and measured the airflow in models (Hahn et al 1993, Keyhani et al 1997), there is surprisingly little work using direct methods to image flows either into or within the nasal cavity. Initial work has begun, however, on the nature of the access of the vapor phase stimulus to the nostrils. Figure 2 shows the use of a schlieren imaging system (DA Kester, GS Settles & LJ Dodson-Dreibelbis, manuscript in preparation) to visualize air flow during the inhalation and exhalation cycle in the dog, based on temperaturedependent differences in refractive index. Dogs are macrosmats, capable of detecting concentrations of certain nitroaromatic compounds as low as 500 parts per trillion (Williams et al 1998). It has been consistently found that the inhalation and exhalation processes in these animals are complex, are not symmetrical, and are modified by the behavioral task that the dog is performing. In this side view (Figure 2, left) one can see how inhalation leads to the constrained flow of high-velocity air being drawn into the nose from the front. In a different animal (Figure 2, right) air is exhaled backward, away from the source. This redirection of flow during different parts of the sniffing cycle is governed by a muscle activated flap on the external nares. These studies are in their early stages, but it is clear that sniffing behavior and the aerodynamics of air flow within the nasal cavity are important components of high performance detection behavior, although the precise effects on olfaction are

5 IMAGING AND OLFACTORY SYSTEM CODING 967 still poorly characterized. It is likely that inhalation and exhalation processes mold the structure of the vapor pulse that is finally presented to the sensory elements and therefore do not simply present to the sensory surface replicas of the spatial and temporal properties of the original stimulus outside the nose. All coding events downstream of odor delivery depend upon this process. A deficiency in many experimental paradigms, often dictated by the constraints of the imaging method, is lack of data from normally sniffing, behaviorally attentive, animals. Another aspect of this process has been evaluated by measuring the influence of delivering odorant stimuli under different flow conditions that, although artificially manipulated, are at least related to flow changes during sniffing behavior (Kent et al 1995, 1996; Youngentob & Kent 1995) (see Figure 4). Access to Receptor Cell Membranes Once the odorant stimulus, which is almost always a complex mixture of compounds, is delivered to the sensory region of the nasal cavity, a number of physical events occur before transduction from chemical to neural information takes place. The stimulus molecules must diffuse through the mucus to reach the cilia of the receptor cells. This diffusion event is still poorly defined and has not been observed directly. It is possible that micro schlieren analysis in OE slices might provide information on this event. Although the mucus can be as much as 30 µm thick, it is likely that many OSN cilia float near the surface, and thus the distances over which diffusion occurs may be short. Interactions of Odorants with Olfactory Receptor Neurons Once through the mucus, the small odorant molecules likely bind to proteinaceous receptors (Buck & Axel 1991) in the plasma membrane of the cilia, triggering one or more transduction cascades that generally lead (in vertebrates) to depolarization of the olfactory cilia and knob. There is much new information on many aspects of these transduction events (see Ache & Zhainazarov 1995, Buck 1996, Krieger & Breer 1999, Schild & Restrepo 1998 for reviews), but details about adaptation, the intrinsic sensitivity of individual receptors, and relationships between biochemical cascades and OSN firing patterns remain undefined. Spatial and temporal consequences of these activation events within the cilia and olfactory knob are seen by the calcium imaging experiments shown in Figure 3a (Leinders-Zufall et al 1998). The top row of images (A E) shows the progression of increases in intracellular Ca ++ taken before, 2, 4, and 16 s after stimulation with cineole in a dissociated salamander OSN. This sequence explicitly demonstrates that odorant stimulation leads first to uniform Ca ++ changes within all the cilia. An increase in Ca ++ then spreads to the knob, dendrite, and soma, decaying in the different compartments with different time courses. As seen in the plots (F H ), the changes in intracellular calcium concentration last longer than responses recorded using electrophysiological methods and longer than the time course of a normal sniff (about one second in amphibia, less in mammals). This

6 968 KAUER WHITE imaging method extends patch clamp recordings from single ciliary membranes (Nakamura & Gold 1987) by clearly showing the degree of involvement and the dynamic spread of the initial response. Distribution of Receptor Types in the Olfactory Epithelium There are perhaps as many as 1000 genes in the mammal that encode the 7 transmembrane G-protein-linked receptor molecules found in the cilia. It is not yet known how many of the potential population of 1000 receptor types are expressed at any one time in a single animal. It appears that each OSN may express only one or a few of each receptor type. Based on in situ hybridization imaging OSNs expressing any one receptor are found in regional areas or zones of the nasal cavity in among other OSNs expressing other receptors. In rodents these zones have complex topographies, but are generally oriented in the anterior-posterior direction as stripes along different regions of the turbinates (see Buck 1996 for review). The task of correlating spectra of odorant responses from dissociated, individual OSNs and from OSNs in situ in the intact epithelium with the distribution of molecular receptor types has just begun (see below). Responses of Individual OSNs Based on many years of in vivo electrical recording, it is well established that single OSNs, in virtually all species studied, typically respond with temporally patterned bursts of action potentials to more than one odorant (e.g. see Gesteland et al 1965, Duchamp-Viret & Duchamp 1997 for review). Different OSNs have different response spectra and different sensitivities within the range of compounds to which they respond. Recent calcium imaging studies in dissociated OSNs in which the expressed receptor has been identified also show different responses to groups of test compounds, sometimes responding to relatively few chemically related stimuli, sometimes responding to broadly divergent chemical structures (Malnic et al 1999, Krautwurst et al 1998, Touhara et al 1999). These kinds of studies have begun to lay the foundation for explicit comparisons between receptor types and the structural features of odorants that bind to them. Other studies have exploited Ca ++ imaging methods for examining details of the transduction process (Noe & Breer 1998, Schild et al 1995, Gomez et al 2000). Calcium imaging of OSNs identified by retrograde labeling from different olfactory bulb glomerular regions has shown that cells projecting to the dorso-medial in contrast to the dorso-lateral bulb not only respond to somewhat different compounds, they also have different breadths of response (see Figure 3b and OB imaging described below) (Bozza & Kauer 1998). A remaining challenge for imaging studies is to assess how groups of simultaneously imaged, individual OSNs respond to odorants in real time when in their normal positions in the OE where the mucus, their relationships with surrounding supporting cells, and their axons are still intact. There are no observations yet available from OSNs that project to bulbar regions other than the optically accessible dorsal surface. A possible solution to

7 IMAGING AND OLFACTORY SYSTEM CODING 969 this problem, if the resolution of the technique is improved, may be to use imaging methods such as functional magnetic resonance imaging (fmri) that are not limited to surface structures and do not require surgical intervention (Yang et al 1998). Differential binding of odor ligands to different receptors is presumably the initial event in odorant discrimination, carrying the first information in the neural encoding process. As noted above, because single OSNs likely express one or few molecular receptor types and because each cell can respond to a number of odorants, the first order neurons likely do not recognize overall odorant structure, rather they recognize some chemical feature shared by the effective ligands. This hypothesis is consistent with observations and data gathered from the olfactory bulb and its essential tenet was suggested many years ago based on chemical analyses and human behavioral responses (Beets 1970, Polak 1973). All the studies at each level of the olfactory pathway that have sought to characterize the odorant responsiveness of individual cells have been confounded by the problem of how to adequately test the universe of all odors ( odor space ) to which a particular experimental animal is sensitive. Because the dimensionality of this space must be defined by behavioral measures, animals such as insects offer an opportunity for choosing odorant stimuli with behavioral relevance (Stopfer et al 1997). Whereas insects offer the possibility of observing behavior and physiology simultaneously, a number of other studies in vertebrates have also focused on correlating behavioral relevance with imaged activity (Coopersmith & Leon 1984, Galizia et al 1999, Kent et al 1995, Johnson & Leon 2000, Woo et al 1996, Youngentob & Kent 1995). Distribution of Responses across the Olfactory Epithelium Early electrophysiological studies in frog showed that responses to single compounds were widely distributed across the mucosa of the surgically exposed OE (Mustaparta 1971). Receptive field mapping in the salamander, also in surgically exposed preparations, showed that single mitral/tufted cells could be activated by punctate stimulation over large OE regions (Kauer & Moulton 1974; see summary in Kauer 1987). This study also provided evidence for widespread sensitivity to an odorant across the OE and showed that axons from distributed OSNs responsive to the same odorant converge onto single bulbar output cells. Other studies in which odorant responses in the amphibian OE were observed directly by voltage-sensitive dye (VSD) imaging (Kent & Mozell 1992) also showed distributed activity over wide areas that varied with odor. None of these experiments, however, examined patterning of OE activity with respect to the aerodynamics of how the odorants were delivered to the mucosal surface. More recent imaging experiments using VSD imaging have explicitly tested the effect of odorant delivery direction and examined the relative contributions of inherent differences in OSN receptor sensitivities compared with differences imposed by air flow direction and with chromatographic effects due to odorant retention in the mucus. Figure 4 (Kent et al 1996) shows the dramatic

8 970 KAUER WHITE differences in the patterns of VSD responses between strongly sorbed odorants (carvone and ethyl acetoacetate) and a weakly sorbed compound (propyl acetate) when the odorants were delivered either tangential (from the right) or normal (from the top) to the OE surface. The top and bottom of the figure show response profiles at high (top) and low (bottom) flow rates. The effect is best seen at the lower flow rate at the bottom. Strongly sorbed odorants delivered tangentially (left column) generate large response gradients, with the larger response close to the odorant input (right). The weakly sorbed odorant, propyl acetate, shows much less of a response gradient. The same odorants delivered from above directly to the OE (middle column), rather than along it, show more uniformly distributed response patterns. The imposed flow rate responses (right column) are generated by subtracting the middle from the left column. These data indicate that there are both response pattern differences generated by intrinsic OSN sensitivities as well as pattern differences imposed by the odorant delivery process itself. These studies suggest that the aerodynamic events, some of which are related to sniffing behavior, and some to the physical properties of the nasal cavity, can be important for understanding how responses occur within populations of distributed OSNs. Activation of sensory receptor cell populations is clearly sensitive to delivery conditions and is a widely distributed event showing complex responses in space across the convolutions of the turbinate structure and in time across regions having differential access to the odorant stream. One could imagine that evolutionary pressures have influenced the distribution of OSNs expressing particular molecular odorant receptors so that they are appropriately located to take advantage of the distribution of odorant due to their chemical properties and to the aerodynamics of the cavity (Mozell et al 1987). Projections of Olfactory Sensory Neurons to the Glomeruli of the Olfactory Bulb Evidence from several experiments now shows that the neighbor relations between OSNs in the OE are not preserved in their projection to the OB (see Mombaerts et al 1996) and are, therefore, unlike the connections between the periphery and central targets in the visual, auditory, and somatosensory systems. OSNs expressing a particular molecular receptor in the left or right nasal cavity, distributed within one of the OE receptor distribution zones, project their axons in a convergent fashion onto (usually) two glomeruli symmetrically situated on either side of the ipsilateral olfactory bulb [see Figure 7 and review by Buck 1996]. It is not known what determines which subset of OSNs from this population projects to one or the other of the two glomeruli. The presence of paired glomeruli is interesting because there are anatomical pathways connecting homologous lateral and medial OB regions (Schoenfeld et al 1985) that might use the information coming to the two glomerular targets to enhance detection or discrimination. It is generally thought that all the OSNs projecting to the two glomeruli share the same response profile, because they express the same receptor. However, at least one study (Malnic et al 1999) has shown that

9 IMAGING AND OLFACTORY SYSTEM CODING 971 dissociated OSNs ostensibly expressing the same receptor can sometimes show different responses to a particular odorant set. Much additional work is needed to relate the response properties of OSN populations expressing particular receptors to the response properties of their glomerular targets (Bozza & Kauer 1998). Imaged Responses in the Olfactory Bulb From the early 2DG experiments to the most recent intrinsic-signal imaging studies, the consistent finding has been that even single odorant compounds generate activity that is widely distributed across the glomerular layer of the OB. Different odorants generate different response patterns, and the patterns appear to relate in size and position to the modularity of glomeruli characteristic of this layer where the OSN axons terminate. In studies achieving spatial resolution adequate to see glomerular-sized structures it appears that different numbers of glomeruli are activated to different degrees by different odorant compounds and probably at different times (see below). It has also been shown that compounds with structural similarities to one another, as in homologous series, activate nearby structures. While here we focus on the application of imaging methods to odor coding in vivo, there are a number of studies carried out in vitro in which other attributes of OB circuit physiology have been examined using imaging approaches (Keller et al 1998; Lam et al 2000; Senseman 1996; Wachowiak & Cohen 1999; Wellis & Kauer 1993, 1994). Imaging studies using 2DG uptake provided one of the first opportunities to examine patterns of distributed, glomerular, activity generated by odor stimulation (Stewart et al 1979, Coopersmith & Leon 1984). Results from this method have helped guide the development of odorant coding hypotheses and have specifically directed experiments that have been pursued using other methods. For example, Bell and colleagues (1987) demonstrated that propionic acid activates a particularly prominent 2DG uptake region in the dorso-medial OB. This finding was then used by Mori and colleagues (Imamura et al 1992, Mori et al 1992) to select a location at which to examine odorant response profiles of single-unit output mitral cells. These single-unit studies were among the first to characterize odorant response spectra of mitral cells in defined OB locations, giving rise to the concept of mitral cell molecular receptive ranges for odorants with structural similarities. The naturally discrete anatomical features of glomerular structures and the localization of odor-generated activity within the glomerular layer have long been thought to represent modules that could provide a basis for an odorant coding scheme (Shepherd 1981). Observations with VSDs showed that odorants stimulate modular-like structures across the layers of the in vivo salamander OB (Kauer & Cinelli 1993). Modules extending through the layers of the rat OB were also suggested by studies using c-fos expression (Guthrie et al 1993). Figure 5a shows increased c-fos expression in modular-like groups of glomeruli and regions of underlying OB layers after exposure of the animal

10 972 KAUER WHITE to peppermint (A), isoamyl acetate (B), and air (C) for 30 min. These c-fos expression patterns were similar to 2DG glomerular patterns generated by the same odors, but in addition, showed broad, flask-shaped regions in the deeper layers related to the glomerular foci, similar to the voltage-sensitive dye patterns in the salamander. The modular nature of OB activity has been further investigated by Johnson & Leon (2000) using quantitative 2DG methods in the rat. Examples of activity patterns generated by four concentrations of two related odorants are shown in Figure 5b. The pseudo-colored images depict the glomerular layer unfolded on its ventral meridian as shown at the top of the figure. These studies also indicated that compounds that are perceived by humans to differ qualitatively with intensity show changes in their activity pattern as concentration is increased (see white arrows in pentanal series). A second odorant, methyl pentanoate, that is not perceived differently at different concentrations did not show qualitative activity pattern differences. Activity patterns also have been examined by imaging fluorescent dyes or intrinsic signals in a number of species in vivo: bees (Joerges et al 1997), zebrafish (Friedrich & Korsching 1998, 1997), salamander (Cinelli et al 1995a,b; Cinelli & Kauer 1995a), and rat (Rubin & Katz 1999, Uchida et al 2000). Two examples are shown in Figure 6. Figure 6a shows presynaptic activity in OSN terminals of zebrafish OB glomeruli labeled by anterograde transport of a Ca ++ indicator dye delivered to the OE (Friedrich & Korsching 1997). Stimulation with various amino acids clearly generates different patterns of activity that are shared by certain glomeruli for a number of the compounds. For example, the region indicated by the white arrow (1) in the upper right Trp panel showed activity to all neutral amino acids; the white arrow (2) in the central Ile panel shows a region with activity to Val, Ile, Leu, and Met; and the white arrow (3) in the lower left Lys panel shows a region with activity to basic amino acids. Consistent with 2DG and c-fos studies, activity for all compounds was found in a number of sites with some overlap. Figure 6b shows data from a study, which used intrinsic signal imaging, asking similar questions about the molecular receptive ranges of glomeruli in the rat OB (Rubin & Katz 1999). Stimulation consisted of a homologous series of n-aliphatic aldehydes with different carbon chain lengths. In (A), the histogram bars represent the glomerular signal intensities at positions shown by the filled circle in the inset. The different aldehyde chain lengths are shown above. Shaded bars are responses to the aldehydes that were >50% of the highest response; open bars are responses <50%. Note that a particular glomerulus responded best to aldehydes with chain lengths similar to the compound that gave the largest response. Furthermore, a number of the glomeruli within the extent of the inset showed similar response ranges. There were, of course, glomeruli in this field that showed no response to these stimuli, and one does not know what other compounds might have generated activity in these glomeruli had they been tested. Figure 6b (B) shows the ranges of aldehyde carbon chain length to which all the glomeruli observable in this experiment responded.

11 IMAGING AND OLFACTORY SYSTEM CODING 973 Another recent study using intrinsic signal imaging in the rat demonstrated two response domains in the dorsal OB that have activity to sets of odorants having different functional groups (Uchida et al 2000). This experiment showed responses similar to those described above for a series of aldehydes. Additional data were also presented that suggested there may be subdivisions within the observed OB domains in which responses not only relate to carbon chain length but also to functional groups and molecular branching patterns. Again, not all glomeruli within the areas observed responded to the compounds tested. Taken together, these findings indicate that odorant representation in the OB has the following characteristics: 1. Monomolecular odorant stimulation leads to patterns of activity that are distributed nonhomogeneously over the glomerular sheet and underlying layers. 2. Higher concentration stimuli give larger patterns of response that may change pattern shape. 3. Activity in the deeper OB layers is often located beneath foci of activity in the glomerular layer and may reflect modules extending through the bulbar layers. 4. Spatially distributed activity patterns probably relate to certain structural features of the stimulus. Such differential patterns undoubtedly exist, and their elucidation has been a major advance in beginning to describe the coding process. 5. Some classes of compounds generate activity in identifiable glomerular regions, but within these domains, not all glomeruli respond to the tested compounds. 6. There are many other regions of the bulb that have not yet been observed because of optical inaccessibility. The larger picture of what actually happens during the second or less that it takes an animal to make a behavioral odor discrimination is, however, likely to be more complex than these patterns suggest. There are several reasons for this. The first is the question of how short-term, temporally patterned, behaviorally relevant responses contribute to the spatial patterns observed by methods that require extended stimulation such as 2DG, c-fos, and imaging of intrinsic signals. The second is the question of whether the patterns are really stationary if observed in real time (Cinelli et al 1995a). For example, does activity actually rise and fall within different areas that are represented by static observation methods that average activity over time? Third, it is difficult to reconcile exclusively spatial encoding of odorant information with studies in which large lesions of the bulb that substantially disrupt regions involved in the spatial responses show relatively little effect on behavioral discrimination (Hudson & Distel 1987, Lu & Slotnick 1998, Slotnick et al 1997). Relatively few imaging studies have used methods that have adequate temporal resolution to observe neuronal events in real time (see Cinelli et al 1995a c). In

12 974 KAUER WHITE vertebrates many electrophysiological studies have consistently shown that activity patterns in the OB measured by both electroencephalogram and single unit methods change over time (e.g. Eeckman & Freeman 1990, Freeman & Di Prisco 1986, Harrison & Scott 1986, Wellis et al 1989, Hamilton & Kauer 1989, Kauer 1974, Meredith & Moulton 1978). In invertebrates attempts to understand temporal patterning in olfactory lobe circuits have been taken a step further by Laurent and his colleagues by comparing single-cell firing patterns with field potential patterns and with behavior (see Stopfer et al 1997, Laurent 1999 for review). Such complex temporal properties likely also occur in vertebrates and probably relate to the relationships among single neuron firing patterns and synchronous firing among populations. fmri methods may be useful for such studies in the future. In addition to caveats relating to temporal and spatial issues, it should be noted that the system by which odorants are encoded and classified by the olfactory system need not coincide with conventional chemical classification related to, for example, functional groups and carbon chain length. Based on the response ranges of individual OSNs and their concentration sensitivities (Duchamp-Viret & Duchamp 1997), it would seem likely that olfactory molecular receptors do not recognize attributes of their cognate ligands via high affinity binding. Individual OSNs appear to be broadly responsive, rather than quite specific. Therefore, it may be unreasonable to expect to be able to define strict pharmacological relationships among ligand candidates and particular molecular receptor types. Concerns about specificity also, of course, always need to be considered in the context of how many odorants are tested, what concentrations were examined, and how long the stimuli were applied. This latter concern is especially important for the sense of smell because the olfactory system adapts so readily. Figure 7 summarizes some of the attributes of the peripheral system as described above. This diagram is based on data from the salamander, but the general principles appear to hold for other vertebrates. OSNs expressing molecular receptors for recognition sites on odorant molecules are distributed within bounded regions across the OE. Axonal projections from OSNs expressing one receptor type converge on glomeruli in the OB. The groups of mitral/tufted, periglomerular, and granule cells related to the activated glomerular regions may form functional modules, a number of which participate in the encoding of odorant molecular structure. This is, of course, a highly simplified diagram, but a number of its features have been revealed by the imaging methods discussed above. CONCLUSIONS The use of methods for imaging neuronal activity has enhanced our understanding of the odor coding process. The results are beginning to generate complementary views that emphasize different aspects of coding in space and time. There is still no single approach that provides all of the ideal properties of noninvasive, high spatial

13 IMAGING AND OLFACTORY SYSTEM CODING 975 and temporal observation. By gathering data using approaches that complement one another and that mitigate each others deficiencies, we begin to assemble a view of olfactory events that can be related to identifiable properties of the odorant stimulus world and to the extraordinary behavioral abilities of animals using this sensory modality. Among the many important research problems that still need to be pursued are (a) elucidation of the detailed molecular interactions between defined olfactory receptor molecules and odorants structure, (b) identification of which anatomical and physiological details in olfactory circuits are required for odor detection and discrimination in real world behaviors, and (c) characterization of the modes by which olfactory information is encoded in higher olfactory centers extending from prepyriform to neocortex. Given the progress so far and the history of the contributions of imaging methods, it is likely that imaging studies will also contribute significantly to the solutions of these problems in the future. ACKNOWLEDGMENTS This work was carried out with support from The National Institutes of Health NIDCD, the Office of Naval Research, and the Defense Advanced Research Projects Agency. We thank Barbara Talamo for critically reading the manuscript. Visit the Annual Reviews home page at LITERATURE CITED Ache BW, Zhainazarov A Dual secondmessenger pathways in olfactory transduction. Curr. Opin. Neurobiol. 5: Adrian ED Sensory messages and sensation. The response of the olfactory organ to different smells. Acta Physiol. Scand. 29:5 14 Ai H, Okada K, Hill ES, Kanzaki R Spatio-temporal activities in the antennal lobe analyzed by an optical recording method in the male silkworm moth Bombyx mori. Neurosci. Lett. 258: Antic S, Major G, Zecevic D Fast optical recordings of membrane potential changes from dendrites of pyramidal neurons. J. Neurophysiol. 82: Beets MGJ The molecular parameters of olfactory response. Pharmacol. Rev. 22:1 34 Bell GA, Laing DG, Panhuber H Odour mixture suppression: evidence for a peripheral mechanism in human and rat. Brain Res. 426:8 18 Blasdel GG, Salama G Voltage-sensitive dyes reveal a modular organization in monkey striate cortex. Nature 321: Bozza TC, Kauer JS Odorant response properties of convergent olfactory receptor neurons. J. Neurosci. 18: Buck LB Information coding in the vertebrate olfactory system. Annu. Rev. Neurosci. 19: Buck LB, Axel R A novel multigene family may encode odorant receptors: a molecular basis for odor recognition. Cell 65: Christensen TA, White J Representation of olfactory information in the brain. In The Neurobiology of Taste and Smell, ed. TE Finger, WL Silver, D Restrepo, pp New York: Wiley-Liss Cinelli AR, Hamilton KA, Kauer JS. 1995a. Salamander olfactory bulb neuronal activity observed by video-rate voltage-sensitive dye imaging. III. Spatio-temporal properties of

14 976 KAUER WHITE responses evoked by odorant stimulation. J. Neurophysiol. 73: Cinelli AR, Kauer JS Voltage-sensitive dyes and functional activity in the olfactory pathway. Annu. Rev. Neurosci. 15: Cinelli AR, Kauer JS. 1995b. Salamander olfactory bulb neuronal activity observed by video-rate voltage-sensitive dye imaging. II. Spatio-temporal properties of responses evoked by electrical stimulation. J. Neurophysiol. 73: Cinelli AR, Neff SR, Kauer JS. 1995c. Salamander olfactory bulb neuronal activity observed by video-rate voltage-sensitive dye imaging. I. Characterization of the recording system. J. Neurophysiol. 73: Cohen LB, Lesher S Optical monitoring of membrane potential: methods of multisite optical measurement. Soc. Gen. Physiol. Ser. B 40:71 99 Coopersmith R, Leon M Enhanced neural response to familiar olfactory cues. Science 225: Duchamp-Viret P, Duchamp A Odor processing in the frog olfactory system. Prog. Neurobiol. 53: Dynes JL, Ngai J Pathfinding of olfactory neuron axons to stereotyped glomerular targets revealed by dynamic imaging in living zebrafish embryos. Neuron 20: Ebner TJ, Chen G Use of voltagesensitive dyes and optical recordings in the central nervous system. Prog. Neurobiol. 46: Eeckman FH, Freeman WJ Correlations between unit firing and EEG in the rat olfactory system. Brain Res. 528: Freeman WJ, Di Prisco GV Relation of olfactory EEG to behavior: time series analysis. Behav. Neurosci. 100: Friedrich RW, Korsching SI Combinatorial and chemotopic odorant coding in the zebrafish olfactory bulb visualized by optical imaging. Neuron 18: Friedrich RW, Korsching SI Chemotopic, combinatorial, and noncombinatorial odorant representations in the olfactory bulb revealed using a voltage-sensitive transducer. J. Neurosci. 18: Galizia CG, Sachse S, Rappert A, Menzel R The glomerular code for odor representation is species specific in the honeybee Apis mellifera. Nat. Neurosci. 2: Gervais R, Kleinfeld D, Delaney KR, Gelperin A Central and reflex neuronal responses elicited by odor in a terrestrial mollusk. J. Neurophysiol. 76: Gesteland RC, Lettvin JY, Pitts WH Chemical transmission in the nose of the frog. J. Physiol. 181: Gomez G, Rawson NE, Cowart B, Lowry LD, Pribitkin EA, Restrepo D Modulation of odor-induced increases in [Ca(2+)](i) by inhibitors of protein kinases A and C in rat and human olfactory receptor neurons. Neuroscience 98: Grinvald A, Lieke E, Frostig RD, Gilbert CD, Wiesel TN Functional architecture of cortex revealed by optical imaging of intrinsic signals. Nature 324: Grynkiewicz G, Poenie M, Tsien RY A new generation of Ca ++ indicators with greatly improved fluorescence properties. J. Biol. Chem. 260: Guthrie KM, Anderson AJ, Leon M, Gall C Odor-induced increases in c-fos mrna expression reveal an antomical unit for odor processing in olfactory bulb. Proc. Natl. Acad. Sci. USA 90: Hahn I, Scherer PW, Mozell MM Velocity profiles measured for airflow through a large-scale model of the human nasal cavity. J. Appl. Physiol. 75: Hamilton KA, Kauer JS Patterns of intracellular potentials in salamander mitral/tufted cells in response to odor stimulation. J. Neurophysiol. 62: Harrison TA, Scott JW Olfactory bulb responses to odor stimulation: analysis of response pattern and intensity relationships. J. Neurophysiol. 56: Hildebrand JG, Shepherd GM Mechanisms of olfactory discrimination: converging evidence for common principles

15 IMAGING AND OLFACTORY SYSTEM CODING 977 across phyla. Annu. Rev. Neurosci. 20: Hudson R, Distel H Regional autonomy in the peripheral processing of odor signals in newborn rabbits. Brain Res. 421:85 94 Imamura K, Mataga N, Mori K Coding of odor molecules by mitral/tufted cells in rabbit olfactory bulb. I. Aliphatic compounds. J. Neurophysiol. 68: Joerges J, Kuttner A, Galizia CG, Menzel R Representations of odours and odour mixtures visualized in the honeybee brain. Nature 387: Johnson BA, Leon M Modular representations of odorants in the glomerular layer of the rat olfactory bulb and the effects of stimulus concentration. J. Comp. Neurol. 422: Kauer JS Response patterns of amphibian olfactory bulb neurones to odour stimulation. J. Physiol. 243: Kauer JS Coding in the olfactory system. In The Neurobiology of Taste and Smell, ed. TE Finger, WL Silver, pp New York: Wiley Kauer JS, Cinelli AR Are there structural and functional modules in the vertebrate olfactory bulb? Microsc. Res. Tech. 24: Kauer JS, Moulton DG Responses of olfactory bulb neurones to odour stimulation of small nasal areas in the salamander. J. Physiol. 243: Keller A, Yagodin S, Aroniadou-Anderjaska V, Zimmer LA, Ennis M, et al Functional organization of rat olfactory bulb glomeruli revealed by optical imaging. J. Neurosci. 18: Kennedy C, Des Rosiers MH, Jehle JW, Reivich M, Sharp FR, Sokoloff L Mapping of functional neural pathways by autoradiographic survey of local metabolic rate with 14C-deoxyglucose. Science 187: Kent PF, Mozell MM The recording of odorant-induced mucosal activity patterns with a voltage-sensitive dye. Journal of Neurophysiology 68: Kent PF, Mozell MM, Murphy SJ, Hornung DE The interaction of imposed and inherent olfactory mucosal activity patterns and their composite representation in a mammalian species using voltage-sensitive dyes. J. Neurosci. 16: Kent PF, Youngentob SL, Sheehe PR Odorant-specific spatial patterns in mucosal activity predict perceptual differences among odorants. Journal of Neurophysiology 74: Keyhani K, Scherer PW, Mozell MM A numerical model of nasal odorant transport for the analysis of human olfaction. J. Theoret. Biol. 186: Kimura T, Toda S, Sekiguchi T, Kawahara S, Kirino Y Optical recording analysis of olfactory response of the procerebral lobe in the slug brain. Learning & Memory 4: Kobal G, Kettenmann B Olfactory functional imaging and physiology. Int. J. Psychophysiol. 36: Krautwurst D, Yau KW, Reed RR Identification of ligands for olfactory receptors by functional expression of a receptor library. Cell 95: Krieger J, Breer H Olfactory reception in invertebrates. Science 286: Lam YW, Cohen LB, Wachowiak M, Zochowski MR Odors elicit three different oscillations in the turtle olfactory bulb. J. Neurosci. 20: LaMantia AS, Pomeroy SL, Purves D Vital imaging of glomeruli in the mouse olfactory bulb. J. Neurosci. 12: Laurent G A systems perspective on early olfactory coding. Science 286: Leinders-Zufall T, Greer CA, Shepherd GM, Zufall F Imaging odor-induced calcium transients in single olfactory cilia: specificity of activation and role in transduction. J. Neurosci. 18: Leveteau J, MacLeod P Olfactory discrimination in the rabbit olfactory glomerulus. Science 153: Levy LM, Henkin RI, Lin CS, Finley A Rapid imaging of olfaction by functional

16 978 KAUER WHITE MRI (fmri): identification of presence and type of hyposmia. J. Comput. Assist. Tomogr. 23: Lieke EE, Frostig RD, Arieli A, Tso DY, Hildesheim R, Grinvald A Optical imaging of cortical activity: real-time imaging using extrinsic dye-signals and high resolution imaging based on slow intrinsicsignals. Annu. Rev. Physiol. 51: Litaudon P, Datiche F, Cattarelli M Optical recording of the rat piriform cortex activity. Prog. Neurobiol. 52: Lu XC, Slotnick BM Olfaction in rats with extensive lesions of the olfactory bulbs: implications for odor coding. Neuroscience 84: Malnic B, Hirono J, Sato T, Buck LB Combinatorial receptor codes for odors. Cell 96: Meredith M, Moulton DG Patterned response to odor in single neurones of goldfish olfactory bulb: influence of odor quality and other stimulus parameters. J. Gen. Physiol. 71: Mombaerts P Seven-transmembrane proteins as odorant and chemosensory receptors. Science 286: Mombaerts P, Wang F, Dulac C, Chao SK, Names A, et al Visualizing an olfactory sensory map. Cell 87: Mori K, Fujita SC, Imamura K, Obata K Immunohistochemical study of subclasses of olfactory nerve fibers and their projections to the olfactory bulb in the rabbit. J. Comp. Neurol. 242: Mori K, Mataga N, Imamura K Differential specificities of single mitral cells in rabbit olfactory bulb for a homologous series of fatty acid odor molecules. J. Neurophysiol. 67: Mori K, Nagao H, Yoshihara Y The olfactory bulb: coding and processing of odor molecule information. Science 286: Moulton DG Spatial patterning of response to odors in the peripheral olfactory system. Physiol. Rev. 56: Mozell MM, Sheehe PR, Hornung DE, Kent PF, Youngentob SL, Murphy SJ Imposed and inherent mucosal activity patterns. Their composite representation of olfactory stimuli. J. Gen. Physiol. 90: Mustaparta H Spatial distribution of receptor responses to stimulation with different odours. Acta Physiol. Scand. 82: Nakamura T, Gold GH A cyclic nucleotide-gated conductance in olfactory receptor cilia. Nature 325: Noe J, Breer H Functional and molecular characterization of individual olfactory neurons. J. Neurochem. 71: Polak EH Multiple profile-multiple receptor site model for vertebrate olfaction. J. Theor. Biol. 40: Rubin BD, Katz LC Optical imaging of odorant representations in the mammalian olfactory bulb. Neuron 23: Sagar SM, Sharp FR, Curran T Expression of c-fos protein in brain: metabolic mapping at the cellular level. Science 240: Schild D, Geiling H, Bischofberger J Imaging of L-type Ca2+ channels in olfactory bulb neurones using fluorescent dihydropyridine and a styryl dye. J. Neurosci. Methods 59: Schild D, Restrepo D Transduction mechanisms in vertebrate olfactory receptor cells. Physiol. Rev. 78: Schoenfeld TA, Marchand JE, Macrides F Topographic organization of tufted cell axonal projections in the hamster main olfactory bulb: an intrabulbar associational system. J. Comp. Neurol. 235: Senseman DM High-speed optical imaging of afferent flow through rat olfactory bulb slices: voltage-sensitive dye signals reveal periglomerular cell activity. J. Neurosci. 16: Shepherd GM The olfactory glomerulus; its significance for sensory processing. In Brain Mechanisms of Sensation, ed. Y Katsuki, R Norgren, M Sto, pp New York: Wiley Siegel MS, Isacoff EY A genetically

17 IMAGING AND OLFACTORY SYSTEM CODING 979 encoded optical probe of membrane voltage. Neuron 19: Slotnick BM, Bell GA, Panhuber H, Laing DG Detection and discrimination of propionic acid after removal of its 2-DG identified major focus in the olfactory bulb: a psychophysical analysis. Brain Res. 762:89 96 Sobel N, Prabhakaran V, Desmond JE, Glover GH, Goode RL, et al Sniffing and smelling: separate subsystems in the human olfactory cortex. Nature 392: Stewart WB, Kauer JS, Shepherd GM Functional organization of rat olfactory bulb analysed by the 2-deoxyglucose method. J. Comp. Neurol. 185: Stopfer M, Bhagavan S, Smith BH, Laurent G Impaired odour discrimination on desynchronization of odour-encoding neural assemblies. Nature 390:70 74 Touhara K, Sengoku S, Inaki K, Tsuboi A, Hirono J, et al Functional identification and reconstitution of an odorant receptor in single olfactory neurons. Proc. Natl. Acad. Sci. USA 96: Erratum Proc. Natl. Acad. Sci. USA 97(7):3782 Uchida N, Takahashi YK, Tanifuji M, Mori K Odor maps in the mammalian olfactory bulb: domain organization and odorant structural features. Nat. Neurosci. 3: Wachowiak M, Cohen LB Presynaptic inhibition of primary olfactory afferents mediated by different mechanisms in lobster and turtle. J. Neurosci. 19: Wellis DP, Kauer JS GABAa and glutamate receptor involvement in dendrodendritic synaptic interactions from salamander olfactory bulb. J. Physiol. 469: Wellis DP, Kauer JS GABAergic and glutamatergic synaptic input to identified granule cells in salamander olfactory bulb. J. Physiol. 475: Wellis DP, Scott JW, Harrison TA Discrimination among odorants by single neurons of the rat olfactory bulb. J. Neurophysiol. 61: Williams M, Johnston JM, Cicoria M, Paletz E, Waggoner LP, et al Canine detection odor signatures for explosives. In Proc. 2nd Annu. Conf. Enforc. Secur. Technol., ed. AT DePersia, JJ Pennella. Boston, MA: Int. Soc. Optic. Eng. Woo CC, Oshita MH, Leon M A learned odor decreases the number of Fosimmunopositive granule cells in the olfactory bulb of young rats. Brain Res. 716: Yang X, Renken R, Hyder F, Siddeek M, Greer CA, et al Dynamic mapping at the laminar level of odor-elicited responses in rat olfactory bulb by functional MRI. Proc. Natl. Acad. Sci. USA 95: Youngentob SL, Kent PF Enhancement of odorant-induced mucosal activity patterns in rats trained on an odorant identification task. Brain Res. 670:82 88 Yousem DM, Maldjian JA, Siddiqi F, Hummel T, Alsop DC, et al Gender effects on odor-stimulated functional magnetic resonance imaging. Brain Res. 818: Zald DH, Pardo JV Functional neuroimaging of the olfactory system in humans. Int. J. Psychophysiol. 36: Zochowski M, Wachowiak M, Falk CX, Cohen LB, Lam YW, et al Imaging membrane potential with voltage-sensitive dyes. Biol. Bull. 198:1 21

18 Figure 1 Multisite monitoring of a spike evoked by shock to the white matter in a pyramidal cell from a slice of rat somatosensory cortex. Cell was filled with the voltage sensitive dye JPW3028. The entire extent of the cell is covered by positioning the photodiode detector array over two positions (see two grids). (A) CCD camera fluorescence image of the cell. (B) Distributions of action potentials taken at the two diode array positions. Fluorescence recordings are normalized to resting light level. (C) Expanded time base action potential signals, scaled to the same height, from diodes at different locations shown in B. (D) Color coded representation of data in B provides information about both spatial and temporal dimensions of the electrical events (color range from blue to red indicates peak of action potential in red; numbers are times in ms after the shock). From Antic et al 1999 with permission.

19 Figure 2 Schlieren images of air flow in two different dogs during inhalation (left) and exhalation (right). Dog at the left is sniffing a small food pellet. Notice the air stream entering the nose from the front (arrow). Dog at the right is exhaling after sniffing from source (probe at right). Notice the air leaving the nose in a backwards direction (arrow) as a result of the movement of a muscular flap at the naris. From Settles 2000 with permission.

20 Figure 3 (a) Confocal measurements of transient increases in intracellular Ca ++ concentration using fluo-3 from cilia and knob of a salamander olfactory sensory neuron (OSN) after a 1 sec pulse of cineole, 300 mm. (A) Phase-contrast image showing several cilia emanating from the knob. (B E) Pseudocolored series (0.5 Hz) of fluorescence images taken before the odor (B), near peak fluorescence at 2 sec (C), 4 sec (D), and 16 sec (E) after the odor pulse. In E the ciliary signal has returned to baseline, but Ca ++ remains high in the knob. (F) Time courses of ciliary Ca ++ transient from cilium labeled by the white arrow in C. The decay time constant (dotted line exponential fit) was 5.3 sec. Letters indicate when frames above were taken. (G) Time courses of odor-induced Ca ++ changes in various OSN compartments. Data are from the neuron shown in A E. (H) Time course and peak amplitude of the ciliary signal depends on odorant concentration. Odor-free Ringer s gives no signal. From Leinders-Zufall et al 1998 with permission. (b) Example of odorant response profiles of individual OSNs projecting to different bulbar regions. Different OSNs (row numbers) responding to various odorant mixtures [column letters A, B, C, D, E, F, OA (organic acids)]. (A) Dorso-lateral projecting OSNs (n = 21). (B) Dorso-medial projecting neurons (n = 20). Blank spaces indicate a stimulus was not tested; dashes indicate no response. The percentage of cells responding to each mixture is below each plot. Cells with the same number are from the same animal and are ordered by increasing breadth of response. Size of filled circles indicates relative response magnitudes. From Bozza & Kauer 1998 with permission.

21 Figure 4 Voltage-sensitive dye activity patterns from an in vitro rat olfactory epithelium (OE). Color-enhanced surface plots for composite, inherent, and imposed activity patterns on the medial surface of the turbinates in response to carvone, ethyl acetoacetate, and propyl acetate when drawn across the surface at either 440 cc/min (top) or 100 cc/min (bottom). Composite patterns are observed when the odorant is delivered from the external naris (EN) side of the recording array. Inherent patterns are observed by stimulating directly down onto the OE from above so there is no tangential flow component. Imposed patterns are generated by subtracting the inherent from the composite responses. The z-axis shows relative response magnitudes, measured as natural logarithms, normalized to the response to amyl acetate puffed in the downward direction. Log values vary from 1.0 to 0.5 as indicated by the color bar to the right. EN, external naris; NP, nasopharynx; D, dorsal; CR, cribriform plate. From Kent et al 1996 with permission.

22 Figure 5 See legend for next page

23 Figure 5 (a) (A C) Pseudocolored autoradiograms of rat olfactory bulb (OB) sections hybridized with 35 S-labeled c-fos crna of animals exposed to peppermint odor (A), isoamyl acetate (B), or clean air (C) for 30 min. Distinct hybridization patterns are seen in glomerular (gl, solid arrows) and granule cell layers (gcl, arrowheads) with odor (A and B), but much less with air (C). From Guthrie et al 1993 with permission. (b) Averaged patterns of 2DG uptake across the glomerular layer exposed to different concentrations of two odorants. Each image is data averaged from 3 5 rats. The orientation of the unfolded glomerular layer is shown in the upper left panel. Z-score values are relative degrees of 2DG uptake. Asterisks indicate concentrations that gave patterns different from air. For pentanal, black arrows denote patterns that were consistent across all concentrations above threshold. White arrows indicate patterns that emerged at higher concentrations. Scale bar = 2 mm. From Johnson & Leon 2000 with permission. Figure 6 (a) Glomerular activity patterns in the zebrafish olfactory bulb. Activity was induced by 18 amino acids (aa s) (10 um) in the aa-respove subregion as measured by changes in fluorescence of calcium green dextran. Glomeruli were loaded by anterograde transport of the dye nsi from the olfactory epithelium. Note each stimulus induces a unique pattern but with some shared active glomeruli. Arrows show glomerular mo- dules responding to all neutral aa s (1); to Val, Ile, Leu, and Met (2); and to basic aa s (3). From Friedrich & Korsching 1997 with permission. (b) Individual rat glomeruli appear to have restricted receptive ranges when tested with a limited set of odorants varying along a single presumed molecular dimension. (A) For four glomeruli in the inset at right, the magnitude of the optically imaged response of each glomerulus is plotted as a percentage of the maximal response. Odorants eliciting >50% of the maximal response were considered effective in activating a glomerulus (filled bars; open bars are glomeruli with <50% response). These four glomeruli were activated by between one and three members of the homologous series. (B) Summary of molecular receptive ranges from 40 glomeruli (n = 5 bulbs in 4 animals) to the homologous series. From Rubin & Katz 1999 with permission.

24 Figure 6 See legend for previous page