194, 1–21 (1994) 1 Printed in Great Britain © The Company ...jeb.biologists.org/content/jexbio/194/1/1.full.pdf · secondary and tertiary targets in the pyriform cortex (lateral

1J. exp. Biol. 194, 1–21 (1994)Printed in Great Britain © The Company of Biologists Limited 1994

REVIEW

TRANSDUCTION DIVERSITY IN OLFACTION

VINCENT E. DIONNE AND ADRIENNE E. DUBIN*

Boston University Marine Program, Marine Biological Laboratory, Woods Hole,MA 02543, USA

SummaryOdors are powerful stimuli that can focus the attention, elicit behaviors (or

misbehaviors) and even resurrect forgotten memories. These actions are directed by thecentral nervous system, but they depend upon the initial transduction of chemical signalsby olfactory receptor neurons. Electrophysiological recordings suggest that the responsesof olfactory receptor neurons to odors are more diverse than was initially believed, beingmediated by effects on several different conductances. Both excitatory and inhibitoryresponses are produced by these effects and some, if not all, odors can affect more thanone component of the membrane conductance. The extent of this diversity is reviewedhere, and its impact on our understanding of odor discrimination is discussed.

Introduction

The olfactory systems of extant animals derive from the earliest chemical sensingsystems. These senses were crucial to survival, providing organisms with an ability tofind food and mates. The modern system displays a structure (Fig. 1) that is remarkablysimilar among different species and phyla, suggesting that this ancient sense dependsupon certain robust, but still obscure, biological principles. In spite of its relatively simplearchitecture, we are still learning how the olfactory system acquires and processessensory information. There are three main reasons for our poor understanding. First, thereare no basic principles for organizing odors in terms of critical functional parameters.Second, except for the peripheral portion of each sensory neuron, most of the olfactorysystem is relatively inaccessible. Third, as discussed below, there is no universalmechanism for transducing odor signals; rather, several mechanisms are used. Thisdiversity of transduction mechanisms may be necessary for olfactory discrimination.Several recent reviews provide additional information (Getchell, 1986; Reed, 1992;Shirley, 1992; Farbman, 1992; Scott, 1991; Kauer, 1991; Anholt, 1993; Trombley andShepherd, 1993).

There is no unanimity of opinion about the mechanisms that underlie odor detectionand transduction. In the recent literature, several mechanisms for odor transduction have

*Present address: Department of Biology, San Diego State University, San Diego, CA 92182, USA.

Key words: olfaction, odor transduction, olfactory receptor neurons, odor modulation.

been proposed, including direct and receptor-mediated indirect effects on ion channelsand changes in membrane fluidity. Unarguably, odors can affect ion channels, membranefluidity and enzyme activity in olfactory receptor neurons (ORNs). But non-olfactorycells also respond to odors, suggesting that not all odor responses mediate physiologicallyrelevant olfactory transduction. The real question is whether any or all of the effectsexamined in vitro underlie odor transduction in vivo. Different investigators have seldomemployed the same criteria to address this issue. We have used the following minimalrequirements that any mechanism must satisfy if it is used for odor transduction. (1) Anodorant should not elicit identical responses from all cells. If each odorant affected allORNs identically, there could be no odor discrimination. (2) The response must occur atodor concentrations encountered in vivo. Responses that are elicited only by highconcentrations of odor may not be mediated by mechanisms that underlie physiologicalodor transduction. (3) Exposure of only the apical membrane of an ORN to odor must besufficient to elicit the response. Only the cilia and dendritic endings of ORNs are exposedto odors in vivo. Tight junctions between the apical end of the dendrite and adjacent cellsocclude odors from the basolateral space around the dendrite and soma. (4) The odor-induced response must affect neuronal excitability. A response that has no effect onexcitability will not be communicated to the central nervous system. This requirementencompasses both intracellular and intercellular pathways, the latter communicated byelectrical coupling (Schwartz Levey et al. 1992) or membrane-permeant chemical agentssuch as nitric oxide (Breer and Shepherd, 1993) and carbon monoxide (Ronnett andSnyder, 1992).

2 V. E. DIONNE AND A. E. DUBIN

mp

lp

Nasalcavity

Olfactorybulb

am

Fig. 1

Detection thresholds for different odorants vary enormously among cells, dependingboth on whether a particular odorant is the optimal stimulus for an individual ORN and onhow the thresholds are measured. Thresholds measured in situ by single-unit recording orin vitro by intracellular and patch-clamp methods range from 10212 to 1025 mol l21

(Sutterlin and Sutterlin, 1971; Suzuki and Tucker, 1971; Getchell, 1974; Caprio, 1978;Silver, 1982; Frings and Lindemann, 1990). This wide range of sensitivity may reflect thenumber of receptors of a particular type on different cells and may let the system monitorodor concentration as well as quality. An even wider range of thresholds has beenreported using more integrative or behavioral assays (Passé and Walker, 1985). Thisbroader range is not surprising, since threshold measurements using the intact olfactorysystem should reflect the most and least sensitive cells, which might not be assayed insingle-cell studies. In air-breathing vertebrates, including humans, sensitivity thresholdsare reported in the range from less than 10213 to 1024 mol l21 with a mean in thepicomolar range (Patte et al. 1975; Moulton, 1977; Amoore, 1982; Slotnick and

3Transduction diversity in olfaction

Fig. 1. Diagram of a vertebrate olfactory system. The vertebrate olfactory system consists ofbilaterally paired sensory neuroepithelia, each projecting to an olfactory bulb and then tosecondary and tertiary targets in the pyriform cortex (lateral pallium, lp), hippocampus(medial pallium, mp), amygdala (am) and other parts of the limbic system. The neuroepitheliaare located in the nasal passages, where they are bathed by mucus and exposed to the externalenvironment. Odors are detected and transduced on the apical ends of olfactory receptorneurons (ORNs) whose somata reside in the epithelia. ORNs extend two neurites from theirsoma; one is a thin dendrite (approximately 1 mm in diameter) which grows to the apicalsurface of the neuroepithelium and extends a cluster of 10–20 cilia or microvilli into themucus. Odors are detected on the apical membranes of the cilia and dendritic ending. Thesecond neurite is an unmyelinated axon which runs from the basal surface of the epitheliumthrough the cribriform plate of the ethmoid bone to the ipsilateral olfactory bulb. There, theORNs make synaptic connections with primary projection neurons (mitral and tufted cells) aswell as inhibitory interneurons (periglomerular short axon neurons) in small neuropils(approximately 100 mm in diameter) termed glomeruli (Scott and Harrison, 1987; Trombleyand Shepherd, 1993). The chemosensitivity of ORNs in vivo varies in terms of their odorselectivity. Each cell responds to only a subset of odorants, and each may be broadly ornarrowly tuned depending on the size of this subset (Sicard and Holley, 1984). ORNs withsimilar chemosensitivities populate particular regions of the olfactory epithelium. A variety oftechniques has been used to examine this feature, including measurement of the transepithelialpotential (Ottoson, 1956; Daval et al. 1970; Mustaparta, 1971; Kauer and Moulton, 1974;MacKay-Sim et al. 1982), fluorescent monitoring of intracellular Ca2+ levels (Hirono et al.1992), optical measurement of membrane potential (Cinelli and Kauer, 1992; Kent andMozell, 1992) and detection of 2-deoxyglucose uptake (Lancet et al. 1981; Slotnick et al.1989). More recently, several laboratories have cloned putative odor receptors belonging tothe superfamily of seven-helix G-protein-coupled receptors (Buck and Axel, 1991; Nef et al.1992; Raming et al. 1993; Ngai et al. 1993b; Selbie et al. 1992). Receptor RNAs have beenmapped by in situ methods in olfactory epithelia (Nef et al. 1992; Ressler et al. 1993; Ngaiet al. 1993a; Strotmann et al. 1992); each receptor clone appears to be expressed in a subset ofjust a few per cent of the ORNs (Ressler et al. 1993), with the subsets partially overlappingone another. In catfish, the subsets are randomly distributed in the epithelium (Ngai et al.1993a,b), but in mice and rats the subsets form patterns with bilateral symmetry (Nef et al.1992; Ressler et al. 1993; Vassar et al. 1993) that are similar among individuals of a species.Drawing by J. A. Dionne and H. Eisthen from Tagaki (1971).

Schoonover, 1984). In fish, thresholds for bile acids are 10211 to 1025 mol l21 (Døvinget al. 1980), whereas those for amino acids appear to be in the nanomolar to hundreds ofmicromolar range (Hara, 1973, 1976; Belghaug and Døving, 1977; Caprio, 1978; Døvinget al. 1980). High concentrations of odorants can elicit various responses even from non-olfactory cells. At concentrations above 10–100 mmol l21, the commonly employedodorants citralva, b-ionone, octanol and ethyl vanillin stimulate adenylyl cyclase andcause dispersal of melanin when applied to the surface of non-neuronal melanophores(Lerner et al. 1988; Potenza and Lerner, 1992). Additionally, a variety of odorants atconcentrations above 10 mmol l21 have been shown to alter membrane fluidity of lipidvesicles and N18 neuroblastoma cells (Kashiwayanagi and Kurihara, 1984, 1985). It isnot clear whether odor concentrations above 10–100 mmol l21 are either normallyencountered by air-breathing animals or required for odor detection, since at roomtemperature the concentration of a saturated vapor of most odorants (vapor pressures<2 mmHg) is less than 100 mmol l21.

The cellular response to odors

Odors elicit responses in ORNs that alter neuronal firing to communicate the presenceof odor to the brain. Both inhibitory and excitatory odor responses have been observed inORNs from vertebrates (Gesteland et al. 1965; O’Connell and Mozell, 1969; Duchampet al. 1974; Revial et al. 1978; Getchell and Shepherd, 1978a,b; Sicard and Holley, 1984;Dionne, 1992) and invertebrates (Derby and Harpaz, 1988; Derby et al. 1988;McClintock and Ache, 1989a,b), but until recently the inhibitory responses weregenerally discounted (discussed in Getchell, 1986; McClintock and Ache, 1989b). It isonly with the use of modern recording methods and protocols that inhibitory odorresponses have been consistently detected. Most olfactory responses appear to beproduced by odor-elicited receptor potentials. Receptor potentials are induced by alteringthe membrane conductance for selected types of ion channels. Both depolarizing andhyperpolarizing receptor potentials have been observed, and both can produce excitationor inhibition depending on the resting state and the magnitude of the voltage change.Channels of more than one type are involved in odor transduction. In addition, some odorresponses appear to occur independently of any receptor potential.

The cyclic-nucleotide-gated receptor potential in vertebrates

In vertebrates, the odor-modulated conductance most commonly associated with anolfactory receptor potential is produced by the cyclic-nucleotide-gated (CNG) cationchannel. These channels, first identified in toad (Bufo marinus) ORNs (Nakamura andGold, 1987), have been widely studied in salamanders, frogs and other species (seebelow). Nevertheless, odor modulation of the CNG cation channel has been demonstratedonly in post-metamorphic amphibians, and the role of the channel in producing a receptorpotential has been clarified only recently. The CNG cation channel in ORNs from thetiger salamander (Ambystoma tigrinum) can be activated by an odor mixture containingamyl acetate, acetophenone and cineole (Firestein et al. 1991a,b, 1993) with or without 2-hexylpyridine (Lowe and Gold, 1991). The cells respond with a depolarizing receptor


potential to high micromolar (Firestein and Werblin, 1989; Firestein et al. 1990) ormillimolar (Lowe and Gold, 1991) concentrations of the mixture preferentially applied totheir cilia, corresponding to the location of the CNG channels. Fifty to sixty per cent ofthe cells respond to these stimuli, a surprisingly high value in the light of the smallpercentage of cells that appear to harbor each putative odor receptor, but fewer cellsrespond to the individual components of the mixture (Firestein et al. 1993). Similarly, inORNs isolated from newts (Cynops pyrrhogaster), cyclic nucleotides activate a cationconductance (Kurahashi, 1990) found primarily in the cilia (Kurahashi, 1989; Kurahashiand Kaneko, 1991); the conductance shows little selectivity among the monovalent,inorganic cations (Li+, Na+, K+, Rb+, Cs+) (Kurahashi, 1990). A similarly non-selectivecation conductance can be activated in about 30 % of the cells by 10 mmol l21 amylacetate (Kurahashi and Shibuya, 1990; Kurahashi, 1989) or a mixture containing severalmillimolar each of amyl acetate, isoamyl acetate and limonene (Kurahashi and Yau,1993). Although it is a concern that many of these studies employed very highconcentrations of odorant, the results otherwise indicate that these were transduction-related olfactory responses.

Whereas the CNG cation channel is about equally permeable to the small monovalentcations, it is only slightly permeable to divalent cations such as Mg2+ and Ca2+

(Kurahashi, 1990; Kolesnikov et al. 1990; Frings et al. 1992; Kolesnikov andKosolapov, 1993). Physiological concentrations of Ca2+ partially block the channel(Nakamura and Gold, 1987; Kolesnikov et al. 1990; Kurahashi, 1990; Suzuki, 1990;Dhallan et al. 1990; Zufall et al. 1991; Frings et al. 1991; Kramer and Siegelbaum,1992; Zufall and Firestein, 1993; Kolesnikov and Kosolapov, 1993). Despite thisblockade, the Ca2+ influx through the CNG cation channel is sufficient to stimulate aCa2+-activated Cl2 conductance that can dominate the cyclic-nucleotide-stimulatedresponse in frog, salamander, newt and rat ORNs (Kurahashi and Yau, 1993; Kleene,1993a; Lowe and Gold, 1993a). The increase in Cl2 conductance, depending onintracellular buffering of Ca2+, appears to be responsible for half (Kurahashi and Yau,1993) or more (Kleene, 1993a; Lowe and Gold, 1993a) of the CNG receptor potential.Studies conducted before the Cl2 dependence of the CNG receptor potential wasrecognized have characterized the potential as a depolarizing response produced by anincrease in a non-selective cation conductance; however, those studies typicallyemployed cation and Cl2 distributions both with equilibrium potentials near 0 mV(Nakamura and Gold, 1987; Firestein and Werblin, 1989; Firestein et al. 1990) orinadvertently obstructed the Cl2 current by omitting Ca2+ from the bath (Kurahashi,1989). The normal intracellular concentration of Cl2 in ORNs is still uncertain for mostspecies including tiger salamanders, newts, frogs and rats. Only in dissociated ORNsfrom the aquatic salamander Necturus maculosus has the Cl2 equilibrium potentialbeen estimated. Its value (245.5±2.5 mV, N=11) (Dubin and Dionne, 1994)corresponds to an intracellular [Cl2] of 23.3±2.5 mmol l21. If this value is typical ofORNs in other vertebrates, a moderate increase in Cl2 conductance would indeed causea depolarizing receptor potential that was excitatory; however, a large conductanceincrease could depolarize the cell enough to inactivate Na+ channels and inhibitelectrical activity. The apparent dependence of the vertebrate CNG receptor potential on


a Ca2+-activated Cl2 conductance and the lack of data on the Cl2 equilibrium potentialwarrant a careful reassessment of its physiological role.

Aside from tiger salamanders and newts, in most vertebrates the data implicating theCNG cation channel in odor transduction are incomplete. CNG cation channels have beenfound in ORNs from rat (Frings et al. 1992; Lowe and Gold, 1993a; Lynch andLindemann, 1994), catfish (Miyamoto et al. 1992b), frog (Suzuki, 1990; Frings andLindemann, 1991; Kolesnikov et al. 1990; Frings et al. 1992; Kleene, 1993b) and carp(Kolesnikov and Kosolapov, 1993) tissues, but these channels have not been shown to beactivated by odors. While the precise subunit composition of the CNG cation channel isstill unknown, a subunit has been cloned from rat (Dhallan et al. 1990), catfish (Gouldinget al. 1992) and bovine (Ludwig et al. 1990) sources. When expressed in Xenopus oocytes(Altenhofen et al. 1991; Goulding et al. 1992) or 293 human embryonic kidney cells(Dhallan et al. 1990), this subunit forms functional channels with properties nearly, butnot completely, identical to those of the native channel. In rod photoreceptors, where ahomologous channel resides (Kaupp, 1991), at least one additional subunit of the CNGchannel has been identified (Chen et al. 1993), suggesting that additional subunits of theolfactory channels may also exist. Fig. 2A illustrates the odor-modulated conductances invertebrates.

Excitatory and inhibitory receptor potentials in lobsters

An odor-evoked, cyclic-nucleotide-responsive conductance has been observed inORNs from lobsters (Panulirus argus); but, in contrast to that in vertebrates, theconductance is K+-selective (Michel et al. 1991; Michel and Ache, 1992), and whether itis directly gated by cyclic nucleotides or modulated by enzymatic action is uncertain. Inlobster ORNs, the cyclic-nucleotide-responsive K+ conductance can be activated by odormixtures and single amino acids applied apically at micromolar concentrations; incultured neurons, the responses are dose-dependent with thresholds of 10–100 nmol l21

(Fadool et al. 1993). The conductance can be activated by an individual odorant in but afraction of odor-sensitive cells. Activation produces a hyperpolarizing receptor potentialthat is inhibitory (Michel et al. 1991). Interestingly, the cyclic-nucleotide-responsive K+

conductance was detected in only about 50 % of cells tested with membrane-permeableanalogs of cyclic adenosine 59-monophosphate (cyclic AMP) or cyclic guanosine 59-monophosphate (cyclic GMP), forskolin or the phosphodiesterase inhibitor isobutylmethylxanthine (IBMX) (Michel and Ache, 1992), suggesting substantial heterogeneityamong receptor cells.

Lobster ORNs also respond to odors with depolarizing receptor potentials that are


Fig. 2. Diverse odor-modulated conductances characterize olfactory transduction in severalspecies. The examples described in the text are illustrated in these three panels. Each paneldepicts the membrane of a cilium or outer dendrite of an ORN with the odor-modulatedconductances illustrated as ion channels. An intracellular mediator or second messenger,when known, is shown for each channel along with a symbol to indicate activation (+) orinhibition (2) of channel activity. The cellular responses elicited by each odor-modulatedconductance are indicated beside each figure. NSC, non-selective cation; cAMP, cyclic AMP;InsP3, inositol 1,4,5-trisphosphate.


B

C

A

Increase[NSC+]

Increase Ca2+-activated Cl−conductance

Depolarization

cAMP

Ca2+

Cl−

InsP3 ? Ca2+

Cations

Increase [K+]

Hyperpolarization

InsP3

+

+Cations

K+

cAMP+

Ca2+Increase [Ca2+]

Increase[NSC+]

Depolarization

Depolarization

Decrease[Cl−]

Hyperpolarization

Cl−−

??

+

+ ?

K+

?−

Decrease[K+]

Increase[NSC+]

Increase[Cl−]

Cations

Tiger salamander, newt, frog, rat

Lobster

Mudpuppy

+

+

Fig. 2

excitatory (Schmiedel-Jakob et al. 1989, 1990), and these responses are mimicked bycultured lobster ORNs (Fadool et al. 1991, 1993; Fadool and Ache, 1992). Like theinhibitory responses, these responses are elicited by complex odor mixtures as well as bysingle amino acids applied to the apical dendrite (Schmiedel-Jakob et al. 1989, 1990), butsingle-channel measurements indicate that the responses are produced by two differentchannel types, a non-selective cation channel and a Ca2+ channel, both of which aredirectly gated by intracellular inositol 1,4,5-trisphosphate (Ins1,4,5P3) (Fadool and Ache,1992; Ache et al. 1993; Ache, 1994). In addition, there is a third type of channel that isdirectly gated by inositol 1,3,4,5-tetrakisphosphate (Ins1,3,4,5P4) that also depolarizesthe cell and can interact in a reciprocal fashion with the Ins1,4,5P3-gated channels eitherto potentiate their activity or to be inhibited by them (Fadool and Ache, 1994).

Although the pathways for excitatory and inhibitory odor responses in lobster ORNsare separate, they occur in the same cell (McClintock and Ache, 1989b; Michel et al.1991; Michel and Ache, 1992; Fadool et al. 1993). Individual neurons can be excited byone odorant and inhibited by another. In addition, the same single odorant can elicitexcitatory and inhibitory responses from different ORNs (Michel et al. 1991; Michel andAche, 1992), and some odorants can elevate both cyclic AMP and Ins1,4,5P3

concentrations (Ache et al. 1993). Thus, neither the receptor neurons nor the odorants canbe subdivided into excitatory and inhibitory types. Such diversity of odor responsessuggests that there may be more than one receptor type for each odorant or that specificreceptors may couple (possibly through different G-proteins) to different transductionpathways in different cells. Fig. 2B illustrates the transduction pathways in lobster ORNs.

Receptor potentials and ‘silent’ responses in Necturus

The multiplicity of odor-elicited responses seen in lobsters is not unique toinvertebrates. Similarly diverse responses were first described using extracellularrecording methods in frogs (Gesteland et al. 1965; O’Connell and Mozell, 1969; Revialet al. 1978) and tiger salamanders (Getchell and Shepherd, 1978a; Sicard and Holley,1984). In ORNs from Necturus maculosus (the mudpuppy), odors can elicit bothdepolarizing and hyperpolarizing receptor potentials with varying ionic dependencies,and odors can also alter neuronal excitability without producing a detectable receptorpotential (Dionne, 1992; Dubin and Dionne, 1993, 1994) (Fig. 2C). The odor-elicitedreceptor potentials satisfy the four criteria for transduction: only a small fraction ofchemosensitive cells appears to respond to any particular odorant; responses are inducedat odor concentrations as low as 10–100 nmol l21; responses are elicited preferentially toodors applied on the ciliated end of the cell (Dionne, 1992); and neuronal excitability isexcited by depolarizing receptor potentials of modest size, but inhibited by largedepolarizing potentials and by hyperpolarizing receptor potentials (Dubin and Dionne,1993). The ionic bases of the receptor potentials in mudpuppy are complex: odorants canincrease or decrease a Cl2 conductance; increase a non-selective cation conductance; ordecrease a specific K+ conductance (Dubin and Dionne, 1993). The same odorant canelicit each of these conductance changes in different cells, and different odorants cancause different types of response in the same cell.

Olfactory responses seen as a change in excitability not accompanied by a receptor


potential or a change of input impedance have been termed ‘silent responses’ (Dionne,1992; Dubin and Dionne, 1994). Silent responses were detected in Necturus ORNs aschanges in the rate of spontaneous firing and as a change in the frequency or duration offiring when a neuron was driven with depolarizing current pulses; the frequency may bemore than doubled or reduced to less than half. The responses appear to satisfy the criteriafor transduction, although the preferential association with apical membrane has not yetbeen tested. Silent odor responses have been detected in cells that also responded todifferent odorants with a receptor potential (Dionne, 1992), suggesting that individualORNs may have several odor transduction pathways. Although the mechanisms thatproduce silent responses have not been fully characterized, they may involve either smalllocal conductance changes that are not detected in whole-cell measurements or changesin the gating properties of specific types of Na+, K+ or Ca2+ channels. In mudpuppyORNs, action potentials can be generated in the dendrite, conferring upon the cell anenhanced sensitivity to small odor-modulated conductances (Dubin and Dionne, 1994).

All chemosensitivity in isolated mudpuppy ORNs washes out rapidly in the normalwhole-cell recording mode (Dionne, 1992; Dubin and Dionne, 1993). This loss ofresponsiveness includes odor-elicited receptor potentials as well as silent responses andsupports the idea that soluble constituents in the cytoplasm constitute part of eachtransduction pathway. Although the intracellular mediators of the odor responses inNecturus have not yet been identified, increases in the Cl2 and cation conductances inthese cells can be elicited by cyclic AMP (A. E. Dubin and V. E. Dionne, unpublishedobservations). Fig. 2C illustrates the transduction pathways in Necturus ORNs.

Diverse odor responses in other species

Catfish

In 20 % of isolated ORNs tested from channel catfish (Ictalurus punctatus), a mixtureof L-arginine, L-alanine and L-norleucine with or without L-glutamate (100 mmol l21

each) activated receptor currents of three different types, based upon voltage- and ion-dependence, time course, reversal potential and sensitivity to amiloride (Miyamoto et al.1992b). In about 40 % of the chemosensitive cells, the response occurred with a 50 mslatency, was reversibly decreased by removal of extracellular Ca2+, and reversed polarityat +23 mV under ionic conditions where the equilibrium potentials for Cl2 (ECl) and non-specific cation conductances (ENSC) were both about 0 mV. In another 30 % of the cells, ashort-latency outwardly rectifying response reversed polarity near 0 mV, was insensitiveto amiloride, but was blocked by forskolin; inward, but not outward, currents wereblocked by extracellular Ba2+. In the remaining 30 % of chemosensitive cells, a short-latency response was amiloride-sensitive and reversed polarity at 215 mV. Miyamotoet al. (1992b) suggested that at least two different conductances in different populationsof cells were activated by the odorants, but the ionic bases of the responses were not fullycharacterized. In a separate study, internal perfusion with either Ins1,4,5P3 or cyclic AMPcaused a transient depolarization of the receptor neurons (Miyamoto et al. 1992a); theresponses were sustained when extracellular Ca2+ was removed or the intracellular Ca2+

buffering capacity was elevated, but otherwise the responses had dissimilarpharmacologies. The authors suggested that Ca2+ influx through Ins1,4,5P3- or cyclic-


AMP-induced conductances curtailed the response by activating a Ca2+-dependent K+

conductance. Ca2+-activated K+ currents have been described in ORNs from catfish(Miyamoto et al. 1992b) and other species (Trotier, 1986; Firestein and Werblin, 1987;Maue and Dionne, 1987; Schild, 1989; Suzuki, 1990; Nevitt and Moody, 1992;McClintock and Ache, 1989a), and L-alanine-stimulated Ins1,4,5P3 production in catfisholfactory cilia is GTP-dependent (Restrepo et al. 1993).

Toad

Excitatory and inhibitory odor responses have been observed in toad (Caudiverberacaudiverbera) ORNs and reported briefly (Bacigalupo et al. 1993). An odor mixturecontaining isovaleric acid, triethylamine and pyrazine elicited an outward(hyperpolarizing) current in ORNs, causing inhibition, whereas a mixture containingcitralva, citronellal and geraniol caused an inward (depolarizing) current and excitation.Three groups of cells were described, one showing the inhibitory response only, one theexcitatory response only, and the other showing a mixed response. In membranes from ratolfactory tissue, the odorants in the inhibitory mixture stimulate the production ofIns1,4,5P3 whereas those in the excitatory mixture stimulate cyclic AMP (Boekhoff et al.1990).

Squid

Various chemical stimuli applied to the olfactory organ of squid (Loligo opalescens)elicit an escape response (Gilly and Lucero, 1992); these responses can be produced bytwo different and opposing effects on the membrane conductance. Several naturallyoccurring compounds, including squid ink, L-Dopa and dopamine, activate the escaperesponse in the animal and reversibly increase the membrane conductance (possibly forK+ or Cl2) in isolated ORNs, causing the cells to hyperpolarize and inhibiting firing.Other compounds that are not natural odorants, including propyl paraben,tetrabutylammonium, tetraethylammonium, 4-aminopyridine and methadone, also elicitthe escape response, but at the same concentrations at which they block K+ channels inthe neurons, causing them to depolarize and increasing their excitability (Lucero et al.1992). Chemosensitivity was detected using the perforated patch method (Horn andMarty, 1988), but was not seen under normal whole-cell recording conditions.

Mechanisms underlying odor transduction

As summarized above, odor transduction utilizes several different pathways, more thanone of which may be found in each ORN, but a single odorant appears to activate only onetransduction pathway in any cell. In some cases, it is clear that odor-elicited effects onneuronal excitability are mediated by second messengers. Is this always the case? Sincenumerous compounds with diverse physico-chemical properties are perceived asodorants, it is entirely possible that more than one mechanism is involved in odortransduction. In this section, after evaluating three classes of mechanisms that have beenproposed to underlie odor transduction, we conclude that the known transductionpathways belong to the class of receptor-coupled, indirect mechanisms.


Direct gating of ion channels

Some odors may directly gate ion channels, acting like various neurotransmitters.Several studies employing reconstitution into artificial membranes reported direct, odor-induced activation of single channels (Vodyanoy and Murphy, 1983; Labarca et al.1988), and in lobster ORNs histamine appears to gate a somatic Cl2 channel directly(McClintock and Ache, 1989c). However, the data fail to demonstrate that odors aretransduced by a direct gating mechanism.

Vodyanoy and Murphy (1983) reconstituted membrane vesicles prepared from a crudehomogenate of rat olfactory epithelium into phosphatidylcholine planar bilayers;nanomolar concentrations of diethyl sulfide or (2)-carvone, but not a saturated aqueoussolution of camphor, activated a K+-selective channel with particularly slow kinetics: theodorant was reported nearly to double the unstimulated mean open time (29 s) of thechannel. Although the response seemed to be specific for the olfactory epithelium, since itwas not observed with vesicles prepared from respiratory epithelia, no similar channel orresponse has been reported in rat ORNs. Later studies reported that diethyl sulfideactivated a membrane conductance only in the presence of ATP and GTP, and that cyclicAMP alone induced a similar but smaller response, suggesting an indirect effect of odoron membrane conductance (Vodyanoy and Vodyanoy, 1987). Using vesicles preparedfrom cilia-enriched membranes of bullfrog (Rana catesbeiana) olfactory epithelia,Labarca et al. (1988) found that nanomolar concentrations of isobutyl methoxy-pyrazine(bell pepper odor) and citralva activated a multistate channel of indeterminate ionselectivity (symmetrical solutions were used). Similar channels were not observed withvesicles from cilia-enriched preparations of respiratory epithelia, but since vesicles wereobtained from crude membrane preparations or preparations partially enriched in cilia,and since ORNs are not the only ciliated cells in these epithelia, the origins of therelatively few channels that were examined are unknown. Furthermore, there is aprecedent for channel properties being altered by reconstitution (Miller, 1987), raising thepossibility that the responses observed in vitro might not occur in vivo. Finally, it cannotbe ruled out that the reconstituted vesicles did not include other proteins that indirectlymediated the response between odor and channel; however, odor-induced effects in theabsence of exogenous ATP, GTP and other second messengers were observed (Labarcaet al. 1988).

McClintock and Ache (1989c) found ion channels on the somata of lobster ORNs thatappeared to be directly gated by histamine. Since histamine applied to lobster ORNselicits a behavioral response, it was considered that the channels might mediate odortransduction. A careful search for histamine-gated channels on dendritic membranes oflobster ORNs revealed none, however (Bayer et al. 1989; McClintock and Ache,1989c), making it unlikely that the somatic-type channels are involved in odortransduction.

Membrane fluidity

Airborne odorants are volatile compounds that dissolve readily in nonpolarenvironments, including lipid membranes. Such compounds can alter both the membranefluidity and membrane potential in liposomes (Kashiwayanagi et al. 1990; Nomura and


Kurihara, 1987), in cultured cells (Kashiwayanagi and Kurihara, 1984, 1985) and indissociated olfactory epithelial cells (Kashiwayanagi et al. 1987). Changes in fluiditycould affect protein–lipid interactions and have been proposed as a mechanism for odortransduction (Nomura and Kurihara, 1989; Kurihara, 1990). Since these responses appearto be independent of specific ions (Shoji and Kurihara, 1991; Kashiwayanagi et al. 1991),it has been proposed that odors alter the membrane potential through an effect on thephase boundary potential rather than by modulating ion conductances. In ORNs, odor-induced changes in membrane fluidity have been reported with thresholds only in the1023 to 1025 mol l21 range (Kashiwayanagi et al. 1987). These values border the range ofodor detection thresholds and cannot account for olfactory transduction under mostconditions. The three criteria not involving odor thresholds (see above) have not beentested. Although odorous compounds can affect the fluidity of membranes, the data do notyet support the hypothesis that this is a mechanism that underlies odor transduction.

Receptor-mediated transduction

The preponderance of recent biochemical and electrophysiological data suggest thatodors modulate neuronal excitability indirectly through receptor proteins linked tosecond-messenger-dependent pathways. Several different types of evidence support thisconclusion.

(1) Washout of odor responses in whole-cell recordings. The whole-cell patch-clampmethod employs an electrode sealed tightly to the cell membrane, but allows freeexchange between the intracellular saline and the pipette solution (Hamill et al. 1981).The chemosensitivity of isolated neurons studied with this method is labile, appearing tofade or wash out, presumably because important intracellular constituents diffuse awayinto the pipette (Trotier, 1986; Maue and Dionne, 1987; Frings and Lindemann, 1988;Dionne, 1992). The loss of chemosensitivity is especially rapid when the recordingpipette is sealed to the dendrite close to the chemosensitive apical membrane, but can beretarded if resistive methods that reduce fluid exchange are used (Dubin and Dionne,1993). Slower washout is seen when the pipette is sealed to the soma relatively far fromthe apical surface, and studies recording from the somata of ORNs with especially longdendrites (e.g. from lobsters) often find no washout (Schmiedel-Jakob et al. 1989; Micheland Ache, 1992). Typically, neither MgATP alone nor in combination with GTP rescuescells from washout (Frings and Lindemann, 1988; Dionne, 1989), but cyclic GMP mayslow the rate of washout (Lischka and Schild, 1993).

(2) The latency between odor application and the cellular response. The response ofORNs to odors occurs with a latency of several hundred milliseconds in adult tigersalamanders (Firestein and Werblin, 1989; Firestein et al. 1990). This would be longenough to involve a multistep pathway in the transduction process, even in poikilothermssuch as these, where enzyme kinetics are likely to be slower than in mammals.

(3) Odor responses depend on G-protein function. G-proteins are heterotrimericproteins that mediate GTP-dependent responses elicited by bioactive ligands and whichact as intermediaries between ligand receptors and targets such as adenylyl cyclase andphospholipase C. A third or more of the known G-protein a-subunits have been detectedin rat olfactory epithelia (Jones and Reed, 1989; Jones et al. 1990; Mania-Farnell and


Farbman, 1990; Shinohara et al. 1992), as have the bg-subunits (Shinohara et al. 1992;Mania-Farnell and Farbman, 1990). Although GS-a expression occurs predominantly innon-neuronal support cells (Jones et al. 1988), GOLF-a expression occurs specifically inORNs and depends on receptor cell innervation of the olfactory bulb (Jones and Reed,1989). Odor-activated, GTP-dependent elevations of cyclic AMP (Pace et al. 1985; Sklaret al. 1986; Shirley et al. 1986; Breer et al. 1990; Boekhoff et al. 1990) and Ins1,4,5P3

(Huque and Bruch, 1986; Bruch and Kalinowski, 1987; Breer et al. 1990; Boekhoff et al.1990) concentrations indicate that G-proteins mediate the odor-elicited stimulation ofsecond-messenger systems. Pertussis and cholera toxins, which ADP-ribosylate differentbut specific G-a subunits and alter their activity, differentially and predictably, affect theodor-sensitive increases in cyclic AMP and Ins1,4,5P3 concentrations in rat olfactorymembranes (Boekhoff et al. 1990). In addition, in cilia-enriched membranes from catfisholfactory tissue, GTP analogues modulate the binding affinities of receptors for L-alanineand L-arginine in a manner suggesting that G-proteins couple to these receptors (Bruchand Kalinowski, 1987). Furthermore the odor-sensitive current is predictably altered bynon-hydrolyzable guanosine nucleotide analogs in tiger salamander ORNs (Firesteinet al. 1991a).

(4) Intracellular compounds and second messengers. The diffusible secondmessengers cyclic AMP, Ins1,4,5P3 and Ca2+ act directly on specific types of ionchannels in ORNs, thereby altering membrane excitability and providing several putativepathways for olfactory transduction. Odor-induced effects on cyclic-nucleotide-sensitivechannels can be mimicked by application of exogenous compounds (Kurahashi, 1990;Firestein et al. 1991a) or by manipulation of endogenous second messenger levels usingan adenylyl cyclase activator (forskolin) or a phosphodiesterase inhibitor (IBMX) (Fringsand Lindemann, 1991; Firestein et al. 1991b; Michel and Ache, 1992). Ins1,4,5P3

directly activates cation channels in lobsters and rats (Fadool and Ache, 1992; Restrepoet al. 1990; Restrepo and Boyle, 1991). Ins1,4,5P3-sensitive channels in excised patchesfrom cultured lobster ORNs are activated when the patch is crammed into the cytoplasmof a different cell and food odor is applied (Fadool and Ache, 1992). Ca2+ activates Cl2

(Kleene and Gesteland, 1991) and K+ (Maue and Dionne, 1987; Firestein and Werblin,1987; Nevitt and Moody, 1992) channels. Local odorant-induced increases inintracellular [Ca2+] at the apical end of intact cultured rat ORNs have been observed usingCa2+ imaging techniques (Restrepo et al. 1993). Second messengers may also alter ionchannel gating indirectly by modulating enzyme activity (Reed, 1992; Wegener et al.1993). Other compounds, including cyclic GMP, diacyl glycerol, Ins1,3,4,5P4, NO andCO, have also been implicated in olfactory transduction (Zufall and Hatt, 1991; Ronnettand Snyder, 1992; Breer et al. 1992; Breer and Shepherd, 1993; Lischka and Schild,1993; Verma et al. 1993; Fadool and Ache, 1994). NO and CO are membrane-permeableand may provide an intercellular communication pathway.

(5) G-protein-coupled receptors. Since odor modulation of second messenger levels isG-protein-dependent, it was widely postulated that odor receptors might belong to theseven-helix family of G-protein-coupled receptors (see Lancet and Ben-Arie, 1993; Nef,1993, for recent reviews). Guided by this idea, Buck and Axel (1991) cloned 18 differentseven-helix receptors from rat olfactory tissue. They proposed that these proteins were


odor receptors, and estimated from screening a genomic library that there may behundreds of structurally similar receptors in olfactory tissue. Others (Selbie et al. 1992;Raming et al. 1993; Ngai et al. 1993b) have extended the cloning to mice, humans, dogsand fish, producing sequences for over 150 different receptors. Despite concerted effortsto demonstrate function of the receptor proteins by expressing them in host cells, only onemember of the family has shown any odor-elicited activity. This member was a ratreceptor protein expressed in an insect cell line; membranes prepared from transfectedcells and stimulated with a mixture of lyral and lilial showed a 1.5- to twofold increase inIns1,4,5P3 levels with a response threshold between 10 and 100 nmol l21 (Raming et al.1993). Whether a similar response is mediated by this protein in vivo is unknown.Although we still do not know whether any, some or all of these receptor proteins areinvolved directly in odor transduction, antibodies to BARK-2 (but not BARK-1), a kinasethat specifically phosphorylates the b-adrenergic G-protein-coupled receptor, prevent therapid desensitization of odor-induced signals in preparations of rat olfactory cilia(Schleicher et al. 1993).

(6) Non-linear summation of odorant and cyclic AMP responses. Non-linear additivityof an odor response and the effect of exogenous second messenger has been observed intiger salamander ORNs with an odor mixture (cineole, isoamyl acetate and 2-hexylpyridine or b-ionone, (2)-methone, amyl salicylate, isoamyl acetate and 2-hexylpyridine) applied focally when cyclic AMP was released intracellularly byphotolysis (Lowe and Gold, 1993a,b). The results resemble the non-linear responsesinduced when two identical odor applications are made in rapid succession (Firesteinet al. 1991a). This is consistent with the hypothesis that the odor response is mediated bythe second messenger, but it is important to note that non-linear summation would beexpected regardless of how the adenylyl cyclase were activated, either by an olfactorytransduction pathway or otherwise by a non-transduction mechanism possibly involvinghigh concentrations of lipophilic odorant (see above).

Physiological implications of multiple transduction pathways

The data reviewed here support the hypothesis that odors are detected anddiscriminated when they activate one or more different transduction pathways in parallelin ORNs. Similar features can be seen in the visual (DeVries and Baylor, 1993), gustatory(Kinnamon and Cummings, 1992) and somatosensory systems (French, 1992) also,suggesting that parallel but different transduction pathways provide an efficient solutionfor processing complex sensory information. The following conclusions can be drawnabout odor transduction. (1) Odors are transduced in ORNs by modulating severaldifferent conductances in different cells from the same animal. In all species studied,single odorants have been shown to affect several components of the membraneconductance. (2) More than one odor-sensitive conductance can be found in some ORNs,and possibly in all. These conductances mediate excitatory as well as inhibitoryresponses. (3) A second messenger can modulate more than one type of ion channel inORNs. (4) Some odorants can modulate more than one second-messenger pathway inORNs, whereas others may affect just a single pathway. In either case, a single odorant


can have different effects on different cells by activating a pathway that diverges prior tothe effector ion channels. (5) Even among ORNs that are sensitive to a particularstimulus, not all cells respond with an increase in the concentration of a particular secondmessenger or by modulation of a particular type of ion channel. There are functionalsubtypes of ORNs.

The variety of conductances that odors modulate allows a varied and graded effect onneuronal excitability; if the effect were simply to increase or decrease excitability, thevariety of transduction pathways would be redundant. Increases as well as decreases inexcitability are caused by hyperpolarizing and depolarizing receptor potentials, byincreases and decreases in the resting conductance and, possibly, by modulation of thekinetics of voltage-activated ion channels. Odors are discriminated by extraction ofcertain relevant parameters from the pattern of odor-elicited activity in peripheral ORNs(Gesteland et al. 1965; Giradot and Derby, 1990; Shirley, 1992). Since odor recognitionprobably requires a comparison of responses across the population of ORNs, increasingthe potential diversity of output from individual cells by allowing graded excitatory andinhibitory responses could sharpen and enhance pattern differences. The pattern ofactivity elicited by a single odor should change when it is presented as one component ofa mixture, since parallel signalling pathways in individual ORNs and cross-talk betweenthe pathways can occur. Determining how the olfactory system discriminates individualodors under these conditions will require an understanding of the critical parametersextracted by higher-order neurons. Together with an emerging appreciation of the overlapof neuronal responses in ORNs in terms of specific odor receptors and convergingtransduction pathways, we may finally account for the chemosensitive abilities of theolfactory system.

ReferencesACHE, B. W. (1994). Towards a common strategy for transducing olfactory information. Seminars Cell

Biol. 5, 55–63.ACHE, B. W., HATT, H., BREER, H., BOEKHOFF, I. AND ZUFALL, F. (1993). Biochemical and physiological

evidence for dual transduction pathways in lobster olfactory receptor neurons. Chem. Senses 18, 523.ALTENHOFEN, W., LUDWIG, J., EISMANN, E., KRAUS, W., BONIGK, W. AND KAUPP, U. B. (1991). Control

of ligand specificity in cyclic nucleotide-gated channels from rod photoreceptors and olfactoryepithelium. Proc. natn. Acad. Sci. U.S.A. 88, 9868–9872.

AMOORE, J. E. (1982). Odor theory and classification. In Fragrance Chemistry (ed. E. T. Theimer), pp.28–76. New York: Academic Press.

ANHOLT, R. R. H. (1993). Molecular neurobiology of olfaction. Crit. Rev. Neurobiol. 7, 1–22.BACIGALUPO, J., MORALES, B., UGARTE, G., DELGADO, R., JORQUERA, O. AND LABARCA, P. (1993).

Electrophysiological studies in toad olfactory receptor neurons. Chem. Senses 18, 525.BAYER, T. A., MCCLINTOCK, T. S., GRUNERT, U. AND ACHE, B. W. (1989). Histamine-induced

modulation of olfactory receptor neurones in two species of lobster, Panulirus argus and Homarusamericanus. J. exp. Biol. 145, 133–146.

BELGHAUG, R. AND DØVING, K. B. (1977). Odour threshold determined by studies of the induced wavesin the olfactory bulb of the char (Salmo alpinus L.). Comp. Biochem. Physiol. 57A, 327–333.

BOEKHOFF, I., TAREILUS, E., STROTMANN, J. AND BREER, H. (1990). Rapid activation of alternativesecond messenger pathways in olfactory cilia from rats by different odorants. EMBO J. 9, 2453–2458.

BREER, H., BOEKHOFF, I. AND TAREILUS, E. (1990). Rapid kinetics of second messenger formation inolfactory transduction. Nature 345, 65–68.

BREER, H., KLEMM, T. AND BOEKHOFF, I. (1992). Nitric oxide mediated formation of cyclic GMP in theolfactory system. Neuroreport 3, 1030–1032.


BREER, H. AND SHEPHERD, G. M. (1993). Implications of the NO/cGMP system for olfaction. TrendsNeurosci. 16, 5–9.

BRUCH, R. C. AND KALINOWSKI, D. L. (1987). Interaction of GTP-binding regulatory proteins withchemosensory receptors. J. biol. Chem. 262, 2401–2404.

BUCK, L. AND AXEL, R. (1991). A novel multigene family may encode odorant receptors: a molecularbasis for odor recognition. Cell 65, 175–187.

CAPRIO, J. (1978). Olfaction and taste in the channel catfish: an electrophysiological study of theresponses to amino acids and derivatives. J. comp. Physiol. 123, 357–371.

CHEN, T. Y., PENG, Y. W., DHALLAN, R. S., AHAMED, B., REED, R. R. AND YAU, K. W. (1993). A newsubunit of the cyclic nucleotide-gated cation channel in retinal rods. Nature 362, 764–767.

CINELLI, A. R. AND KAUER, J. S. (1992). Voltage-sensitive dyes and functional activity in the olfactorypathway. A. Rev. Neurosci. 15, 321–351.

DAVAL, G., LEVETEAU, J. AND MACLEOD, P. (1970). EOG local et discrimination olfactive chez lagrenouille. J.Physiol., Paris 62, 477–488.

DERBY, C. D., GIRARDOT, M.-N. AND HARPAZ, S. (1988). Inhibition in the periphery: occurrence inolfaction and gustatory receptor cells of aquatic crustaceans, correlation with mixture suppression andeffect on quality coding. Chem. Senses 13, 683–684.

DERBY, C. D. AND HARPAZ, S. (1988). Physiology of chemoreceptor cells in the legs of the freshwaterprawn, Macrobrachium rosenbergii. Comp. Biochem. Physiol. A 90, 85–91.

DEVRIES, S. H. AND BAYLOR, D. A. (1993). Synaptic circuitry of the retina and olfactory bulb. Neuron 10(Suppl.),139–149.

DHALLAN, R. S., YAU, K. W., SCHRADER, K. A. AND REED, R. R. (1990). Primary structure and functionalexpression of a cyclic nucleotide-activated channel from olfactory neurons. Nature 347, 184–187.

DIONNE, V. E. (1989). Odor detection and discrimination: can isolated olfactory receptor neurons smell?In Chemical Senses (ed. J. G. Brand, J. H. Teeter, R. H. Cagan and M. R. Kare), pp. 415–426. NewYork, Basel: Marcel Dekker, Inc.

DIONNE, V. E. (1992). Chemosensory responses in isolated olfactory receptor neurons from Necturusmaculosus. J. gen. Physiol. 99, 415–433.

DØVING, K. B., SELSET, R. AND THOMMESEN, G. (1980). Olfactory sensitivity to bile acids in salmonidfishes. Acta physiol. scand. 108, 123–131.

DUBIN, A. E. AND DIONNE, V. E. (1993). Modulation of Cl2, K+ and nonselective cation conductances bytaurine in olfactory receptor neurons of the mudpuppy Necturus maculosus. J. gen. Physiol. 101,469–485.

DUBIN, A. E. AND DIONNE, V. E. (1994). Action potentials and chemosensitive conductances in thedendrites of olfactory neurons suggest new features for odor transduction. J. gen. Physiol. 103,181–201.

DUCHAMP, A., REVIAL, M. F., HOLLEY, A. AND MACLEOD, P. (1974). Odor discrimination by frogolfactory receptors. Chem. Sens. Flav. 1, 213–233.

FADOOL, D. A. AND ACHE, B. W. (1992). Plasma membrane inositol 1,4,5-trisphosphate-activatedchannels mediate signal transduction in lobster olfactory receptor neurons. Neuron 9, 907–918.

FADOOL, D. A. AND ACHE, B. W. (1994). Ligand and ion specificity of IP3-gated channels in lobsterolfactory receptor cells. In Proceedings of the International Symposium on Olfaction and Taste XI(ed. K. Kurihara, H. Ogawa and N. Suzuki). Tokyo: Springer-Verlag (in press).

FADOOL, D. A., MICHEL, W. C. AND ACHE, B. W. (1991). Sustained primary culture of lobster (Panulirusargus) olfactory receptor neurons. Tissue & Cell 23, 719–732.

FADOOL, D. A., MICHEL, W. C. AND ACHE, B. W. (1993). Odor sensitivity of cultured lobster olfactoryreceptor neurons is not dependent on process formation. J. exp. Biol. 174, 215–233.

FARBMAN, A. I. (1992). Cell Biology of Olfaction. Cambridge: Cambridge University Press.FIRESTEIN, S., DARROW, B. AND SHEPHERD, G. M. (1991a). Activation of the sensory current in

salamander olfactory receptor neurons depends on a G protein-mediated cyclic AMP secondmessenger system. Neuron 6, 825–835.

FIRESTEIN, S., PICCO, C. AND MENINI, A. (1993). The relation between stimulus and response in olfactoryreceptor cells of the tiger salamander. J. Physiol., Lond. 468, 1–10.

FIRESTEIN, S., SHEPHERD, G. M. AND WERBLIN, F. S. (1990). Time course of the membrane currentunderlying sensory transduction in salamander olfactory receptor neurones. J. Physiol., Lond. 430,135–158.


FIRESTEIN, S. AND WERBLIN, F. S. (1987). Gated currents in isolated olfactory receptor neurons of thelarval tiger salamander. Proc. natn. Acad. Sci. U.S.A. 84, 6292–6296.

FIRESTEIN, S. AND WERBLIN, F. S. (1989). Odor-induced membrane currents in vertebrate-olfactoryreceptor neurons. Science 244, 79–82.

FIRESTEIN, S., ZUFALL, F. AND SHEPHERD, G. M. (1991b). Single odor-sensitive channels in olfactoryreceptor neurons are also gated by cyclic nucleotides. J. Neurosci. 11, 3565–3572.

FRENCH, A. S. (1992). Mechanotransduction. A. Rev. Physiol. 54, 135–152.FRINGS, S., BENZ, S. AND LINDEMANN, B. (1991). Current recording from sensory cilia of olfactory

receptor cells in situ. II. Role of mucosal Na+, K+ and Ca2+ ions. J. gen. Physiol. 97, 725–747.FRINGS, S. AND LINDEMANN, B. (1988). Odorant response of isolated olfactory receptor cells is blocked

by amiloride. J. Membr. Biol. 105, 233–243.FRINGS, S. AND LINDEMANN, B. (1990). Single unit recording from olfactory cilia. Biophys. J. 57,

1091–1094.FRINGS, S. AND LINDEMANN, B. (1991). Current recording from sensory cilia of olfactory receptor cells in

situ. I. The neuronal response to cyclic nucleotides. J. gen. Physiol. 97, 1–16.FRINGS, S., LYNCH, J. W. AND LINDEMANN, B. (1992). Properties of cyclic nucleotide-gated channels

mediating olfactory transduction. Activation, selectivity and blockage. J. gen. Physiol. 100, 45–67.GESTELAND, R. C., LETTVIN, J. Y. AND PITTS, W. H. (1965). Chemical transmission in the nose of the

frog. J. Physiol., Lond. 181, 525–559.GETCHELL, T. V. (1974). Unitary responses in frog olfactory epithelium to sterically related molecules at

low concentrations. J. gen. Physiol. 64, 241–261.GETCHELL, T. V. (1986). Functional properties of vertebrate olfactory receptor neurons. Physiol. Rev. 66,

772–818.GETCHELL, T. V. AND SHEPHERD, G. M. (1978a). Responses of olfactory receptor cells to step pulses of

odour at different concentrations in the salamander. J. Physiol., Lond. 282, 521–540.GETCHELL, T. V. AND SHEPHERD, G. M. (1978b). Adaptive properties of olfactory receptors analysed

with odour pulses of varying durations. J. Physiol., Lond. 282, 541–560.GILLY, W. F. AND LUCERO, M. T. (1992). Behavioral responses to chemical stimulation of the olfactory

organ in the squid Loligo opalescens. J. exp. Biol. 162, 209–229.GIRADOT, M. N. AND DERBY, C. D. (1990). Peripheral mechanisms of olfactory discrimination of

complex mixtures by the spiny lobster: no cell types for mixtures but different contribution of the cellsto the across neuron patterns. Brain Res. 513, 225–236.

GOULDING, E. H., NGAI, J., KRAMER, R. H., COLICOS, S., AXEL, R., SIEGELBAUM, S. A. AND CHESS, A.(1992). Molecular cloning and single-channel properties of the cyclic nucleotide-gated channel fromcatfish olfactory neurons. Neuron 8, 45–58.

HAMILL, O. P., MARTY, A., NEHER, E., SAKMANN, B. AND SIGWORTH, F. S. (1981). Improved patch-clamptechniques for high resolution current recording from cells and cell-free membrane patches. PflügersArch. 391, 85–100.

HARA, T. J. (1973). Olfactory responses to amino acids in rainbow trout, Salmo gairdneri. Comp.Biochem. Physiol. 44A, 407–416.

HARA, T. J. (1976). Structure–activity relationships of amino acids in fish olfaction. Comp. Biochem.Physiol. 54A, 31–36.

HIRONO, J., SATO, T., TONOIKE, M. AND TAKEBAYASHI, M. (1992). Simultaneous recording of [Ca2+]iincreases in isolated olfactory receptor neurons retaining their original spatial relationship in intacttissue. J. Neurosci. Meth. 42, 185–194.

HORN, R. AND MARTY, A. (1988). Muscarinic activation of ionic currents by a new whole cell recordingmethod. J. gen. Physiol. 92, 145–159.

HUQUE, T. AND BRUCH, R. C. (1986). Odorant- and guanine nucleotide-stimulated phosphoinositideturnover in olfactory cilia. Biochem. biophys. Res. Commun. 137, 36–42.

JONES, D. T., BARBOSA, E. AND REED, R. R. (1988). Expression of G-protein alpha subunits in ratolfactory neuroepithelium: candidates for olfactory signal transduction. Cold Spring Harb. Symp.quant. Biol. 53, 349–353.

JONES, D. T., MASTERS, S. B., BOURNE, H. R. AND REED, R. R. (1990). Biochemical characterization ofthree stimulatory GTP-binding proteins. The large and small forms of Gs and the olfactory-specific G-protein, Golf. J. biol. Chem. 265, 2671–2676.

JONES, D. T. AND REED, R. R. (1989). Golf: an olfactory neuron specific-G protein involved in odorantsignal transduction. Science 244, 790–795.


KASHIWAYANAGI, M., HORIUCHI, M. AND KURIHARA, K. (1991). Differential ion dependence of frogolfactory responses to various odorants. Comp. Biochem. Physiol. A 100, 287–292.

KASHIWAYANAGI, M. AND KURIHARA, K. (1984). Neuroblastoma cell as model for olfactory cell:mechanism of depolarization in response to odorants. Brain Res. 293, 251–258.

KASHIWAYANAGI, M. AND KURIHARA, K. (1985). Evidence for nonreceptor odor discrimination usingneuroblastoma cells as a model for olfactory cells. Brain Res. 359, 97–103.

KASHIWAYANAGI, M., SAI, K. AND KURIHARA, K. (1987). Cell suspensions from porcine olfactorymucosa. Changes in membrane potential and membrane fluidity in response to various odors. J. gen.Physiol. 89, 443–457.

KASHIWAYANAGI, M., SUENAGA, A., ENOMOTO, S. AND KURIHARA, K. (1990). Membrane fluidity changesof liposomes in response to various odorants. Complexity of membrane composition and variety ofadsorption sites for odorants. Biophys. J. 58, 887–895.

KAUER, J. S. (1991). Contributions of topography and parallel processing to odor coding in the vertebrateolfactory pathway. Trends Neurosci. 14, 79–85.

KAUER, J. S. AND MOULTON, D. G. (1974). Responses of olfactory bulb neurones to odour stimulation ofsmall nasal areas in the salamander. J. Physiol., Lond. 243, 717–737.

KAUPP, U. B. (1991). The cyclic nucleotide-gated channels of vertebrate photoreceptors and olfactoryepithelium. Trends Neurosci. 14, 150–157.

KENT, P. F. AND MOZELL, M. M. (1992). The recording of odorant-induced mucosal activity patternswith a voltage-sensitive dye. J. Neurophysiol. 68, 1804–1819.

KINNAMON, S. C. AND CUMMINGS, T. A. (1992). Chemosensory transduction mechanisms in taste. A. Rev.Physiol. 54, 715–731.

KLEENE, S. J. (1993a). Origin of the chloride current in olfactory transduction. Neuron 11, 123–132.KLEENE, S. J. (1993b). The cyclic nucleotide-activated conductance in olfactory cilia: effects of

cytoplasmic Mg2+ and Ca2+. J. Membr. Biol. 131, 237–243.KLEENE, S. J. AND GESTELAND, R. C. (1991). Calcium-activated chloride conductance in frog olfactory

cilia. J. Neurosci. 11, 3624–3629.KOLESNIKOV, S. S. AND KOSOLAPOV, A. V. (1993). Cyclic nucleotide-activated channels in carp olfactory

receptor cells. Biochim. biophys. Acta 1150, 63–72.KOLESNIKOV, S. S., ZHAINAZAROV, A. B. AND KOSOLAPOV, A. V. (1990). Cyclic nucleotide-activated

channels in the frog olfactory receptor plasma membrane. FEBS Lett. 266, 96–98.KRAMER, R. H. AND SIEGELBAUM, S. A. (1992). Intracellular Ca2+ regulates the sensitivity of cyclic

nucleotide-gated channels in olfactory receptor neurons. Neuron 9, 897–906.KURAHASHI, T. (1989). Activation by odorants of cation-selective conductance in the olfactory receptor

cell isolated from the newt [published erratum appears in J. Physiol., Lond. (1990) 424, 561–562].J. Physiol., Lond. 419, 177–192.

KURAHASHI, T. (1990). The response induced by intracellular cyclic AMP in isolated olfactory receptorcells of the newt. J. Physiol., Lond. 430, 355–371.

KURAHASHI, T. AND KANEKO, A. (1991). High density cyclic AMP-gated channels at the ciliarymembrane in the olfactory receptor cell. Neuroreport 2, 5–8.

KURAHASHI, T. AND SHIBUYA, T. (1990). Ca2+-dependent adaptive properties in the solitary olfactoryreceptor cell of the newt. Brain Res. 515, 261–268.

KURAHASHI, T. AND YAU, K.-W. (1993). Co-existence of cationic and chloride components in odorant-induced current of vertebrate olfactory receptor cells. Nature 363, 71–74.

KURIHARA, K. (1990). Molecular mechanisms of reception and transduction in olfaction and taste. Jap. J.Physiol. 40, 305–324.

LABARCA, P., SIMON, S. A. AND ANHOLT, R. R. (1988). Activation by odorants of a multistate cationchannel from olfactory cilia. Proc. natn. Acad. Sci. U.S.A. 85, 944–947.

LANCET, D. AND BEN-ARIE, N. (1993). Olfactory receptors. Curr. Biol. 3, 668–674.LANCET, D., KAUER, J. S., GREER, C. A. AND SHEPHERD, G. M. (1981). High resolution 2-deoxyglucose

localization in the olfactory epithelium. Chem. Senses 6, 343–349.LERNER, M. R., REAGAN, J., GYORGYI, T. AND ROBY, A. (1988). Olfaction by melanophores: what does it

mean? Proc. natn. Acad. Sci. U.S.A. 85, 261–264.LISCHKA, F. W. AND SCHILD, D. (1993). Effects of nitric oxide upon olfactory receptor neurones in

Xenopus laevis. Neuroreport 4, 582–584.LOWE, G. AND GOLD, G. H. (1991). The spatial distributions of odorant sensitivity and odorant-induced

currents in salamander olfactory receptor cells. J. Physiol., Lond. 442, 147–168.


LOWE, G. AND GOLD, G. H. (1993a). Non-linear amplification by calcium-dependent chloride channelsin olfactory receptor cells. Nature 366, 283–286.

LOWE, G. AND GOLD, G. H. (1993b). Contribution of the ciliary cyclic nucleotide-gated conductance toolfactory transduction in the salamander. J. Physiol., Lond. 462, 175–196.

LUCERO, M. T., HORRIGAN, F. T. AND GILLY, W. F. (1992). Electrical responses to chemical stimulationof squid olfactory receptor cells. J. exp. Biol. 162, 231–249.

LUDWIG, J., MARGALIT, T., EISMANN, E., LANCET, D. AND KAUPP, U. B. (1990). Primary structure ofcyclic AMP-gated channel from bovine olfactory epithelium. FEBS Lett. 270, 24–29.

LYNCH, J. W. AND LINDEMANN, B. (1994). Cyclic nucleotide-gated channels of rat olfactory receptorcells: divalent cations control the sensitivity to cyclic AMP. J. gen. Physiol. 103, 87–106.

MACKAY-SIM, A., SHAMAN, P. AND MOULTON, D. G. (1982). Topographic coding of olfactoryquality: Odorant specific patterns of epithelial responsivity in the salamander. J. Neurophysiol. 48,584–596.

MANIA-FARNELL, B. AND FARBMAN, A. I. (1990). Immunohistochemical localization of guaninenucleotide-binding proteins in rat olfactory epithelium during development. Devl Brain Res. 51,103–112.

MAUE, R. A. AND DIONNE, V. E. (1987). Patch-clamp studies of isolated mouse olfactory receptorneurons. J. gen. Physiol. 90, 95–125.

MCCLINTOCK, T. S. AND ACHE, B. W. (1989a). Ionic currents and ion channels of lobster olfactoryreceptor neurons. J. gen. Physiol. 94, 1085–1099.

MCCLINTOCK, T. S. AND ACHE, B. W. (1989b). Hyperpolarizing receptor potentials in lobster olfactoryreceptor cells: implications for transduction and mixture suppression. Chem. Senses 14, 637–647.

MCCLINTOCK, T. S. AND ACHE, B. W. (1989c). Histamine directly gates a chloride channel in lobsterolfactory receptor neurons. Proc. natn. Acad. Sci. U.S.A. 86, 8137–8141.

MICHEL, W. C. AND ACHE, B. W. (1992). Cyclic nucleotides mediate an odor-evoked potassiumconductance in lobster olfactory receptor cells. J. Neurosci. 12, 3979–3984.

MICHEL, W. C., MCCLINTOCK, T. S. AND ACHE, B. W. (1991). Inhibition of lobster olfactory receptorcells by an odor-activated potassium conductance. J. Neurophysiol. 65, 446–453.

MILLER, R. C. (1987). Ion Channel Reconstitution. New York: Plenum Press.MIYAMOTO, T., RESTREPO, D., CRAGOE, E. J. AND TEETER, J. H. (1992a). IP3- and cyclic AMP-induced

responses in isolated olfactory receptor neurons from the channel catfish. J. Membr. Biol. 127,173–183.

MIYAMOTO, T., RESTREPO, D. AND TEETER, J. H. (1992b). Voltage-dependent and odorant-regulatedcurrents in isolated olfactory receptor neurons of the channel catfish. J. gen. Physiol. 99, 505–530.

MOULTON, D. G. (1977). Minimum odorant concentrations detectable by the dog and their implicationsfor olfactory receptor sensitivity. In Chemical Signals in Vertebrates (ed. D. Muller-Schwarz andM. M. Mozell), pp. 455–464. New York, London: Plenum Press.

MUSTAPARTA, H. (1971). Spatial distribution of receptor responses to stimulation with different odours.Acta physiol. scand. 82, 154–166.

NAKAMURA, T. AND GOLD, G. H. (1987). A cyclic nucleotide-gated conductance in olfactory receptorcilia. Nature 325, 442–444.

NEF, P. (1993). Early events in olfaction: diversity and spatial patterns of odorant receptors. ReceptorsChannels 1, 259–266.

NEF, P., HERMANS-BORGMEYER, I., ARTIERES-PIN, H., BEASLEY, L., DIONNE, V. E. AND HEINEMANN, S. F.(1992). Spatial pattern of receptor expression in the olfactory epithelium. Proc. natn. Acad. Sci.U.S.A. 89, 8948–8952.

NEVITT, G. A. AND MOODY, W. J. (1992). An electrophysiological characterization of ciliated olfactoryreceptor cells of the coho salmon Oncorhynchus kisutch. J. exp. Biol. 166, 1–17.

NGAI, J., CHESS, A., DOWLING, M. M., NECLES, N., MACAGNO, E. R. AND AXEL, R. (1993a). Coding ofolfactory information: topography of odorant receptor expression in the catfish olfactory epithelium.Cell 72, 667–680.

NGAI, J., DOWLING, M. M., BUCK, L., AXEL, R. AND CHESS, A. (1993b). The family of genes encodingodorant receptors in the channel catfish. Cell 72, 657–666.

NOMURA, T. AND KURIHARA, K. (1987). Liposomes as a model for olfactory cells: changes in membranepotential in response to various odorants. Biochemistry, N.Y. 26, 6135–6140.

NOMURA, T. AND KURIHARA, K. (1989). Similarity of ion dependence of odorant responses between lipidbilayer and olfactory system. Biochim. biophys. Acta 1005, 260–264.


O’CONNELL, R. J. AND MOZELL, M. M. (1969). Quantitative stimulation of frog olfactory receptors.J. Neurophysiol. 32, 51–63.

OTTOSON, D. (1956). Analysis of the electrical activity of the olfactory epithelium. Acta physiol. scand.35 (Suppl. 122), 1–83.

PACE, U., HANSKI, E., SALOMON, Y. AND LANCET, D. (1985). Odorant-sensitive adenylate cyclase maymediate olfactory reception. Nature 316, 255–258.

PASSÉ, D. H. AND WALKER, J. C. (1985). Odor psychophysics in vertebrates. Neurosci. BioBehav. Rev. 9,431–467.

PATTE, F., ETCHETO, N. AND LAFFORT, P. (1975). Selected and standardized values of suprathresholdodor intensities for 110 substances. Chem. Sens. Flav. 1, 283–307.

POTENZA, M. N. AND LERNER, M. R. (1992). A rapid quantitative bioassay for evaluating the effects ofligands upon receptors that modulate cyclic AMP levels in a melanophore cell line. Pigment Cell Res.5, 372–378.

RAMING, K., KRIEGER, J., STROTMANN, J., BOEKHOFF, I., KUBICK, S., BAUMSTARK, C. AND BREER, H.(1993). Cloning and expression of odorant receptors. Nature 361, 353–356.

REED, R. R. (1992). Signaling pathways in odorant detection. Neuron 8, 205–209.RESSLER, K. J., SULLIVAN, S. L. AND BUCK, L. B. (1993). A zonal organization of odorant receptor gene

expression in the olfactory epithelium. Cell 73, 597–609.RESTREPO, D., BOEKHOFF, I. AND BREER, H. (1993). Rapid kinetic measurements of second messenger

fromation in olfactory cilia from channel catfish. Am. J. Physiol. 264, C906–C911.RESTREPO, D. AND BOYLE, A. G. (1991). Stimulation of olfactory receptors alters regulation of [Ca2+]i in

olfactory neurons of the catfish (Ictalurus punctatus). J. Membr. Biol. 120, 223–232.RESTREPO, D., MIYAMOTO, T., BRYANT, B. P. AND TEETER, J. H. (1990). Odor stimuli trigger influx of

calcium into olfactory neurons of the channel catfish. Science 249, 1166–1168.REVIAL, M. F., DUCHAMP, A. AND HOLLEY, A. (1978). Odour discrimination by frog olfactory receptors:

a second study. Chem. Sens. Flav. 3, 7–21.RONNETT, G. V. AND SNYDER, S. H. (1992). Molecular messengers of olfaction. Trends. Neurosci. 15,

508–513.SCHILD, D. (1989). Whole-cell currents in olfactory receptor cells of Xenopus laevis. Expl Brain Res. 78,

223–232.SCHLEICHER, S., BOEKHOFF, I., ARRIZA, J., LEFKOWITZ, R. J. AND BREER, H. (1993). A beta-adrenergic

receptor kinase-like enzyme is involved in olfactory signal termination. Proc. natn. Acad. Sci. U.S.A.90, 1420–1424.

SCHMIEDEL-JAKOB, I., ANDERSON, P. A. V. AND ACHE, B. W. (1989). Whole cell recording fromlobster olfactory receptor cells: responses to current and odor stimulation. J. Neurophysiol. 61,994–1000.

SCHMIEDEL-JAKOB, I., MICHEL, W. C., ANDERSON, P. A. V. AND ACHE, B. W. (1990). Whole cellrecording from lobster olfactory receptor cells: multiple ionic basis for the receptor potential. Chem.Senses 15, 397–405.

SCHWARTZ LEVEY, M., CINELLI, A. R. AND KAUER, J. S. (1992). Intracellular injection of vital dyes intosingle cells in the salamander olfactory epithelium. Neurosci. Lett. 140, 265–269.

SCOTT, J. W. (1991). Central processing of olfaction. J. Physiol., Paris 39, 593–600.SCOTT, J. W. AND HARRISON, T. A. (1987). The olfactory bulb: anatomy and physiology. In

Neurobiology of Taste and Smell (ed. T. E. Finger and W. L. Silver), pp. 151–178. New York: JohnWiley and Sons.

SELBIE, L. A., TOWNSEND-NICHOLSON, A., IISMAA, T. P. AND SHINE, J. (1992). Novel G protein-coupledreceptors: a gene family of putative human olfactory receptor sequences. Molec. Brain Res. 13,159–163.

SHINOHARA, H., KATO, K. AND ASANO, T. (1992). Differential localization of G proteins, Gi and Go, inthe olfactory epithelium and the main olfactory bulb of the rat. Acta anat. 144, 167–171.

SHIRLEY, S. G. (1992). Olfaction. Int. Rev. Neurobiol. 33, 1–53.SHIRLEY, S. G., ROBINSON, C. J., DICKINSON, K., AUJLA, R. AND DOOD, G. (1986). Olfactory adenylate

cyclase of the rat. Stimulation by odorants and inhibition by Ca2+. Biochem. J. 240, 605–607.SHOJI, T. AND KURIHARA, K. (1991). Sensitivity and transduction mechanisms of responses to general

odorants in turtle vomeronasal system. J. gen. Physiol. 98, 909–919.SICARD, G. AND HOLLEY, A. (1984). Receptor cell responses to odorants: similarities and differences

among odorants. Brain Res. 292, 283–296.


SILVER, W. L. (1982). Electrophysiological responses from the peripheral olfactory system of theAmerican eel, Anguilla rosttata. J. comp. Physiol. A 148, 379–388.

SKLAR, P. B., ANHOLT, R. R. H. AND SNYDER, S. H. (1986). The odor-sensitive adenylate cyclase ofolfactory receptor cells. J. biol. Chem. 261, 15538–15543.

SLOTNICK, B. M., PANHUBER, H., BELL, G. A. AND LAING, D. G. (1989). Odor-induced metabolic activityin the olfactory bulb of rats trained to detect propionic acid vapor. Brain Res. 500, 161–168.

SLOTNICK, B. M. AND SCHOONOVER, F. W. (1984). Olfactory thresholds in unilaterally bulbectomizedrats. Chem. Senses 9, 325–340.

STROTMANN, J., WARNER, I., KRIEGER, J., RAMING, K. AND BREER, H. (1992). Expression of odorantreceptors in spatially restricted subsets of chemosensory neurones. Neuroreport 3, 1053–1056.

SUTTERLIN, A. M. AND SUTTERLIN, N. (1971). Electrical responses of the olfactory epithelium of Atlanticsalmon (Salmo salar). J. Fish. Res. Bd Can. 28, 565–572.

SUZUKI, N. (1990). Single cyclic nucleotide-activated ion channel activity in olfactory receptor cell somamembrane. Neurosci. Res. Suppl. 12, S113–S126.

SUZUKI, N. AND TUCKER, D. (1971). Amino acids as olfactory stimuli in freshwater catfish, Italurus catus(Linn.). Comp. Biochem. Physiol. A. 40, 399–404.

TAGAKI, S. F. (1971). Degeneration and regeneration of the olfactory epithelium. In Handbook ofSensory Physiology, vol. IV (ed. L. M. Beidler), pp. 76–94. Berlin, Heidelberg, New York: Springer.

TROMBLEY, P. Q. AND SHEPHERD, G. M. (1993). Synaptic transmission and modulation in the olfactorybulb. Curr. Opin. Neurobiol. 3, 540–547.

TROTIER, D. (1986). A patch-clamp analysis of membrane currents in salamander olfactory receptorcells. Pflügers Arch. 407, 589–595.

VASSAR, R., NGAI, J. AND AXEL, R. (1993). Spatial segregation of odorant receptor expression in themammalian olfactory epithelium. Cell 74, 309–318.

VERMA, A., HIRSCH, D. J., GLATT, C. E., RONNETT, G. V. AND SNYDER, S. H. (1993). Carbon monoxide: aputative neuronal messenger. Science 259, 381–384.

VODYANOY, V. AND MURPHY, R. B. (1983). Single-channel fluctuations in bimolecular lipid membranesinduced by rat olfactory epithelial homogenates. Science 220, 717–719.

VODYANOY, V. AND VODYANOY, I. (1987). ATP and GTP are essential for olfactory response. Neurosci.Lett. 73, 253–258.

WEGENER, J. W., BOEKHOFF, I., TAREILUS, E. AND BREER, H. (1993). Olfactory signalling in antennalreceptor neurones of the locust (Locusta migratoria). J. Insect Physiol. 39, 153–163.

ZUFALL, F. AND FIRESTEIN, S. (1993). Divalent ions block the cyclic nucleotide-gated channel ofolfactory receptor neurons. J. Neurophysiol. 69, 1758–1768.

ZUFALL, F. AND HATT, H. (1991). Dual activation of a sex pheromone-dependent ion channel from insectolfactory dendrites by protein kinase C activators and cyclic GMP. Proc. natn. Acad. Sci. U.S.A. 88,8520–8524.

ZUFALL, F., SHEPHERD, G. M. AND FIRESTEIN, S. (1991). Inhibition of the olfactory cyclic nucleotidegated ion channel by intracellular calcium. Proc. R. Soc. Lond. B 246, 225–230.


Documents

194, 1–21 (1994) 1 Printed in Great Britain © The Company ...jeb.biologists.org/content/jexbio/194/1/1.full.pdf · secondary and tertiary targets in the pyriform cortex (lateral