Mechanisms of odorant receptor gene choice in Drosophila and vertebrates

Molecular and Cellular Neuroscience 41 (2009) 101–112

Contents lists available at ScienceDirect

Molecular and Cellular Neuroscience j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y m c n e

Review

Mechanisms of odorant receptor gene choice in Drosophila and vertebrates Stefan H. Fuss a, Anandasankar Ray b,⁎ a b

Department of Molecular Biology and Genetics, Bogazici University, 34342 Istanbul, Turkey Department of Entomology, University of California, 3401 Watkins Drive, Riverside, CA 92521, USA

a r t i c l e

i n f o

Article history: Received 20 February 2009 Accepted 27 February 2009 Available online 19 March 2009 Keywords: Odorant receptor gene Regulation

a b s t r a c t Odorant receptors are encoded by extremely large and divergent families of genes. Each receptor is expressed in a small proportion of neurons in the olfactory organs, and each neuron in turn expresses just one odorant receptor gene. This fundamental property of the peripheral olfactory system is widely conserved across evolution, and observed in vertebrates, like mice, and invertebrates, like Drosophila, despite their olfactory receptor gene families being evolutionarily unrelated. Here we review the progress that has been made in these two systems to understand the intriguing and elusive question: how does a single neuron choose to express just one of many possible odorant receptors and exclude expression of all others? © 2009 Elsevier Inc. All rights reserved.

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . Zonal expression of vertebrate OR genes . . . . . . . . . . . Mechanisms of zonal expression in mammals . . . . . . Class I OR genes . . . . . . . . . . . . . . . . . . Patch OR genes . . . . . . . . . . . . . . . . . . . Class II OR genes . . . . . . . . . . . . . . . . . . Regulation by transcription factors. . . . . . . . . . Zonal expression in Drosophila. . . . . . . . . . . . . . . . Mechanisms of zonal expression in Drosophila . . . . . . Cis-regulation . . . . . . . . . . . . . . . . . . . Regulation by transcription factors. . . . . . . . . . Neuron-specific expression in vertebrates . . . . . . . . . . Co-expression of two OR genes . . . . . . . . . . . . . Mono-allelic expression . . . . . . . . . . . . . . . . . No evidence for recombination based mechanisms . . . . No evidence for trans-chromosomal mechanisms . . . . . Receptor switching . . . . . . . . . . . . . . . . . . . Negative-feedback regulation . . . . . . . . . . . . . . Neuron-specific expression in Drosophila . . . . . . . . . . . Co-ordination of Or gene expression in paired neurons of a Lack of negative-feedback regulation. . . . . . . . . . . Mechanisms of co-expression of Or genes . . . . . . . . Punctate expression patterns . . . . . . . . . . . . . . Cis-regulatory code for neuron-specific expression . . . . Transcription factor code for neuron-specific expression . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . sensillum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

102 102 103 103 104 105 105 105 106 106 106 106 107 107 107 107 107 108 108 108 109 109 109 109 109 109 111 111

Abbreviations: LCR, locus control region; OB, olfactory bulb; OE, olfactory epithelium; OR, vertebrate olfactory receptor; Or, insect odor receptor; ORN, insect olfactory receptor neuron; OSN, vertebrate olfactory sensory neuron. ⁎ Corresponding author. Fax: +1 951 827 3086. E-mail address: [email protected] (A. Ray). 1044-7431/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.mcn.2009.02.014

102

S.H. Fuss, A. Ray / Molecular and Cellular Neuroscience 41 (2009) 101–112

Introduction A fundamental principle in the organization of the olfactory system that is conserved from insects to mammals is the precise functional specification of sensory receptor cells. Individual neurons choose to express only one (and sometimes two) odorant receptor from a genomic repertoire that can be as large as 1500 genes (Buck, 2000; Hallem et al., 2006). Olfactory sensory neurons send axonal projections to an olfactory processing center in the brain where axons of cells expressing the same receptor converge onto one or a few stereotypical glomeruli. In the peripheral olfactory system expression of each receptor is restricted to a subregion of the sensory surface with different olfactory receptors expressed in distinct but overlapping domains (Fig. 1). Although other types of sensory neurons may also follow a ‘one receptor rule’ (Mazzoni et al., 2004), the regulatory challenge confronted by the olfactory system represents an extreme amongst gene regulation problems (Mombaerts, 2004). Similarly, B- and Tlymphocytes of the immune system also face the problem of selecting a single antigen receptor from a multitude of possibilities. It has been several years since the identification of olfactory receptor genes (N18 years in mammals, and N11 years in Drosophila), yet little is known about the mechanisms regulating their expression. Here we review the progress made in understanding the elusive problem of odorant receptor gene choice in vertebrates and Drosophila. Various models that invoke either stochastic or deterministic mechanisms have been proposed to underlie olfactory receptor gene choice. At least five basic mechanisms, which are not mutually

exclusive, are conceivable (Fig. 2). (1) Olfactory neurons could be specified in a “one-to-one” manner by specific transcription factors that are expressed in a distinct neuronal cell type and drives expression of only one receptor. However, such a model simply transfers the problem of odorant receptor gene choice up to the level of establishing specific expression patterns for transcription factors. (2) Receptor expression could be specified by a ‘combinatorial’ code of transcription factors, expressed in overlapping domains or combinations of gradients, rather than a battery of unique regulatory factors for each receptor. (3) Limiting factors may stochastically initiate expression of a single or a few receptors per cell; expression of a functional receptor may then exclude expression of others by a ‘negative-feedback’ mechanism. (4) Exclusive receptor expression could be controlled by a single locus control region (LCR), present either in cis or trans, which stochastically selects and activates the promoter of only one out of several odorant receptor promoters in an individual neuron. (5) Lastly, DNA recombination events could lead to expression of a single receptor by juxtaposing the promoter of a chosen receptor with an enhancer region. We will discuss the experiments that have explored these models. Zonal expression of vertebrate OR genes The family of vertebrate olfactory receptor (OR) genes was initially discovered based on the prediction that they may encode seventransmembrane G-protein coupled receptors (Buck and Axel, 1991). With ORs comprising up to 3% of genes in mammals, they constitute the largest gene family in vertebrate genomes, where they are

Fig. 1. Olfactory receptor expression. (A) In the mouse olfactory epithelium individual ORs are expressed in distinct banded expression zones (top) indicated by different colors. The numbers refer to the classical 4 zone distinction, p denotes the patch region, an exception from the otherwise elongated parallel organization of zones (adapted from Vassalli et al., 2002). Within a given region of the OE multiple ORs are expressed in distinct neuronal cell population (bottom) as shown for MOR28 and MOR230-1 (taken from Serizawa et al., 2003). (B) Representation of Drosophila head, labeled for expression of Or22a (blue, large basiconic), Or47a (yellow, small basiconic), Or23a (magenta, trichoid), and Or71a (green, palp basiconic) (taken from Ray et al., 2008). The two Drosophila olfactory organs are marked, the 3rd antennal segment and the maxillary palp, which express distinct, nonoverlapping sets of OR genes. The seven Or genes expressed in the maxillary palp are distributed in three distinct functional types of sensilla. The schematic (bottom) shows the positions of nuclei of neurons that have been differentially stained for one receptor from each of the three sensilla classes. The three receptors appear randomly distributed within the maxillary palp (adapted from Ray et al., 2008).

S.H. Fuss, A. Ray / Molecular and Cellular Neuroscience 41 (2009) 101–112

103

Fig. 2. Various models for patterning of one-receptor-per-neuron in the olfactory system: deterministic (top) and stochastic (bottom).

typically arranged in clusters spread out over several chromosomes. The size of the repertoire ranges from around one hundred in fish (Alioto and Ngai, 2005; Niimura and Nei, 2005) to more than a thousand in mammals. (Zhang et al., 2007a; Zhang and Firestein, 2007). Olfactory sensory neurons (OSNs) expressing a given OR gene are restricted to defined expression domains or ‘zones’ within the olfactory epithelium (OE). These zones appear as multiple parallel stripes oriented along the anteroposterior axis of the complex mammalian OE (Fig. 1A) (Ressler et al., 1993; Vassar et al., 1993) but a similar restriction can also be observed in the form of concentric rings in zebrafish (Weth et al., 1996) and overlapping patches in the flat tiger salamander OE (Marchand et al., 2004). Historically, four zones with limited overlap were described in mouse and rat (Fig. 1A) (Ressler et al., 1993; Vassar et al., 1993). However, recent studies characterizing multiple ORs simultaneously indicate that OR expression patterns, although restricted, are not limited to four zones, but occupy multiple zones that overlap considerably with each other and appear to be arranged continuously along the dorso-ventral axis of the OE (Iwema et al., 2004; Miyamichi et al., 2005; Tsuboi et al., 2006). In its most extreme notion, one could conceive of a unique zone for every single OR. A curious exception to the parallel zone organizations is seen with the members of the MOR262 (OR37) family, which are expressed in a narrow ‘patch’ in the center of the turbinates (Fig. 1A) (Strotmann et al., 1994). Thus, a given OR is generally confined to one coherent zone, yet a multitude of ORs are expressed in overlapping patterns within a given zone of the OE (Fig. 1A). The distribution of ORs within a zone has been described as random or ‘punctate’ (Mombaerts, 2004) as it lacks obvious aggregates of OSNs expressing the same OR (Baier and Korsching, 1994; Ressler et al., 1993; Vassar et al., 1993). Remarkably, zones are discontinuous along the surface of the OE, such that there are multiple occurrences of the same zone if the OE is unrolled, suggesting that zone organization is established early during ontogenetic development before evagination of the turbinates. OSN axons project to the olfactory bulb (OB) in a ‘zone-to-zone’ fashion, such that the dorso-ventral position of OSNs expressing a given OR dictates the dorso-ventral position of their cognate

glomerulus in the OB (Wang et al., 1998). Contrary to initial reports claiming that ORs related by sequence are primarily expressed in the same zone (Kubick et al., 1997; Ressler et al., 1993; Vassar et al., 1993), recent data show that members of the same subfamily can be expressed in different zones, sometimes covering almost threequarters of the OE (Miyamichi et al., 2005). However similarities in expression profiles that are observed by and large for subfamily members may be a consequence of their evolutionary origin from gene duplication events that included regulatory sequences controlling their spatial expression. Mechanisms of zonal expression in mammals How is zonal OR expression established during development? Two phylogenetically distinct classes of ORs can be found in vertebrate genomes. The class I ORs are supposed to be evolutionary more ancient and are the predominant type in fish (Alioto and Ngai, 2005; Ngai et al., 1993; Niimura and Nei, 2005), while class II ORs have undergone a massive expansion in mammals and comprise 90% of the more than 1000 OR genes in mouse and rat (Young et al., 2002; Zhang and Firestein, 2002). Interestingly, all but 2 of the ∼ 150 mouse class I ORs are restricted to the dorsal OE, while class II ORs occupy the entire OE (Miyamichi et al., 2005; Tsuboi et al., 2006; Zhang et al., 2004). Accordingly, class I-expressing OSNs project to a coherent glomerular domain in the dorsal OB (Kobayakawa et al., 2007). Class I OR genes Class I ORs reside in one large cluster on mouse chromosome 7 which is only interrupted by the β-globin locus (Bulger et al., 1999; Tsuboi et al., 2006; Zhang et al., 2007a). Therefore, class I genes provide a unique opportunity to study a large set of genomically linked OR genes that are coordinately expressed in a single zone. Bioinformatic approaches identified unique arrangements of O/E(Olf1/Early B-cell factor) and homeodomain-like sites within class I promoters (Hoppe et al., 2006). Although identical binding sites are also present in class II promoters (Hoppe et al., 2003; Vassalli et al., 2002), O/E-like sites appear to be more numerous and to be more concentrated around transcription start sites in class I promoters (Hoppe et al., 2006) (Fig. 3A). The genomic loci of the atypical genes

104

S.H. Fuss, A. Ray / Molecular and Cellular Neuroscience 41 (2009) 101–112

Fig. 3. A summary of proposed mechanisms of zonal and monogenic OR expression in vertebrates. (A) Zonal restriction of OR gene expression may be determined by specific combinations or arrangements of regulatory sequences in OR promoters. Class I ORs contain multiple repeats of O/E- (red boxes) and homeodomain-like (purple circles) clustered around their transcription start sites, whereas ‘patch’ ORs display multiple, but conserved transcription factor binding sites. These unique arrangements of promoter elements are not seen in class II ORs and may direct their expression patterns in the dorsal OE and the patch region (p), respectively. For class II ORs unique distance arrangements between O/E- and homeodomain-like sites may be involved to instruct their expression in continuous and overlapping patterns along the dorso-ventral axis (I to IV denote the classical 4 zones). (B) A model for monoallelic and monogenic expression at the MOR28 cluster locus. A single LCR (H-region, red dot) regulates expression of downstream OR genes (3 out of 7 shown). Methylation and silencing of one copy in some OSNs will result in monoallelic expression of ORs from the second allele. In subsequent steps a single OR is chosen for expression by the exclusive interaction of the H-region with a single OR promoter. OR protein expression elicits an unknown feedback signal to lock in stable expression. The signal could act positively to drive differentiation of OSNs or to stabilize the actively expressed promoter. Alternatively, inactive promoters or their respective LCRs could be permanently silenced by negative feedback. An involvement of the OR coding sequence itself has also been demonstrated.

MOR35-1 and MOR41-1 are embedded amongst other class I ORs, yet they are expressed more ventrally (Tsuboi et al., 2006). Remarkably, their promoter structures share a higher degree of similarity with class II promoters (Hoppe et al., 2006). A recent finding supports the idea that OSNs may be committed by lineage and that expression of class I ORs in the dorsal OE could be dictated by the identity of a particular OSN type (Bozza et al., 2009). However, the molecular mechanisms underlying the identity of different OSN subtypes are unknown.

Patch OR genes Candidate regulatory motifs have also been identified for the twelve members of the MOR262 subfamily that are exclusively expressed in the patch (Hoppe et al., 2003, 2000). A total of six highly conserved motifs were identified in their putative promoters, which are also conserved in the promoters of recently identified patch genes from unrelated OR subfamilies (Hoppe et al., 2006; Pyrski et al., 2001), suggesting that they impart patch-specific expression (Fig. 3A). The transcription factors Lhx2, O/E-2, Ptx-1, BEN, Alx-3 and AP-2b,

S.H. Fuss, A. Ray / Molecular and Cellular Neuroscience 41 (2009) 101–112

were shown to bind to the alleged MOR262-6 promoter in vitro and their recognition sites match conserved motifs identified by the bioinformatics approach (Hoppe et al., 2003). Class II OR genes Class II OR genes are the predominant class in higher vertebrates and are distributed in multiple clusters in the genome (e.g. 43 in the mouse). However, there does not seem to be a correlation between OR expression profiles and genomic organization as genes from the same cluster can be expressed in different zones. Nevertheless, specific LCRs could direct zonal expression in a subset of cluster ORs. The H-region on mouse chromosome 14 is a 2.1 kb sequence, which is evolutionarily conserved between mouse and human and is located 70–90 kb upstream of the MOR28 cluster (Lane et al., 2002). It has been shown to regulate expression of seven OR genes from the cluster, three of which are expressed in the ventral OE, while the remaining four genes are expressed dorsally (Fuss et al., 2007; Nishizumi et al., 2007; Serizawa et al., 2003). Thus, a single LCR can control expression of ORs at different ends of the OE with non-overlapping patterns (Fig. 3B), indicating that LCRs may be important for OR gene choice but not for specifying expression in a unique zone. In contrast to the long-range cis-regulation of the MOR28 cluster, transgenes carrying short genomic sequences upstream of OR genes can, in many cases, reproduce expression patterns that are indistinguishable from the endogenous OR (Lewcock and Reed, 2004; Qasba and Reed, 1998; Rothman et al., 2005; Vassalli et al., 2002; Zhang et al., 2007a). Sequences as short as 161 nucleotides can be sufficient to dictate appropriate OR gene expression, suggesting shortrange rather than long-range control by cis-acting elements (Rothman et al., 2005). However, these studies also find that genomic context of transgene integration and copy numbers may influence expression patterns (Qasba and Reed, 1998; Rothman et al., 2005; Zhang et al., 2007b). The compact nature of OR regulation that has been observed in most cases may have facilitated the vast evolutionary expansion of the OR gene family by gene duplication (Zhang and Firestein, 2002). Zonal restriction does not depend on the OR protein. Promoter transgenes lacking a functional OR coding sequence can be expressed in a pattern resembling endogenous OR expression (Lewcock and Reed, 2004; Qasba and Reed, 1998; Serizawa et al., 2003). In addition, OR genes can be ectopically expressed in inappropriate zones by swapping of OR coding sequences (Feinstein and Mombaerts, 2004; Wang et al., 1998). These observations suggest that receptordependent neuronal activity or developmental feedback from the OB do not play a major role in patterning OR expression. Regulation by transcription factors Comparative analyses of OR upstream regions have identified conserved O/E- ((Y)3CA(R)4) and homeodomain-like (TATTXX) sites in close proximity to each other to be present in virtually every OR promoter (Hoppe et al., 2003; Michaloski et al., 2006; Vassalli et al., 2002) as well as in the functional core of the H-region (Hirota and Mombaerts, 2004; Nishizumi et al., 2007). Among other factors, the LIM homeodomain protein Lhx2 was shown to bind to homeodomainlike sites in the M71 promoter (Hirota and Mombaerts, 2004). Mice deficient in Lhx2 expression show defects in OSN development (Hirota and Mombaerts, 2004; Kolterud et al., 2004). Interestingly, OSNs expressing class II ORs are completely abolished in Lhx2 deficient mice, while class I-expressing OSNs are largely spared (Hirota et al., 2007). Since Lhx2 deficient mice are embryonically lethal it is difficult to determine whether Lhx2 functions exclusively on OSN development or if it also has a direct role on OR expression. However, given the uniform expression across the OE (Hirota and Mombaerts, 2004), Lhx2 by itself is unlikely to impart a spatial profile on OR expression. In a recent report the transcription factor Emx2, which can also bind to homeodomain sites (Hirota et al., 2007) was also shown to affect OR expression (McIntyre et al., 2008). While up to 75% of ORs may be

105

absent in Emx2 mutant mice, other ORs were expressed in a higher than usual number of OSNs, perhaps to compensate for the severe reduction in OR expression. Interestingly, both OR classes and ORs from all zones were equally affected. In contrast to Lhx2, which affects the formation of mature and immature OSNs, Emx2 deficient mice show a specific reduction of mature OSNs, suggesting that it may play a more direct role in regulating OR gene expression than Lhx2. O/E-like binding sites are present not only in OR promoters but also in the promoters of genes ubiquitously expressed in OSNs such as OMP (Buiakova et al., 1994; Kudrycki et al., 1993), Golf, ACIII, and OCNC1 (Wang and Reed, 1993). Therefore, O/E-like sites by themselves may be important for OE expression, but not sufficient to instruct zonal expression. A unique arrangement of multiple O/E-like sites has been observed in class I OR promoters, however, which could constitute a specific instruction (Hoppe et al., 2006). Unfortunately, the most comprehensive study of 198 different OR promoters failed to compare the distribution of O/E- and homeodomain-like sites with expression profiles of OR genes (Michaloski et al., 2006). Therefore, it remains speculative if zone expression could be governed to a large extent by specific arrangements of O/E- and homeodomain-like sites in class II genes (Fig. 3A), or whether additional unknown factors are required. Regulatory elements in the M71 promoter have been studied in some detail (Rothman et al., 2005). Individual mutations of either the O/E- or homeodomain-like site consistently resulted in ventralization of expression, while the combined mutations of both sites abolished expression in transgenes. Interestingly however, when identical mutations were introduced by gene-targeting at the endogenous locus, the combined mutation did not abolish expression completely but also resulted in a ventralization of the expression pattern, suggesting that the most proximal O/E and homeodomain sites may play a more important role in specifying the relative dorsal–ventral position. In summary, zonal expression of OR genes may, at least in part, be established by unique combinations of regulatory sequences in OR gene promoters (Fig. 3A). The bioinformatic analyses of class I and patch OR promoters have provided several zone-specific elements that can be put to a test in the future and it will be exiting to see if they can be validated experimentally in transgenic studies. While there is a growing list of transcription factors (Norlin et al., 2001) and markers (Norlin and Berghard, 2001; Schoenfeld and Knott, 2002) that are expressed in zonal fashion or in gradients across the OE, they have yet to be tested experimentally. The continuity and the overlap of OR expression profiles imply that these patterns could be established by morphogen gradients or combinations of transcription factor gradients in the OE (Miyamichi et al., 2005) rather than by one-to-one expression of zone-specific factors. Zonal expression in Drosophila The olfactory system of Drosophila melanogaster provides a particularly tractable system to address the problem of odor receptor (Or) gene regulation because of its numerical simplicity as compared to that of vertebrates. Moreover the olfactory system has been described in great detail at both the cellular and molecular levels (Dahanukar et al., 2005; Hallem et al., 2006; Vosshall and Stocker, 2007) and despite having a simpler olfactory system, it is still striking in complexity and precision of organization, posing a challenge to understand. Odorants are detected by two types of olfactory organs: the 3rd segment of the antenna, and the maxillary palp (Fig. 1B). The surface of the antenna is covered by ∼ 500 sensory hairs called sensilla, and the surface of the maxillary palps is covered by ∼ 60 sensilla. Each sensillum harbors the dendrites of 1 to 4 (usually 2) olfactory receptor neurons (ORNs), thus there are ∼ 2000–3000 ORNs per fly (Shanbhag et al., 1999; Stocker, 1994). Based on their morphological characteristics and location, the sensilla can be divided into five classes; large basiconic (thumblike), small basiconic, coeloconic (short conical), and

106

S.H. Fuss, A. Ray / Molecular and Cellular Neuroscience 41 (2009) 101–112

trichoid (long tapering) sensilla that are found on the antenna, while the maxillary palp is covered by palp basiconics. The Drosophila Or genes were discovered in 1999 and comprise a family of 60 genes (Clyne et al., 1999b; Gao and Chess, 1999; Vosshall et al., 1999). Although they were initially identified by virtue of having a seven trans-membrane domain structure like their vertebrate counterparts, their trans-membrane organization shows a unique inside–out orientation. Therefore the insect odor receptor superfamilies belong to a unique family of proteins that are evolutionarily unrelated to any known proteins including vertebrate ORs or GPCRs in general. Analysis of Or expression demonstrated that a subset of Or genes are expressed in the antenna, while a different subset of seven Ors are exclusively expressed in the maxillary palps (Clyne et al., 1999b; Vosshall et al., 1999). Within the antenna, different Or genes segregate in different regions along the proximal–distal and dorsal–ventral axes reminiscent of zones in the vertebrate OE (Fig. 1B) (Vosshall et al., 1999). As in mammals, the zonal expression patterns are conserved between individual flies suggesting the existence of regulatory mechanisms that specify these patterns. In order to map Or gene expression with high resolution, transgenic fly lines containing Or upstream regions of variable lengths were analyzed (Couto et al., 2005; Dobritsa et al., 2003; Fishilevich and Vosshall, 2005; Gao et al., 2000; Goldman et al., 2005; Vosshall et al., 2000). It was found that relatively short regions of upstream DNA were necessary and sufficient to drive appropriate expression. Interestingly, for every Or promoter construct, reporter gene expression is restricted to one of the five morphological classes of sensilla (Fig. 1B). While the maxillary palps are covered exclusively by basiconic sensilla, in the antenna large basiconics, small basiconics, and trichoids are arranged progressively in a dorso-medial to ventrolateral pattern within specific spatial domains, and coeloconics are spread broadly (Shanbhag et al., 1999). Or gene expression in a specific morphological class of sensilla thus automatically restricts their expression to one of five zones in the olfactory organs. Interestingly, the different morphological classes of sensilla arise ontogenetically from the action of different combinations of proneural, and helix– loop–helix transcription factors. The proneural transcription factor atonal is required for the development of the palp basiconic and the antennal coeloconic sensilla (Gupta and Rodrigues, 1997), while amos is required for the development of the antennal basiconic and trichoid sensilla (Goulding et al., 2000; zur Lage et al., 2003). These observations suggest that the specification of Or gene expression within one of these five zones may be a consequence of developmental programs triggered by the activity of specific proneural genes. A novel family of odor receptors have recently been identified called ionotropic receptors (IRs) that are related to ionotropic glutamate receptors and respond to ammonia and phenylacetaldehyde (Benton et al., 2009). Interestingly the IRs are specifically expressed in the coeloconic sensilla, that arise from atonal positive precursors. Members of the Or gene family, with one exception, are expressed in the basiconic and trichoid sensilla and not in the coeloconics. Mechanisms of zonal expression in Drosophila Or upstream regulatory DNA can direct appropriate expression regardless of their genomic integration site, indicating that they do not require the long-range regulatory context of the endogenous locus. By using varying lengths of upstream genomic sequences it was demonstrated that 500 bps or less immediately upstream of the ATG, are necessary and sufficient to drive the appropriate expression patterns (Ray et al., 2007). Cis-regulation In order to uncover mechanisms that could impart zone-specific expression, the regulatory regions of the seven Or genes expressed in

Fig. 4. A summary of mechanisms of Or gene choice in Drosophila. A combinatorial code of positive and negative regulatory elements can dictate Or gene expression at the level of organ, as well as the level of individual neurons present within the maxillary palps (adapted from Ray et al., 2007). Dyad-1 promotes expression in the maxillary palp, while the Oligo-1 represses expression in the antenna. Asymmetric division leads to formation of the two neighboring neurons within a sensillum. A neuron-specific box (blue) promotes expression in 2 specific classes of neurons in the maxillary palp, while another neuron-specific box (orange), suppresses expression in one of the two classes thus restricting expression of the Or to one class of neuron.

the maxillary palp were analyzed to identify shared motifs that could be binding sites for zone-specific transcription factors (Ray et al., 2007). Two motifs called Dyad1 (CTA(N)9TAA) and Oligo1(CTTATAA) were identified to be highly overrepresented in promoters of maxillary palp Or genes. Mutations in the Dyad1 sequence resulted in the loss of expression suggesting that Dyad1 was required for expression in the maxillary palps. Mutation of the Oligo1 sequence, on the other hand, resulted in misexpression of maxillary palp Or genes in the antenna (Fig. 4). This strongly suggests that the presence of the Oligo1 motif specifically represses misexpression of palp Or genes presumably by binding of a transcriptional repressor present in the antennal ORNs. Thus these two cis-regulatory sites act together to ensure maxillary palp specific expression. It remains to be seen whether similar types of conserved cis-elements play a role in the specification of Or genes that are expressed in the other morphological classes of sensilla on the antenna. Regulation by transcription factors A number of transcription factors such as pdm3, lozenge, and scalloped have been identified to affect Or gene expression by forward genetics and candidate gene approaches, yet only the function of POU domain transcription factors acj6 and pdm3 have been examined in both organs (Ray et al., 2007, 2008; Tichy et al., 2008). Interestingly, flies mutant for both acj6 and pdm3 lack expression of specific subsets of Or genes in the antenna and the palps, while expression of other Or genes are unaffected (Clyne et al., 1999b; Komiyama et al., 2004). Organ-specific transcription factors that bind to the Dyad1 and Oligo1 elements await experimental identification. Neuron-specific expression in vertebrates Vertebrate ORs are not only restricted in their spatial expression profiles but also at the level of expression by individual OSNs. Each OSN is thought to express only a single gene from the entire OR repertoire (Bozza et al., 2002; Malnic et al., 1999; Mombaerts, 2004). This phenomenon is commonly referred to as ‘one neuron – one receptor’ hypothesis. Since expression of individual ORs are detected in approximately 0.05 to 0.2% of OSNs, it was suggested that each of

S.H. Fuss, A. Ray / Molecular and Cellular Neuroscience 41 (2009) 101–112

the 1300 ORs are expressed in specific non-overlapping subpopulation of OSNs to adequately cover all OSNs (Nef et al., 1992; Ressler et al., 1993; Vassar et al., 1993). Furthermore, single-cell RT-PCR usually amplifies a single OR transcript per OSN supporting the hypothesis (Chess et al., 1994; Malnic et al., 1999). However, the general validity of the ‘one neuron – one receptor’ hypothesis is hard to prove (for a critical review see Mombaerts, 2004). First, the size of the OR repertoire makes it impractical to investigate systematic co-expression of all possible OR combinations. Secondly, single-cell RT-PCR studies only comprise ‘snap-shots’ on a very small proportion of OSNs and even so, often fail to amplify any OR gene (Malnic et al., 1999). Co-expression of two OR genes Violations of the hypothesis have been reported. Systematic coexpression is observed for three linked zebrafish OR genes (Sato et al., 2007) and for two unlinked rat ORs (Rawson et al., 2000). Coexpression of mouse OR genes, particularly in immature OSNs, has been reported in the septal organ (Tian and Ma, 2008). Co-expression of the SR1 receptor with eight other ORs can be as high as 2% in newborn animals, but drops significantly to a basal rate of 0.2% around 4 weeks postnatally. High rates of co-expression can be maintained or restored in animals that are deprived of sensory input suggesting some form of activity-dependent elimination of co-expressing OSNs (Tian and Ma, 2008). Mono-allelic expression Surprisingly an OSN expresses either the maternal or the paternal allele of an OR gene but not both. Thus, OR genes are subject to random monoallelic expression (Chess et al., 1994; Gimelbrant et al., 2007; Ishii et al., 2001). In contrast to monogenic expression, monoallelic expression has been demonstrated rigorously, (Li et al., 2004; Mombaerts, 1996); but see (Serizawa et al., 2000), (Shykind et al., 2004). Allelic exclusion extends beyond endogenous ORs and can also be seen among transgenic alleles in most OSNs (Ebrahimi et al., 2000; Serizawa et al., 2000). The combination of monogenic and monoallelic expression has been termed ‘singular’ expression (Vassalli et al., 2002) but it is not known whether they arise from a common mechanism or from two independent processes. A cell could first choose an OR that is competent for expression within a given region of the OE, and subsequently select one allele for expression. Alternatively, monogenic and monoallelic expression could be directly linked if the OR choice mechanism treats both alleles from every OR independently. Thus a cell would choose amongst 2000 alleles, instead of 1000 ORs. Finally the choice of an allele could precede monogenic expression, particularly if the mechanism of OR gene choice would execute at the level of OR clusters that are controlled by LCRs. In this scenario one parental copy of the cluster is selectively rendered competent for expression prior to the selection of an OR from the cluster (Fig. 3B). Interestingly, methylation of one allele of the H-region has been observed, which could support such a mechanism (Lomvardas et al., 2006). OR gene loci replicate asynchronously, but in a coordinate fashion with other monoallelically expressed loci on the same chromosome (Singh et al., 2003). Asynchronous replication is a hallmark of the X chromosome, imprinted genes and genes that are subject to random monoallelic expression (Gimelbrant et al., 2007; Goldmit and Bergman, 2004) and is supposed to be involved in the random selection of one allele. However, in contrast to X-inactivation and imprinted genes, an early and irreversible silencing of OR alleles is not likely since both alleles can be expressed if one allele fails to produce a functional OR protein (Feinstein et al., 2004; Lewcock and Reed, 2004; Serizawa et al., 2003; Shykind et al., 2004).

107

No evidence for recombination based mechanisms Both the function and diversity of the olfactory system bears close resemblance to the immune system where diversity is generated by the process of V(D)J recombination. In order to understand whether OR choice involves similar DNA recombination events in OSNs, two independent groups successfully cloned mice from mature OSN nuclei (Eggan et al., 2004; Li et al., 2004). Surprisingly, OE from cloned mice expressed the full OR repertoire, arguing against irreversible DNA recombination as a mechanism for the process of OR gene choice (Fig. 2). By analogy, B-lymphocytes in animals cloned from mature B-cells nuclei are monoclonal as they express only one type of antibody (Hochedlinger and Jaenisch, 2002). Gene conversion is a type of recombination during which a copy of the expressed sequence is translocated into an expression locus and it does not necessarily require irreversible DNA changes (Palmer and Brayton, 2007). However, the consequence of such a mechanism would be the presence of a third copy of the expressed OR gene, which was not observed for the few OR genes that were directly tested (Ishii et al., 2001; Lomvardas et al., 2006). No evidence for trans-chromosomal mechanisms Another intriguing OR gene choice mechanism was proposed that is based on inter-chromosomal interactions between a single active copy of the H-region on chromosome 14 and various OR promoters residing on different chromosomes (Lomvardas et al., 2006). Since the H-region was presumed to physically interact with only one promoter at a time, such a process would thus ultimately ensure singular OR expression. Although this mechanism seemed very attractive at the time, subsequently two independent studies disputed the proposed trans-regulatory function (Fuss et al., 2007; Nishizumi et al., 2007). Deletion of the H-region or its functional core resulted in a complete loss of expression of three genes from the downstream MOR28 cluster but did not abolish expression of other OR genes outside the cluster, either on the same, or on any other chromosome. These observations indicated that the H-region does not play any trans-regulatory role, but rather is a long-range cis-regulatory site coordinating the expression of specific ORs. However, the observed co-localization of the H-region and the expressed OR locus in the nucleus (Lomvardas et al., 2006) suggests the existence of an active nuclear site that could potentially recruit OR promoters to coordinate their expression and it is conceivable that the H sequence bears a high affinity for this site. Receptor switching The apparent random distribution of OR expression within a zone (Fig. 1A) has fueled the idea that the mechanism of OR gene choice is stochastic as well (Shykind et al., 2004). However, the equal competence of multiple OR genes to be chosen by the same OSN requires subsequent mechanisms to stabilize this choice and prevent random switching of OR expression in mature OSNs that have established stable synaptic connections. Lineage tracing experiments using Cre-recombinase expression from the MOR28 and P2 locus show, however, that switching of OR expression can occur in immature OSNs (Shykind et al., 2004). In the case of MOR28, about 10% of OSNs labeled by the lineage tracer were found to be positive for other ORs from the same zone, but not for Cre-recombinase suggesting that these OSNs at some point expressed the MOR28-Cre allele but had now switched. Interestingly, 10% of switching cells expressed the wildtype MOR28 allele in heterozygous animals suggesting that at the time of OR choice, both alleles have equal competence for expression, as do other ORs. However, the results leave unresolved whether two different OR loci can be transcriptionally active at the same time. Contradictory results were obtained for the M71 locus where switching could not be detected (Li et al., 2004).

108

S.H. Fuss, A. Ray / Molecular and Cellular Neuroscience 41 (2009) 101–112

Negative-feedback regulation How is switching, however, precluded in most OSNs? A ‘negativefeedback’ signal has been proposed that prevents continuation of the OR choice process as soon as a functional OR is expressed (Lewcock and Reed, 2004; Serizawa et al., 2003; Shykind et al., 2004). OSNs that fail to express a functional OR protein, either because the expressed OR is a pseudogene or the coding sequence has been deleted experimentally do not receive the feedback signal and continue to choose other ORs for expression (Fig. 3B). The set of ORs available for second choice may not be random as judged by the unique but distinct glomerular projection patterns of OSNs expressing different OR deletions (Bozza et al., 2009; Feinstein and Mombaerts, 2004; Serizawa et al., 2003; Shykind et al., 2004). Interestingly, second choice appears to be restricted to the class of ORs that are initially expressed by the OSN: cells expressing a non-functional class I locus exclusively express other class I ORs as second choice, while OSNs expressing a non-functional class II locus make their second choices among other class II ORs (Bozza et al., 2009). Thus OR gene choice may be determined by lineage, at least for the class I/class II distinction. For the class II ORs which cover all zones of the OE, ORs residing within the same zone were supposed to be targets for second choice. Yet, it remains unresolved whether these OR subsets could as well be predetermined by different OSN lineages with different prevalence for certain ORs. Since LCRs play a role in expression of certain ORs, second choice made by an OSN could also be restricted to genes controlled by the same LCR, which appears to be true for at least one locus in zebrafish (Sato et al., 2007). This hypothesis is also supported by the observation that cells expressing a non-functional OR from a M4 promoter transgene that has been directionally integrated close to the M71 locus by gene-targeting, co-express M71 in a high number of cells (Lewcock and Reed, 2004). On the other hand, a systematic analysis of transgenic mice carrying a non-functional MOR28 revealed that a variety of unlinked ORs can be expressed, with only 5% of OSNs expressing the three linked genes that are strictly dependent on the Hregion (Serizawa et al., 2003). In contrast, vomeronasal neurons expressing a deleted V1R vomeronasal receptor choose to express V1R genes from the same cluster but not from the parental copy of the cluster that carries the deletion (Roppolo et al., 2007). It is tempting to speculate that the subset of ORs expressed as second choice in OR deletions bears the key to the mystery of OR gene choice. However, second choice may not be a mere recapitulation of the regular OR gene choice process. The signal that emanates from the expressed OR and prevents further choices remains equally enigmatic. Two mechanisms are conceivable: one that stabilizes transcription from the active OR promoter or LCR at the expense of silent OR promoters, or maturation of OSNs beyond a stage at which OR gene choice is possible (Fig. 3B). The latter is in agreement with the observation that switching occurs predominantly in immature OSNs (Shykind et al., 2004). It is unlikely that odorant-induced activity plays a role in signaling OR gene choice. Mice deficient in various components of the odorant signaling cascade such as Gαolf (Belluscio et al., 1998), OCNC1 (Zhao and Reed, 2001; Zheng et al., 2000) or ACIII (Wong et al., 2000), which are generally anosmic, do not show grossly abnormal OR expression. Similarly, deletion of the G-protein binding motif in a transgenic I7 receptor did not perturb its expression but prevented axonal convergence of transgenic OSNs (Imai et al., 2006). Yet, these experiments do not exclude alternative OR-mediated signaling pathways. Astonishingly, negative-feedback regulation is not limited to OR promoters and extends to expression from synthetic promoters that express an OR coding sequence as well (Fleischmann et al., 2008; Nguyen et al., 2007). Using the bipartite tetracycline transactivator system to drive OR expression from a TetO-promoter, it was observed that OR expression from the transgene was repressed by endogenous

ORs. This surprising observation suggests that the OR coding sequence itself could be a target of the feedback signal. OR protein sequences exhibit several highly conserved sequence motifs (Zhang and Firestein, 2002) that could be targets for the feedback signal but only the G-protein binding motif has yet been scrutinized experimentally. Conversely, transgenic expression of an OR protein from the ubiquitous promoter can block endogenous OR expression across the OE if it precedes onset of endogenous OR expression (Fleischmann et al., 2008; Nguyen et al., 2007). These results suggest that the OR first expressed by an OSN prevents expression of other ORs in most OSNs and thereby ensures monogenic OR expression. Neuron-specific expression in Drosophila Although the receptor proteins are evolutionarily unrelated, Drosophila has evolved regulation strategies that by and large appear to ensure a similar outcome as mice: one-receptor-per-neuron. Drosophila Or genes are restricted to one of the five types of sensilla and are thus restricted to one of five overlapping spatial zones. Detailed functional analysis revealed that these sensilla within a zone could be further classified into subtypes based on their odor response profiles. The ∼60 sensilla on the maxillary palps could be divided into three functional types, each containing a specific pair of neurons with distinct odor responses (de Bruyne et al., 1999). In a similar way, multiple functional types could be described for the large and small basiconic sensilla (de Bruyne et al., 2001), trichoid sensilla (Clyne et al., 1997; van der Goes van Naters and Carlson, 2007), and coeloconic sensilla (Yao et al., 2005). Each subtype was found to contain stereotyped combinations of 1–4 neurons that respond to characteristic odorants, suggesting that specific combinations of Or genes were expressed by the ORNs from each sensillar subtype. Expression of individual Or genes in each of the distinct functional classes of neurons were concomitantly mapped using a variety of different strategies including analysis of Or mutants (Dobritsa et al., 2003; Elmore et al., 2003), electrophysiological recordings from ORNs expressing GFP or a cell-death gene driven by individual Or promoters (Goldman et al., 2005), functional mapping of receptors using an invivo expression system (Hallem et al., 2004), and double label analysis of receptor expression (Couto et al., 2005; Fishilevich and Vosshall, 2005). Using these approaches the complete map of Or gene expression in individual classes of olfactory neurons was established. With the exception of the non-canonical Or83b gene which encodes a required co-receptor (Benton et al., 2006; Larsson et al., 2004) that is co-expressed in most ORNs, individual Or expression is restricted to one functional class of neuron. The odor response spectrum of a neuron class depends primarily on the expressed Or gene (Hallem et al., 2004) and paired neurons within each sensillum express a stereotypic combination of Or genes. Functionally identical sensilla appear to be stochastically distributed among the confines of a zone (Fig. 1B), reminiscent of vertebrate zonal restriction and random OR distribution. However important differences in OR regulation between Drosophila and vertebrates exist. Co-ordination of Or gene expression in paired neurons of a sensillum The events generating the stereotyped pairing of Ors expressed within the same sensilla have been investigated in some detail. A typical sensillum contains one pair of neurons and three supporting cells that make up the shaft, the socket and the sheath. Recent studies using mutants for components of the Notch signaling pathway demonstrate that asymmetric cell divisions specify the two types of neurons within a sensillum (Endo et al., 2007). It was noticed that in mastermind mutants, that have low-Notch activity, both neurons express only one of the two Or genes that is usually expressed in the sensillum. Conversely, in numb mutants, with high-Notch activity, both neurons express the other Or of the pair. Furthermore, over-

S.H. Fuss, A. Ray / Molecular and Cellular Neuroscience 41 (2009) 101–112

109

expression of mastermind across the maxillary palp leads to the ectopic expression of one of the Or genes in both neurons in a sensillum (presumably the high-Notch receptor) (Ray et al., 2007). This asymmetric division process where one sibling adopts a high and the other low-Notch fate could trigger a cascade of downstream events that leads to the ultimate precisely coordinated expression of the appropriate Or gene.

in the maxillary palp revealed an even distribution of all three types with little spatial segregation (Fig. 1A). This random pattern is reminiscent of the pattern observed in mammalian olfactory systems. Yet, how is such a precise and seemingly stochastic expression pattern achieved without the contribution of negative-feedback signaling?

Lack of negative-feedback regulation In mammals, the expression of a functional receptor can inhibit the expression of other receptors in the same neuron (Fleischmann et al., 2008; Nguyen et al., 2007; Lewcock and Reed, 2004; Serizawa et al., 2003; Shykind et al., 2004). However, three observations argue against a feedback mechanism in Drosophila ORNs. First, Or-promoter constructs are co-expressed along with the endogenous Or (Dobritsa et al., 2003; Goldman et al., 2005), suggesting that the endogenous Or is unable to repress expression of transgenic alleles. Second, flies containing an empty Or locus contain neurons that do not respond to odor stimulation but still connect to their cognate glomerulus (Dobritsa et al., 2003). Thus, secondary Or choice is also unlikely to occur in flies. And finally, ectopic expression of an Or gene during early development, before onset of the endogenous Or, does not preclude expression of endogenous Ors (Ray et al., 2007).

Evolutionary conservation of gene-specific cis-regulatory elements across 12 Drosophila species was employed to unveil candidate regulatory motifs in the promoters of maxillary palp Or genes that direct expression in specific neuronal subtypes (Ray et al., 2008). Mutational analysis in several Or regulatory regions revealed that individual Or promoters contained a positive cis-regulatory element necessary for expression. Interestingly, lack of a different class of candidate cis-regulatory elements in transgenic promoter constructs resulted in misexpression in 1 or 2 additional classes of neurons. These observations suggest that activating transcription factors specific for each Or gene can be present more broadly in different neuronal classes, yet Or expression is restricted to a particular neuronal class by repression of transcription in inappropriate neurons. Thus each Or contains a unique combinatorial code of positive and negative regulation that can specify the precise neuron-specific expression patterns of the individual Or genes in the maxillary palps (Fig. 4).

Mechanisms of co-expression of Or genes

Transcription factor code for neuron-specific expression

It is generally believed that the majority of mammalian olfactory neurons express only a single receptor per neuron (Mombaerts, 2004). In contrast, expression mapping of the entire Or gene family in Drosophila revealed that at least 20% of ORN subtypes express two canonical receptors (Couto et al., 2005; Fishilevich and Vosshall, 2005; Goldman et al., 2005). In the majority of cases the two co-expressed Ors are the closest homologues and may have arisen by recent gene duplication events. The observed conservation of duplicated regulatory regions, in combination with the lack of an exclusion mechanism such as “negative feedback”, may be responsible for co-expression of these genes by the same neurons (Ray et al., 2007). In addition, a number of co-expressing Or homologues are tightly clustered in the genome, sometimes with less than 100 nucleotides separating their coding regions (Couto et al., 2005; Fishilevich and Vosshall, 2005; Goldman et al., 2005). For some of them, alternatively spliced transcripts that share common 5′-exons and rare bi-cistronic messengers were detected that caused the apparent co-expression (Ray et al., 2007). In a special case, two unlinked Or genes Or33c and Or85e that are not close homologs, are expressed together in the same neuron type. It has been shown that the upstream regulatory regions of these genes contain two shared cis-regulatory motifs that are not present in the regulatory regions of any other Or genes expressed in the maxillary palp (Ray et al., 2007). Transgenes containing a mutation in one of these motifs (AGTTTTTA) were not expressed, suggesting that the motif is required for expression. Thus in addition to zone-specific elements that specify expression in the maxillary palp, neuronspecific cis-motifs may direct co-expression of two Or genes in the same neuron.

The identification of transcription factors that bind to the identified sequence motifs discussed above awaits further experimentation. However in parallel studies, forward genetic and candidate gene approaches have shown that the POU domain transcription factors, acj6 and pdm3 are required for Or gene expression (Clyne et al., 1999a,b; Komiyama et al., 2004; Tichy et al., 2008). Acj6 is expressed in all maxillary palp neurons and is required for the expression of a subset of Or genes in the palps, as well as in the antenna (Clyne et al., 1999a,b; Komiyama et al., 2004). On the other hand, pdm3 is expressed in a subset of the ORN classes and is required for the expression of a very limited subset of Or genes, one (out of 7 tested) in the maxillary palp, and one (out of 8 tested) in the antenna (Tichy et al., 2008). Binding sites for the AML-1/Runt-like transcription factor lozenge were also predicted to be present in the upstream and downstream regulatory region of some Or genes (Ray et al., 2007). Interestingly, those maxillary palp Or genes that contain binding sites for lozenge show reduced expression levels in lozenge mutant flies. Furthermore, complementary sets of Or genes appear to be regulated by acj6 and lozenge. The TEA domain transcription factor scalloped has also been identified to play a role as repressor. Or59c is usually expressed in the A neuron of the pb3 sensilla type found on the maxillary palps. In scalloped mutant flies, Or59c is additionally mis-expressed in the neighboring B neuron as well. Thus scalloped, which is expressed in the pb3B neuron class, represses the expression of Or59c and thus refines its expression to the pb3A neuron. (Ray et al., 2008).

Punctate expression patterns The majority of olfactory neurons in Drosophila however express just one canonical receptor per neuron. Since the two Or genes expressed within the two neurons present in one sensillum are paired stereotypically, labeling of just one Or from each sensilla type reveals the distribution of that subtype within a zone. Triple labeling experiments (Ray et al., 2008) and electrophysiological recordings (de Bruyne et al., 1999) to study the distribution of sensillar subtypes

Cis-regulatory code for neuron-specific expression

Conclusions The function of the olfactory system is to provide an organism with the ability to detect and to discriminate between a vast and diverse set of volatile chemical cues from the environment. Fascinatingly, insects and vertebrates have developed olfactory systems that share many anatomical and functional similarities and employ related strategies to process olfactory information. For instance, both systems are equipped with a multitude of distinct neuronal cell populations that are specified by their selective expression of one or a few olfactory receptors, which in turn impart responsiveness to odorants. Neurons that express the same receptor converge onto one or a few glomeruli

110

S.H. Fuss, A. Ray / Molecular and Cellular Neuroscience 41 (2009) 101–112

as they project to the OB or antennal lobe, respectively. These anatomical similarities between the fly and vertebrate olfactory system are particularly fascinating since the molecules critical for olfactory function, their olfactory receptors, are evolutionary unrelated, suggesting that convergent evolutionary trends have shaped similar olfactory systems in such disparate species. A comparison of the different processes reveals that both systems may use certain related strategies by establishing defined spatial expression zones. Yet, they utilize very different strategies to specify receptor gene expression in individual neurons. The most striking similarity appears to be the potential use of deterministic mechanisms to restrict expression of receptor genes within defined spatial zones. This restriction may help to simplify the process of neuronal specification by reducing the repertoire of receptor genes from which a sensory neuron in any given part of the sensory surface can choose. The most striking differences can be seen in the strategies that lie at the heart of the problem: how does the olfactory system specify expression of a single receptor per neuron? It appears that both systems have come up with different solutions to this problem. Drosophila largely employs deterministic principles through specific combinations of cis-elements and transcription factors in addition to novel use of the Notch-dependent asymmetric cell division pathway to generate diversity in receptor gene expression. Vertebrates on the other hand have developed a yet unknown powerful stochastic choice process in combination with feedback regulation in order to cope with the greatly increased number of possible choices. In general, constrains on the regulation of receptor gene choice appear to be more restrictive in vertebrates than in Drosophila. Coexpression of two odorant receptors in the same neuron seems to be a rare occurrence in vertebrates, but a lot more common in Drosophila. This could be a reflection of the vastly expanded receptor repertoire found in vertebrates. Drosophila expresses only around 43 Ors in the adult animal, while mammals have to coordinate the expression of up to 1500 OR genes that have arisen by numerous gene duplication and subsequent diversification events. While increasing the number of olfactory receptors may make the system receptive to a larger range of odorants, the concomitant anatomical separation of uniquely specified OSN populations may greatly enhance the ability to discriminate between different stimuli (Fleischmann et al., 2008). The acquisition of multiple new OR genes in mammals may have required their concomitant expression in novel and distinct neuronal populations, in order to maintain appropriate discriminatory power. These critical requirements may have led vertebrates to adopt additional mechanisms to ensure monospecificity in OSNs, including mechanisms that ensure the exclusion of a potentially polymorphic allele of the same OR. In Drosophila it has been shown that the precise and stereotypical patterning of the olfactory system, whereby a single ORN expresses a unique receptor, depends largely on a combinatorial code of transcriptional activation and repression (Fig. 4). Individual receptors can interpret this sophisticated code using unique combinations of cis-regulatory modules that can read these instructions. The pattern of Or gene expression is well conserved across a number of evolutionarily related Drosophila species. This conservation has been exploited to unravel the cis-regulatory code for most of the Ors expressed in the maxillary palps. Based on these findings a model for Or gene choice has been proposed (Fig. 4) which is a modification of the combinatorial coding model (Fig. 2): a combinatorial code of activators and repressors that are expressed in overlapping domains, can precisely set up restricted ORN types, in which a given Or with the appropriate combination of cis-modules can be expressed. There appear to be progressive levels of specification that successively simplify the complex regulatory problem in Drosophila and reduce it to a manageable level. First, ∼ 43 of 60 Ors are expressed in the adult fly, while the remaining ones are expressed in the larval stage. Second, expression of specific subsets of Or genes are restricted

to the morphological classes of sensilla such that only a small subset of Or genes (7–15) are required to be specified for in each class. Third, Notch-dependent asymmetric cell divisions further restricts the choice of an Or gene within neighboring neurons of the same sensillum, thus only a relatively small number of distinct progenitor populations have to be specified within each morphological type of sensilla. The final output of this developmental program, specific expression of one or two Or genes in a defined sensory neuron, is thought to be achieved through an intricate interplay of multiple positive and negative regulatory events. Therefore, the process of receptor gene expression in Drosophila appears to be largely deterministic, while it may be a combination of deterministic and stochastic processes in vertebrates. In vertebrates, the subset of olfactory receptors that can be expressed in any part of the OE may be specified by deterministic processes using characteristic cis-regulatory motifs present in OR promoters (Fig. 3A). However the ultimate choice between several ORs that may be equally competent for expression in a given OSN, appears to be governed by a stochastic process. Several conserved cisregulatory motifs that could potentially instruct zonal expression in vertebrates have begun to be discovered for ORs expressed in unique spatial patterns such as patch and class I ORs. A similar knowledge of conserved motifs, however, is missing for the majority of ORs, particularly those that are expressed in continuous and overlapping patterns across the OE. Vertebrate OR choice appears to be controlled by short-range cisacting mechanisms in some cases, while it employs long-range control in others, such that no coherent picture emerges. So far, the H-region directing expression from the MOR28 cluster is the sole LCR that has been functionally identified in mammals. Multiple LCRs may be present in a zebrafish OR cluster (Nishizumi et al., 2007) and a second LCR has been predicted to control expression of the distal genes in the MOR28 cluster (Fuss et al., 2007). We can not comprehend whether this type of regulation is the norm or an exception until other longrange regulators can be identified for other OR clusters. The overlap of the MOR28 cluster with the T-cell receptor α locus may demand an unusual type of chromatin remodeling that is not required for other OR loci. A widespread function of LCRs acting as crucial intermediates in a stepwise selection of OR gene expression, however, would reduce the complexity of the OR choice problem at least by one or two orders of magnitude, depending on the number of LCRs that are present within each cluster. The mechanisms that render one out of many OR promoters (or one out of a few LCRs) competent for expression, remain elusive. The best understood sequence of events leading to exclusive OR expression has been established for the regulation at the H-MOR28 locus (Fig. 3B). An early step in this process may be the silencing of one copy of the cluster by methylation (Lomvardas et al., 2006). Since the Hregion operates strictly in cis (Fuss et al., 2007, Nishizumi et al., 2007), this directly ensures monoallelic expression of all coordinately regulated ORs. Second, since the H-region is believed to interact with only one OR promoter from the cluster at a time, it selects a single OR from a larger repertoire (Fig. 3A). The outcome would be the monogenic expression of one cluster OR (Serizawa et al., 2003). Third, expression of a functional OR protein then elicits an unknown signal that finalizes the process of OR gene choice. This signaling mechanism remains enigmatic to date, but could operate through epigenetic changes in silent or active OR promoters, or act to promote the terminal differentiation of OSNs and subsequent shutdown of the OR choice cascade similar to events in lymphocytes (Fig. 3B). However, it has been observed that ORs from both parental alleles can be expressed if expression from one allele fails to produce a functional OR protein. This raises the question, whether the methylation step and concomitant silencing of one copy would occur subsequent to OR gene choice. In order to account for switching of alleles, methylation would thus be a part of the feedback signal rather than an early step in

S.H. Fuss, A. Ray / Molecular and Cellular Neuroscience 41 (2009) 101–112

OR choice. Lastly, aberrantly specified OSNs that express more than a single OR or ORs from a different zone may potentially be eliminated by negative selection by their inability to establish appropriate synaptic connections in the OB (Mombaerts, 2004). Despite the progress made in recent years important questions regarding the process of odorant receptor choice remains a mystery. How is low-frequency activation of select ORs and zonal restriction achieved in vertebrates? How much does zonal restriction contribute to receptor gene choice by simplifying the choice process? How widespread is the presence of LCRs in receptor gene clusters? The nature and the pathway of the feedback signal remain equally enigmatic. Which signaling molecules participate in “negative feedback”? What type of proteins bind to the various cis-regulatory elements or to LCRs and regulate expression of receptors? While we have made considerable progress in understanding details about the rules followed by the sophisticated process of olfactory receptor gene choice, the underlying molecular mechanisms remain largely elusive. Acknowledgments S.H.F acknowledges past and ongoing grant support by NIH (1R03DC007975-01A1), the Bogazici University Research Fund (BAP, 07HB111) and TÜBITAK (107T760). A.R was supported by University of California, Riverside. The authors are grateful to Anupama Dahanukar for critically reading the manuscript. References Alioto, T.S., Ngai, J., 2005. The odorant receptor repertoire of teleost fish. BMC Genomics 6, 173. Baier, H., Korsching, S., 1994. Olfactory glomeruli in the zebrafish olfactory system form an invariant pattern and are identifiable across animals. J. Neurosci. 14, 219–230. Belluscio, L., Gold, G.H., Nemes, A., Axel, R., 1998. Mice deficient in G(olf) are anosmic. Neuron 20, 69–81. Benton, R., Sachse, S., Michnick, S.W., Vosshall, L.B., 2006. Atypical membrane topology and heteromeric function of Drosophila odorant receptors in vivo. PLoS Biol. 4, e20. Benton, R., Vannice, K.S., Gomez-Diaz, C., Vosshall, L.B., 2009. Variant ionotropic glutamate receptors as chemosensory receptors in Drosophila. Cell 136, 149–162. Bozza, T., Feinstein, P., Zheng, C., Mombaerts, P., 2002. Odorant receptor expression defines functional units in the mouse olfactory system. J. Neurosci. 22, 3033–3043. Bozza, T., Vassalli, A., Fuss, S., Zhang, J.J., Weiland, B., et al., 2009. Mapping of class I and class II odorant receptors to glomerular domains by two distinct types of olfactory sensory neurons in the mouse. Neuron 61, 220–233. Buck, L.B., 2000. The molecular architecture of odor and pheromone sensing in mammals. Cell 100, 611–618. Buck, L., Axel, R., 1991. A novel multigene family may encode odorant receptors: a molecular basis for odor recognition. Cell 65, 175–187. Buiakova, O.I., Krishna, N.S., Getchell, T.V., Margolis, F.L., 1994. Human and rodent OMP genes: conservation of structural and regulatory motifs and cellular localization. Genomics 20, 452–462. Bulger, M., van Doorninck, J.H., Saitoh, N., Telling, A., Farrell, C., et al., 1999. Conservation of sequence and structure flanking the mouse and human beta-globin loci: the beta-globin genes are embedded within an array of odorant receptor genes. Proc. Natl. Acad. Sci. U. S. A. 96, 5129–5134. Chess, A., Simon, I., Cedar, H., Axel, R., 1994. Allelic inactivation regulates olfactory receptor gene expression. Cell 78, 823–834. Clyne, P., Grant, A., O'Connell, R., Carlson, J.R., 1997. Odorant response of individual sensilla on the Drosophila antenna. Invertebr. Neurosci. 3, 127–135. Clyne, P.J., Certel, S.J., de Bruyne, M., Zaslavsky, L., Johnson, W.A., Carlson, J.R., 1999a. The odor specificities of a subset of olfactory receptor neurons are governed by Acj6, a POU-domain transcription factor. Neuron 22, 339–347. Clyne, P.J., Warr, C.G., Freeman, M.R., Lessing, D., Kim, J.H., Carlson, J.R., 1999b. A novel family of divergent seven-transmembrane proteins: candidate odorant receptors in Drosophila. Neuron 22, 327–338. Couto, A., Alenius, M., Dickson, B.J., 2005. Molecular, anatomical, and functional organization of the Drosophila olfactory system. Curr. Biol. 15, 1535–1547. Dahanukar, A., Hallem, E.A., Carlson, J.R., 2005. Insect chemoreception. Curr. Opin. Neurobiol. 15, 423–430. de Bruyne, M., Clyne, P.J., Carlson, J.R., 1999. Odor coding in a model olfactory organ: the Drosophila maxillary palp. J. Neurosci. 19, 4520–4532. de Bruyne, M., Foster, K., Carlson, J., 2001. Odor coding in the Drosophila antenna. Neuron 30, 537–552. Dobritsa, A.A., van der Goes van Naters, W., Warr, C.G., Steinbrecht, R.A., Carlson, J.R., 2003. Integrating the molecular and cellular basis of odor coding in the Drosophila antenna. Neuron 37, 827–841. Ebrahimi, F.A.W., Edmondson, J., Rothstein, R., Chess, A., 2000. YAC transgene-mediated olfactory receptor gene choice. Dev. Dyn. 217, 225–231.

111

Eggan, K., Baldwin, K., Tackett, M., Osborne, J., Gogos, J., et al., 2004. Mice cloned from olfactory sensory neurons. Nature 428, 44–49. Elmore, T., Ignell, R., Carlson, J.R., Smith, D.P., 2003. Targeted mutation of a Drosophila odor receptor defines receptor requirement in a novel class of sensillum. J. Neurosci. 23, 9906–9912. Endo, K., Aoki, T., Yoda, Y., Kimura, K., Hama, C., 2007. Notch signal organizes the Drosophila olfactory circuitry by diversifying the sensory neuronal lineages. Nat. Neurosci. 10, 153–160. Feinstein, P., Mombaerts, P., 2004. A contextual model for axonal sorting into glomeruli in the mouse olfactory system. Cell 117, 817–831. Feinstein, P., Bozza, T., Rodriguez, I., Vassalli, A., Mombaerts, P., 2004. Axon guidance of mouse olfactory sensory neurons by odorant receptors and the beta 2 adrenergic receptor. Cell 117, 833–846. Fishilevich, E., Vosshall, L.B., 2005. Genetic and functional subdivision of the Drosophila antennal lobe. Curr. Biol. 15, 1548–1553. Fleischmann, A., Shykind, B.M., Sosulski, D.L., Franks, K.M., Glinka, M.E., et al., 2008. Mice with a “monoclonal nose”: perturbations in an olfactory map impair odor discrimination. Neuron 60, 1068–1081. Fuss, S.H., Omura, M., Mombaerts, P., 2007. Local and cis effects of the H element on expression of odorant receptor genes in mouse. Cell 130, 373–384. Gao, Q., Chess, A., 1999. Identification of candidate Drosophila olfactory receptors from genomic DNA sequence. Genomics 60, 31–39. Gao, Q., Yuan, B., Chess, A., 2000. Convergent projections of Drosophila olfactory neurons to specific glomeruli in the antennal lobe. Nat. Neurosci. 3, 780–785. Gimelbrant, A., Hutchinson, J.N., Thompson, B.R., Chess, A., 2007. Widespread monoallelic expression on human autosomes. Science 318, 1136–1140. Goldman, A.L., Van der Goes van Naters, W., Lessing, D., Warr, C.G., Carlson, J.R., 2005. Coexpression of two functional odor receptors in one neuron. Neuron 45, 661–666. Goldmit, M., Bergman, Y., 2004. Monoallelic gene expression: a repertoire of recurrent themes. Immunol. Rev. 200, 197–214. Goulding, S.E., zur Lage, P., Jarman, A.P., 2000. amos, a proneural gene for Drosophila olfactory sense organs that is regulated by lozenge. Neuron 25, 69–78. Gupta, B.P., Rodrigues, V., 1997. Atonal is a proneural gene for a subset of olfactory sense organs in Drosophila. Genes Cells 2, 225–233. Hallem, E.A., Ho, M.G., Carlson, J.R., 2004. The molecular basis of odor coding in the Drosophila antenna. Cell 117, 965–979. Hallem, E.A., Dahanukar, A., Carlson, J.R., 2006. Insect odor and taste receptors. Annu. Rev. Entomol. 51, 113–135. Hirota, J., Mombaerts, P., 2004. The LIM-homeodomain protein Lhx2 is required for complete development of mouse olfactory sensory neurons. Proc. Natl. Acad. Sci. U. S. A. 101, 8751–8755. Hirota, J., Omura, M., Mombaerts, P., 2007. Differential impact of Lhx2 deficiency on expression of class I and class II odorant receptor genes in mouse. Mol. Cell. Neurosci. 34, 679–688. Hochedlinger, K., Jaenisch, R., 2002. Monoclonal mice generated by nuclear transfer from mature B and T donor cells. Nature 415, 1035–1038. Hoppe, R., Weimer, M., Beck, A., Breer, H., Strotmann, J., 2000. Sequence analyses of the olfactory receptor gene cluster mOR37 on mouse chromosome 4. Genomics 66, 284–295. Hoppe, R., Frank, H., Breer, H., Strotmann, J., 2003. The clustered olfactory receptor gene family 262: genomic organization, promotor elements, and interacting transcription factors. Genome Res. 13, 2674–2685. Hoppe, R., Breer, H., Strotmann, J., 2006. Promoter motifs of olfactory receptor genes expressed in distinct topographic patterns. Genomics 87, 711–723. Imai, T., Suzuki, M., Sakano, H., 2006. Odorant receptor-derived cAMP signals direct axonal targeting. Science 314, 657–661. Ishii, T., Serizawa, S., Kohda, A., Nakatani, H., Shiroishi, T., et al., 2001. Monoallelic expression of the odorant receptor gene and axonal projection of olfactory sensory neurones. Genes Cells 6, 71–78. Iwema, C.L., Fang, H., Kurtz, D.B., Youngentob, S.L., Schwob, J.E., 2004. Odorant receptor expression patterns are restored in lesion-recovered rat olfactory epithelium. J. Neurosci. 24, 356–369. Kobayakawa, S., Miike, K., Nakao, M., Abe, K., 2007. Dynamic changes in the epigenomic state and nuclear organization of differentiating mouse embryonic stem cells. Genes Cells 12, 447–460. Kolterud, A., Alenius, M., Carlsson, L., Bohm, S., 2004. The Lim homeobox gene Lhx2 is required for olfactory sensory neuron identity. Development 131, 5319–5326. Komiyama, T., Carlson, J.R., Luo, L.Q., 2004. Olfactory receptor neuron axon targeting: intrinsic transcriptional control and hierarchical interactions. Nature. Neuroscience. 7, 819–825. Kubick, S., Strotmann, J., Andreini, I., Breer, H., 1997. Subfamily of olfactory receptors characterized by unique structural features and expression patterns. J. Neurochem. 69, 465–475. Kudrycki, K., Stein-Izsak, C., Behn, C., Grillo, M., Akeson, R., Margolis, F.L., 1993. Olf-1binding site: characterization of an olfactory neuron-specific promoter motif. Mol. Cell. Biol. 13, 3002–3014. Lane, R.P., Roach, J.C., Lee, I.Y., Boysen, C., Smit, A., et al., 2002. Genomic analysis of the olfactory receptor region of the mouse and human T-cell receptor alpha/delta loci. Genome Res. 12, 81–87. Larsson, M.C., Domingos, A.I., Jones, W.D., Chiappe, M.E., Amrein, H., Vosshall, L.B., 2004. Or83b encodes a broadly expressed odorant receptor essential for Drosophila olfaction. Neuron 43, 703–714. Lewcock, J.W., Reed, R.R., 2004. A feedback mechanism regulates monoallelic odorant receptor expression. Proc. Natl. Acad. Sci. U. S. A. 101, 1069–1074. Li, J.S., Ishii, T., Feinstein, P., Mombaerts, P., 2004. Odorant receptor gene choice is reset by nuclear transfer from mouse olfactory sensory neurons. Nature 428, 393–399.

112

S.H. Fuss, A. Ray / Molecular and Cellular Neuroscience 41 (2009) 101–112

Lomvardas, S., Barnea, G., Pisapia, D.J., Mendelsohn, M., Kirkland, J., Axel, R., 2006. Interchromosomal interactions and olfactory receptor choice. Cell 126, 403–413. Malnic, B., Hirono, J., Sato, T., Buck, L.B., 1999. Combinatorial receptor codes for odors. Cell 96, 713–723. Marchand, J.E., Yang, X., Chikaraishi, D., Krieger, J., Breer, H., Kauer, J.S., 2004. Olfactory receptor gene expression in tiger salamander olfactory epithelium. J. Comp. Neurol. 474, 453–467. Mazzoni, E.O., Desplan, C., Celik, A., 2004. ‘One receptor’ rules in sensory neurons. Dev. Neurosci. 26, 388–395. McIntyre, J.C., Bose, S.C., Stromberg, A.J., McClintock, T.S., 2008. Emx2 stimulates odorant receptor gene expression. Chem. Senses 33, 825–837. Michaloski, J.S., Galante, P.A., Malnic, B., 2006. Identification of potential regulatory motifs in odorant receptor genes by analysis of promoter sequences. Genome Res. 16, 1091–1098. Miyamichi, K., Serizawa, S., Kimura, H.M., Sakano, H., 2005. Continuous and overlapping expression domains of odorant receptor genes in the olfactory epithelium determine the dorsal/ventral positioning of glomeruli in the olfactory bulb. J. Neurosci. 25, 3586–3592. Mombaerts, P., 1996. Targeting olfaction. Curr. Opin. Neurobiol. 6, 481–486. Mombaerts, P., 2004. Odorant receptor gene choice in olfactory sensory neurons: the one receptor–one neuron hypothesis revisited. Curr. Opin. Neurobiol. 14, 31–36. Nef, P., Hermans-Borgmeyer, I., Artieres-Pin, H., Beasley, L., Dionne, V., Heinemann, S., 1992. Spatial pattern of receptor expression in the olfactory epithelium. Proc. Natl. Acad. Sci. U. S. A. 89, 8948–8952. Ngai, J., Dowling, M., Buck, L., Axel, R., Chess, A., 1993. The family of genes encoding odorant receptors in the channel catfish. Cell 72, 657–666. Nguyen, M.Q., Zhou, Z., Marks, C.A., Ryba, N.J., Belluscio, L., 2007. Prominent roles for odorant receptor coding sequences in allelic exclusion. Cell 131, 1009–1017. Niimura, Y., Nei, M., 2005. Evolutionary dynamics of olfactory receptor genes in fishes and tetrapods. Proc. Natl. Acad. Sci. U. S. A. 102, 6039–6044. Nishizumi, H., Kumasaka, K., Inoue, N., Nakashima, A., Sakano, H., 2007. Deletion of the core-H region in mice abolishes the expression of three proximal odorant receptor genes in cis. Proc. Natl. Acad. Sci. U. S. A. 104, 20067–20072. Norlin, E.M., Berghard, A., 2001. Spatially restricted expression of regulators of G-protein signaling in primary olfactory neurons. Mol. Cell. Neurosci. 17, 872–882. Norlin, E.M., Alenius, M., Gussing, F., Hagglund, M., Vedin, V., Bohm, S., 2001. Evidence for gradients of gene expression correlating with zonal topography of the olfactory sensory map. Mol. Cell. Neurosci. 18, 283–295. Palmer, G.H., Brayton, K.A., 2007. Gene conversion is a convergent strategy for pathogen antigenic variation. Trends Parasitol. 23, 408–413. Pyrski, M., Xu, Z., Walters, E., Gilbert, D.J., Jenkins, N.A., et al., 2001. The OMP-lacZ transgene mimics the unusual expression pattern of OR-Z6, a new odorant receptor gene on mouse chromosome 6: implication for locus-dependent gene expression. J. Neurosci. 21, 4637–4648. Qasba, P., Reed, R.R., 1998. Tissue and zonal-specific expression of an olfactory receptor transgene. J. Neurosci. 18, 227–236. Rawson, N.E., Eberwine, J., Dotson, R., Jackson, J., Ulrich, P., Restrepo, D., 2000. Expression of mRNAs encoding for two different olfactory receptors in a subset of olfactory receptor neurons. J. Neurochem. 75, 185–195. Ray, A., van der Goes van Naters, W., Shiraiwa, T., Carlson, J.R., 2007. Mechanisms of odor receptor gene choice in Drosophila. Neuron 53, 353–369. Ray, A., van der Goes van Naters, W., Carlson, J.R., 2008. A regulatory code for neuronspecific odor receptor expression. PLoS Biol. 6, e125. Ressler, K., Sullivan, S., Buck, L., 1993. A zonal organization of odorant receptor gene expression in the olfactory epithelium. Cell 73, 597–609. Roppolo, D., Vollery, S., Kan, C.D., Luscher, C., Broillet, M.C., Rodriguez, I., 2007. Gene cluster lock after pheromone receptor gene choice. EMBO J. 26, 3423–3430. Rothman, A., Feinstein, P., Hirota, J., Mombaerts, P., 2005. The promoter of the mouse odorant receptor gene M71. Mol. Cell. Neurosci. 28, 535–546. Sato, Y., Miyasaka, N., Yoshihara, Y., 2007. Hierarchical regulation of odorant receptor gene choice and subsequent axonal projection of olfactory sensory neurons in zebrafish. J. Neurosci. 27, 1606–1615. Schoenfeld, T.A., Knott, T.K., 2002. NADPH diaphorase activity in olfactory receptor neurons and their axons conforms to a rhinotopically-distinct dorsal zone of the hamster nasal cavity and main olfactory bulb. J. Chem. Neuroanat. 24, 269–285. Serizawa, S., Ishii, T., Nakatani, H., Tsuboi, A., Nagawa, F., et al., 2000. Mutually exclusive expression of odorant receptor transgenes. Nat. Neurosci. 3, 687–693. Serizawa, S., Miyamichi, K., Nakatani, H., Suzuki, M., Saito, M., et al., 2003. Negative feedback regulation ensures the one receptor–one olfactory neuron rule in mouse. Science 302, 2088–2094.

Shanbhag, S., Muller, B., Steinbrecht, A., 1999. Atlas of olfactory organs of Drosophila melanogaster. 1. Types, external organization, innervation and distribution of olfactory sensilla. Int. J. Insect Morphol. Embryol. 28, 377–397. Shykind, B.M., Rohani, S.C., O'Donnell, S., Nemes, A., Mendelsohn, M., et al., 2004. Gene switching and the stability of odorant receptor gene choice. Cell 117, 801–815. Singh, N., Ebrahimi, F.A., Gimelbrant, A.A., Ensminger, A.W., Tackett, M.R., et al., 2003. Coordination of the random asynchronous replication of autosomal loci. Nat. Genet. 33, 339–341. Stocker, R., 1994. The organization of the chemosensory system in Drosophila melanogaster: a review. Cell Tissue Res. 275, 3–26. Strotmann, J., Wanner, I., Helfrich, T., Beck, A., Breer, H., 1994. Rostro-caudal patterning of receptor-expressing olfactory neurones in the rat nasal cavity. Cell Tissue Res. 278, 11–20. Tian, H., Ma, M., 2008. Differential development of odorant receptor expression patterns in the olfactory epithelium: a quantitative analysis in the mouse septal organ. Dev. Neurobiol. 68, 476–486. Tichy, A.L., Ray, A., Carlson, J.R., 2008. A new Drosophila POU gene, pdm3, acts in odor receptor expression and axon targeting of olfactory neurons. J. Neurosci. 28, 7121–7129. Tsuboi, A., Miyazaki, T., Imai, T., Sakano, H., 2006. Olfactory sensory neurons expressing class I odorant receptors converge their axons on an antero-dorsal domain of the olfactory bulb in the mouse. Eur. J. Neurosci. 23, 1436–1444. van der Goes van Naters, W., Carlson, J.R., 2007. Receptors and neurons for fly odors in Drosophila. Curr. Biol. 17, 606–612. Vassalli, A., Rothman, A., Feinstein, P., Zapotocky, M., Mombaerts, P., 2002. Minigenes impart odorant receptor-specific axon guidance in the olfactory bulb. Neuron 35, 681–696. Vassar, R., Ngai, J., Axel, R., 1993. Spatial segregation of odorant receptor expression in the mammalian olfactory epithelium. Cell 74, 309–318. Vosshall, L.B., Stocker, R.F., 2007. Molecular architecture of smell and taste in Drosophila. Annu. Rev. Neurosci. 30, 505–533. Vosshall, L.B., Amrein, H., Morozov, P.S., Rzhetsky, A., Axel, R., 1999. A spatial map of olfactory receptor expression in the Drosophila antenna. Cell 96, 725–736. Vosshall, L., Wong, A., Axel, R., 2000. An olfactory sensory map in the fly brain. Cell 102, 147–159. Wang, M., Reed, R., 1993. Molecular cloning of the olfactory neuronal transcription factor Olf-1 by genetic selection in yeast. Nature 364, 121–126. Wang, F., Nemes, A., Mendelsohn, M., Axel, R., 1998. Odorant receptors govern the formation of a precise topographic map. Cell 93, 47–60. Weth, F., Nadler, W., Korsching, S., 1996. Nested expression domains for odorant receptors in zebrafish olfactory epithelium. Proc. Natl. Acad. Sci. U. S. A. 93, 13321–13326. Wong, C.C., Poon, W.H., Tsim, T.Y., Wong, E.Y., Leung, M.S., 2000. Gene expressions during the development and sexual differentiation of the olfactory bulb in rats. Brain Res. Dev. Brain Res. 119, 187–194. Yao, C.A., Ignell, R., Carlson, J.R., 2005. Chemosensory coding by neurons in the coeloconic sensilla of the Drosophila antenna. J. Neurosci. 25, 8359–8367. Young, J.M., Friedman, C., Williams, E.M., Ross, J.A., Tonnes-Priddy, L., Trask, B.J., 2002. Different evolutionary processes shaped the mouse and human olfactory receptor gene families. Hum. Mol. Genet. 11, 535–546. Zhang, X., Firestein, S., 2002. The olfactory receptor gene superfamily of the mouse. Nat. Neurosci. 5, 124–133. Zhang, X., Firestein, S., 2007. Comparative genomics of odorant and pheromone receptor genes in rodents. Genomics 89, 441–450. Zhang, X., Rogers, M., Tian, H., Zou, D.J., Liu, J., et al., 2004. High-throughput microarray detection of olfactory receptor gene expression in the mouse. Proc. Natl. Acad. Sci. U. S. A. 101, 14168–14173. Zhang, X., De la Cruz, O., Pinto, J.M., Nicolae, D., Firestein, S., Gilad, Y., 2007a. Characterizing the expression of the human olfactory receptor gene family using a novel DNA microarray. Genome Biol. 8, R86. Zhang, Y.Q., Breer, H., Strotmann, J., 2007b. Promotor elements governing the clustered expression pattern of odorant receptor genes. Mol. Cell. Neurosci. 36, 95–107. Zhao, H., Reed, R.R., 2001. X inactivation of the OCNC1 channel gene reveals a role for activity-dependent competition in the olfactory system. Cell 104, 651–660. Zheng, C., Feinstein, P., Bozza, T., Rodriguez, I., Mombaerts, P., 2000. Peripheral olfactory projections are differentially affected in mice deficient in a cyclic nucleotide-gated channel subunit. Neuron 26, 81–91. zur Lage, P.I., Prentice, D.R., Holohan, E.E., Jarman, A.P., 2003. The Drosophila proneural gene amos promotes olfactory sensillum formation and suppresses bristle formation. Development 130, 4683–4693.