LUSH Odorant-Binding Protein Mediates Chemosensory Responses to Alcohols in Drosophila melanogaster

Corresponding author: Dean P. Smith, Department of Pharmacology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75235-9111., smith15{at}utsw.swmed.edu (E-mail).

Communicating editor: T. F. C. MACKAY

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The molecular mechanisms mediating chemosensory discrimination in insects are unknown. Using the enhancer trapping approach, we identified a new Drosophila mutant, lush, with odorant-specific defects in olfactory behavior. lush mutant flies are abnormally attracted to high concentrations of ethanol, propanol, and butanol but have normal chemosensory responses to other odorants. We show that wild-type flies have an active olfactory avoidance mechanism to prevent attraction to concentrated alcohol, and this response is defective in lush mutants. This suggests that the defective olfactory behavior associated with the lush mutation may result from a specific defect in chemoavoidance. lush mutants have a 3-kb deletion that produces a null allele of a new member of the invertebrate odorant-binding protein family, LUSH. LUSH is normally expressed exclusively in a subset of trichoid chemosensory sensilla located on the ventral-lateral surface of the third antennal segment. LUSH is secreted from nonneuronal support cells into the sensillum lymph that bathes the olfactory neurons within these sensilla. Reintroduction of a cloned wild-type copy of lush into the mutant background completely restores wild-type olfactory behavior, demonstrating that this odorant-binding protein is required in a subset of sensilla for normal chemosensory behavior to a subset of odorants. These findings provide direct evidence that odorant-binding proteins are required for normal chemosensory behavior in Drosophila and may partially determine the chemical specificity of olfactory neurons in vivo.

INSECTS like Drosophila detect odorants with chemosensory hairs or "sensilla" located primarily on the third antennal segment (STOCKER 1994 ). The sensilla are hollow, fluid-filled structures encasing the olfactory neuron dendrites of one to four olfactory neurons, and therefore provide for anatomical segregation of olfactory neurons. In Drosophila, these sensilla fall into three distinct morphological classes: basiconic, coeloconic, and trichoid (STOCKER 1994 ; RIESGO-ESCOVAR et al. 1997 ). All three classes are thought to mediate olfactory responses. Odor molecules pass through pores or grooves within the cuticle of the sensilla where they enter the sensillum lymph bathing the olfactory neuron dendrites (STEINBRECHT 1969 ; ALTNER and PRILLINGER 1980 ; RIESGO-ESCOVAR et al. 1997 ). Extracellular recordings of the odor-induced electrical activity from different regions of the Drosophila antenna reveal different regions have differential sensitivity to specific odorants (SIDDIQI 1987 ; AYER and CARLSON 1992 ; DUBIN et al. 1995 ). However, the molecular mechanisms that confer odor specificity to olfactory neurons in insects are not understood (reviewed in SMITH 1996 ).

In both vertebrates and insects, primary olfactory neurons activated by odorants make their first synapses in the central nervous system where odorant information is processed in complex neural networks called glomeruli (reviewed in SHEPHERD and GREER 1990 ). Olfactory information is subsequently delivered to higher brain centers and ultimately perceived as odor. Odorant perception can dramatically influence animal behaviors ranging from attraction to food sources and avoidance of noxious compounds to mediation of reproductive cues (reviewed in HALPERN 1987 ; BARGMANN et al. 1990 ; HALL 1994 ; PFEIFFER and JOHNSTON 1994 ; ROELOFS 1995 ).

In Drosophila, each of the approximately 2000 antennal olfactory neurons project their axons directly to the bilateral antennal lobes, the Drosophila equivalent of the olfactory bulbs. Each neuron synapses exclusively in one of the 35 glomeruli, either ipsilaterally or bilaterally (STOCKER et al. 1983 ; STOCKER 1994 ). Antennal lobe output is routed to higher brain structures including the mushroom bodies where memory is thought to be consolidated (DAVIS et al. 1995 ). Different odorants produce different patterns of glomerular activation in Drosophila antennal lobes (RODRIGUES and BUCHNER 1984 ; RODRIGUES 1988 ). Flies injected with ³H-labeled 2-deoxyglucose and exposed to repetitive odorant pulses are labeled in antennal lobe glomeruli, and the labeling pattern is different upon exposure to different odorants (RODRIGUES 1988 ). Similar results have been observed in the vertebrate olfactory bulb (CINELLI et al. 1995 ). Therefore, there is likely to be a correlation between the odorant specificity of the olfactory neurons and the pattern of glomerular activity in both vertebrate and Drosophila olfactory systems.

One family of proteins with potential to influence chemosensory discrimination is invertebrate odorant-binding proteins (OBPs). OBPs are produced by vertebrate and arthropod chemosensory systems where they are secreted from nonneuronal support cells into the fluid that bathes the olfactory neuron dendrites. Odorants have been shown to bind directly to these proteins in both mammals and insects (VOGT and RIDDIFORD 1981 ; PELOSI et al. 1982 ; PEVSNER et al. 1985 ; PEVSNER et al. 1990 ; DU and PRESTWICH 1995 ). In insects, members of the invertebrate OBP family are low-molecular-weight, chemosensory-specific proteins with six conserved cysteine residues. Unlike vertebrate odorant-binding proteins that are members of the lipocalin transport family (FLOWER 1996 ), the invertebrate proteins constitute a unique protein family. In Drosophila, the six previously identified invertebrate OBP members have surprisingly low sequence similarity and are expressed in different, overlapping zones of chemosensory sensilla. This is consistent with these proteins performing an odor-specific function (MCKENNA et al. 1994 ; PIKIELNY et al. 1994 ). Moth pheromone-binding protein members of this family have been shown to bind directly to pheromone with chemical selectivity indicating members of this family interact directly with odorant molecules (DU and PRESTWICH 1995 ). No mutants defective for any odorant-binding protein gene have been previously described; therefore the in vivo function of these proteins is unknown. Possible functions include solubilizing or concentrating odorants in the sensillum lymph, or mediating odorant removal (reviewed in PELOSI 1994 ). We report here the identification and characterization of lush, a gene encoding a new member of the invertebrate odorant-binding protein family in Drosophila. LUSH is expressed in a subset of trichoid sensilla and is required for normal olfactory behavior responses to a small subset of chemically related odorants. Our results support models in which odorant-binding proteins participate in determining the chemical specificity of olfactory neurons in Drosophila.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Drosophila stocks, generation of enhancer trap lines, lush mutants:
Flies carrying the Tau-LacZ P element were obtained from John Thomas and Chris Callahan (Salk Institute). TAU is a microtubule-binding protein that localizes the fusion protein to axons when expressed in neurons (CALLAHAN and THOMAS 1994 ). Genetic crosses were carried out under standard laboratory conditions using balancer stocks (LINDSLEY and ZIMM 1992 ). After isogenic strains were created, ~2500 lines carrying novel insertions were generated as follows: y w; +/+; y⁺ 2,3 Sb/TM2 males carrying the 2,3 activated transposase (ROBERTSON et al. 1988 ) were crossed to w/w; In(2LR)O, Cy p[tau-LacZ, w+]/In2LR Gla; +/+ virgin females. In(2LR)O, Cy and TM2 are balancer chromosomes. Single In(2LR)O, Cy p[tau-LacZ, w+]/+; y⁺ 2,3 Sb/+ males were recovered and mated to 5 attached-X virgin females in individual vials. Single males carrying white⁺(w⁺), but not the n(2LR)O, Cy p[tau-LacZ, w+] or y⁺ 2,3 Sb chromosomes were recovered and used to establish stable strains carrying novel P-element integrations on the X, second, or third chromosomes by crossing each male to 10 attached-X females. Only one male was isolated from each vial to ensure independent insertion events were recovered and screened for LacZ expression (see below).

lush mutants were generated by mobilizing the P element from the ET249 stock (the line carrying the enhancer trap element with trichoid sensillum-specific LacZ expression) by crossing to flies carrying a stable source of transposase (ROBERTSON et al. 1988 ) and recovering third chromosomes that had lost the w⁺ gene over a balancer chromosome. We recovered approximately 300 independent third chromosomes missing the w⁺ eye color marker contained within the P element. Homozygous strains for most of these chromosomes were generated, and genomic DNA was prepared and screened using the polymerase chain reaction with primers specific to the lush coding sequence (see below).

ß-Galactosidase expression:
Enhancer trap lines were screened for ß-galactosidase expression in adult heads as previously described (RIESGO-ESCOVAR et al. 1992 ), except that staining reactions were performed at 25° for 4 hr. Lines with LacZ expression restricted to the chemosensory structures of the head were retested to confirm the staining pattern, and LacZ expression was simultaneously examined in the body. Larvae were stained as described RIESGO-ESCOVAR et al. 1992 . To examine LacZ expression in tissue sections, 10-µm-thick frozen sections were fixed for 10 min in 1% glutaraldehyde (EM Grade, EM Science), washed in PBS, and stained as described above.

DNA, RNA, sequencing and PCR:
Genomic DNA flanking the P-element insertion was cloned by plasmid rescue as described by PIRROTTA 1986 . Genomic DNA was prepared as described by LIS et al. 1983 . Library screening, restriction mapping, and mRNA isolation were performed as described in MANIATIS et al. 1982 . Appendage cDNA was prepared from mRNA using a reverse transcription kit (Invitrogen, San Diego, CA) using appendages isolated as described by Oliver (OLIVER and PHILIPS 1970 ). Hybridizations were performed at 65° in 750 mM NaCl, 100 mM NaH₂PO₄ (pH 6.8), 75 mM sodium citrate, 0.4% Ficoll, and 0.5% sodium dodecyl sulfate. Filters were washed in 0.2x SSC (1x SSC is 150 mM NaCl, 15 mM sodium citrate) and 0.5% SDS at 65°. Sequence analysis was performed using an ABI automated sequencer (ABI Adv. Biotechnologies, Columbia, MD). PCR reactions to identify lush mutants were performed using the method of SAIKI et al. 1985 with oligonucleotides 5' GAAGCTTGTAGGGATACG and 5' TTAAGGCCACATGAACTG. PCR conditions were 94° for 30 sec, 50° for 30 sec, and 72° for 2 min, repeated for 35 cycles. Control primers specific to unlinked sequences were included in each PCR reaction to control for presence of template DNA.

In situ hybridization to polytene chromosomes and tissue sections:
Polytene chromosomes were prepared from salivary glands of late third instar larvae of the Oregon R wild-type strain and hybridized as described by LANGER-SOFER et al. 1982 . DNA fragments to be mapped were labeled with [bio-16]dUTP (Enzo Biochemicals) by nick translation. Signal detection was performed with streptavidin-conjugated horseradish peroxidase (Enzo Biochemicals) and diaminobenzidine.

Generation of antiserum, immunofluorescence, Western blotting:
Rabbit polyclonal antiserum was raised to a six histidine-tagged LUSH protein expressed in bacteria. Serum was affinity purified on LUSH Affi-gel columns (Bio-Rad, Richmond, CA) according to the instructions of the manufacturer. Immunofluorescence was performed as described in (SMITH et al. 1991 ). Western Blots were performed as described in STAMNES et al. 1991 except that antibodies were detected using ECL kits (Amersham, Arlington Heights, IL). Forty antennae equivalents were run per lane. Canton-S and w¹¹¹⁸ were used as controls.

P-element-mediated DNA transformations:
Drosophila transformations were carried out as described by KARESS and RUBIN 1984 . Transposase DNA was used at a concentration of 200 µg/ml and sample DNA was used at 1 mg/ml. The rescue fragment used to restore wild-type lush function extended from the left end of 249 to the first BamHI site (see Figure 2).

View larger version (84K): In this window In a new window Download PPT slide   Figure 1. Expression of Tau-LacZ in ET249. (A) Male (left) and female ET249 adult heads were stained for ß-galactosidase activity. LacZ expression is visible on the ventral-lateral surface of each third antennal segment (arrows). No staining is present in the brain, thorax or abdomen, wings or legs. No olfactory neurons are stained. (B) Frozen tissue section through ET249 antenna stained for ß-galactosidase activity. ß-Galactosidase is associated with support cells in a subset of trichoid sensilla. (C) ß-Galactosidase activity is also present in the larval olfactory organ in ET249, the antennomaxillary complex. (D) LUSH protein is expressed in the same region as ß-galactosidase in ET249 flies. Immunofluorescence image of frozen tissue section of wild-type fly reacted with affinity-purified LUSH antiserum. Note secretion of LUSH protein into the sensillum lymph of the trichoid sensilla (arrow).

View larger version (36K): In this window In a new window Download PPT slide   Figure 2. lush mutant flies have abnormal behavioral responses to ethanol, propanol, and butanol in olfactory trap assays. (A) Olfactory trap data comparing ET249 (control) and lush mutant responses to short-chain alcohols. Bars represent average number of flies entering traps containing the substances noted. Averages represent a minimum of 10 experiments. Standard errors are depicted above the bars. Asterisk denotes statistical significance between the means of control (ET249) and lush mutant flies (two-tailed t-test, independent samples). Concentrations noted are the initial concentration of odorant mixed in agarose. lush mutants have normal responses to methanol and isopropanol, but are more likely to enter traps containing concentrated propanol than are control flies. (B) lush mutant flies are more likely to enter traps containing high levels of ethanol in a dose-dependent manner. The sensitivity of lush mutants for ethanol is not significantly different from controls at low concentrations (1:500). Wild-type responses are restored to the mutants by the lush transgene (hatched bars). (C) lush mutants are defective for chemoavoidance of concentrated ethanol. Wild-type and lush flies are equally attracted to traps containing 1% yeast extract. Wild-type and lush mutants expressing a transgenic lush gene (rescue) are significantly less likely to enter traps containing the same amount of yeast when it is mixed with 25% ethanol (P < 0.001, t-test, independent samples). lush mutants are defective for this avoidance response, are significantly more likely to enter the traps containing the mixture than are the controls, and are significantly more likely to enter the traps containing the mixture than traps containing yeast alone (P < 0.001). Asterisk denotes significant differences between yeast and yeast + ethanol for each genotype.

Olfactory behavioral assays:
Isogenized w¹¹¹⁸ flies were the parental background for all experiments. w¹¹¹⁸ or ET249 flies were chosen as olfactory normal controls for testing lush mutants to minimize differences in genetic background that are well known to influence olfactory behavioral responses (ALCORTA and RUBIO 1988 ; ALCORTA and RUBIO 1989 ; MONTE et al. 1989 ; DUBIN et al. 1995 ). ET249 flies express LUSH at normal levels indicating the P element does not disrupt expression of this gene.

Olfactory trap assays were performed essentially as described in (WOODARD et al. 1989 ) except that 5 male and 5 female flies were tested in each plate. No sex-specific differences in olfactory behavior were observed in lush mutants. One- to three-day-old flies were tested and a minimum of 100 flies (10 plates) were tested for each odorant concentration and genotype. A total of 10 µl of the diluted odorant was vortexed with 1 ml of 1% agarose at 45°, and 100-µl aliquots were distributed to 10 traps on ice to rapidly solidify the agarose. The concentrations noted in the figures are the concentration within the agarose. The actual odorant concentrations in air are significantly less. Odorants were obtained from Aldrich Chemical (Milwaukee, WI) and were the highest purity available. Differences in the means were tested for significance using t-tests for independent samples and ANOVA was used for comparison of more than two means (Statistica Software; StatSoft Inc., Tulsa, OK).

Electroantennograms:
Extracellular recordings of electrical responses of the antenna were obtained essentially as described by DUBIN et al. 1995 using an EX-1 single channel extracellular amplifier (Degan, Minneapolis, MN) and MacAdios II hardware and Superscope software (GW Instruments, Somerville, MA) with an automated odorant delivery system (ALCORTA 1991 ) providing 1-sec odorant pulses. The recording electrode was placed on the ventral lateral surface of the antenna, and the ground electrode was placed in the brain through the vertex of the head.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ET249: an enhancer trap line expressing LacZ exclusively in the chemosensory system:
LUSH was identified using the enhancer detection approach (BELLEN et al. 1989 ) as a gene expressed exclusively in the olfactory organs. Briefly, single P-transposable elements (P elements) modified to express a tau-LacZ fusion reporter gene (CALLAHAN and THOMAS 1994 ) were randomly inserted in the genome. Expression of the reporter fusion gene is dependent on enhancer elements acquired upon integration and can mimic the temporal and spatial expression pattern of individual genes located at or near the integration site (BELLEN et al. 1989 ). Because the LacZ gene is fused in frame to the gene encoding the microtubule-associated protein tau, expression of the reporter gene in neurons results in LacZ staining of axonal projections when expressed in these cells (CALLAHAN and THOMAS 1994 ).

We generated several thousand lines of flies carrying stable, independent P-element insertions. Members of each line were screened for reporter gene expression restricted to the chemosensory structures. ET249 was one of several lines with adult LacZ expression restricted to a subset of chemosensory sensilla on the third antennal segment (Figure 1A). Olfactory neuron axons, visible in other enhancer trap lines expressing tau-LacZ in olfactory neurons, were not stained in ET249 indicating expression of tau-LacZ was restricted to support cells. To more precisely identify the cells that were expressing LacZ in the antennae of ET249 flies, we stained frozen tissue sections from these structures. LacZ expression was prominent in cells associated with trichoid sensilla on the ventral-lateral surface of the third antennal segment (Figure 1B). Based on their relative position in the epithelium, the support cells expressing LacZ in ET249 flies are trichogen support cells that secrete the sensillum lymph that bathes the olfactory neuron dendrites (KEIL and STEINBRECHT 1984 ; HARTENSTEIN and POSAKONY 1989 ; RAY and RODRIGUES 1995 ). We examined third-instar larvae for LacZ expression and found expression restricted to the larval olfactory organs, the antennomaxillary complex (Figure 1C). We mapped the P-element insertion to position 76C on the third chromosome (data not shown). No previously identified olfactory mutants have been mapped to this genomic region.

P-element excision mutants have abnormal olfactory behavior to a subset of odorants:
To create loss-of-function mutations in the putative chemosensory-specific gene identified by the ET249 P element, we generated small deletions at the P-element integration site by mobilizing the transposon from ET249 flies and recovering chromosomes from which excision had occurred (for example see SASS et al. 1993 ). We identified five putative deletions based on the absence of a PCR product using primers specific to DNA sequences flanking the P element. The largest deletion eliminated 3 kb of genomic DNA flanking the P element. This lesion completely removed the lush transcription unit (see below). Flies homozygous for this deletion are viable and fertile and were named lush mutants (see below).

To compare olfactory discrimination between lush and wild-type adults, we employed the olfactory trap assay (WOODARD et al. 1989 ). Briefly, 10 wild-type or mutant flies were placed in a petri plate with a single odorant trap, and the number of flies within the trap was determined after a set time period (see MATERIALS AND METHODS). We screened a panel of 60 simple volatile organic compounds at different concentrations to test for differences in distribution between control and lush flies. Table 1 shows results for representative odorants tested. Odorants were tested at 1:1000 and 1:4 dilutions in agarose. As expected from the restricted expression pattern of LacZ in a subset of sensilla, the majority of the compounds attract similar proportions of wild-type and lush flies, indicating there is no global olfactory defect associated with the deletion. However, odor-specific defects in chemosensory behavior are observed in lush flies when challenged with three chemically related odors. We observe a significant increase in the number of mutant flies in traps containing high concentrations of ethanol, propanol, and butanol compared to control flies. Their responses to a variety of other alcohols are not different from those of wild type (Figure 2A). Interestingly, the apparent increased attraction of lush flies for ethanol, propanol, and butanol is specific to high-odorant concentrations. Figure 2B reveals the dose-dependent, abnormal attraction of lush mutants for ethanol. The extent of the attraction of lush flies to yeast extract, ethyl acetate, and low concentrations of ethanol is similar to that of wild-type. However, the mutant flies display an abnormal attraction to traps containing high concentrations of ethanol (1:100, 1:4; Figure 3B). ET249 flies (that carry the P element but not the deletion) and the w¹¹¹⁸ strain from which these lines were derived have normal chemosensory responses to these alcohols, as do third-instar larvae from lush flies (data not shown). We named this deletion mutant "lush" to reflect their increased affinity for ethanol-rich environments. We conclude that lush flies have odor-specific defects in chemosensory discrimination and are abnormally attracted to high concentrations of a subset of odorants including ethanol, propanol, and butanol.

View larger version (24K): In this window In a new window Download PPT slide   Figure 3. lush mutants deactivate ethanol-sensitive neurons normally. lush mutants and control flies terminate ethanol-induced olfactory responses with similar time courses. The time to deactivate the response from peak to 75% return-to-baseline was determined for lush mutants, wild-type controls (ET249), and lush mutants carrying two wild-type copies of the lush gene (rescue). Recordings from extracellular responses were measured and averaged. SEM is depicted by the error bars. Responses from a minimum of five flies were measured for each genotype for each odor. There is no significant difference in time required to deactivate the ethanol response in lush and control flies. EA, ethyl acetate.

Wild-type flies have endogenous mechanisms to avoid concentrated ethanol that are defective in lush mutants:
The increased likelihood of lush mutant flies entering traps containing high concentrations of these alcohols could result from either increased attraction to these odorants or a defect in avoidance of high concentrations of these compounds. If there is a defect in chemoavoidance to ethanol in lush mutants, we should be able to demonstrate this behavioral response in wild-type flies. To determine if wild-type flies have endogenous mechanisms to avoid high-ethanol environments that are defective in the mutants, we tested the effects of mixing ethanol with yeast extract, a strong chemoattractant. Figure 2C shows that wild-type flies are attracted to dilute yeast extract (left panel, open bars). However, when the same amount of yeast is mixed with concentrated ethanol, wild-type flies are significantly less likely to enter these traps (compare open bars). Therefore, the presence of high levels of ethanol reduces attraction for yeast in wild-type flies. This demonstrates that there is an active avoidance mechanism in wild-type flies that is stimulated by high concentrations of ethanol. lush mutants are equally attracted to yeast compared to wild-type flies (filled bar, left graph) but are defective for the avoidance behavioral response (Figure 2C, filled bars). In fact the lush mutants are significantly more attracted to the mixture of yeast and concentrated ethanol than to yeast alone.

One model that could explain the increased affinity of lush mutants for high concentrations of alcohol is a specific defect in active avoidance behavior to ethanol mediated through the lush gene product. In an alternative model, the same phenotype could arise if the lush gene product were required to deactivate or desensitize neurons mediating chemoattraction. If this latter model is correct, lush mutants should have a delay in termination of the ethanol-induced electrical responses compared to control flies. In an attempt to distinguish between these models, we recorded electroantennograms (EAG) using a computer-triggered odorant delivery system (ALCORTA 1991 ) to analyze the electrical responses of the ethanol-sensitive olfactory neurons in wild-type and lush mutant flies. We observed no significant difference between wild-type and lush mutants in amplitude or the time required to deactivate the response to 75% of baseline over a wide range of ethanol concentrations (see Figure 3). These results argue against a defect in adaptation or deactivation in neurons mediating attraction that would produce prolonged ethanol-induced electrical responses in lush mutants.

Deletion mutants are missing the lush gene: a new member of the invertebrate odorant-binding protein family:
We characterized the genomic DNA at the P-element insertion site in ET249 flies to define the gene responsible for the abnormal chemosensory responses in lush mutant flies. Genomic DNA was prepared from ET249 flies, and the DNA flanking the P-element insertion site was cloned by plasmid rescue (PIRROTTA 1986 ). We recovered 2.5 kb of genomic DNA flanking the P-element insertion site and used these sequences to isolate genomic and cDNA clones (Figure 4). Two transcription units were mapped to the region of the P-element integration site, one of which mimics the expression profile of LacZ in ET249 and is specifically deleted in the mutant (see below).

View larger version (18K): In this window In a new window Download PPT slide   Figure 4. Map of 76C genomic region. The ET249 P element integrated into the genomic region corresponding to 249. In ET249 flies, the P element integrated into the R1-RV fragment. The lower panel depicts the restriction fragments completely (-) or partially (+/-) deleted in lush mutants. The genomic structure of the lush cDNA is shown below the 249 map with an expanded view of the lush locus. The position of the ET249 P element, the lush gene, and the ash-1 gene (TRIPOULAS et al. 1994 ) are depicted above 249. All cloned sequences were confirmed to map to position 76C on the polytene chromosome.

We determined the entire nucleotide sequence of a putative lush cDNA and ~3 kb of genomic DNA flanking the P-element insertion site. The P element inserted 373 base pairs downstream from the polyadenylation site of the lush transcription unit (Figure 4) and did not disrupt expression of this gene (data not shown). This is consistent with the observation that ET249 flies avoid concentrated ethanol (Figure 2). The predicted protein encoded by the lush gene is 153 residues in length with a series of hydrophobic residues near the N terminus typical of a signal sequence (VON HEIJNE 1986 ). Database comparison with previously identified proteins revealed significant homology (24% overall identity) with OS-F/PB-PRP3 (MCKENNA et al. 1994 ; PIKIELNY et al. 1994 ), a Drosophila member of the invertebrate odorant-binding protein family (Figure 5). LUSH shares all features of this protein family including a signal sequence to direct polypeptides to the secretory pathway, chemosensory-specific expression pattern, and six conserved cysteine residues with the spacing between cysteines 2 and 3 and 5 and 6 completely conserved in all members.

View larger version (81K): In this window In a new window Download PPT slide   Figure 5. Alignment of LUSH with other members of the invertebrate odorant-binding protein family. LUSH aligned with four moth and four Drosophila members of the invertebrate odorant-binding protein family. Conserved cysteines are denoted with an asterisk above the alignment. Lush, LUSH; PB-PRP1, Drosophila PBP related protein 1 (PIKIELNY et al. 1994 ); PB-PRP2, Drosophila PBP related protein 2 (PIKIELNY et al. 1994 ); PB-PRP3, Drosophila PBP related protein 3 (MCKENNA et al. 1994 ; PIKIELNY et al. 1994 ); PB-PRP5, Drosophila PBP related protein 5 (PIKIELNY et al. 1994 ); PBP-1, pheromone-binding protein from moth A. polyphemus (RAMING et al. 1989 ); PBP-2, pheromone-binding protein from Manduca sexta (GYORGYI et al. 1988 ); PBP-6 and PBP-8, general odorant-binding proteins from Manduca sexta (VOGT et al. 1991 ).

Rabbit polyclonal antiserum was raised to bacterially expressed LUSH protein for direct examination of the expression of this protein in wild-type and lush flies (see MATERIALS AND METHODS). Affinity-purified anti-LUSH antibodies recognize protein in accessory cells of trichoid sensilla on the ventral-lateral portion of the third-antennal segment in wild-type males and females, in a pattern identical to LacZ expression in ET249 (Figure 1D). In contrast to the LacZ that is localized to the support cell cytoplasm in ET249 flies, LUSH protein was clearly present within the shafts of the trichoid sensilla, confirming it was secreted into the sensillum lymph (compare Figure 1B and Figure 1D). No labeling of olfactory neurons was observed. Western blots of antennal extracts from wild-type and lush mutant flies probed with anti-LUSH antiserum revealed that the mutants are completely defective for LUSH expression (Figure 6, LUSH). Southern blot analysis of lush mutant DNA confirmed that the 3-kb deletion removes the entire protein-coding region of the lush gene (Figure 4, lower panel). This suggests that loss of this odorant-binding protein gene is responsible for the chemosensory defects in the lush mutants.

View larger version (52K): In this window In a new window Download PPT slide   Figure 6. Expression of LUSH protein in wild-type flies, deletion mutants, and deletion mutants transformed with a wild-type lush gene. Anti-LUSH antiserum recognizes a 14-kD LUSH protein in control antennae (Canton S and w1118) that is absent in the lush mutants (LUSH). Mutants transformed with two copies of the wild-type lush gene are restored for LUSH expression (Rescue). LUSH is overexpressed in flies carrying six copies of lush (3xRescue).

To prove the chemosensory defects associated with lush mutants are due entirely and specifically to loss of LUSH protein, we introduced a cloned wild-type copy of this gene into the mutant flies by germ-line transformation (see MATERIALS AND METHODS). Expression of a lush transgene under control of its own promoter in the mutant background restores LUSH expression to normal levels (Figure 6, rescue). Furthermore, the transgene completely restores wild-type olfactory behavioral responses to the lush mutants (Figure 2B and Figure C, striped bars). Therefore, the abnormal chemoattraction to high levels of alcohol associated with the deletion results specifically from loss of LUSH. Flies carrying six lush genes overexpress LUSH in the trichoid sensilla (see Figure 6) and behave indistinguishably from controls in response to ethanol (data not shown) suggesting that the levels of LUSH are not the rate-limiting component of this behavior. We conclude that lush mutants have defective chemosensory responses to a subset of odorants resulting from loss of a single odorant-binding protein in the sensillum lymph of a small subset of trichoid chemosensory sensilla.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

LUSH has the hallmark features of a member of the invertebrate odorant-binding protein family. These features include chemosensory-specific expression, the presence of a signal sequence for secretion from the nonneuronal support cells in which it is expressed, and the presence of six cysteine residues with conserved spacing, especially between cysteines 2 and 3 and 5 and 6 where the spacing is absolutely conserved in all members. The conservation of these cysteines suggests that the members of this family share a common disulfide bonding pattern that may impart a similar tertiary structure (RAMING et al. 1990 ).

We have shown that LUSH is secreted into the sensillum lymph (Figure 1D). Electron microscopy studies have previously demonstrated secretion of moth pheromone-binding proteins and the Drosophila OS-E and OS-F into the sensillum lymph of the sensilla in which they are expressed (STEINBRECHT 1996 ; HEKMAT-SCAFE et al. 1997 ). It is likely, therefore, that all members of this family function in the sensillum lymph. Six other members of the invertebrate OBP family have been identified in Drosophila through the use of differential screening methods (MCKENNA et al. 1994 ; PIKIELNY et al. 1994 ). However, there are no corresponding mutants that specifically disrupt these genes to provide insight into the in vivo function of these proteins. To our knowledge, lush is the first odorant-binding protein mutant described for any organism. The specific olfactory defects associated with the lush mutant provide the first direct evidence that a member of this protein family is required for normal olfactory behavior.

Specific features of the arthropod chemosensory system not present in mammalian systems may allow OBPs to play a unique role in chemosensory discrimination. Unlike mammals, whose olfactory cilia are bathed in a common overlying fluid, most arthropods (including insects) have compartmentalized their olfactory neurons into sensilla. Segregation of individual or small groups of olfactory neurons in separate compartments provides the opportunity to independently regulate the composition of the fluid bathing the olfactory neuron dendrites. Indeed, the seven Drosophila members identified to date are expressed in specific subsets of sensilla, and none are expressed in all sensilla (MCKENNA et al. 1994 ; PIKIELNY et al. 1994 ; HEKMAT-SCAFE et al. 1997 ). More than one odorant-binding protein can be expressed within a single sensillum (HEKMAT-SCAFE et al. 1997 ). Differential expression of a family of odorant-binding proteins, therefore, is a feasible mechanism for influencing the chemical specificity of the olfactory neurons within those sensilla, perhaps by regulating access of odorants to the neuron. The defective olfactory behavior associated with the lush mutant is consistent with this idea.

How does a protein secreted into the fluid that bathes olfactory neuron dendrites (but is not synthesized by olfactory neurons) affect chemosensory behavior? First, the olfactory defects observed in lush mutants do not arise from the loss or global disruption of the function of the support cells that secrete LUSH. These cells appear morphologically normal in the mutants and are able to secrete other members of the invertebrate OBP family normally (M.-S. KIM and D. P. SMITH, unpublished results). Therefore, LUSH is not required for the presence, determination of cell fate, or functioning of the support cells in which it is expressed. Similarly, LUSH is not expressed in the antennal lobe, the central nervous system, or the motor pathways, indicating it does not mediate chemosensory information processing or efferent behavioral responses to odorants. This narrows the site of action of LUSH to effects on the primary olfactory neurons.

Given their location in the sensillum lymph, the fact that moth pheromone-binding proteins bind directly to pheromone odorant (DU and PRESTWICH 1995 ), and that lush mutants have odor-specific defects in olfactory behavior, LUSH probably modulates the activity of primary olfactory neurons in the trichoid sensilla through a mechanism involving a direct interaction with odorants in the sensillum lymph. For example, LUSH could regulate odorant concentration, transport, or metabolism within the lymph. While the exact mechanism by which LUSH affects chemosensory behavior is not known, our data provide important clues about the in vivo role of these proteins. First, ethanol is very soluble in both aqueous and lipid environments. It is unlikely, therefore, that LUSH functions simply to solubilize these alcohols in the sensillum lymph. Furthermore, we think it is unlikely that LUSH simply removes these odorants from the lymph. If LUSH removed ethanol from the sensillum lymph, and the olfactory neurons in the trichoid sensilla mediated attraction, increased attraction could result from the persistence of these alcohols in the lymph. However, if odorant removal is the sole function of LUSH, we would expect to find increased sensitivity to alcohol in lush mutants because they would have increased alcohol levels in the sensillum lymph compared to normal flies at low alcohol concentrations. lush mutants have normal sensitivity to low levels of ethanol (Figure 2B). Finally, LUSH could act by desensitizing the trichoid olfactory neurons mediating attraction to alcohol, perhaps by acting as a neuronal receptor antagonist when bound to alcohol. However, in the absence of LUSH this model predicts that olfactory neurons will be active over a longer time course than wild-type controls, and we do not observe this in extracellular recordings from the antenna (Figure 3). However, LUSH-dependent olfactory neurons may be a small fraction of the ethanol-sensitive neurons, and their contribution to the EAG may not be detectable.

We think the most likely possibility is that LUSH is required to activate a small subset of olfactory neurons in the trichoid sensilla that specifically mediate chemoavoidance. Olfactory neurons specific for chemoavoidance are well documented in Caenorhabditis elegans (BARGMANN and HORVITZ 1991 ; BARGMANN et al. 1993 ; TROEMEL et al. 1997 ). For example, LUSH might concentrate or prevent the rapid metabolism of these alcohols in the sensillum lymph thus increasing the steady-state concentration of these odorants in the trichoid sensillum lymph of wild-type flies. This could trigger activation of olfactory neurons mediating avoidance and altering the perception of ethanol so it "smells bad." Pheromone-binding proteins may perform a similar role in sensitizing chemosensory neurons to pheromone in moths (VOGT et al. 1985 ; KAISSLING 1998 ). If true, this model predicts that these LUSH-dependent olfactory neurons would not be activated by ethanol in lush mutants, but would be activated in wild-type flies. However, we see no significant differences in the EAG recordings, again perhaps because these neurons are not detectable. Alternatively, LUSH may affect olfactory behavior by regulating processes that occur on a slower time scale apparent in chemosensory behavior assays, but not EAG recordings. Additional experiments will be required to identify the exact biochemical function of the LUSH protein and the behavioral specificity of the chemosensory neurons within the trichoid sensilla.

Each of the seven members of the Drosophila odorant-binding protein family are expressed in specific zones on the surface of the antenna. Therefore, there is a topographic map on the surface of the antenna defined by zones of odorant-binding protein expression. We have shown that LUSH is required for normal chemosensory responses to specific odorants. This implies a correlation between the odorant-binding protein expression zone and the odor specificity of olfactory neurons. Previous workers have demonstrated a relationship between odorant sensitivity and position on the surface of the antenna (SIDDIQI 1987 ; AYER and CARLSON 1992 ; DUBIN et al. 1995 ) and these zones of sensitivity could correspond to odorant-binding protein expression zones. Cobalt backfilling experiments labeling the projections of the olfactory neurons from the LUSH expression zone (the ventral-lateral surface) revealed these olfactory neurons synapse primarily in only 2 of the 35 anatomically identified glomeruli in the antennal lobe, VA-1 and DA-1 (STOCKER et al. 1983 ). It will be interesting to determine if one or both of these glomeruli specifically function in chemosensory avoidance, and if neurons associated with other odorant-binding protein zones project to common subsets of glomeruli. The lush expression zone overlaps several other Drosophila odorant-binding proteins, specifically PB-PRP-1 and PB-PRP-3/OS-F (MCKENNA et al. 1994 ; PIKIELNY et al. 1994 ). When mutants defective for these gene products become available, it will be important to determine if they have defective avoidance responses, but to a different subset of odorants. Similarly, we predict mutations in OBPs expressed in the other classes of sensilla will have defective attraction to a subset of odorants.

Our data implicate members of the invertebrate odorant-binding protein family in odorant discrimination in Drosophila. However, neuronal receptors are also likely to contribute to chemosensory discrimination in vivo. Seven transmembrane receptors mediate odorant responses in C. elegans (SENGUPTA et al. 1996 ) and in vertebrates (BUCK and AXEL 1991 ; ZHOU et al. 1998 ). In Drosophila the dGq-3 heterotrimeric G protein -subunit is expressed in the dendritic portion of a subset of olfactory neurons, consistent with a role in transducing a subset of odorant responses through seven transmembrane receptors (TALLURI and SMITH 1995 ). Furthermore, insects often package several olfactory neurons within a single sensillum, and there is evidence to suggest these neurons are not functionally identical. Analysis of the electrical responses of Antherea polyphemus moths revealed that two neurons in the same pheromone-sensitive sensillum responded preferentially either to the pheromone acetate or to the aldehyde (GANJIAN et al. 1978 ). This suggests that the olfactory neurons within a sensillum are not functionally identical, and these differences probably correspond to differential expression of receptor proteins on the dendritic surface of the neuron. Therefore, we suggest that chemical specificity of olfactory neurons in Drosophila results from a combination of interaction of odorants with odorant-binding proteins in the sensillum lymph and the specificity of receptor proteins present on the olfactory neurons. A diverse family of odorant-binding proteins could enable a relatively small family of neuronal receptors to respond differentially to a broad range of compounds through a combinatorial mechanism. For example, two neurons in different sensilla expressing the same broadly tuned neuronal receptor might be activated by different subsets of odorants if each sensillum expressed a different repertoire of binding proteins. Elucidation of the relative size of the receptor and binding protein families and determination of their spatial expression relationships will provide further insight into the question of insect chemoreception.

Finally, it should be noted that the alcohols that induce abnormal chemosensory reponses in lush mutants are biologically relevant odorants for Drosophila melanogaster. In nature, fruit flies feed and deposit eggs on fermenting plant materials in which ethanol is the most abundant short-chain alcohol (MCKECHNIE and MORGAN 1982 ; VAN DELDEN 1982 ). The ability to detect ethanol is important for chemotaxis toward food sources. However, adult flies are also susceptible to intoxication and death in high ethanol environments (CHAKIR et al. 1993 ). Therefore, there is a selective advantage for the ability to avoid environments with dangerously high alcohol concentrations, and LUSH is required for this response. The behavioral response to alcohol, therefore, reflects a finely tuned olfactory response.

	ACKNOWLEDGMENTS

We thank Lisa Montgomery for excellent technical assistance, Romayleh Behrouzi for EAGs, Al Gilman and Leon Avery for helpful discussions, Helmut Kramer for use of injection facilities, and Charles Zuker, Leon Avery, and Steve Wasserman for helpful comments on the manuscript. This work was supported by the National Institutes of Health grant DC-02539-2. Sequence of the lush gene is available in GenBank accession no. AF001621.

Manuscript received March 30, 1998; Accepted for publication June 23, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALCORTA, E., 1991  Characterization of the electroantennogram in Drosophila melanogaster and its use for identifying olfactory capture and transduction mutants. J. Neurophysiol. 65:702-714[Abstract/Free Full Text].

ALCORTA, E. and J. RUBIO, 1988  Genetical analysis of intrapopulational variation in olfactory responses in Drosophila melanogaster. Heredity 60:7-14.

ALCORTA, E. and J. RUBIO, 1989  Intrapopulational variation of olfactory responses in Drosophila melanogaster. Behav. Genet. 19:285-299[Medline].

ALTNER, H. and L. PRILLINGER, 1980  Ultrastructure of invertebrate chemo-, thermo-, and hygroreceptors and its functional significance. Int. Rev. Cytol. 67:69-139.

AYER, R. J. and J. CARLSON, 1992  Olfactory physiology in the Drosophila antenna and maxillary palp: acj6 distinguishes two classes of odorant pathways. J. Neurobiol. 23:965-982[Medline].

BARGMANN, C. I. and H. R. HORVITZ, 1991  Chemosensory neurons with overlapping functions direct chemotaxis to multiple chemicals in C. elegans. Neuron 7:729-742[Medline].

BARGMANN, C. I., J. H. THOMAS, and H. R. HORVITZ, 1990  Chemosensory cell function in the behavior and development of Caenorhabditis elegans. Cold Spring Harbor Symp. Quant. Biol. 55:529-538[Abstract/Free Full Text].

BARGMANN, C. I., E. HARTWEIG, and H. R. HORVITZ, 1993  Odorant-selective genes and neurons mediate olfaction in C. elegans. Cell 74:515-527[Medline].

BELLEN, H. J., C. J. O'KANE, C. WILSON, U. GROSSNIKLAUS, and R. K. PEARSON et al., 1989  P-element-mediated enhancer detection: a versatile method to study development in Drosophila. Genes Dev. 3:1288-1300[Abstract/Free Full Text].

BUCK, L. and R. AXEL, 1991  A novel multigene family may encode odorant receptors: a molecular basis for odor recognition. Cell 65:175-187[Medline].

CALLAHAN, C. A. and J. B. THOMAS, 1994  Tau-beta-galactosidase, an axon-targeted fusion protein. Proc. Natl. Acad. Sci. USA 91:5972-5976[Abstract/Free Full Text].

CHAKIR, M., O. PERIDY, P. CAPY, E. PLA, and J. R. DAVID, 1993  Adaptation to alcoholic fermentation in Drosophila: a parallel selection imposed by environmental ethanol and acetic acid. Proc. Natl. Acad. Sci. USA 90:3621-3625[Abstract/Free Full Text].

CINELLI, A. R., K. A. HAMILTON, and J. S. KAUER, 1995  Salamander olfactory bulb neuronal activity observed by video rate, voltage-sensitive dye imaging. III. Spatial and temporal properties of responses evoked by odorant stimulation. J. Neurophysiol. 73:2053-2071[Abstract/Free Full Text].

DAVIS, R., J. CHERRY, B. DAUWALDER, P. HAN, and E. SKOULAKIS, 1995  The cyclic AMP system and Drosophila learning. Mol. Cell. Biochem. 149:271-278.

DU, G. and G. D. PRESTWICH, 1995  Protein structure encodes the ligand binding specificity in pheromone binding proteins. J. Am. Chem. Soc. 34:8726-8732.

DUBIN, A. E., N. L. HEALD, B. CLEVELAND, J. R. CARLSON, and G. L. HARRIS, 1995  Scutoid mutation of Drosophila melanogaster specifically decreases olfactory responses to short chain acetate esters and ketones. J. Neurobiol. 28:214-233[Medline].

FLOWER, D. R., 1996  The lipocalin protein family—structure and function. Biochem. J. 318:1-14.

GANJIAN, I., M. J. PETTEI, K. NAKANISHI, and K.-E. KAISSLING, 1978  A photoaffinity-labelled insect sex pheromone for the moth Antheraea polyphemus. Nature 271:157-158.

GYORGYI, T. K., A. J. ROBY-SHEMKOVITZ, and M. R. LERNER, 1988  Characterization and cDNA cloning of the pheromone-binding protein from the tobacco hornworm, Manduca sexta: a tissue-specific developmentally regulated protein. Proc. Natl. Acad. Sci. USA 85:9851-9855[Abstract/Free Full Text].

HALL, J. C., 1994  The mating of a fly. Science 264:1702-1714[Abstract/Free Full Text].

HALPERN, M., 1987  The organization and function of the vomeronasal system. Annu. Rev. Neurosci. 10:325-362[Medline].

HARTENSTEIN, V. and J. W. POSAKONY, 1989  Development of adult sensilla on the wing and notum of Drosophila melanogaster. Development 107:389-405[Abstract].

HEKMAT-SCAFE, D., R. STEINBRECHT, and J. R. CARLSON, 1997  Coexpression of two odorant binding homologs in Drosophila: implications for olfactory coding. J. Neurosci. 17:1616-1624[Abstract/Free Full Text].

KAISSLING, K.-E., 1998 Olfactory transduction in moths. I. Generation of receptor potentials and nerve impulses, II. Extracellular transport, deactivation and degradation of stimulus molecules, pp. 93–112 in From Structure to Information in Sensory Systems, edited by C. TADDEI-FERRETTI and C. MUSIO. World Scientific, Singapore/New Jersey/London/Hong Kong.

KARESS, R. E. and G. M. RUBIN, 1984  Analysis of P transposable element functions in Drosophila. Cell 38:135-146[Medline].

KEIL, T. A., and R. A. STEINBRECHT, 1984 Mechanosensitive and olfactory sensilla of insects, pp. 477–516 in Insect Ultrastructure, edited by R. C. KING and H. AKAI. Plenum Press, New York.

LANGER-SOFER, P. R., M. LEVINE, and D. C. WARD, 1982  Immunological method for mapping genes on Drosophila polytene chromosomes. Proc. Natl. Acad. Sci. USA 79:4381-4385[Abstract/Free Full Text].

LINDSLEY, D. L., and G. ZIMM, 1992 The genome of Drosophila melanogaster. Academic Press, San Diego.

LIS, J. T., J. A. SIMON, and C. A. SUTTON, 1983  New heat shock puffs and beta-galactosidase activity resulting from transformation of Drosophila with an hsp70-LacZ hybrid gene. Cell 35:403-410[Medline].

MANIATIS, T., E. F. FRITSCH and J. SAMBROOK, 1982 Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

MCKECHNIE, S. W. and P. MORGAN, 1982  Alcohol dehydrogenase polymorphism of Drosophila melanogaster: aspects of alcohol and temperature variation in the larval environment. Aust. J. Biol. Sci. 35:85-93.

MCKENNA, M. P., D. S. HEKMAT-SCAFE, P. GAINES, and J. R. CARLSON, 1994  Putative pheromone-binding proteins expressed in a subregion of the olfactory system. J. Biol. Chem. 269:1-8[Free Full Text].

MONTE, P., C. WOODARD, R. AYER, M. LILLY, and H. SUN et al., 1989  Characterization of the larval olfactory response in Drosophila and its genetic basis. Behav. Genet. 19:267-283[Medline].

OLIVER, D. V. and J. P. PHILIPS, 1970  Technical note. Dros. Info. Serv. 45:58.

PELOSI, P., 1994  Odorant-binding proteins. Crit. Rev. Biochem. Mol. Biol. 29:199-228[Medline].

PELOSI, P., N. E. BALDACCINI, and A. M. PISANELLI, 1982  Identification of a specific olfactory receptor for 2-isobutyl-3-methoxypyrazine. Biochem. J. 201:245-248[Medline].

PEVSNER, J., R. R. TRIFILETTI, S. M. STRITTMATER, and S. H. SNYDER, 1985  Isolation and characterization of an olfactory protein for odorant pyrazines. Proc. Natl. Acad. Sci. USA 82:3050[Abstract/Free Full Text].

PEVSNER, J., V. HOU, A. M. SNOWMAN, and S. H. SNYDER, 1990  Odorant-binding protein. Characterization of ligand binding. J. Biol. Chem. 265:6118-6125[Abstract/Free Full Text].

PFEIFFER, C. A. and R. E. JOHNSTON, 1994  Hormonal and behavioral responses of male hamsters to females and female odors: roles of olfaction, the vomeronasal system, and sexual experience. Physiol. Behav. 55:129-138[Medline].

PIKIELNY, C. W., G. HASAN, F. ROUYER, and M. ROSBASH, 1994  Members of a family of Drosophila putative odorant-binding proteins are expressed in different subsets of olfactory hairs. Neuron 12:35-49[Medline].

PIRROTTA, V., 1986 Cloning Drosophila genes, pp. 83–110 in Drosophila: A Practical Approach, edited by D. B. ROBERTS. IRL Press, Oxford.

RAMING, K., J. KREIGER, and H. BREER, 1989  Molecular cloning of an insect pheromone-binding protein. FEBS Lett. 256:215-218[Medline].

RAMING, K., J. KRIEGER, and H. BREER, 1990  Primary structure of a pheromone-binding protein from Antheraea pernyi: homologies with other ligand carrying proteins. J. Comp. Physiol. B 160:503-509[Medline].

RAY, K. and V. RODRIGUES, 1995  Cellular events during development of the olfactory sense organs in Drosophila melanogaster. Dev. Biol. 167:426-438[Medline].

RIESGO-ESCOVAR, J., C. WOODWARD, P. GAINES, and J. CARLSON, 1992  Development and organization of the Drosophila olfactory system: an analysis using enhancer traps. J. Neurobiol. 23:947-964[Medline].

RIESGO-ESCOVAR, J. R., W. B. PIEKOS, and J. R. CARLSON, 1997  The Drosophila antenna: ultrastructural and physiological studies in wildtype and lozenge mutants. J. Comp. Physiol. A 180:151-160[Medline].

ROBERTSON, H. M., C. R. PRESTON, R. W. PHILLIS, D. M. JOHNSON-SCHLITZ, and W. K. BENZ et al., 1988  A stable source of P element transposase in Drosophila melanogaster. Genetics 118:461-470[Abstract/Free Full Text].

RODRIGUES, V., 1988  Spatial coding of olfactory information in the antennal lobe of Drosophila melanogaster. Brain Res. 453:299-307[Medline].

RODRIGUES, V. and E. BUCHNER, 1984  [³H] 2-deoxyglucose mapping of odor-induced neuronal activity in the antennal lobes of Drosophila melanogaster. Brain Res. 324:374-378[Medline].

ROELOFS, W. L., 1995  Chemistry of sex attraction. Proc. Natl. Acad. Sci. USA 92:44-49[Abstract/Free Full Text].

SAIKI, R. K., S. SCHARF, F. FALOONA, K. B. MULLIS, and G. T. HORN et al., 1985  Enzymatic amplification of ß-Globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 230:1350-1354[Abstract/Free Full Text].

SASS, G. L., J. D. MOHLER, R. C. WALSH, L. J. KALFAYAN, and L. L. SEARLES, 1993  Structure and expression of hybrid dysgenesis-induced alleles of the ovarian tumor (otu) gene in Drosophila melanogaster. Genetics 133:253-263[Abstract].

SENGUPTA, P., J. H. CHOU, and C. I. BARGMANN, 1996  odr-10 encodes a seven transmembrane domain receptor required for responses to the odorant diacetyl. Cell 84:899-909[Medline].

SHEPHERD, G. M., and C. A. GREER, 1990 Olfactory bulb, pp. 133–169 in The Synaptic Organization of the Brain, edited by G. M. SHEPHERD. Oxford University Press, New York.

SIDDIQI, O., 1987  Neurogenetics of olfaction in Drosophila melanogaster. Trends Neurosci. 3:137-142.

SMITH, D. P., 1996  Olfactory mechanisms in Drosophila melanogaster. Curr. Opin. Neurobiol. 6:500-505[Medline].

SMITH, D. P., R. RANGANATHAN, R. W. HARDY, J. MARX, and T. TSUCHIDA et al., 1991  Photoreceptor deactivation and retinal degeneration mediated by a photoreceptor-specific protein kinase C. Science 254:1478-1484[Abstract/Free Full Text].

STAMNES, M. A., B.-H. SHIEH, L. CHUMAN, G. L. HARRIS, and C. S. ZUKER, 1991  The cyclophilin homolog ninaA is a tissue-specific integral membrane protein required for the proper synthesis of a subset of Drosophila rhodopsins. Cell 65:219-227[Medline].

STEINBRECHT, R. A., 1969 Comparative morphology of olfactory receptors, pp. 3–21 in Olfaction and Taste III, edited by C. PFAFFMAN. Rockefeller University Press, New York.

STEINBRECHT, R. A., 1996  Are odorant binding proteins involved in odorant discrimination? Chem. Senses 21:719-727[Abstract/Free Full Text].

STOCKER, R. F., 1994  The organization of the chemosensory system in Drosophila melanogaster: a review. Cell Tissue Res. 275:3-26[Medline].

STOCKER, R. F., R. N. SINGH, M. SCHORDERET, and O. SIDDIQI, 1983  Projection patterns of different types of antennal sensilla in the antennal glomeruli of Drosophila melanogaster. Cell Tissue Res. 232:237-248[Medline].

TALLURI, S. and D. P. SMITH, 1995  Identification of a Drosophila G protein subunit (dGq-3) expressed in chemosensory cells and in central neurons. Proc. Natl. Acad. Sci. USA 92:11475-11479[Abstract/Free Full Text].

TRIPOULAS, N. A., E. HERSPERGER, D. LA JEUNESSE, and A. SHEARN, 1994  Molecular genetic analysis of the Drosophila melanogaster gene absent, small or homeotic discs1 (ash1). Genetics 137:1027-1038[Abstract].

TROEMEL, E. R., B. E. KIMMEL, and C. I. BARGMANN, 1997  Reprogramming chemotaxis responses: Sensory neurons define olfactory preferences in C. elegans. Cell 91:161-169[Medline].

VAN DELDEN, W., 1982  The alcohol dehydrogenase polymorphism in Drosophila melanogaster: Selection at an enzyme locus. Evol. Biol. 15:187-222.

VOGT, R. G. and L. M. RIDDIFORD, 1981  Pheromone binding and inactivation by moth antennae. Nature 293:161-163[Medline].

VOGT, R. J., L. M. RIDDIFORD, and G. D. PRESTWICH, 1985