Articles, Cellular/Molecular

Reduced Odor Responses from Antennal Neurons of Gqα, Phospholipase Cβ, and rdgA Mutants in Drosophila Support a Role for a Phospholipid Intermediate in Insect Olfactory Transduction

Pinky Kain, Tuhin Subra Chakraborty, Susinder Sundaram, Obaid Siddiqi, Veronica Rodrigues and Gaiti Hasan
Journal of Neuroscience 30 April 2008, 28 (18) 4745-4755; https://doi.org/10.1523/JNEUROSCI.5306-07.2008

Abstract

Mechanisms by which G-protein-coupled odorant receptors transduce information in insects still need elucidation. We show that mutations in the Drosophila gene for Gqα (dgq) significantly reduce both the amplitude of the field potentials recorded from the whole antenna in responses to odorants as well as the frequency of evoked responses of individual sensory neurons. This requirement for Gqα is for adult function and not during antennal development. Conversely, brief expression of a dominant-active form of Gqα in adults leads to enhanced odor responses. To understand signaling downstream of Gqα in olfactory sensory neurons, genetic interactions of dgq were tested with mutants in genes known to affect phospholipid signaling. dgq mutant phenotypes were further enhanced by mutants in a PLCβ (phospholipase Cβ) gene, plc21C. Interestingly although, the olfactory phenotype of mutant alleles of diacylglycerol kinase (rdgA) was rescued by dgq mutant alleles. Our results suggest that Gqα-mediated olfactory transduction in Drosophila requires a phospholipid second messenger the levels of which are regulated by a cycle of phosphatidylinositol 1,4-bisphosphate breakdown and regeneration.

Introduction

In all animals studied, odorant detection is initiated by interaction of volatiles with seven transmembrane domain receptors. In vertebrates, these receptors are coupled to the olfactory G-protein (Golf), which activates adenyl cyclase III and raises cAMP levels in sensory neurons resulting in the activation of a cyclic nucleotide gated ion channel followed by the opening of a chloride channel (Ache and Young, 2005). In contrast to this well established mechanism, the events that follow odor reception in invertebrates are still rather controversial. In insects, the role of G-protein-coupled sensory transduction is confounded by the fact that the highly conserved OR83b receptor heterodimerizes with conventional receptors (ORs) and the Manduca OR/OR83b dimer directly confers ligand stimulated nonselective channel activity in heterologous systems (Nakagawa et al., 2005). Cell culture and in vivo evidence that the Drosophila Or83b exhibits an unusual membrane topology with the proposed cytoplasmic G-protein binding region lying on the extracellular side has also been provided (Benton et al., 2006). These findings beg the analysis of the role of G-protein-coupled signaling during olfactory transduction in invertebrates.

Previous evidence supports a role for cAMP (Martin et al., 2001; Gomez-Diaz et al., 2004) and inositol 1,4,5-trisphosphate (InsP3) as second messengers downstream of receptor activation in arthropods (Krieger and Breer, 1999). More recently, changes in the levels of membrane-bound phospholipids have been shown to alter the primary response to odorants in crustaceans (Zhainazarov et al., 2004). These studies together favor the possibility of differential activation of dual transduction systems involving both Gsα (via cAMP) and Gqα (via InsP3 and other phospholipids) during olfactory sensing (Ache and Young, 2005). Moreover, expression of Gqα RNA was demonstrated in a subset of olfactory sensory neurons (OSNs) (Talluri et al., 1995) and its knockdown by targeted RNA interference (RNAi) resulted in behavioral deficits (Kalidas and Smith, 2002).

In the present study, we directly tested the role of Gqα in olfactory transduction through the electrophysiological measurements of antennal neuron responses in several classes of mutants in the Drosophila gene for Gqα (dgq). Null mutants in dgq are recessive lethal as expected from previous expression studies (Ratnaparkhi et al., 2002). Hence to measure olfactory responses from the adult olfactory organs of dgq null alleles, we used the mosaic analysis with repressible cell marker (MARCM) method (Lee and Luo, 1999) to produce green fluorescent protein (GFP) marked clones of dgq null antennal neurons. A developmental contribution of Gqα was ruled out by adult-specific rescue of the physiological deficits observed with a Gqα transgene. To understand signaling mechanisms downstream of Gqα, the interaction of dgq mutants was tested with existing mutants and RNAi lines, predicted to affect phospholipid intermediates.

Materials and Methods

Fly stocks.

The dgq221c allele was generated by P-element excision of the P{SUPor-P}Galpha49B[KG07290] strain carrying an insertion in the 5′-untranslated region (UTR) of dgq; the dgq1370 allele was generated by ethyl methane sulfonate mutagenesis. Mutations were verified by their failure of complementation with Df(2R)vg-C, which uncovers the dgq locus and subsequent molecular analysis. GqGAL4 and UASdgqα3 stocks were provided by Sonal Srikanth [National Centre for Biological Sciences (NCBS), Bangalore, India] and Neha Agrawal (NCBS); details of the RNAi construct for dgq (UASGq1F1) and UASAcGq transgenics are published (Ratnaparkhi et al., 2002; Banerjee et al., 2006). plc21C insertion allele (plc21CP319) and the deficiency Df(2L)p60A, which uncovers plc21C, were obtained from S. Leevers (Cancer Research Institute, London, UK) (Weinkove et al., 1999). rdgA1 and rdgA3 stocks were provided by Dr. R. Padinjat (The Babraham Institute, Cambridge, UK) (Hardie et al., 2002). Tub-PGAL80ts, Tub-PGAL80ts (McGuire et al., 2003) recombinant was provided by Albert Chiang (NCBS). y,w; P{ry+t7.2 = neoFRT}42B/CyO, Tub-GAL80 FRT42B, ey-Flp; hsGAL4t/CyO, and Df(2R)vg-C/CyO were obtained from the Drosophila Stock Center (Bloomington, IN). Or22aGAL4 and Or83bGAL4 strains were kindly provided by Leslie Vosshall (Rockefeller University, New York, NY) (Vosshall et al., 2000). RNAi lines for rdgA and plc21C were obtained from the VDRC stock center. Stocks were grown on standard cornmeal medium at 25°C. White prepupae [0 h after puparium formation (APF)] were collected and allowed to develop on moist filter paper. Wild-type pupae take ∼100 h to eclose when grown at 25°C in our laboratory.

DNA sequencing.

The dgq gene was amplified by the PCR from genomic DNA from dgq1370 larvae, cloned into a TOPO vector (Invitrogen, Carlsbad, CA) and sequenced by standard methods. The background strain used for generating dgq1370 was sequenced in the same region to verify that the mutation was not a result of a background polymorphism.

Reverse transcription and PCRs.

Total RNA was isolated from 200 adult antennas following standard procedures. Primers complementary to exon 2 (forward primer, 5′-GGTCACCTCATCGAGCAAGC-3′) and exon 3 (reverse primer, 5′-GAGCAAGTCTGCATGCAACGG-3′) were used for reverse transcription and PCR analysis (RT-PCR) of plcβ21C, according to published procedures (Nair et al., 2007).

Clonal analysis.

Clones of dgq mutants were generated using the MARCM method (Lee and Luo, 1999). dgq221c and dgq1370 were recombined onto the FRT42B chromosome. Females of genotype FRT42B, dgq221C/CyOGFP; ey-Flp/TM6Tb and FRT42B, dgq1370/CyOGFP; ey-Flp/TM6Tb were crossed to Tub-GAL80 FRT42B/CyO; OR83bGAL4, UAS2XEGFP/TM6Tb. Non-CyOGFP, non-Tb flies were used for electroantennogram (EAG) recordings and the presence of clones was monitored by expression of GFP in the third antennal segment. To visualize projections of OSNs into the antennal lobe, Or22aGAL4; FRT42B dgq221c/CyO; ey-Flp/TM6Tb and Or22aGAL4; FRT42B, dgq1370/CyO; ey-Flp/TM6Tb were mated with Tub-GAL80 FRT42B/CyO; UASNsybGFP/TM6Tb males. Non-CyO, non-Tb flies were dissected and stained using anti-GFP and mAbnc82.

Cuticle preparation.

Adult antennas were removed and placed in a drop of Faure's solution (34% v/v chloral hydrate, 13% v/v glycerol, and 20 mg/ml gum-arabic) and allowed to clear overnight at 70°C. Sensilla were counted after projecting Nomarski images from the microscope onto the video monitor. The mean and SD of sensillar numbers were estimated from at least five antennas for each genotype.

Immunohistochemistry.

Dissection and antibody staining of pupal antenna and brain as well as whole mounts of the adult brain were performed as described previously (Jhaveri et al., 2000). Fifteen micrometer frozen sections through the fly head were fixed in 4% paraformaldehyde in PBS for 30 min, washed extensively, and blocked in PBS containing 5% normal goat serum, 0.1% bovine serum albumin, and 0.03% Triton X-100 for 1 h at room temperature. Where mAbnc82 was to be used, the protocol of Laissue et al. (1999) was followed. The primary antibodies used were anti-Gq (1:200; SC-3912; Santa Cruz Biotechnology, Santa Cruz, CA), rabbit anti-GFP (1:10,000; Invitrogen), mAb22C10 (1:5; Developmental Studies Hybridoma Bank, Iowa City, IA), mAbnc82 (1:10; Developmental Studies Hybridoma Bank), and mouse anti-GFP (1:20; Sanmar Chemicals, Bangalore, India). Secondary antibodies were anti-mouse and anti-rabbit IgG conjugated to either Alexa 488 or Alexa 568 (1:200; from Invitrogen). Labeled samples were mounted in 70% glycerol and examined in the Bio-Rad (Hercules, CA) Radiance 2000 or an Olympus (Tokyo, Japan) FV1000, at 1 μm intervals; data were processed using Image J, Confocal Assistant 4.2, and Adobe Photoshop 5.5.

For immunochemical staining of adult antenna, the A3 segments were chopped in chilled PBS and fixed in 4% paraformadehyde with 3% Triton X-100 on ice for 5 h followed by three washes of 1 h each with PBS containing 2% Triton X-100, initially at room temperature and then at 4°C for 2 d. Samples were incubated for 3 d at 4°C in 1:5 dilution of mAb22C10 and anti-GFP (1:10,000). Antennas were washed three times for 1 h each in PBS containing 0.1% Triton X-100 after primary antibody incubation and treated with 1:200 dilution of anti-mouse and anti-rabbit IgG secondary antibodies conjugated to either Alexa 488 or Alexa 568, for 1 d at 4°C. After washes with PBS containing 0.1% Triton X-100, samples were mounted in anti-fading agent Vectashield (Vector Laboratories, Burlingame, CA), and examined in an Olympus FV1000 confocal microscope.

Electroantennogram recordings.

Electrophysiological recordings were performed as described previously (Ayer and Carlson, 1991). Two- to 6-d-old female flies were immobilized on ice and wedged into an appropriately trimmed tapering ends of a yellow pipette tip allowing the head to protrude. Low melting point (58°C) myristic acid wax was used to hold the head onto the thorax. Glass capillaries (<1 μm in diameter) filled with 0.8% NaCl and connected to a Ag/AgCl2 wire were used as electrodes. The reference electrode was inserted into the head capsule, whereas the recording electrode was placed in the basiconica rich region of the third antennal segment and connected to a high-impedance preamplifier (Electro 705; WPI, Sarasota, FL). Signals were viewed on a two-channel oscilloscope (TDS320; Tektronix, Beaverton, OR). Odorants, ethyl acetate (10−4), butanol (10−2), propionic acid (10−2), benzaldehyde (10−3), and iso amyl acetate (10−2), of the highest purity available (Sigma-Aldrich, St. Louis, MO), were freshly diluted in liquid paraffin and used for odor delivery. Odor pulses were delivered into a constant stream of air for 2 s. An interval of 5 min was allowed after each odorant pulse to minimize effects attributable to adaptation. The peak-to-trough amplitude of the response was measured after each pulse of odor stimulation. When different genotypes were being tested, the unpaired t test was used to calculate significance. A value of p < 0.01 was taken to be significant.

Single-unit recordings.

The reference electrode was inserted in the funicular tip of the antenna, and the recording electrode (<1 μm in diameter) was inserted into the base of an ab2 basiconic sensillum and connected to a high-impedance dual-channel preamplifier (model 750; WPI); identity of ab2 was ascertained by the characteristic spontaneous spike rate and by its location on the antenna. Signals were viewed on a two-channel oscilloscope (Tektronix TDS320). Odor pulses of 500 ms duration separated with an interstimulus interval of 10 s were given using an eight port custom-built olfactometer. Each sensillum was stimulated three times with 10−6, 10−5, and 10−4 dilution of ethyl acetate (Sigma-Aldrich), respectively, prepared fresh in liquid paraffin. The spike frequency was calculated using Labview 7.1 (National Instruments, Austin, TX) from an average of three trials. Recordings were obtained from numbers as described in the individual figure legends; minimally these were from six or more sensilla, taken from five or more flies.

Results

Mutants in the Drosophila dgq gene

The dgq gene is located at 49B on the second chromosome and encodes the only known functional Gq-like α subunit of heterotrimeric G-proteins in Drosophila. The gene encodes at least two functional isoforms, which arise by alternative splicing. Among these, dgqα1 is expressed primarily in the adult eye (Lee et al., 1990), whereas dgqα3 mRNA is expressed widely through development including the adult head and appendages (Talluri et al., 1995; Ratnaparkhi et al., 2002). Based on RNA expression analysis and behavioral studies with an RNAi construct (Kalidas and Smith, 2002), it has been suggested that a Gqα isoform might function downstream of olfactory G-protein-coupled receptors in antennal sensory neurons. To test this idea directly, we generated mutations in the dgq gene by excision of a P-element in the 5′-UTR of dgq, and by ethyl methane sulfonate mutagenesis. Molecular analysis demonstrated that the P-excision allele, dgq221c, carries a lesion in the 5′ end of dgq spanning the translation start site in exon 3, thus rendering it null. Exons located 3′ to exon 3 are intact (Banerjee et al., 2006). Both dgq221c and dgq1370 (induced by chemical mutagenesis) are lethal as homozygotes and as heterozygotes with Df(2R)vg-C, which uncovers the dgq locus. The lethality in dgq1370 fails to complement that of dgq221c. Homozygous dgq1370 mutants produce a transcript with a G→A mutation at base pair 1933 (Fig. 1). This would result in a change from arginine (AGA) to lysine (AAA) at residue 207, which lies in the switch II helix region that is highly conserved among α subunits of all heterotrimeric G-proteins (Fig. 1). Structural studies show that this region undergoes a conformational switch on GTP binding, which is considered essential for effector function (Noel et al., 1993). Specifically, the arginines at positions 204 and 207 of Dgqα3 correspond to Arg205 and Arg208 of Gαi (Mixon et al., 1995). Both these residues have been shown to provide important stabilizing contacts after GTP binding with glutamates located further downstream. Presumably, the mutation in dgq1370 destabilizes the GTP-bound state of Gq and renders it effectively dead by preventing activation of the downstream effector. Normal levels of the mutant protein are present in dgq1370 homozygous larvae (Fig. 1).

Figure 1.
Figure 1.

Identification of a single base pair change in dgq1370. G→A change is found in genomic DNA isolated from dgq1370. As a consequence, a highly conserved arginine residue in the switch region of Gqα is mutated to lysine. The Gqα sequences from Caenorhabditis elegans (EGL-30), human, mouse, and Drosophila are compared. Protein levels of Gqα3 are not changed significantly in larvae of the genotype dgq1370/Df(vg-C) compared with dgq1370/± larvae.

Olfactory receptor neurons lacking dgq function exhibit reduced stimulus-evoked activity

Animals homozygous for the null alleles, dgq221c and dgq1370, die as either first- or early second-instar larvae. To study the odorant response from adult antennas, we generated animals with homozygous mutant clones on the antenna using the MARCM method (Lee and Luo, 1999). Mitotic recombination at the FRT42B target site was induced using Flipase (Flp) driven by the eyeless (ey) promoter, which has been previously demonstrated to generate clones mostly in visual and olfactory sensory neurons (OSNs) and only minimally in the central brain (Zhang et al., 2003; Jefferis et al., 2004). Mutant clones identified by GFP driven by the OSN-specific “driver” Or83bGAL4 (Larsson et al., 2004) were obtained in >80% of progeny of the appropriate genotypes (FRT42B, dgqnull/FRT42B, tubPGAL80; ey-Flp/OR83bGAL4, UAS2XEGFP) (Fig. 2A). Animals with strong GFP fluorescence in their antennas were chosen for the experiment, whereas animals with no visible GFP expression (i.e., no mutant clones) were control for effects attributable to background genotypes (Fig. 2A). Confocal sectioning of antenna bearing large clones, revealed significant mixing of marked and nonmarked cells suggesting that ey-FLP induces FRT-mediated recombination at multiple times during antennal development. The olfactory response was measured after pulsed odor delivery. EAGs (Fig. 2A) were obtained after each successive pulse of a concentration of each of five different odorants (Fig. 2B–F). Odorant concentrations used were chosen such that they lay within the range of the log linear response to that specific odorant, when tested on wild-type flies (data not shown). The response of wild-type (CS) and control flies bearing no detectable GFP-positive cells gave a comparable range of amplitudes for each of the odors tested (p > 0.05 in all cases except benzaldehyde and isoamyl acetate where p > 0.01; symbols in different colors in Fig. 2B–F represent different individuals). Antennas bearing clones of homozygous dgq221c or dgq1370 tissue showed a response reduced to 3–10 mV compared with 8–16 mV in the controls (a comparison of the median responses of mutant and normal individuals showed significant difference; p < 0.001 in all cases). We ascertained that the mutation did not affect the morphology of the antenna by counting the three anatomically characterized sensilla (trichoidea, coeloconica, and basiconica); their numbers appeared normal in antenna with large clones homozygous for dgq221c and dgq1370 (supplemental Fig. S1, available at www.jneurosci.org as supplemental material). Antennal sensory neurons and their projections also appear normal as judged by immunohistochemical staining with mAb22C10 and anti-GFP shown in Figure 5, J and K.

Figure 2.
Figure 2.

Electroantennograms from dgq-null clones have reduced amplitudes. A, The MARCM technique was used to generate large GFP-positive clones in the third antennal segment from flies of the genotype FRT42Bdgqnull/FRT42BGAL80; ey-Flp/Or83bGAL4, UAS2XEGFP. A 1 μm confocal section through the third antennal segment (A3) reveals the presence of clonal neurons (in green). Representative EAG traces from clonal antenna as well as internal controls (animals of the same genotype but with no visible GFP fluorescence in the third antennal segment) are shown. The diagrams indicate clones in each case (green shading). B–F, Response profile of dgqnull clones and control siblings to a spectrum of odorants. Amplitudes of responses obtained from individual antennas of different genotypes are shown in colored symbols. In B–F, CS is the wild-type strain; control, FRT42Bdgqnull/FRT42BGAL80; ey-Flp/Or83bGAL4, UAS2XEGFP (with undetectable GFP positive clones); dgq221c, FRT42Bdgq221c/FRT42BGAL80; ey-Flp/Or83bGAL4,UAS2XEGFP; and dgq1370, FRT42Bdgq1370/FRT42BGAL80; ey-Flp/Or83bGAL4,UAS2XEGFP with large GFP mutant clones (see example in A); rescue indicates animals with large dgqnull clones and a UASdgqα3 transgene (UASdgqα3/+; FRT42Bdgq221c/FRT42BGAL80; ey-Flp/Or83bGAL4, UAS2XEGFP or UASdgqα3/+; FRT42Bdgq1370/FRT42BGAL80; ey-Flp/Or83bGAL4, UAS2XEGFP). Odor pulses are of 2 s duration followed by a gap of 5 min. Differences between the amplitudes for control and rescue strains are not significant for any of the odorant tested (p > 0.1). Both dgqnull strains are significantly different from the corresponding control strain (p < 0.001). G–K, A ubiquitously expressed temperature-sensitive GAL80 transgene was introduced in the rescue strain in B–F (TubGAL80tsx2/UASdgqα3; FRT42Bdgqnull/FRT42BGAL80; ey-Flp/Or83bGAL4, UAS2XEGFP). Animals reared throughout at 18°C are shown as control 1 and control 2 in G–K. Recordings from 15 animals showed a range of responses from 2 to 10 mV as shown by the clones in B–F. Experimental animals were shifted to 29°C immediately after eclosion for 5–6 d. dgq221c or dgq1370 mutant animals without the rescue transgene (TubGAL80tsx2/+; FRT42Bdgqnull/FRT42BGAL80; ey-Flp/Or83bGAL4, UAS2XEGFP) reared at 29°C showed similar responses as control animals grown at 18°C. Rescue animals (TubGAL80tsx2/UASdgqα3; FRT42Bdgqnull/FRT42BGAL80;ey-Flp/Or83bGAL4, UAS2XEGFP) showed higher responses and rescued the phenotype.

To confirm that lower EAG responses were indeed attributable to mutations in the dgq gene, a UASdgqα3 transgene (Banerjee et al., 2006) was expressed in dgq mutants using the GAL4/UAS system (Brand and Perrimon, 1993). In these clones, dgqα3 was coexpressed with GFP by Or83bGAL4. Animals bearing dgqnull antennal clones in the background of Or83bGAL4;UAS-dgqα3 responded with higher amplitudes in EAG recordings (Fig. 2B–F, data marked as rescue) (the median responses showed significant difference of p < 0.005, for all cases). To test whether the requirement for Dgq in odor sensing was sufficient during adulthood, we introduced a ubiquitously expressed temperature-sensitive GAL80 transgene (TubGAL80ts) (McGuire et al., 2003) into the rescue strains discussed above and grew them at permissive temperature (18°C) to silence GAL4 activity. Experimental animals were shifted to nonpermissive temperature (29°C) a few hours before eclosion to induce GAL4 expression and maintained at this temperature until tested, whereas controls were maintained at 18°C throughout. As shown in Figure 2, G–K (data marked as rescue), the UASdgqα3 transgene rescues reduced EAG amplitudes even when expression begins in mature OSNs (by inactivation of the TubGAL80tsx2 transgene at 29°C very late in pupal development; p < 0.001). EAG amplitudes remain low in animals grown at 18°C (Fig. 2G–K, control 1 and control 2) or in animals transferred to 29°C as late pupae, but without the UASdgqα3 transgene (Fig. 2G–K, dgqnull controls at 29°C). Thus, normal function of the dgq gene (and most probably the dgqα3 splice variant) is necessary in adults for normal activity at the olfactory receptor neurons in response to odor stimuli.

It is known that a mutation in dgq affecting the visual splice variant (dgqα1) compromises the light response as measured by electroretinograms (ERGs) (Scott et al., 1995). As expected, ey-Flp also generated clones of dgq nulls in the retina (data not shown). As an independent confirmation of the clonal analysis, we measured ERGs from flies with visible antennal clones and found that responses were reduced to the same level as seen with the dgq1 allele. Moreover, dgq221c/dgq1 and dgq1370/dgq1 flies are viable and show compromised phototransduction, although not to the same extent as shown previously by dgq1/Df flies (data not shown) (Scott et al., 1995). The residual visual response observed in dgq221c/dgq1 and dgq1370/dgq1 animals may arise from different genetic backgrounds of the three dgq alleles tested. Because the dgq1 allele affects only the visual splice variant of dgq (Scott et al., 1995; Ratnaparkhi et al., 2002), neither dgq221c/dgq1 nor dgq1370/dgq1 flies exhibited any defects in EAG measurements (data not shown). These data together with EAG recordings suggest that light and odor transduction in Drosophila exploit common mechanisms albeit using distinct transcripts of the dgq gene.

Odor responses in antennas are sensitive to levels of Gq

If Gqα is indeed stimulated by odor-bound receptors, it is expected that expression of a constitutively active Gqα subunit (AcGq3) (Ratnaparkhi et al., 2002) could affect the EAG response. To measure the effect of AcGq3 expression on olfactory transduction without compromising possible antennal development, a hsGAL4 strain (hsGAL4t) that has minimal basal expression at 25°C (Banerjee et al., 2006) was used. Flies of the genotype hsGAL4t/UASAcGq3 (and appropriate control genotypes) were reared at 18°C to minimize AcGq3 expression during pupal development. It has been noted that all flies including wild-type controls produced smaller EAG responses when reared at 18°C compared with responses generated at 25°C (O. Siddiqi, unpublished observations) (Fig. 3A–D) [e.g., ethyl acetate (EA) at 10−4 induces an EAG of ∼12 mV at 25°C but only ∼6 mV at 18°C]. To induce AcGq3 expression, 4- to 5-d-old animals were heat shocked at 37°C for either 15 or 30 min. These flies, as well as similarly pulsed control genotypes, were allowed to recover for 1 h at 25°C and EAG responses were stimulated with three concentrations of ethyl acetate and propionic acid, representing basiconic and coeloconic hairs, respectively. Expression of AcGq3 significantly enhanced the amplitude of the EAG response at all concentrations of the two odorants tested (Fig. 3A,B) (p < 0.001). The enhancement was greater in flies that received a heat shock for 30 min (p < 0.0001). The AcGq3 transgene has a Q203L mutation (Ratnaparkhi et al., 2002), which in other Gα subunits, including Dgqα1, has been shown to slow down, but not abolish, their GTPase activity (Lee et al., 1994). Expression of this protein in OSNs enhances the normal EAG response, suggesting when Gqα molecules remain GTP-bound and therefore active for longer times after stimulation, there is greater activation of downstream signaling components. EAG responses were also tested from flies expressing AcGq3 after a 1 h heat shock, in which they appear reduced or similar to control values (Fig. 3C) (p ≫ 0.1). An increased sensitivity of response at low levels of induction followed by a drop at higher levels is not unexpected because prolonged Gqα activation would be expected to deplete the membrane-bound lipid substrate phosphatidylinositol 4,5-bisphosphate (PIP2) or lead to an adaptation of the transduction machinery.

Figure 3.
Figure 3.

Odor responses in antennas are sensitive to levels of Gqα in sensory neurons. A–C, Flies of the genotype w1118;hsGAL4t/UASAcGq and control animals CS, w1118;hsGAL4t/+ and w1118;UASAcGq/+ were reared at 18°C throughout development. These animals exhibit a general reduction in EAG amplitudes compared with animals reared at 25°C (D). To induce AcGq expression, 4- to 5-d-old experimental and control animals were subjected to 37°C temperatures for 15 min (A), 30 min (B), or 1 h (C). EAG responses were measured at 10−6 to 10−4 dilution of ethyl acetate (EA) and 10−4 to 10−2 propionic acid (PA). Expression of AcGq significantly enhanced the amplitude of the EAG response at all concentrations of the two odorants tested (A, p < 0.001; B, p < 0.0001) for heat shocks of 15 and 30 min. No enhancement of response was observed after a 1 h heat shock (C, p > 0.1). E, Expression of an RNAi construct for dgq in the antenna reduces response to odorants. Representative EAG traces for ethyl acetate (10−4) in controls (GqGAL4/+, UASGq1F1/+ and OR83bGAL4, UAS2XEGFP/+) and experimental animals (GqGAL4/UASGq1F1 and UASGq1F1/+; OR83bGAL4, UAS2XEGFP/+) are shown. F, Average amplitude of EAG responses for butanol (10−2) (BUT), ethyl acetate (10−4) (EA), propionic acid (10−2) (PA), benzaldehyde (10−3) (BENZ), and iso-amyl acetate (10−2) (IAA) in animals of genotypes shown in E. Bars represent mean ± SEM (n = 10) for each genotype (p < 0.0001 for both genotypes). G, Dose responses for ethyl acetate and propionic acid. Recordings from controls (CS, GqGAL4/+, UASGq1F1/+ and OR83bGAL4, UAS2XEGFP/+) and experimental animals (GqGAL4/UASGq1F1 and UASGq1F1/+; OR83bGAL4, UAS2XEGFP/+) are plotted. Response of the experimental animals is significantly different from controls (p < 0.0001).

Next, we measured EAG responses in flies with reduced Gqα levels by expressing a Gqα RNAi construct (UASGq1F1) (Banerjee et al., 2006; Xia and Tully, 2007) in a majority of OSNs (Or83bGAL4) and in the expression domain of dgq using a dgqGAL4 driver. This line was constructed by fusing 3.9 kb of dgq upstream sequence plus 3.4 kb of the 5′-untranslated region with the GAL4 coding sequence. A significant reduction (4–5 mV) was observed in the amplitude of EAGs recorded from both Or83b>GqRNAi and dgq>GqRNAi strains compared with the individual parental strains (p < 0.0001) for all the odorants tested (Fig. 3E,F). The EAG response to two odorants, ethyl acetate and propionic acid (representative of basiconic and coeloconic hairs, respectively), was also reduced at higher dilutions (Fig. 3G) (p < 0.0001). Together with data from mutant clones, these results strongly support the idea that a dgq gene product transduces signals downstream of multiple odorant receptors in Drosophila antennal neurons.

Ab2a neurons from basiconic sensilla of dgqnull clonal antennas show a reduced sensitivity to ethyl acetate

Our results show that antennas mutant for dgq alleles exhibit reduced electrical activity to all odors tested. To confirm that the reduced EAG amplitudes in dgq clonal antennas were indeed attributable to a change in the response from individual sensory neurons, we measured the response to ethyl acetate from the well studied ab2 class of large basiconic hairs in animals of the genotype FRT42Bdgqnull/FRT42BGAL80; ey-Flp/Or83bGFP. Ab2 sensilla are innervated by two OSNs of which the ab2a neuron expresses the Or59b receptor (Couto et al., 2005), which responds strongly to ethyl acetate (de Bruyne et al., 2001). We interrogated wild-type cells by stimulation with pulses of three dilutions of ethyl acetate and observed a dose-dependent increase in firing frequency (de Bruyne et al., 2001). We analyzed recordings obtained from single ab2 sensillum, which lay within the regions of the antenna bearing large GFP clones. Morphologically similar clones were examined by confocal sectioning and found to have 256 ± 28 GFP-positive neurons (based on counts from five antennas). Because each antenna possesses ∼1200 OSNs, we expected a 20–25% probability of finding a mutant neuron when sampled at random. We recorded from 18 ab2 sensilla present on antennas of the genotype FRT42Bdgq1370/FRT42BGAL80; ey-Flp/Or83bGFP. Among these, we obtained four sensilla in which the firing frequency of ab2a in response to pulses of ethyl acetate rose only minimally above the background firing rate when the odorant was delivered at dilutions of 10−6 and 10−5 (Fig. 4A). Their response at 10−4 dilution was reduced by ∼50% compared with the controls (see below). As expected from the mosaic nature of the mutant antennas, in each of these animals a second ab2 hair responded normally to all concentrations of ethyl acetate (Fig. 4B). The remaining sensilla gave responses corresponding to controls as expected for neurons that are not included within the homozygous mutant patch (data not shown). Similar results were obtained with dgq221c mutant sensilla (supplemental Fig. S2A, available at www.jneurosci.org as supplemental material). In contrast all (11 of 11) ab2 sensilla responded with a normal increase of the firing frequency on stimulation with increasing concentrations of ethyl acetate, when antennas with large GFP clones were taken from the control genotype FRT42B/FRT42BGAL80; ey-Flp/Or83bGFP (supplemental Fig. S2B, available at www.jneurosci.org as supplemental material). These data firmly implicate Gqα signaling as one of the mechanisms directly used for odor sensing in OSNs. The presence of a significant residual response even in the complete absence of Gqα (as in the null clones) does suggest that other mechanisms must exist within the same cell. The presence of two or more transduction mechanisms coupled differentially to a single odorant receptor might help explain some of the observed complexities of odorant receptor responses (Hallem et al., 2004; Hallem and Carlson, 2006).

Figure 4.
Figure 4.

Odor response from single antennal neurons is reduced in Gqα mutants. A, Single-unit recordings were measured from the ab2 class of large basiconic hairs in animals of the genotype FRT42Bdgq1370/FRT42BGAL80; ey–Flp/Or83bGAL4, UAS2XEGFP for different concentrations of ethyl acetate. Mean ± SEM from four mutant cells (squares; n = 4) and controls from the same antennas (see text; circles) at 10−6 to 10−4 dilution of ethyl acetate (A). The response of mutants cells is significantly different from controls (p < 0.0001). B, C, Representative traces for control and dgq1370 sensilla. The bar above each trace indicates the period of odor delivery (500 ms). In the control strains GqGAL4/+ and UASGq1F1/+, both dilutions of ethyl acetate elicit an increased frequency of firing from the “a” neuron. The frequency of response of ab2a is strongly reduced in GqGAL4/UASGq1F1 compared with controls. D, Mean ± SEM of the change in ab2a neuron spike frequency on odor stimulation from the indicated genotypes. A minimum of eight recordings were evaluated for each genotype. The firing frequency from GqGAL4/UASGq1F1 animals is significantly different from controls at both concentrations of ethyl acetate (p < 0.0001).

Single-unit recordings were also obtained from dgqGAL4>UASGq1F1 animals. In control flies of the genotypes GqGAL4/+ and UASGq1F1/+, the firing frequency of neuron A in ab2 sensilla increased visibly on the delivery of a 500 ms pulse of ethyl acetate (10−4 and 10−5 dilutions). This response was conspicuously reduced in ab2 hairs from flies of the genotype GqGAL>UASGq1F1 (Fig. 4D,C, representative hairs).

The results we describe in dgq mutants could, in principle, arise from alterations in cellular metabolism because of changes in levels of lipid intermediates. However, this seems unlikely because the amplitudes and spontaneous neuronal firing rates of wild-type and mutant OSNs are indistinguishable from each other (p ≪ 0.001) indicating that in dgq mutant OSNs action potentials can fire normally (supplemental Fig. S2C,D, available at www.jneurosci.org as supplemental material).

Gq is expressed in a majority of antennal sensilla

Functional analysis shows that a Gqα isoform, possibly Dgqα3, is required in multiple antennal sensory neurons for their response to a range of odorants. Although the presence of dgqα3 RNA has been demonstrated in adult antennas, characterization of Gqα protein expression in developing and adult antennas has not been done (Talluri et al., 1995; Ratnaparkhi et al., 2002). Broad expression of Gqα in developing and adult antennal sensory neurons was confirmed by the following experiments. Initially, we looked at dgqGAL4-driven expression of UAS2EXGFP, which begins as early as 18 h APF in the differentiated neurons within the third antennal segment (Fig. 5A) and continues at 36 h APF when the projections into the antennal nerve become prominent (Fig. 5B, asterisk). Early expression of Gqα in the antennal system at 18 and 36 h APF was further confirmed by immunohistochemical staining with an antibody that recognizes the Dgqα3 isoform (Fig. 5D,E) (Ratnaparkhi et al., 2002). Expression continues in adults as seen by immunoreactivity to frozen sections of the antenna (Fig. 5F). Careful examination of confocal sections through dgqGAL4>GFP antenna stained with antibodies against GFP revealed a strong expression in a majority (but probably not all) OSNs (Fig. 5C). We further confirmed that the large basiconic sensilla that were analyzed in the single unit recordings described above were included in the dgq expression domain. From the broad pattern of expression observed in Figure 5, C and F, it seems likely that Gqα is also present in non-neural support cells surrounding the OSNs.

Figure 5.
Figure 5.

Gqα is expressed in OSNs of developing and adult antenna. A–F, At 18 h (A, D) and 36 h (B, E) APF, expression of dgq is detected in the third antennal segment by reporter expression in the dgqGAL4>UAS2XEGFP (A–C) as well as by anti-Dgq immunostaining (D–F). Expression is detected in neuronal processes (* in A, B, and arrows in D, E) as well as in some non-neuronal cells. Dgq expression is detected in 15 μm frozen sections of adult antenna visualized with anti-Dgq antibody (F). GFP expression is also seen in the third (A3) antennal segments of the dgqGAL4>UAS2EXGFP adults (C). The inset in C shows a region of the antenna enriched for sensilla basiconica (G) dgq221c OSNs are deficient for Dgq protein. Frozen sections (15 μm) of antennas of FRT42Bdgq221c/FRT42BGAL80; ey-Flp/Or83bGAL4, UAS2XEGFP double labeled with anti-Dgq and anti-GFP. H, Merge of 6 × 1 μm confocal stacks. Pixel intensities in the red channel within individual cells were measured in multiple 1 μm sections. Cells labeled with GFP (dgq221c/dgq221c) showed a mean value of 3.63 ± 1 (n = 5), whereas the unmarked cells (dgq221c/+) gave a value of 34.6 ± 2 (n = 5; p < 0.0001). Representative images (demarcated in the white boxes in G) of a dgq221c homozygous neuron (H) and a control dgq221c/+ neuron (I) are shown for both red and green channels. J, dgqnull adult antennas show normal OSN morphology. Antennas from FRT42Bdgq1370/FRT42BGAL80; ey-Flp/Or83bGAL4, UAS2XEGFP adults with large GFP-positive clones were stained for mAb22C10 (red) and anti-GFP (green). A magnified view of the region marked by a white box in J is shown in K.

To confirm absence of Gqα in dgq221c clonal OSNs, frozen sections (15 μm) of antennas were double labeled with anti-Dgqα3 and anti-GFP and analyzed by confocal microscopy (Fig. 5G). Figure 5G is a merge of several confocal stacks to demonstrate the presence of the clones. We selected sections covering single GFP-labeled neurons and quantified the pixel intensities resulting from anti-Dgqα3 immunoreactivity in each cell (Fig. 5G, boxed-in areas). A representative cell (Fig. 5H) demonstrates that the level of staining in homozygous mutant cells (marked with GFP) was no different from background levels. Nonclonal cells (not marked with GFP) show significant levels of Dgqα3 immunoreactivity (Fig. 5I). This result demonstrates that dgq clonal OSNs are indeed deficient for Dgq protein. Loss of Dgq protein, however, does not affect their morphology or axonal projections as observed by double labeling with a neuron-specific antibody mAb22C10 (Fig. 5J,K). Moreover, dgqnull mutant OSNs target correctly to the cognate antennal glomeruli (supplemental Fig. S3, available at www.jneurosci.org as supplemental material).

plc21C functions downstream of dgq

In the canonical receptor and G-protein-coupled signaling pathway, the enzyme phospholipase Cβ (PLCβ) is activated on GTP binding to Gqα, which cleaves the membrane bound phospholipid, PIP2, to generate soluble InsP3 and membrane-bound diacylglycerol (DAG). The Drosophila genome has two genes encoding for PLCβ -norpA (Bloomquist et al., 1988) and plc21C (Shortridge et al., 1991). In a previous study, we found that dgq221c and plc21CP319 mutant alleles have synergistic effects on larval viability and adult flight (Banerjee et al., 2006), indicating that PLCβ21C can function together with Drosophila Gqα. The plc21CP319 mutant allele is a hypomorph, with a P-insert in the first intron, and is homozygous viable, whereas plc21Cp60a is a small deficiency that removes the 5′ end of plc21C plus a neighboring essential gene p60 (Weinkove et al., 1999). A reduction of plc21C RNA levels in plc21CP319 homozygotes and plc21Cp60a heterozygotes was confirmed by RT-PCR (Fig. 6C). UASplc21C557 is a recently generated UASRNAi strain reported as specific for plc21C (http://stockcenter.vdrc.at/).

Figure 6.
Figure 6.

Trans-heterozygote interactions between alleles of dgq and plc21C. A, Average amplitudes of EAGs for five odorants are shown for the indicated genotypes. plc21Cp60A/+ (p60A/+), plc21CP319+ (P319/+), as well as dgq221c/+ and dgq1370/+ show a small reduction in response compared with the CS controls (p < 0.001). Responses of trans-heterozygote combinations p60A/dgq221c, p60A/dgq1370, P319/dgq221c, P319/dgq1370, and p60A/P319 were significantly different from heterozygous controls (p < 0.0001 in all cases). The observed reduction in response of the trans-heterozygotes was significantly different (p < 0.01 in all cases) from the expected drop if this resulted from the lowered response of the two heterozygotes. Reduction was also observed in RNAiplc21C557 (UASRNAiplc21C557/+; OR83bGAL4/+), RNAiplc21C557/RNAiGq (UASRNAiplc21C557/UASGq1F1; OR83bGAL4/+), and GqGAL4/RNAiplc21C557 animals compared with control UASRNAiplc21C557/+ animals (p < 0.0001). Symbols represent the mean ± SEM (n = 10 for all genotypes). B, Dose responses for ethyl acetate and propionic acid for p60A/P319 and P319/P319 and controls plc21Cp60A/+ (p60A/+), plc21CP319+ (P319/+). Values are means ± SEM (n = 10); p < 0.0001. C, RT-PCRs from antennal RNA isolated from CS, plc21CP319 heterozygotes and homozygotes, and heterozygotes of the plcβ21C deficiency, plc21Cp60a/+. RNA levels were normalized by amplifying rp49 from the same samples. D, The change in spike frequency of ab2a neurons to varying concentrations of ethyl acetate is shown. Animals in which OSNs express RNAi for Plc21C (RNAiplc21C557/+; Or83bGAL4/+) show a significant reduction (p < 0.001) in response at all odor concentrations compared with the controls (RNAiplc21C557/+). Each trace is the average response (±SEM) of nine sensilla from five individuals.

EAG responses of plc21CP319 homozygotes and plc21CP319/p60A combinations were reduced to <6 mV for all chemicals tested (Fig. 6A). The EAG responses to ethyl acetate (from basiconic sensilla) and propionic acid (from coeloconic sensilla) were found to be reduced at several concentrations (Fig. 6B) (p < 0.0001). plc21C/+ and dgq/+ heterozygotes resulted in a small but significant decrease in EAG responses compared with wild-type (CS) controls (decrease is ∼2.5 mV, p < 0.01 in the case of plc21C/+, and ∼3 mV, p < 0.0005 for dgq/+).

A combination of plc21C/dgq resulted in a 7–9 mV reduction in EAG responses compared with wild-type controls (Fig. 6A). This reduction is significantly greater than the value expected to arise from a mere additive effect of plc21C/+ and dgq/+ heterozygotes (for statistical validation, see legend to Fig. 6A). Although other explanations for this haploinsufficent interaction are formally possible, we favor the explanation that PLCβ21C interacts with Gqα in normal olfactory transduction. This observation was supported by the significantly reduced EAG response to all five odorants obtained in an RNAi-driven OSN-specific downregulation of plc21C, and by downregulation of plc21c in the dgq expression domain (GqGAL4/RNAiplc21c557) (Fig. 6A). EAG responses of the plc21C RNAi strain were reduced further on introduction of UASGq1F1 (RNAiGq) (Fig. 6A). In the double mutant dgq-plc21C RNAi strain, EAG responses were reduced to ∼2 mV (which is below the single RNAi lines) but are still not abolished completely. Moreover, expression of plc21C RNAi in the OSNs resulted in a significant reduction of the spike frequency of ab2a neurons on stimulation by ethyl acetate (Fig. 6D) (p < 0.001). Thus, odor transduction in ab2a neurons requires both Gqα and Plc21C.

Because small, but significant, odor responses remain in all dgq and plc21C mutant genotypes tested, these data further highlight the likelihood of other signaling mechanisms that collaborate with Gqα during odor detection. As reported previously, EAG responses recorded from a null allele for the second PLCβ, norpAP24, appear close to normal for the five odorants tested (Riesgo-Escovar et al., 1995) (data not shown). The introduction of norpAP24 in plc21C/dgq animals had no additional effect in reducing EAG amplitudes (data not shown). norpA thus has no significant effect on antennal physiology in contrast to its proposed effect on signaling in OSNs located on the maxillary palp (Riesgo-Escovar et al., 1995) and the major role played by this gene in visual transduction (Bloomquist et al., 1988).

Genetic interaction of dgq mutant alleles with rdgA, a DAG kinase mutant

The results so far show that signaling downstream of olfactory receptors in the Drosophila antennas requires Dgqα3 and PLCβ21C. To identify which of the two second messengers generated by Gqα activation of PLCβ21C, DAG or InsP3, are required for the electrical response to odorants, we tested mutants that function in the two arms of the pathway. The rdgA gene codes for an ATP-dependant DAG kinase that converts DAG to phosphatidic acid, which in turn is a precursor for PIP2 formation (Inoue et al., 1989). Homozygotes for two rdgA mutant alleles rdgA1 and rdgA3 show reduced (∼4 mV) EAG responses to all five odorants tested (Fig. 7A), whereas heterozygotes appear close to normal (∼10 mV). rdgA1 homozygotes also show reduced EAG responses to multiple concentrations of two odorants tested, ethyl acetate and propionic acid (Fig. 7B). Interestingly, unlike the effect of rdgA mutants on ERG termination kinetics (Garcia-Murillas et al., 2006), we do not observe slower termination of the EAG response in rdgA mutants (Fig. 7D). Moreover, unlike eyes of 1-d-old rdgA1/rdgA1 adults, which exhibit strong retinal degeneration, both rdgA alleles have normal antennal morphology even 6 d after eclosion (data not shown). Thus, the reduced EAGs observed are not a consequence of antennal neuron degeneration.

Figure 7.
Figure 7.

Trans-heterozygote combinations of dgq and rdgA mutant alleles exhibit rescue of reduced EAGs observed in rdgA mutants. A, rdgA1 (severe allele) and rdgA3 (weaker allele) homozygotes show reduced (∼4 mV) EAG responses to all five odorants tested compared with heterozygotes (rdgA1/+ and rdgA3/+); p < 0.0001. B, Dose responses of rdgA1/+ and rdgA1 homozygotes to ethyl acetate and propionic acid. The response of rdgA1/rdgA1 animals are reduced significantly (p < 0.0001) at all odor concentrations. C, Reduction of Gqα levels can rescue EAG amplitudes of rdgA mutant homozygotes. Genotypes with strongest rescue are marked with asterisks. rdgA1; RNAiGq (rdgA1/rdgA1; GqGAL4/UASGq1F1) and rdgA3; RNAiGq (rdgA3/rdgA3; GqGAL4/UASGq1F1) rescues rdgA mutants comparable with dgq mutants. Expression of two different RNAi constructs in OSNs RNAirdgA100 (UASRNAirdgA100/OR83bGAL4) and RNAirdgA024 (UASRNAirdgA024/+; OR83bGAL4/+) showed lower responses compared with control animals, UASRNAirdgA100/+ and UASRNAirdgA024/+ (p < 0.0001). Expression of RNAi constructs for both Gqα and rdgA in OSNs RNAiGq; RNAirdgA100 (UASGq1F1/+; UASRNAirdgA100/OR83bGAL4) and RNAiGq; RNAirdgA024 (UASRNAirdgA024/UASGq1F1; OR83bGAL4/+) showed higher responses (p < 0.0001) compared with UASRNAirdgA100/OR83bGAL4 and UASRNAirdgA024/+; OR83bGAL4/+. D, Representative EAG traces for genotypes tested in A and C.

Next, the EAG response was tested after introducing a single mutant copy of dgq in rdgA homozygotes (rdgA/rdgA;dgq/+). Interestingly, we found that reduction of Dgq levels to 50% by introduction of a single copy of the dgqnull allele rescued the EAG responses of rdgA homozygotes bringing them back to near normal (Fig. 7C). The extent of rescue varies with specific dgq and rdgA mutant allele combinations. Strongest effects were seen in rdgA1/rdgA1; dgq1370/+ animals (p < 0.0001). Reducing dgq function by expression of the RNAi construct UASGq1F1 also rescued the EAG defects produced by both rdgA mutant homozygous alleles. The rescue of EAG responses in rdgA mutants by dgq mutants was further verified by testing RNAi lines specific for rdgA and dgq in appropriate combinations (Fig. 7C). EAG responses were significantly higher in the presence of RNAi constructs for both rdgA and dgq (p < 0.0001) (Fig. 7C) compared with the response from individual RNAi lines (Fig. 3E,F for Gqα RNAi). A possible explanation for these data is that failure of conversion of DAG back to phosphatidic acid and PIP2 through rdgA encoded DAG kinase compromised the generation of an odor-evoked response but does not cause degeneration of OSNs. Reduction in the signaling flux through Gqα (by lower Dgq levels in dgqnull/+ genotypes or by Gqα RNAi expression) compensates for this defect and rescues the olfactory defect of rdgA (see Discussion).

Because there is a significant body of data from studies with other organisms suggesting that InsP3 is a second messenger in invertebrate olfactory transduction (Ache and Young, 2005), we measured EAG responses from single copies of the two dgq alleles with two different mutations of the InsP3 receptor as heterozygotes (dgqnull/+; itprEMS/+). itprsv35 is a null (Joshi et al., 2004), whereas itprka901 can act as a gain-of-function allele (Srikanth et al., 2006). The effect of these alleles on antennal olfactory responses has not been investigated before. No significant change in the response to any chemical was observed in the eight strains tested (data not shown). These data are in agreement with a previous study from Drosophila, in which EAGs measured from the antennas of viable alleles for the InsP3 receptor were normal (Deshpande et al., 2000). Thus, primary olfactory responses from the Drosophila antennas do not require the InsP3 receptor but are dependent on a DAG kinase (rdgA).

Discussion

We have shown that mutations in the G-protein Gqα result in reduced sensitivity of Drosophila antennal sensory neurons to several different chemical stimuli. Genetic analysis suggested that its downstream effector is phospholipase Cβ encoded by the plc21C gene. PLCβ enzymes catalyze breakdown of the membrane-bound lipid PIP2 to generate DAG and InsP3. Reduced EAG amplitudes in DAG kinase mutants (rdgA) and their rescue by lowering Gqα levels, argue that the extent of PIP2 depletion after Gqα and PLCβ activation is responsible for this arm of olfactory transduction. Altered PIP2 levels have been shown to gate or modulate membrane conductances in other contexts (Zhang et al., 2003). The presence of a residual olfactory response in antennal neurons that are null for dgq supports existence of an additional transduction mechanism.

Olfactory transduction in Drosophila: single or multiple pathways?

Odorant detection in animals occurs through a diverse class of G-protein-coupled receptors (Buck and Axel, 1991). In mammalian olfactory sensory neurons, it appears that all odorant receptors activate a single G-protein Golf and generate one second messenger, cAMP (Jones and Reed, 1989; Belluscio et al., 1998). In Drosophila dgq mutants, we find a reduced response to all five odorants tested, indicating that multiple odorant receptors activate Gqα. Ethyl acetate, isoamyl acetate, and benzaldehyde stimulate receptors located in the basiconic sensilla, whereas butanol and propionic acid activate neurons innervating the coeloconic sensilla (Hallem et al., 2004; Couto et al., 2005; Yao et al., 2005). It has been proposed that the basiconic sensilla primarily detect food odors and the coeloconics express receptors that respond to certain specialized odors (Laissue and Vosshall, 2008). Our results suggest that these functional distinctions are not based on a change in signal transduction machinery.

The residual odor-induced response observed in Gqα mutant antennas and in single ab2a neurons argues for the presence of an alternate transduction mechanism(s) in OSNs. Odorants could stimulate either other G-proteins, which alter cyclic nucleotide levels (Gomez-Diaz et al., 2004), or a direct odor-gated ion channel, as in heteromers of Manduca OR/OR83b (Nakagawa et al., 2005) or both. Precisely how the alternate mechanism of odor transduction interacts with Gqα-mediated transduction described here requires additional study. The strong genetic evidence presented here, supporting a role for a Gqα-mediated signaling cascade during olfactory transduction, needs to be reconciled with recent data suggesting that Drosophila odorant receptors do not conform to the structure or topology typical of other G-protein-coupled receptors (Benton et al., 2006).

Based on the quality and strength of the odor stimulus it receives, each odorant receptor could stimulate different transduction pathways to varying extents. We propose that firing frequency of the OSN is an integration of this information, perhaps by cross-sensitization or desensitization of membrane conductances gated by the individual pathways. Interestingly, it has been demonstrated that a single odorant receptor (e.g., Or59b), present within the same neuron can mediate both excitatory and inhibitory responses as a consequence of stimulation by two different odorants (Hallem et al., 2004). Precedents for multiple signaling pathways in odor detection within a single organism are present in invertebrates (Ache and Young, 2005), although their existence within the same olfactory neuron has not been demonstrated so far.

Similarities of signaling intermediates in Drosophila olfaction and vision

The requirement of phospholipase C for odor-induced activity of OSNs in the maxillary palp highlights an overlap in signaling mechanisms during olfaction and vision (Riesgo-Escovar et al., 1995). Our data further highlight parallels in these two sensory systems. In Drosophila photoreceptors, signal amplification begins with activation of a number of G-protein molecules by a single photoisomerised rhodopsin molecule, followed by activation of one PLC molecule for each G-protein (Hardie et al., 2002). Although the current methods of measuring olfactory responses in Drosophila do not allow direct measurement of signal amplification, the enhanced EAG amplitudes observed in AcGq3-expressing animals and synergism between dgq and plc21C mutant alleles observed here supports the idea of signal amplification by Gqα.

The rdgA gene has a role in both amplification and termination of the visual response (Hardie et al., 2002). Our observation that rdgA1 and rdgA3 homozygotes exhibit reduced EAG responses, in the absence of antennal neuron degeneration, suggest that the rate of DAG turnover to generate other intermediates leading to PIP2 is important for membrane depolarization of OSNs. The absence of antennal degeneration predicts a low level of basal PLCβ activity in OSNs, and perhaps strong adaptive mechanisms that prevent activation of membrane channels in normal odor-rich environments. This needs to be investigated further. The fact that reduced EAGs can be rescued by a single copy of dgq mutant alleles, supports the existence of a cycle of phosphoinositide turnover in OSNs, in which reduced levels of Gqα help to slow down this cycle and thus balance the rate of PIP2 regeneration in rdgA mutants. The observed suppression appears qualitatively similar to the interaction of rdgA and dgq during the amplification phase of Drosophila visual transduction in which rdgA mutants enhance the sensitivity of dgqα1 mutants as measured by the amplitude of quantum bumps (Hardie et al., 2002). Finally, either PIP2 depletion, or a lipid intermediate presumably gates a depolarizing membrane channel. The identity of this channel can only be conjectured on at this stage. However, based on studies from other G-protein-coupled lipid signaling pathways, a transient receptor potential (TRP) family channel might be a possible candidate. Among this family, the two most well studied members in Drosophila are TRP and TRPL (TRP-like), both of which are enriched in photoreceptors. A previous study on the olfactory responses of trp mutants ruled out a role for this gene in primary sensory transduction of odor stimuli (Stortkuhl et al., 1999). Several members of this family exist in Drosophila and need to be investigated for their role in olfactory transduction.

Footnotes

  • This work was supported by funding from the National Centre for Biological Sciences (NCBS) and the Department of Biotechnology (G.H.), the National Institute on Drug Abuse-associated Supplement for International Collaboration Grant DA15495 (V.R.), and the Indian Council for Medical Research (P.K., V.R.). We acknowledge the Centre for Nanotechnology (Department of Science and Technology No. SR/S5/NM-36/2005) for the Olympus confocal microscope. We thank Prabhat Tiwari, Keshava Subramanya, H. Krishnamurthy, and the NCBS imaging facility for help with confocal imaging and Sonia Sen for providing us with protocols for adult antennal staining.

  • Correspondence should be addressed to Gaiti Hasan, National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bellary Road, Bangalore 560065, India. gaiti{at}ncbs.res.in

References

  1. ↵
    1. Ache BW,
    2. Young JM
    (2005) Olfaction: diverse species, conserved principles. Neuron 48:417–430.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Ayer RK Jr.,
    2. Carlson J
    (1991) acj6: a gene affecting olfactory physiology and behavior in Drosophila. Proc Natl Acad Sci USA 88:5467–5471.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Banerjee S,
    2. Joshi R,
    3. Venkiteswaran G,
    4. Agrawal N,
    5. Srikanth S,
    6. Alam F,
    7. Hasan G
    (2006) Compensation of inositol 1,4,5-trisphosphate receptor function by altering sarco-endoplasmic reticulum calcium ATPase activity in the Drosophila flight circuit. J Neurosci 26:8278–8288.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Belluscio L,
    2. Gold GH,
    3. Nemes A,
    4. Axel R
    (1998) Mice deficient in G(olf) are anosmic. Neuron 20:69–81.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Benton R,
    2. Sachse S,
    3. Michnick SW,
    4. Vosshall LB
    (2006) Atypical membrane topology and heteromeric function of Drosophila odorant receptors in vivo. PLoS Biol 4:e20.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Bloomquist BT,
    2. Shortridge RD,
    3. Schneuwly S,
    4. Perdew M,
    5. Montell C,
    6. Steller H,
    7. Rubin G,
    8. Pak WL
    (1988) Isolation of a putative phospholipase C gene of Drosophila, norpA, and its role in phototransduction. Cell 54:723–733.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Brand AH,
    2. Perrimon N
    (1993) Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118:401–415.
    OpenUrlAbstract
  8. ↵
    1. Buck L,
    2. Axel R
    (1991) A novel multigene family may encode odorant receptors: a molecular basis for odor recognition. Cell 65:175–187.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Couto A,
    2. Alenius M,
    3. Dickson BJ
    (2005) Molecular, anatomical, and functional organization of the Drosophila olfactory system. Curr Biol 15:1535–1547.
    OpenUrlCrossRefPubMed
  10. ↵
    1. de Bruyne M,
    2. Foster K,
    3. Carlson JR
    (2001) Odor coding in the Drosophila antenna. Neuron 30:537–552.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Deshpande M,
    2. Venkatesh K,
    3. Rodrigues V,
    4. Hasan G
    (2000) The inositol 1,4,5-trisphosphate receptor is required for maintenance of olfactory adaptation in Drosophila antennae. J Neurobiol 43:282–288.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Garcia-Murillas I,
    2. Pettitt T,
    3. Macdonald E,
    4. Okkenhaug H,
    5. Georgiev P,
    6. Trivedi D,
    7. Hassan B,
    8. Wakelam M,
    9. Raghu P
    (2006) lazaro encodes a lipid phosphate phosphohydrolase that regulates phosphatidylinositol turnover during Drosophila phototransduction. Neuron 49:533–546.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Gomez-Diaz C,
    2. Martin F,
    3. Alcorta E
    (2004) The cAMP transduction cascade mediates olfactory reception in Drosophila melanogaster. Behav Genet 34:395–406.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Hallem EA,
    2. Carlson JR
    (2006) Coding of odors by a receptor repertoire. Cell 125:143–160.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Hallem EA,
    2. Ho MG,
    3. Carlson JR
    (2004) The molecular basis of odor coding in the Drosophila antenna. Cell 117:965–979.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Hardie RC,
    2. Martin F,
    3. Cochrane GW,
    4. Juusola M,
    5. Georgiev P,
    6. Raghu P
    (2002) Molecular basis of amplification in Drosophila phototransduction: roles for G protein, phospholipase C, and diacylglycerol kinase. Neuron 36:689–701.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Inoue H,
    2. Yoshioka T,
    3. Hotta Y
    (1989) Diaclyglycerol kinase defect in a Drosophila retinal degeneration mutant rdgA. J Biol Chem 264:5996–6000.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Jefferis GS,
    2. Vyas RM,
    3. Berdnik D,
    4. Ramaekers A,
    5. Stocker RF,
    6. Tanaka NK,
    7. Ito K,
    8. Luo L
    (2004) Developmental origin of wiring specificity in the olfactory system of Drosophila. Development 131:117–130.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Jhaveri D,
    2. Sen A,
    3. Reddy GV,
    4. Rodrigues V
    (2000) Sense organ identity in the Drosophila antenna is specified by the expression of the proneural gene atonal. Mech Dev 99:101–111.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Jones DT,
    2. Reed RR
    (1989) Golf: an olfactory neuron specific-G protein involved in odorant signal transduction. Science 244:790–795.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Joshi R,
    2. Venkatesh K,
    3. Srinivas R,
    4. Nair S,
    5. Hasan G
    (2004) Genetic dissection of itpr gene function reveals a vital requirement in aminergic cells of Drosophila larvae. Genetics 166:225–236.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Kalidas S,
    2. Smith DP
    (2002) Novel genomic cDNA hybrids produce effective RNA interference in adult Drosophila. Neuron 33:177–184.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Krieger J,
    2. Breer H
    (1999) Olfactory reception in invertebrates. Science 286:720–723.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Technau GM
    1. Laissue PP,
    2. Vosshall LB
    (2008) in Brain development in Drosophila melanogaster, The olfactory sensory map in Drosophila, ed Technau GM (Springer, New York), pp 102–114.
  25. ↵
    1. Laissue PP,
    2. Reiter C,
    3. Hiesinger PR,
    4. Halter S,
    5. Fischbach KF,
    6. Stocker RF
    (1999) Three-dimensional reconstruction of the antennal lobe in Drosophila melanogaster. J Comp Neurol 405:543–552.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Larsson MC,
    2. Domingos AI,
    3. Jones WD,
    4. Chiappe ME,
    5. Amrein H,
    6. Vosshall LB
    (2004) Or83b encodes a broadly expressed odorant receptor essential for Drosophila olfaction. Neuron 43:703–714.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Lee T,
    2. Luo L
    (1999) Mosaic analysis with a repressible cell marker for studies of gene function in neuronal morphogenesis. Neuron 22:451–461.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Lee YJ,
    2. Dobbs MB,
    3. Verardi ML,
    4. Hyde DR
    (1990) dgq: a drosophila gene encoding a visual system-specific G alpha molecule. Neuron 5:889–898.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Lee YJ,
    2. Shah S,
    3. Suzuki E,
    4. Zars T,
    5. O'Day PM,
    6. Hyde DR
    (1994) The Drosophila dgq gene encodes a G alpha protein that mediates phototransduction. Neuron 13:1143–1157.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Martin F,
    2. Charro MJ,
    3. Alcorta E
    (2001) Mutations affecting the cAMP transduction pathway modify olfaction in Drosophila. J Comp Physiol A Neuroethol Sens Neural Behav Physiol 187:359–370.
    OpenUrlCrossRefPubMed
  31. ↵
    1. McGuire SE,
    2. Le PT,
    3. Osborn AJ,
    4. Matsumoto K,
    5. Davis RL
    (2003) Spatiotemporal rescue of memory dysfunction in Drosophila. Science 302:1765–1768.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Mixon MB,
    2. Lee E,
    3. Coleman DE,
    4. Berghuis AM,
    5. Gilman AG,
    6. Sprang SR
    (1995) Tertiary and quaternary structural changes in Gi alpha 1 induced by GTP hydrolysis. Science 270:954–960.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Nair S,
    2. Agrawal N,
    3. Hasan G
    (2007) Homeostasis of glutamate neurotransmission is altered in Drosophila inositol 1,4,5-trisphosphate receptor mutants. Invert Neurosci 7:137–147.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Nakagawa T,
    2. Sakurai T,
    3. Nishioka T,
    4. Touhara K
    (2005) Insect sex-pheromone signals mediated by specific combinations of olfactory receptors. Science 307:1638–1642.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Noel JP,
    2. Hamm HE,
    3. Sigler PB
    (1993) The 2.2 Å crystal structure of transducin-alpha complexed with GTP gamma S. Nature 366:654–663.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Ratnaparkhi A,
    2. Banerjee S,
    3. Hasan G
    (2002) Altered levels of Gq activity modulate axonal pathfinding in Drosophila. J Neurosci 22:4499–4508.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Riesgo-Escovar J,
    2. Raha D,
    3. Carlson JR
    (1995) Requirement for a phospholipase C in odor response: overlap between olfaction and vision in Drosophila. Proc Natl Acad Sci USA 92:2864–2868.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Scott K,
    2. Becker A,
    3. Sun Y,
    4. Hardy R,
    5. Zuker C
    (1995) Gq alpha protein function in vivo: genetic dissection of its role in photoreceptor cell physiology. Neuron 15:919–927.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Shortridge RD,
    2. Yoon J,
    3. Lending CR,
    4. Bloomquist BT,
    5. Perdew MH,
    6. Pak WL
    (1991) A Drosophila phospholipase C gene that is expressed in the central nervous system. J Biol Chem 266:12474–12480.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    1. Srikanth S,
    2. Banerjee S,
    3. Hasan G
    (2006) Ectopic expression of a Drosophila InsP(3)R channel mutant has dominant-negative effects in vivo. Cell Calcium 39:187–196.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Stortkuhl KF,
    2. Hovemann BT,
    3. Carlson JR
    (1999) Olfactory adaptation depends on the Trp Ca2+ channel in Drosophila. J Neurosci 19:4839–4846.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    1. Talluri S,
    2. Bhatt A,
    3. Smith DP
    (1995) Identification of a Drosophila G protein alpha subunit (dGq alpha-3) expressed in chemosensory cells and central neurons. Proc Natl Acad Sci USA 92:11475–11479.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Vosshall LB,
    2. Wong AM,
    3. Axel R
    (2000) An olfactory sensory map in the fly brain. Cell 102:147–159.
    OpenUrlCrossRefPubMed
  44. ↵
    1. Weinkove D,
    2. Neufeld TP,
    3. Twardzik T,
    4. Waterfield MD,
    5. Leevers SJ
    (1999) Regulation of imaginal disc cell size, cell number and organ size by Drosophila class I(A) phosphoinositide 3-kinase and its adaptor. Curr Biol 9:1019–1029.
    OpenUrlCrossRefPubMed
  45. ↵
    1. Xia S,
    2. Tully T
    (2007) Segregation of odor identity and intensity during odor discrimination in Drosophila mushroom body. PLoS Biol 5:e264.
    OpenUrlCrossRefPubMed
  46. ↵
    1. Yao CA,
    2. Ignell R,
    3. Carlson JR
    (2005) Chemosensory coding by neurons in the coeloconic sensilla of the Drosophila antenna. J Neurosci 25:8359–8367.
    OpenUrlAbstract/FREE Full Text
  47. ↵
    1. Zhainazarov AB,
    2. Spehr M,
    3. Wetzel CH,
    4. Hatt H,
    5. Ache BW
    (2004) Modulation of the olfactory CNG channel by Ptdlns(3,4,5)P3. J Membr Biol 201:51–57.
    OpenUrlCrossRefPubMed
  48. ↵
    1. Zhang H,
    2. Craciun LC,
    3. Mirshahi T,
    4. Rohacs T,
    5. Lopes CM,
    6. Jin T,
    7. Logothetis DE
    (2003) PIP(2) activates KCNQ channels, and its hydrolysis underlies receptor-mediated inhibition of M currents. Neuron 37:963–975.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Email
Print
View Full Page PDF
Citation Tools
Respond to this article
Reduced Odor Responses from Antennal Neurons of Gqα, Phospholipase Cβ, and rdgA Mutants in Drosophila Support a Role for a Phospholipid Intermediate in Insect Olfactory Transduction
Pinky Kain, Tuhin Subra Chakraborty, Susinder Sundaram, Obaid Siddiqi, Veronica Rodrigues, Gaiti Hasan
Journal of Neuroscience 30 April 2008, 28 (18) 4745-4755; DOI: 10.1523/JNEUROSCI.5306-07.2008
Twitter logo Facebook logo Mendeley logo

Jump to section

Responses to this article

Respond to this article

Jump to comment:

No eLetters have been published for this article.