Ionotropic Receptors Mediate Drosophila Oviposition ... - Cell Press

Article

Ionotropic Receptors Mediate Drosophila Oviposition Preference through Sour Gustatory Receptor Neurons Graphical Abstract

Authors Yan Chen, Hubert Amrein

Correspondence [email protected]

In Brief Acids are abundant in fruit, a common food source. Chen and Amrein show that fruit flies have dedicated sour taste neurons in taste sensilla and require two ionotropic receptors (IR25a and IR76b) to sense acids. Moreover, they find that preference of females to oviposit on acidcontaining food is mediated by these taste neurons and receptors.

Highlights d

Drosophila have taste neurons in sensilla of legs dedicated to sour taste

d

Sour taste is mediated by ionotropic receptor 25a (IR25a) and IR76b

d

Activation of sour GRNs is dependent on IR25a and IR76b

d

Female oviposition preference on acid-containing food is mediated by sour GRNs

Chen & Amrein, 2017, Current Biology 27, 2741–2750 September 25, 2017 ª 2017 Elsevier Ltd. http://dx.doi.org/10.1016/j.cub.2017.08.003

Current Biology

Article Ionotropic Receptors Mediate Drosophila Oviposition Preference through Sour Gustatory Receptor Neurons Yan Chen1 and Hubert Amrein1,2,* 1Department of Molecular and Cellular Medicine, College of Medicine, Texas A&M University Health Science Center, College Station, TX 77843, USA 2Lead Contact *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cub.2017.08.003

SUMMARY

Carboxylic acids are present in many foods, being especially abundant in fruits. Yet, relatively little is known about how acids are detected by gustatory systems and whether they have a potential role in nutrition or provide other health benefits. Here we identify sour gustatory receptor neurons (GRNs) in tarsal taste sensilla of Drosophila melanogaster. We find that most tarsal sensilla harbor a sour GRN that is specifically activated by carboxylic and mineral acids but does not respond to sweet- and bitter-tasting chemicals or salt. One pair of taste sensilla features two GRNs that respond only to a subset of carboxylic acids and high concentrations of salt. All sour GRNs prominently express two Ionotropic Receptor (IR) genes, IR76b and IR25a, and we show that both these genes are necessary for the detection of acids. Furthermore, we establish that IR25a and IR76b are essential in sour GRNs of females for oviposition preference on acid-containing food. Our investigations reveal that acids activate a unique set of taste cells largely dedicated to sour taste, and they indicate that both pH/proton concentration and the structure of carboxylic acids contribute to sour GRN activation. Together, our studies provide new insights into the cellular and molecular basis of sour taste. INTRODUCTION The ability to sense and discriminate between different food chemicals is critical for fitness and survival of most animals. Species from evolutionarily distant animal phyla, such as vertebrates and arthropods, can exhibit surprisingly similar taste preferences [1]. For example, humans and Drosophila melanogaster exhibit strong appetitive responses to dietary sugars, which constitute a major energy resource, and they are generally repulsed by chemicals that taste bitter to humans [1]. In contrast to sweetand bitter-tasting chemicals, acids elicit varied behavioral responses, depending on their structure and concentration [2]. Strong acids, which damage the membrane integrity of most cells, elicit repulsion, but moderately acidic carboxylic

acids—abundant in fruits—are regularly ingested by frugivores and generalist feeders and are well tolerated by many insects and mammals. For example, lemon and lime can contain up to 8% citric acid [3], and the juice of many other fruits, including apples, grapes, and many types of berries, are moderately to strongly acidic (pH between 3 and 6). In mice and humans, acids are sensed by sour taste cells, interspersed among sweet-, bitter-, umami-, and salt-sensing cells in the taste buds of the tongue [4, 5]. Numerous candidate receptors and channels were found to be expressed in sour taste cells, such as ACCN1, HCN1/4, and PKD1L3/PKD2L1, albeit none has been shown to be essential for the perception of acids [4]. In contrast to mammals, no cellular entity or molecular receptors for sour taste in Drosophila have emerged to date. Given that most Drosophila species, including D. melanogaster, feed largely on fruits, we hypothesized that fruit flies are able to detect and sense acids. The fly gustatory system is composed of hundreds of taste sensilla stereotypically arranged across multiple appendages, including the labellum and the legs. In the labellum, the majority of sensilla (referred to as large [l] or small [s] sensilla) harbor four gustatory receptor neurons (GRNs), whereas intermediate (i) sensilla contain only two GRNs [4, 6]. Based on electrophysiological recordings of taste sensilla and expression studies of Gustatory receptor (Gr) genes, it has been proposed that each GRN is dedicated to a specific taste modality, with the exception of bitter taste GRNs, which are also activated by high concentrations of salt. Specifically, three of the four GRNs of s-type sensilla are dedicated to sweet, bitter, and water taste, respectively. The two i-type-associated GRNs are responsive to bitter- and sweettasting chemicals each, and finally three of the four GRNs of l-type sensilla have been associated with sweet, water, and low-salt taste, respectively [7–12]. Activation profiles of GRNs in sensilla on the legs have been characterized even less, and have focused mostly on the fifth tarsal segment. Two of the four GRNs in these sensilla respond to sugars and bitter-tasting chemicals, respectively [13–15], whereas the roles of the remaining two neurons are unknown. Numerous studies have provided direct evidence that responses to bitter chemicals and sugars are mediated by members of the Gr protein family. Specifically, sweet GRNs express eight closely related sugar Gr genes, members of a large gene family, and were shown to mediate behavioral and cellular responses to specific sugars [11, 12, 16–19], whereas bitter GRNs express many of the remaining Gr genes, several of which have been shown to mediate bitter taste responses [8, 10, 20, 21]. Pickpocket 28 (Ppk28), a protein

Current Biology 27, 2741–2750, September 25, 2017 ª 2017 Elsevier Ltd. 2741

related to the mammalian degenerin/epithelial sodium channel protein family, is expressed in the water-sensing GRNs of labellar s-type sensilla and required for behavioral responses to water solutions of low molarity [9]. Last, ionotropic receptor 76b (IR76b) was shown to be necessary for low-salt and amino acid taste [7, 22]. Notably, IR76b is part of a second large chemoreceptor gene family evolutionarily related to ionotropic glutamate receptors (iGluRs). Members of the IR gene family were first reported in olfactory neurons of the antenna [23], but subsequently a largely distinct subgroup of IR genes was shown to be expressed in many taste neurons [24]. Here we identify sour GRNs in tarsal taste sensilla of Drosophila. Using Ca2+ imaging experiments, we show that these neurons respond to mineral and carboxylic acids in a pH-dependent as well as a structure-dependent manner. Interestingly, we find that IR76b and IR25a are co-expressed in sour taste neurons and required for acid taste responses. We also show that oviposition preference for acidic food is mediated by the tarsal sour GRNs and dependent on both IR76b and IR25a. These observations, together with previous studies linking IR proteins to the detection of salt, amino acids, and odorants as well as monitoring temperature and sensing humidity [7, 22, 23, 25–28], led us to propose that these two IR proteins are part of multimodal receptors involved in up to five different sensory pathways. RESULTS

IR76b-QF, but neither by Gr33a-GAL4 nor Gr64f-GAL4 (Figure 1B), suggesting that IR76b-QF labeled an additional neuron (referred to as IR76bonly neuron) representing yet another taste modality. Intriguingly, the IR76bonly neuron was also recognized by an antibody raised against IR25a [23] (Figure 1C), suggesting that these two IRs might co-operate in this GRN to sense a specific group of chemicals. We next examined the projections of IR76b GRNs located in the tarsi. Axons of bitter GRNs and Gr43a expressing sweet GRNs terminate in the subesophageal zone of the brain, whereas axons of most sweet GRNs project to and terminate in the thoracic ganglion [18, 29]. Interestingly, the axons of most IR76bonly tarsal GRNs terminate in the thoracic ganglion as well, in areas distinct from those occupied by termini of sweet GRNs (Figure 1D). We note that additional IR76bonly-specific termini are found between the pro- and mesathoracic ganglion (Figures 1D and 1E), a region that receives input from sensory neurons located in the anterior wing margin [30]. Taken together, these observations indicate that tarsal taste sensilla harbor an IR76bonly GRN that is distinct from sweet and bitter GRNs. Last, both IR25a and IR76b are also broadly expressed in the labial palp, but co-expression is less pronounced than in tarsi, with IR25a-GAL4 labeling more GRNs than IR76b-QF (Figure S1A). Moreover, expression overlap between IR76b-GAL4 and markers for sweet/bitter GRNs is more complex, as many Gr33a-GAL4- and Gr64f-GAL4-positive neurons were not labeled by IR76b-QF (Figures S1B and S1C).

IR76b and IR25a Are Prominently Expressed in Nonsweet/Non-bitter GRNs The current dogma posits that Drosophila taste sensilla share a basic set of taste neurons, dedicated to sugar, bitter/high salt, low salt, and water [6]. However, the lack of marker genes for many taste neurons, especially those not representing sweet and bitter taste, has limited our ability to functionally analyze the GRN repertoire comprehensively. To identify novel GRN markers, we focused our attention on the IR gene family, numerous members of which are expressed in the gustatory system [23]. Examination of expression data from microarray analyses indicated that IR25a and IR76b are enriched and highly expressed in taste neurons [9], and respective reporter transgenes revealed that these two genes are indeed broadly expressed in the taste system (see below). We therefore used IR-GAL4 and IR-QF transgenes to investigate the cellular expression profile of IR76b and IR25a in more detail, focusing on the most distal segment of the tarsi for three reasons. First, its anatomical location establishes this segment as the initial transmitter of chemosensory cues received by tarsi. Second, the number of sensilla on this segment is smaller and sensillum structure is more homogeneous compared to sensilla on the palps, and third, tarsal GRNs are amenable to single-cell Ca2+ imaging, whereas this method has not yet been successfully applied to labellar GRNs [14]. Our analysis showed that IR76b-QF and IR25a-GAL4 are co-expressed, labeling up to three neurons in three pairs of taste sensilla (5b, 5s, and 5v; Figure 1A). When IR76b-QF was combined with Gr33a-GAL4 and Gr64f-GAL4, cell markers for bitter and sweet GRNs, respectively [10, 12, 19], we found broad expression overlap (Figure 1B). However, one GRN in both the 5b and 5s sensilla was labeled by

IR76b-Specific GRNs in the Tarsal Taste Sensilla Are Tuned to Acids To characterize the response profiles of IR76bonly GRNs, we performed Ca2+ imaging experiments on tarsal taste sensilla [14]. To facilitate identification of the IR76bonly GRNs in each sensillum, we silenced GCaMP6 expression in sweet and bitter GRNs using GAL80 suppressor transgenes (Figures 1C and 2A). The response profiles for the 5b- and 5s-associated GRNs were very similar: stimulation with acids induced concentrationdependent Ca2+ increases, whereas stimulation with sucrose, salt, or various bitter compounds elicited no responses (Figure 2B). In contrast to 5b and 5s, which only harbor a single IR76b/IR25a-specific GRN, the 5v sensillum harbors two such neurons (Figure 1D), both of which showed narrow response profiles limited to glycolic and malic acid and high concentrations of salt (Figure 2B, bottom panel). Together, our experiments establish that taste sensilla of the fifth tarsal segment differ in their composition of GRNs, harboring a novel, previously not characterized neuron that is largely (5v) or specifically (5b and 5s) tuned to acids. With the exception of the 5s sensilla (which are only found in the foretarsi), positions of all other sensilla on fifth segments are conserved in all three pairs of legs [13], and Ca2+ imaging of the respective IR76b-specific GRNs showed similar response profiles. Because IR76b and IR25a are also expressed in sweet and bitter GRNs (Figure 1B), we tested whether these neurons respond to acids. Consistent with previous studies [31], neither the sweet (Gr64f-GAL4) nor the bitter (Gr33a-GAL4) GRNs showed evidence for Ca2+ increases when stimulated with the various acid compounds (Figures S2A and S2B). We note that some bitter-sensing GRNs on the labellum were previously reported to respond to acids [32].

2742 Current Biology 27, 2741–2750, September 25, 2017

Figure 1. IR25a-Gal4 and IR76b-Gal4 Are Co-expressed in a GRN Not Expressing Markers for Bitter and Sweet Neurons (A) Immunostaining of the fifth tarsal segment of a w1118/w1118; IR25a-GAL4/IR76b-QF UAS-mCD8GFP; QUAS-mtdTomato-3XHA/+ fly. IR25a and IR76b are broadly co-expressed in a large number of tarsal GRNs. Inset: schematic drawing of chemosensory sensilla in the fifth tarsal segment. (B) Immunostaining of the fifth tarsal segment of a w1118/w1118; IR76b-QF UAS-mCD8GFP/Gr33a-GAL4 Gr64f-GAL4; QUAS-mtdTomato-3XHA/+ fly. Bitter and sweet GRNs express IR76b, but each sensillum harbors one (or two) additional GRN(s) that is not labeled by bitter and sweet Gr-GAL4 drivers. (A and B) Numbers indicate the average number of labeled GRNs of the three taste sensilla: 5b and 5s+5v (mean ± SEM; 6 % n % 8). Neurons of the 5a sensillum (dotted oval) respond to pheromones and were excluded in the count. Arrows indicate non-sweet and non-bitter neurons. (C) Endogenous IR25a, visualized by anti-IR25a antibody, and IR76b-GAL4 are co-expressed in IR76bonly GRNs of the foreleg. Genotype: w1118/w1118; IR76bGAL4 Gr66a-LexA/UAS-GCaMP6m; Gr64f-LexA/LexAop-GAL80. Note that the bitter GRN of the 5b sensillum (*) does not express Gr66-LexA. (D) Immunostaining of the ventral nerve cord in the thorax of a w1118/w1118; IR76b-QF UAS-mCD8GFP/Gr33a-GAL4 Gr64f-GAL4; QUAS-mtdTomato-3XHA/+ fly. Tarsal projections of GRNs expressing IR76b (red) terminate predominantly in the thoracic ganglion, whereas most tarsal projections of GRNs expressing bitter Gr genes and a subset of GRNs expressing sweet Gr genes (green) bypass the ganglion to terminate in the subesophageal zone (SEZ). f, foreleg; h, hindleg; m, midleg; w, wing. (E) Projections of tarsal and wing IR76bonly GRNs predominantly terminate in the thoracic ganglion. Genotype: w1118/w1118; IR76b-GAL4 Gr66a-LexA/UASGCaMP6m; Gr64f-LexA/LexAop-GAL80. GRNs of the 5a sensillum (dotted oval) were not counted in (A) and (B), as these neurons are thought to sense pheromones [13, 15]. All scale bars represent 10 mm. See also Figure S1.

Current Biology 27, 2741–2750, September 25, 2017 2743

Figure 2. IR76bonly GRNs Associated with 5b and 5s Sensilla Specifically Respond to Many Carboxylic and Mineral Acids Ca2+ imaging was performed on the fifth segment of forelegs of females with the genotype w1118/w1118; IR76b-GAL4, Gr66a-LexA/UAS-GCaMP6m; Gr64f-LexA/ lexAop-GAL80 (IR76bonly). (A) Schematic image of chemosensory sensilla in the fifth segment of tarsi (left), and a live image of an IR76bonly neuron after stimulation with 100 mM acetic acid (middle). Relative fluorescence change (DF) is visualized (right). Scale bars represent 10 mm. Asterisks indicate bitter GRNs in 5b sensilla, which do not express Gr66a-LexA. (legend continued on next page)


Dose-dependent responses to specific acid ligands indicate that the proton concentration is a major determinant for the level of GRN activation. However, acid structure also contributes to the excitability of sour GRNs (Figure 2B). For example, 100 mM acetic acid (pH 2.88) elicits a significantly stronger response than the more acidic 10 mM citric acid (pH 2.62) or 10 mM HCl (pH 2.0) (Figure S3A). Similarly, 10 mM glycolic acid (pH 2.93) generates a stronger response than either 10 mM citric or tartaric acid (pH 2.62 and 2.52) or 10 mM HCl (pH 2.0) (Figure S3B). To further investigate the contribution of acid structure and proton concentration, we conducted Ca2+ imaging experiments with buffered solutions of decreasing pH (5.4 to 3) but constant conjugate base concentration (100 mM), using mixtures of acetic or citric acids and their respective sodium salts (Figure 2C). Because the sour GRNs of the 5v sensillum also responded to NaCl, the analysis was restricted to sour GRNs of the 5b and 5s sensilla. Ca2+ responses increased as the pH decreased, indicating that the proton concentration is a main determinant for the strength of GRN activation. Interestingly, no significant responses were observed for either 100 mM buffered solutions at pH 5 or higher, indicating that the conjugate base alone is not sufficient to activate the sour GRN in the absence of a threshold proton concentration. Sour GRNs Mediate Acid Responses through IR76b- and IR25a-Containing Receptor Complexes We wondered whether IR76b and IR25a are necessary in taste neurons to mediate sour taste responses. We first performed Ca2+ imaging experiments on tarsi of homozygous mutant IR76b flies (Figure 3A), and we observed that responses to both carboxylic and mineral acids were completely abolished in GRNs of such flies. Importantly, expressing a UAS-IR76b transgene in the sour GRNs in the mutant background restored the Ca2+ responses (Figure 3A). Of note, IR76b is also required in the two sour GRNs of the 5v for responses to NaCl. We next investigated the requirement for IR25a in the sour taste neuron. Similar to IR76b mutants, IR25a mutants completely lost responses to all acids tested, and these responses were fully rescued in the presence of a UAS-IR25a transgene, expressed under the control of IR25a-GAL4 (Figure 3B). Taken together, these experiments show that both IR25a and IR76b are necessary for cellular responses to acids in sour taste GRNs. The phenotypes described for IR25a and IR76b mutant flies are most consistent with roles as components in a sour taste receptor. However, other possibilities, such as functions in neural specification, neural differentiation, or cell viability, cannot be formally excluded. We note, however, that GRNs lacking IR76b or IR25a have normal gross morphology, and sweet GRNs of

IR76b or IR25a mutant flies respond normally to sugars yet fail to respond to fatty acid ligands (J.-E. Ahn, Y.C., and H.A., unpublished data). Last, KCl, which activates potassium channels found in most central and peripheral neurons, elicits similar Ca2+ response in sour GRNs of IR76b mutant and wild-type flies (Figure 3A), indicating that GRNs lacking IR76b are healthy and capable of responding to external ligands. Together, these data strongly support roles for IR25a and IR76b as components of a multimeric sour taste receptor. Oviposition Preference on Acid-Containing Food Is Mediated by Sour GRNs and IR25a and IR76b When given a choice, Drosophila females prefer to lay their eggs on acetic-acid-containing food [33–35]. Acids also enhance palatability and consumption of sugars contaminated with bitter compounds [31], but this process is mediated by bitter GRNs and might involve extracellular interactions of taste ligands with odorant-binding proteins [36]. Therefore, we focused our investigations for establishing a role for acids in behavior on oviposition preference, for which neither the cellular nor the molecular basis is known. Previous studies have shown that females exhibit a strong oviposition preference for acetic-acid-containing food [34]. We confirmed these observations using sugaragar as substrate, finding that females strongly prefer to lay eggs on acid-containing sugar-agar as opposed to sugar-agar alone, a preference that was not restricted to acetic acid but included several other acids (Figure 4A; Figure S4A). Consistent with previous reports [33, 34], acids alone do not affect the number of eggs laid but simply induce a shift in preference toward acid-containing sugar-agar (Figure S4B). To determine which GRNs mediate oviposition preference, we expressed the inward-rectifying K+ channel Kir2.1 under the control of various GAL4 drivers, thereby inactivating specific GRN subtypes (Figure 4B). Inactivation of all IR76b-expressing GRNs or IR76bonly GRNs abolished acid preference. In contrast, simultaneous inactivation of sweet and bitter GRNs had no effect. These findings imply that the sour GRNs mediate female oviposition preference. We next sought to determine the contribution of different taste organs to egg-laying preference. In previous studies, Tsh-Gal80 and Otd-nls:FLP were successfully used to specifically suppress Gr gene expression in either the labellum or legs, respectively [37, 38]. However, both Tsh-Gal80 and Otd-nls:FLP appear to be expressed in bitter and sweet GRNs, respectively, leaving expression unaffected in a subset of IR76b-GAL4-expressing GRNs (Figure S5). We therefore performed oviposition preference assays using females in which specific taste organs were surgically removed (Figure 4C). Overall, preference for acidcontaining sugar-agar was significantly reduced, regardless of which taste structures were removed. However, ablation of

(B) Ca2+ response (DF/F [%]) of IR76bonly GRNs after stimulation with the indicated ligands. Response profiles are similar for the 5b- (top) and 5s- (middle) associated sour GRNs (i.e., IR76bonly). 5v sensilla (bottom) harbor two similarly tuned IR76bonly GRNs that respond only to malic and glycolic acid (at high concentration), as well as to 500 mM NaCl. Representative traces to 100 mM citric acid are shown to the right of each graph. An amino acid (aa) mix (40 mM) contains all 20 amino acids (each at 2 mM). Acetic acid concentrations: 10, 100, and 500 mM; citric and tartaric acid concentrations: 1, 10, and 100 mM; malic and glycolic acid concentrations: 10 mM; HCl concentrations: 1, 10, 25, and 50 mM. Data are shown as the mean, and error bars indicate SEM; Student’s t test, *p < 0.05, **p < 0.01, ***p < 0.001; 6 % n % 11; den, denatonium benzoate, lob, lobeline, quin, quinine, suc, sucrose. (C) Ca2+ response of IR76bonly GRNs in 5b and 5s sensilla stimulated with 100 mM acetate and citrate buffered solutions. Data are shown as means, and error bars indicate the SEM. 6 % n % 10. See also Figures S2 and S3.


foretarsi had a significantly stronger effect than removing midlegs or hindlegs or elimination of the labial palps. Surgical manipulations had less of an effect when acetic acid was replaced with citric acid, and only removal of the forelegs led to a significant reduction in egg-laying preference (Figure 4C). Taken together, these observations strongly suggest that sour GRNs residing in the forelegs are the main mediator of oviposition preference on acid-containing substrates. To determine whether oviposition preference is dependent on IR76b and IR25a function, we examined the distribution of eggs laid by mutant females. Indeed, both types of females lost preference for acid-containing sugar-agar completely (Figure 5). Restoring expression of IR76b in all GRNs rescued oviposition preference (Figure 5A, lane 5) and, importantly, the function of IR76b was not required in sweet and bitter GRNs (Figure 5A, lane 7). Surprisingly, however, restoring expression of IR25a only marginally improved oviposition preference (Figure 5B, lane 5), which was unexpected, because sour GRN responses of such flies were fully restored (Figure 3B). We wondered whether lack of oviposition rescue was related to the more basic role of IR25a compared to IR76b. Specifically, IR25a is expressed in many more olfactory neurons than IR76b, serves as a core subunit of olfactory receptor complexes [23, 39], and also functions as a critical component of IR-based hygro sensors and temperature sensors [25, 26, 28]. For example, broad and high IR25a expression might lead to the formation of novel receptor complexes in neurons of these sensory systems and interfere or counteract the ability of sour GRNs to mediate oviposition preference on acid-containing sugar-agar. To further explore this possibility, we performed rescue experiments of IR25a mutant flies by complementing IR25a function in IR76bexpressing neurons. Indeed, these females showed strong oviposition preference on citric-acid-containing sugar-agar but still failed to rescue oviposition preference on acetic-acid-containing sugar-agar (Figure 5B, lane 8). Because volatile acetic acid can be sensed by olfactory sensory neurons through IR complexes [40], we suspected that ectopic IR25a expression in IR76b-GAL4 olfactory neurons might interfere with acetic acid oviposition preference. Indeed, when antennae were removed, oviposition on acetic-acid-containing sugar-agar was also restored in such females (Figure 5B, lane 9). Together,

Figure 3. Activation of Sour GRNs Is Dependent on Both IR76b and IR25a Function Ca2+ response (DF/F [%]) upon stimulation with the indicated acids in control (A and B), homozygous IR76b mutant (A), homozygous IR25a mutant (B), and the respective rescue flies (A and B). (A) Absence of IR76b leads to a complete loss of sour GRN responses, which is rescued in the presence of an IR76b transgene, expressed under the control of IR76b-GAL4. Genotypes: w1118/w1118; IR76b-GAL4 Gr66a-LexA/ UAS-GCaMP6m; Gr64f-LexA/LexAop-Gal80 (IR76b control); w1118/w1118; IR76b-GAL4 Gr66a-LexA/UAS-GCaMP6m; Gr64f-LexA IR76b2/IR76b1


LexAop-GAL80 (IR76b mutant); and w1118/w1118; IR76b-GAL4 Gr66a-LexA/ UAS-IR76b UAS-GCaMP6m; Gr64f-LexA IR76b2/IR76b1 LexAop-GAL80 (IR76b rescue). (B) Absence of IR25a leads to a complete loss of sour GRN responses, which is rescued in the presence of an IR25a transgene, expressed under the control of IR25a-GAL4. Note that because the IR25a-Gal4 and Gr66a-LexA transgenes and IR25a2 are all located on the second chromosome, rescue experiments could not be carried out with the GAL80 suppressor transgenes. Consequently, only the 5b sensillum was analyzed, because the neurons in the 5v and 5s sensilla are in too close a proximity for unequivocal identification. Genotypes: w1118/w1118; IR25a-GAL4/+; UAS-GCaMP6m/+ (IR25a control); w1118/ w1118; IR25a-GAL4 IR25a2/IR25a2; UAS-GCaMP6m/+ (IR25a mutant); and w1118/w1118; IR25a-Gal4 IR25a2/IR25a2 UAS-IR25a; UAS-GCaMP6m/+ (IR25a rescue). Concentrations used: 100 mM for acetic acid, citric acid, and tartaric acid; 10 mM for glycolic acid and malic acid; 50 mM for HCl; 50 mM and 500 mM for NaCl; and 100 and 500 mM for KCl. All data are shown as the mean, and error bars indicate SEM; one-way ANOVA with post hoc Bonferroni correction, different letters indicate a significant difference with p < 0.05; 6 % n % 12.

Figure 4. Oviposition Preference Is Mainly Mediated by Tarsal Sour GRNs (A) Representative example of female oviposition preference on acid-containing sugar-agar: the majority of eggs are laid on the acid-containing half of the plate (left). w1118 females have a strong, dosedependent preference for oviposition on acidcontaining sugar-agar (right) (see also Figure S4). (B) Silencing of all IR76b-GAL4 GRNs or sour (i.e., IR76bonly) GRNs abolishes acid preference, whereas silencing only the sweet and bitter GRNs has no effect. Genotypes: w1118/w1118 (WT); w1118/ w1118; UAS-Kir2.1/+ (UAS control); IR76b-GAL4/ UAS-Kir2.1 (silencing IR76b GRNs); w1118/w1118; IR76b-GAL4 Gr66a-LexA/UAS-Kir2.1; Gr64f-LexA/ LexAop-GAL8o (silencing sour GRNs); and w1118/ Gr33a-GAL4 Gr64f-GAL4/UAS-Kir2.1 w1118; (silencing bitter/sweet GRNs). (C) All major taste organs contribute to oviposition preference for acetic acid (left), but removal of the forelegs has a more severe effect than removal of any other taste organ. The loss of oviposition preference for citric acid after removal of any specific taste organ is less pronounced, and significant reduction is only observed when the forelegs are removed. Genotype: w1118/w1118 (see also Figure S5). Data are shown as the mean, and error bars indicate SEM; Student’s t test for (A), ***p < 0.001; 20 % n % 30. One-way ANOVA with post hoc Bonferroni correction for (B) and (C), different letters indicate significant difference with p < 0.05; 23 % n % 39. O.I., oviposition index.

these data show that oviposition preference requires the function of both IR25a and IR76b. DISCUSSION In mammals, acids activate a specific set of sour taste cells, presumably through a proton channel, the identity of which is not yet known [41, 42]. In this paper, we identified taste sensory neurons in the tarsi of Drosophila that are narrowly tuned to acids, implying that soluble, non-volatile chemicals have critical roles in the chemosensation of fruit flies. We further show that acid detection is mediated by the ionotropic receptors IR76b and IR25a, and behavioral studies have established that both sour GRNs and these IR proteins are essential for oviposition preference. To our knowledge, no sour taste receptors have yet been identified in any animal and, thus, IR76b and IR25a represent the first subunits of such a taste receptor. Refinement of the Drosophila Chemosensory System The complexity of the Drosophila chemosensory system has long been underappreciated. Comprehensive electrophysiological and Ca2+ imaging studies of taste sensilla revealed that both bitter as well as sweet GRNs have distinct tuning profiles [8, 12–15, 43, 44]. This observation is consistent with extensive expression analyses, which showed that different bitter GRNs express overlapping, but not identical, sets of bitter Gr genes [8, 13]. Likewise, sweet GRNs of distinct taste sensilla fall into different subclasses characterized by their sugar response profile [12], which is mediated by the specific subset of sugar Gr genes they express [19]. However, sweet GRNs were recently

implicated in the taste of fatty acids [45], challenging the notion of an exclusive, labeled line model in the taste system of the fly. The identification of taste neurons tuned exclusively to acids and two components of a sour taste receptor, IR25a and IR76b, significantly broadens the perceptive range of the Drosophila taste system. Sour Taste Receptors The IR gene family expanded during invertebrate evolution, presumably from a common ancestor gene of the iGluR gene family [23, 46]. Of the 60 Drosophila IR genes, 14 encode subunits of multimeric receptors in olfactory neurons of coleoconic sensilla in the antenna [23, 39, 47], each expressing up to four different members that include either IR8a or IR25a. These two IRs are the closest relatives to iGluRs, with whom they share overall structure that includes an amino-terminal domain (ATD) absent in all other IR proteins [46]. Based on genetic and electrophysiological studies, Abuin and colleagues proposed that IRs function as tetrameric ligand-gated ion channels with heterogeneous ion-conductance properties, composed either of IR8a or IR25a, and two additional ‘‘antennal’’ IRs that provide odorant specificity [39]. Based on the work presented here, we suggest that IR25a also serves as a core subunit in multimeric ligand-gated taste receptor channels for acids. Furthermore, we propose that IR76b is a co-receptor of this channel along with one additional IR subunit, which together convey ligand-binding specificity. This additional IR might be found among members of the IR20a clade, which are expressed in small subsets of taste neurons [24], albeit none of the available GAL4 lines was found to be expressed in the sour GRNs of tarsi (data not shown). We note that either one or the Current Biology 27, 2741–2750, September 25, 2017 2747

Figure 5. Female Oviposition Preference for Acid-Containing Food Is Mediated by IR76b and IR25a (A) Females lacking IR76b lost preference for oviposition on acid-containing sugar-agar, a phenotype that can be rescued by expressing IR76b under the control of IR76b-GAL4. Note that oviposition preference is also restored when IR76b is only expressed in sour GRNs (lane 7). Genotypes: IR76b+ control (w1118/w1118) (lane 1); w1118/w1118; IR76b1/IR76b1 (lane 2); w1118/w1118; IR76b-GAL4/+; IR76b1/IR76b1 (lane 3); w1118/w1118; UAS-IR76b/+; IR76b1/IR76b1 (lane 4); w1118/w1118; IR76b-GAL4/UAS-IR76b; IR76b1/IR76b1 (lane 5); w1118/w1118; IR76b-GAL4 Gr66a-LexA/+; Gr64f-LexA IR76b2/ R76b1 LexAop-GAL8o (lane 6); and w1118/w1118; IR76b-GAL4 Gr66a-LexA/UAS-IR76b; Gr64f-LexA IR76b2/ R76b1 LexAop-GAL8o (lane 7). (B) Females lacking IR25a lost preference for oviposition on acid-containing sugar-agar. This phenotype was not rescued by expression of a UAS-IR25a gene under the control of the IR25a-GAL4 driver (lane 5), but it was rescued when expressed under the control of IR76b-GAL4 (lanes 8 and 9; see text). Genotypes: IR25a+ control (w1118/w1118) (lane 1); w1118/w1118; IR25a2/IR25a2 (lane 2); w1118/w1118; IR25a-GAL4/+; IR25a2/IR25a2 (lane 3); w1118/w1118; UAS-IR25a/+; IR25a2/ IR25a2 (lane 4); w1118/w1118; IR25a-GAL4 IR25a2/IR25a2 UAS-IR25a (lane 5); w1118/w1118; IR76b-GAL4 IR25a2/IR25a2 (lane 6); w1118/w1118; IR76b-GAL4 IR25a2/ IR25a2 antenna removed (lane 7); w1118/w1118; IR76b-GAL4 IR25a2/IR25a2 UAS-IR25a (lane 8); and w1118/w1118; IR76b-GAL4 IR25a2/IR25a2 UAS-IR25a antenna removed (lane 9). All data are shown as the mean, and error bars indicate SEM; one-way ANOVA with post hoc Bonferroni correction, different letters indicate significant difference with p < 0.05; 20 % n % 39.

other ATD-containing IR protein (IR25a and the olfactory neuronspecific IR8a) is present in most IR-based chemoreceptor complexes characterized thus far [26, 28, 39]. A more basic core function for IR25a as opposed to IR76b is also suggested by rescue experiments in IR25a mutant flies. Specifically, limited expression of UAS-IR25a under the control of IR76ba-GAL4 in antenna-less females was competent to rescue oviposition 2748 Current Biology 27, 2741–2750, September 25, 2017

preference completely for both acids we tested, whereas wider expression of UAS-IR25a under control of IR25a-GAL4 appeared to counteract restored sour taste function, most likely due to the generation receptors in other sensory neurons (Figure 5B). A potential mechanism by which sour GRNs are activated by acids might involve proton (i.e., hydronium) translocation. Our

Ca2+ imaging experiments using HCl or buffered carboxylic acid solutions have shown that protons are necessary and sufficient for activating sour GRNs, whereas the presence of the conjugate carboxylic base is not (Figures 2 and 3). How are these observations compatible with the finding that strong acids are less potent activators of sour GRNs than weak acids (Figure 2; Figure S3)? We suggest that GRN responses to strong, completely dissociated acids are mediated by translocation of free protons (i.e., H3O+) across the channel pore. In contrast, only a small fraction of weak, carboxylic acids is dissociated in the aqueous taste lymph, and a large putative proton pool remains available in non-dissociated form. Non-dissociated carboxylic acids might bind to the IR complex at the channel pore, providing a source of additional protons through dissociation. However, our data are also compatible with a more complex proton transport mechanism that might include additional channels associated with the IR complex. Interestingly, a proton channel has been proposed to be responsible for activation of sour taste cells in the mouse taste bud [41, 42].

SUPPLEMENTAL INFORMATION Supplemental Information includes five figures and can be found with this article online at http://dx.doi.org/10.1016/j.cub.2017.08.003. AUTHOR CONTRIBUTIONS Y.C. and H.A. designed and analyzed the experiments. Y.C. performed all experiments and assisted H.A. in writing the paper. ACKNOWLEDGMENTS We thank Drs. Craig Montell, Kristin Scott, David Anderson, and Yuh Nung Jan for fly strains, members of the Amrein lab for suggestions on experiments, and Dr. Sitcheran for comments on the manuscript. This work was supported by a grant from the NIH-NIDCD (R01DC005606-12). Received: March 2, 2017 Revised: June 29, 2017 Accepted: August 1, 2017 Published: September 7, 2017 REFERENCES

IRs as Multimodal Receptors Although IR proteins have been studied mainly in the olfactory system, they have emerged as critical sensors in numerous other processes, including temperature sensing in larvae and humidity sensing in the antenna [25, 28, 48], and several genes of the IR20a clade are expressed in tarsi and labial palps of the fly, presumably with roles in taste and/or pheromone perception [24]. Last, IR76b has been implicated in attractive low-salt sensing mediated by labellar taste neurons [7], polyamines in taste pegs [27], and amino acid taste in larvae [22]. Interestingly, the sour neuron of one of the tarsal sensilla (5v) is not only activated by acids but also by high concentration of NaCl in an IR76bdependent manner (Figure 3). Our behavioral analyses have shown that sour GRNs on the foretarsi and IR25a/IR76b are essential for oviposition preference on agar that contains acids, which are common ingredients found in natural egg-laying sites (fruits). Acids suppress bacterial and microbial growth [49], and hence their presence may have direct benefits on larval fitness. Last, sour taste neurons are fully functional in adult male flies, and thus it is likely that both males and females can perceive sour taste for purposes other than oviposition. STAR+METHODS Detailed methods are provided in the online version of this paper and include the following: d d d d

d

KEY RESOURCES TABLE CONTRACT FOR REAGENT AND RESOURCE SHARING EXPERIMENTAL MODEL AND SUBJECT DETAILS B Flies husbandry and strains METHOD DETAILS B Expression Analysis and Immunohistochemistry B Chemicals B Acetate and Citrate Buffer B Calcium Imaging B Two-choice Oviposition Assay B Surgical manipulations QUANTIFICATION AND STATISTICAL ANALYSIS

1. Yarmolinsky, D.A., Zuker, C.S., and Ryba, N.J. (2009). Common sense about taste: from mammals to insects. Cell 139, 234–244. 2. Ramos Da Conceicao Neta, E.R., Johanningsmeier, S.D., and McFeeters, R.F. (2007). The chemistry and physiology of sour taste—a review. J. Food Sci. 72, R33–R38. 3. Penniston, K.L., Nakada, S.Y., Holmes, R.P., and Assimos, D.G. (2008). Quantitative assessment of citric acid in lemon juice, lime juice, and commercially-available fruit juice products. J. Endourol. 22, 567–570. 4. Liman, E.R., Zhang, Y.V., and Montell, C. (2014). Peripheral coding of taste. Neuron 81, 984–1000. 5. Roper, S.D. (2007). Signal transduction and information processing in mammalian taste buds. Pflugers Arch. 454, 759–776. 6. Amrein, H., and Thorne, N. (2005). Gustatory perception and behavior in Drosophila melanogaster. Curr. Biol. 15, R673–R684. 7. Zhang, Y.V., Ni, J., and Montell, C. (2013). The molecular basis for attractive salt-taste coding in Drosophila. Science 340, 1334–1338. 8. Weiss, L.A., Dahanukar, A., Kwon, J.Y., Banerjee, D., and Carlson, J.R. (2011). The molecular and cellular basis of bitter taste in Drosophila. Neuron 69, 258–272. 9. Cameron, P., Hiroi, M., Ngai, J., and Scott, K. (2010). The molecular basis for water taste in Drosophila. Nature 465, 91–95. 10. Moon, S.J., Lee, Y., Jiao, Y., and Montell, C. (2009). A Drosophila gustatory receptor essential for aversive taste and inhibiting male-to-male courtship. Curr. Biol. 19, 1623–1627. 11. Jiao, Y., Moon, S.J., Wang, X., Ren, Q., and Montell, C. (2008). Gr64f is required in combination with other gustatory receptors for sugar detection in Drosophila. Curr. Biol. 18, 1797–1801. 12. Dahanukar, A., Lei, Y.T., Kwon, J.Y., and Carlson, J.R. (2007). Two Gr genes underlie sugar reception in Drosophila. Neuron 56, 503–516. 13. Ling, F., Dahanukar, A., Weiss, L.A., Kwon, J.Y., and Carlson, J.R. (2014). The molecular and cellular basis of taste coding in the legs of Drosophila. J. Neurosci. 34, 7148–7164. 14. Miyamoto, T., Chen, Y., Slone, J., and Amrein, H. (2013). Identification of a Drosophila glucose receptor using Ca2+ imaging of single chemosensory neurons. PLoS ONE 8, e56304. 15. Meunier, N., Marion-Poll, F., Rospars, J.P., and Tanimura, T. (2003). Peripheral coding of bitter taste in Drosophila. J. Neurobiol. 56, 139–152. 16. Jiao, Y., Moon, S.J., and Montell, C. (2007). A Drosophila gustatory receptor required for the responses to sucrose, glucose, and maltose identified by mRNA tagging. Proc. Natl. Acad. Sci. USA 104, 14110–14115.


17. Slone, J., Daniels, J., and Amrein, H. (2007). Sugar receptors in Drosophila. Curr. Biol. 17, 1809–1816. 18. Miyamoto, T., Slone, J., Song, X., and Amrein, H. (2012). A fructose receptor functions as a nutrient sensor in the Drosophila brain. Cell 151, 1113–1125. 19. Fujii, S., Yavuz, A., Slone, J., Jagge, C., Song, X., and Amrein, H. (2015). Drosophila sugar receptors in sweet taste perception, olfaction, and internal nutrient sensing. Curr. Biol. 25, 621–627. 20. Shim, J., Lee, Y., Jeong, Y.T., Kim, Y., Lee, M.G., Montell, C., and Moon, S.J. (2015). The full repertoire of Drosophila gustatory receptors for detecting an aversive compound. Nat. Commun. 6, 8867. 21. Moon, S.J., Ko¨ttgen, M., Jiao, Y., Xu, H., and Montell, C. (2006). A taste receptor required for the caffeine response in vivo. Curr. Biol. 16, 1812–1817. 22. Croset, V., Schleyer, M., Arguello, J.R., Gerber, B., and Benton, R. (2016). A molecular and neuronal basis for amino acid sensing in the Drosophila larva. Sci. Rep. 6, 34871. 23. Benton, R., Vannice, K.S., Gomez-Diaz, C., and Vosshall, L.B. (2009). Variant ionotropic glutamate receptors as chemosensory receptors in Drosophila. Cell 136, 149–162. 24. Koh, T.W., He, Z., Gorur-Shandilya, S., Menuz, K., Larter, N.K., Stewart, S., and Carlson, J.R. (2014). The Drosophila IR20a clade of ionotropic receptors are candidate taste and pheromone receptors. Neuron 83, 850–865. 25. Ni, L., Klein, M., Svec, K.V., Budelli, G., Chang, E.C., Ferrer, A.J., Benton, R., Samuel, A.D., and Garrity, P.A. (2016). The ionotropic receptors IR21a and IR25a mediate cool sensing in Drosophila. eLife 5, e13254. 26. Knecht, Z.A., Silbering, A.F., Ni, L., Klein, M., Budelli, G., Bell, R., Abuin, L., Ferrer, A.J., Samuel, A.D., Benton, R., and Garrity, P.A. (2016). Distinct combinations of variant ionotropic glutamate receptors mediate thermosensation and hygrosensation in Drosophila. eLife 5, e17879. 27. Hussain, A., Zhang, M., U¨c¸punar, H.K., Svensson, T., Quillery, E., Gompel, N., Ignell, R., and Grunwald Kadow, I.C. (2016). Ionotropic chemosensory receptors mediate the taste and smell of polyamines. PLoS Biol. 14, e1002454. 28. Enjin, A., Zaharieva, E.E., Frank, D.D., Mansourian, S., Suh, G.S., Gallio, M., and Stensmyr, M.C. (2016). Humidity sensing in Drosophila. Curr. Biol. 26, 1352–1358. 29. Wang, Z., Singhvi, A., Kong, P., and Scott, K. (2004). Taste representations in the Drosophila brain. Cell 117, 981–991. 30. Kwon, J.Y., Dahanukar, A., Weiss, L.A., and Carlson, J.R. (2014). A map of taste neuron projections in the Drosophila CNS. J. Biosci. 39, 565–574. 31. Chen, Y., and Amrein, H. (2014). Enhancing perception of contaminated food through acid-mediated modulation of taste neuron responses. Curr. Biol. 24, 1969–1977. 32. Charlu, S., Wisotsky, Z., Medina, A., and Dahanukar, A. (2013). Acid sensing by sweet and bitter taste neurons in Drosophila melanogaster. Nat. Commun. 4, 2042. 33. Gou, B., Liu, Y., Guntur, A.R., Stern, U., and Yang, C.H. (2014). Mechanosensitive neurons on the internal reproductive tract contribute to egg-laying-induced acetic acid attraction in Drosophila. Cell Rep. 9, 522–530. 34. Joseph, R.M., Devineni, A.V., King, I.F., and Heberlein, U. (2009). Oviposition preference for and positional avoidance of acetic acid provide a model for competing behavioral drives in Drosophila. Proc. Natl. Acad. Sci. USA 106, 11352–11357. 35. Eisses, K.T. (1997). The influence of 2-propanol and acetone on oviposition rate and oviposition site preference for acetic acid and ethanol of Drosophila melanogaster. Behav. Genet. 27, 171–180.


36. Jeong, Y.T., Shim, J., Oh, S.R., Yoon, H.I., Kim, C.H., Moon, S.J., and Montell, C. (2013). An odorant-binding protein required for suppression of sweet taste by bitter chemicals. Neuron 79, 725–737. 37. Andrews, J.C., Ferna´ndez, M.P., Yu, Q., Leary, G.P., Leung, A.K., Kavanaugh, M.P., Kravitz, E.A., and Certel, S.J. (2014). Octopamine neuromodulation regulates Gr32a-linked aggression and courtship pathways in Drosophila males. PLoS Genet. 10, e1004356. 38. Thoma, V., Knapek, S., Arai, S., Hartl, M., Kohsaka, H., Sirigrivatanawong, P., Abe, A., Hashimoto, K., and Tanimoto, H. (2016). Functional dissociation in sweet taste receptor neurons between and within taste organs of Drosophila. Nat. Commun. 7, 10678. 39. Abuin, L., Bargeton, B., Ulbrich, M.H., Isacoff, E.Y., Kellenberger, S., and Benton, R. (2011). Functional architecture of olfactory ionotropic glutamate receptors. Neuron 69, 44–60. 40. Ai, M., Blais, S., Park, J.Y., Min, S., Neubert, T.A., and Suh, G.S. (2013). Ionotropic glutamate receptors IR64a and IR8a form a functional odorant receptor complex in vivo in Drosophila. J. Neurosci. 33, 10741–10749. 41. Bushman, J.D., Ye, W., and Liman, E.R. (2015). A proton current associated with sour taste: distribution and functional properties. FASEB J. 29, 3014–3026. 42. Chang, R.B., Waters, H., and Liman, E.R. (2010). A proton current drives action potentials in genetically identified sour taste cells. Proc. Natl. Acad. Sci. USA 107, 22320–22325. 43. Yavuz, A., Jagge, C., Slone, J., and Amrein, H. (2014). A genetic tool kit for cellular and behavioral analyses of insect sugar receptors. Fly (Austin) 8, 189–196. 44. Marella, S., Fischler, W., Kong, P., Asgarian, S., Rueckert, E., and Scott, K. (2006). Imaging taste responses in the fly brain reveals a functional map of taste category and behavior. Neuron 49, 285–295. 45. Masek, P., and Keene, A.C. (2013). Drosophila fatty acid taste signals through the PLC pathway in sugar-sensing neurons. PLoS Genet. 9, e1003710. 46. Croset, V., Rytz, R., Cummins, S.F., Budd, A., Brawand, D., Kaessmann, H., Gibson, T.J., and Benton, R. (2010). Ancient protostome origin of chemosensory ionotropic glutamate receptors and the evolution of insect taste and olfaction. PLoS Genet. 6, e1001064. 47. Ai, M., Min, S., Grosjean, Y., Leblanc, C., Bell, R., Benton, R., and Suh, G.S. (2010). Acid sensing by the Drosophila olfactory system. Nature 468, 691–695. 48. Chen, C., Buhl, E., Xu, M., Croset, V., Rees, J.S., Lilley, K.S., Benton, R., Hodge, J.J., and Stanewsky, R. (2015). Drosophila Ionotropic Receptor 25a mediates circadian clock resetting by temperature. Nature 527, 516–520. 49. Carpenter, C.E., and Broadbent, J.R. (2009). External concentration of organic acid anions and pH: key independent variables for studying how organic acids inhibit growth of bacteria in mildly acidic foods. J. Food Sci. 74, R12–R15. 50. Thistle, R., Cameron, P., Ghorayshi, A., Dennison, L., and Scott, K. (2012). Contact chemoreceptors mediate male-male repulsion and male-female attraction during Drosophila courtship. Cell 149, 1140–1151. 51. Asahina, K., Watanabe, K., Duistermars, B.J., Hoopfer, E., Gonzalez, C.R., Eyjolfsdottir, E.A., Perona, P., and Anderson, D.J. (2014). Tachykinin-expressing neurons control male-specific aggressive arousal in Drosophila. Cell 156, 221–235. 52. Gomori, G. (1955). Preparation of buffers for use in enzyme studies. Methods Enzymol. 1, 138–146.

STAR+METHODS KEY RESOURCES TABLE

REAGENT or RESOURCE

SOURCE

IDENTIFIER

Antibodies Chicken polyclonal anti- GFP

Thermo Fisher Scientific

PA1-86341, RRID: AB_931091

Rat polyclonal anti- HA

Covance

MMS-101P, RRID: AB_2314672

Rabbit polyclonal anti- IR25a

[23]

N/A

Mouse monoclonal anti- nc82

DSHB

nc82, RRID: AB_2314866

Alexa 488 conjugated goat anti- chicken


A11039, RRID: AB_2534096

Alexa 568 conjugated goat antimouse


A11004, RRID: AB_2534072

Alexa 555 conjugated goat anti- rabbit


A32732, RRID: AB_2633281

Cy3-conjugated goat anti- rat

Jackson Immunoresearch

112-165-072

Sucrose

Sigma-Aldrich

S0389

Denatonium benzoate

Sigma-Aldrich

D5765

Lobeline hydrochloride

Tokyo Chemical Industry

L0096

Chemicals, Peptides, and Recombinant Proteins

Quinine hydrochloride

Sigma-Aldrich

8221940025

Alanine

Sigma-Aldrich

A7627

Arginine

Sigma-Aldrich

A5006

Asparagine

Sigma-Aldrich

A0884

Aspartic acid

Sigma-Aldrich

A9256

Cysteine

Sigma-Aldrich

C7352

Glutamine

Sigma-Aldrich

G3126

Glutamic acid

Sigma-Aldrich

G8415

Glycine

Sigma-Aldrich

G1251

Histidine

Sigma-Aldrich

H8000

Isoleucine

Sigma-Aldrich

I2752

Leucine

Sigma-Aldrich

L8000

Lysine

Sigma-Aldrich

L5501

Methionine

Sigma-Aldrich

M9625

Phenylalanine

Sigma-Aldrich

P2126

Proline

Sigma-Aldrich

P0380

Serine

Sigma-Aldrich

S4500

Threonine

Sigma-Aldrich

T8625

Tryptophan

Sigma-Aldrich

93659

Tyrosine

Sigma-Aldrich

T3754

Valine

Sigma-Aldrich

V0500

Citric acid

Sigma-Aldrich

251275

Glycolic acid

Sigma-Aldrich

124737

Malic acid

Sigma-Aldrich

M8304

Acetic acid

EMD

AX0073

Hydrochloric acid

EMD

HX0603

Tartaric acid

Spectrum Chemical MFG

T1009

Sodium chloride

Sigma-Aldrich

S9888

Potassium chloride

EMD

PX1405

Sodium acetate

Fisher Scientific

S209-3

Sodium citrate

EMD

SX0445

Cornmeal

Genesee

66-101 (Continued on next page)

Current Biology 27, 2741–2750.e1–e4, September 25, 2017 e1

Continued REAGENT or RESOURCE

SOURCE

IDENTIFIER

Drosophila agar, type II

Genesee

62-103

Yeast extract

Genesee

62-106

Malt extract

Alternative Beverage

MUN-UL

Propionic acid

VWR

TCP0500-500mL

Streptomycin

Sigma-Aldrich

S6501

Tegosept

Sigma-Aldrich

PHR1012

Instant active yeast

Lesaffre

15850

D. melanogaster: w1118

Bloomington Drosophila Stock Center

BDSC: 3605; FlyBase: FBst0003605

D. melanogaster: w*; IR76b1


BDSC: 51309; Flybase: FBst0051309

D. melanogaster: w*; IR76b2



D. melanogaster: w*; P{IR76b-GAL4.1.5}2



D. melanogaster: w*; P{IR76b-QF.1.5}

[7]

Flybase: FBtp0085487

D. melanogaster: w*; P{UAS-IR76b.Z}2/CyO; TM2/TM6B, Tb1



D. melanogaster: w*; TI{TI}IR25a1/CyO



D. melanogaster: w*; P{IR25a-GAL4.A}236.1; TM2/TM6B, Tb1



D. melanogaster: w*; M{UAS-IR25a.attB}

[39]

Flybase: FBal0249355

D. melanogaster: w*; P{Gr64f-GAL4.9.7}5/CyO; MKRS/TM2



D. melanogaster: w*; TI{LexA::VP16}Gr64fLexA

[14]

Flybase: FBti0168176

D. melanogaster: w*; TI{GAL4}Gr33aGAL4



D. melanogaster: w*; Gr66a-LexA/CyO; TM2/TM6B

[50]

Flybase: FBal0277069

D. melanogaster: y1 w*; P{UAS-mCD8::GFP.L}LL5



D. melanogaster: y1 w*; P{UAS-mCD8.mRFP.LG}18a



D. melanogaster: w1118; P{20XUAS-IVS-GCaMP6m}attP40



D. melanogaster: w*; P{LexAop-Gal80.T}

[50]


D. melanogaster: w*; P{UAS-HsapyKCNJ2.EGFP}1



Experimental Models: Organisms/Strains

D. melanogaster: w*; p{Otd-nls:FLPo}

[51]


D. melanogaster: w*; P{GAL80}tshGAL80

[37]

Flybase: FBti0114123

D. melanogaster: y1 w1118; P{QUAS-mtdTomato-3xHA}14/CyO



Software and Algorithms NIS-Elements

Nikon

N/A

SPSS

IBM

N/A

Other Nikon Eclipse Ti inverted microscope

Nikon

N/A

Nikon A1 confocal microscope

Nikon

N/A

35mM Glass bottom dish

MatTek

P35G-0-10-C

60X15 mM Petri dish

Falcon

353004

e2 Current Biology 27, 2741–2750.e1–e4, September 25, 2017

CONTRACT FOR REAGENT AND RESOURCE SHARING Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Hubert Amrein ([email protected]). EXPERIMENTAL MODEL AND SUBJECT DETAILS Flies husbandry and strains All flies were cultured on common corn meal food at 25 C on a 12 hr light/dark cycle. Standard fly food, contained in 1.5 L of water, is composed of 10.88 g of agar, 78 g of corn meal, 165 g malt extract, 41.25 g of yeast extract, 4.69 g of propionic acid, 0.28 g of streptomycin and 2.11 g of tegosept. A detailed list of fly strains used for this paper is provided in the Key Source Table. METHOD DETAILS Expression Analysis and Immunohistochemistry Four to seven day old flies were anesthetized with CO2: Forelegs were cut between the 3th and 4th segment with a razor blade, while labella were dissected with forceps. For brains and ventral nerve cords, whole, abdominally punctured flies were used. Dissected tissues/whole flies were fixed in PBS with 4% paraformaldehyde for 30 min at room temperature, and then rinsed twice with PBST (PBS+0.3% triton) solution and washed with PBST for 1 hr. Brains and ventral nerve cords were then dissected for antibody staining. For antibody staining, tissue preparations were incubated in PBST+10% horse serum at 4 C overnight. Preparations were rinsed twice and washed for 1 hr with PBST. Secondary antibodies were added to PBST+10% horse serum, and then incubated at room temperature for 3 hr. Secondary antibodies were removed, tissue rinsed twice and washed for 3 hr with PBST. Preparations were mounted on slides with Vector Shield, covered with a coverslip and sealed with nail polish. Slides were stored at 4 C until examination by confocal microscopy. Primary antibodies: Chicken polyclonal anti- GFP (dilution at 1:2000, Thermo Fisher Scientific), rat polyclonal anti- HA (dilution at 1:200, Covance), rabbit polyclonal anti- IR25a (dilution at 1:200) [23] and mouse monoclonal antinc82 (dilution at 1:50, DSHB). Secondary antibodies: Alexa 488 conjugated goat anti- chicken (dilution at 1:500, Thermo Fisher Scientific) and Alexa 555 conjugated goat anti- rabbit (dilution at 1:500, Thermo Fisher Scientific), Alexa 568 conjugated goat antimouse (dilution at 1:500, Thermo Fisher Scientific), and Cy3 conjugated goat anti- rat (dilution at 1:300, Jackson Immunoresearch). Images were acquired using a Nikon A1 confocal microscope and NIS element acquisition and analysis package. Chemicals A detailed list of chemicals used in these studies is provided in the Key Resources Table. Acetate and Citrate Buffer To produce 100 mM acetate and citrate buffers, 100 mM acetic or citric acid was mixed with 100 mM sodium acetate or 100 mM sodium citrate, respectively, at ratios described by Gomori [52] until the solutions reached the pH of 3.0, 3.6, 4.2, 4.8 and 5.4, respectively. Calcium Imaging Foreleg of females of indicated genotypes were prepared as follows: After cutting below the fermur, the lower part of the tibia and the first three tarsal segments were dipped in silicone oil (MatTek Corp) to prevent leakage and dehydration and placed on double-sided scotch tape and stuck to a glass bottom dish (see also [14]). The preparation was fixed by covering it with 1% agarose, so that only the fourth and fifth tarsal segments were exposed. The preparation was covered with 100 mL of water. Imaging was initiated by adding 100 mL of test solutions (2x of the final concentration of the indicated ligand) using a pipette. Images were acquired every 500 ms, 20 frames before application (10 s) and 90 frames after application (45 s) of ligand. Each preparation was tested with 2-4 different compounds. Imaging was performed with a Nikon eclipse Ti inverted microscope using a Nikon 20x water objective and a Lumen 200 light source (Prior Scientific Inc). Samples were excited at 488 nm (metal halide lamp), and emitted light was collected through a 515-555 nm filter. Data acquisition was performed with NIS-Elements software (Nikon). To calculate max DF/F %, measurements were taken in the cell bodies. Adjacent regions were used to determine background auto fluorescence. Average of ten frames taken immediately before the application of ligand was defined as a baseline. Max DF/F % represented the highest value within 30 s after ligand application. Two-choice Oviposition Assay two-choice oviposition assay was modified from Joseph et al. [34], by replacing corn meal food with 1% agar containing 100 mM sucrose (sugar-agar). Plastic petri dishes (60 X15 mm, Falcon), one half containing sugar-agar and the other half containing sugaragar plus acid, were used as oviposition platform. About 40 to 60 newly eclosed flies (half males, half females) were transferred to fresh vials with standard food, plus instant yeast and kept at 25 C for 4-7 days; flies were anesthetized on ice and transferred into plastic cups (perforated with holes), which were covered with the oviposition platforms. Flies were let to lay eggs at room temperature

Current Biology 27, 2741–2750.e1–e4, September 25, 2017 e3

(22 ± 1 C) for 12 to 14 hr in the dark. Female flies in each plastic cup and eggs on each side of the plate were counted. Oviposition index (O.I.) was calculated as: (number of eggs on sucrose-acid mixture – number of eggs on sucrose)/total number of eggs. Surgical manipulations Surgical manipulations were carried out on 2 to 3 day-old females, which were anesthetized with CO2 for removal of respective appendages. A razor blade or forceps were used for removing the tip of labellum and the tarsi/3rd antennal segment, respectively. Females were let to recover for 4 to 5 days on standard food with instant yeast in the presence of males, before they were used in the oviposition assays. QUANTIFICATION AND STATISTICAL ANALYSIS Statistical analysis was carried out in SPSS. Data shown in graphs indicate ‘‘mean’’ and error bar represent ‘‘standard error of the mean’’ (sem). n represents the number of preparations in immunostatinng experiments, the number of neurons in Ca2+ imaging experiments and the number of plates in two-choice oviposition preference assays. Two-tailed unpaired Student’s t test was performed to compare two different groups of samples (Figures 2B and 4A; Figures S2, S3, and S4), while one-way ANOVA with Bonferroni correction was performed to compare multiple groups of samples (Figures 3, 4B, 4C, and 5).

e4 Current Biology 27, 2741–2750.e1–e4, September 25, 2017