
==== Front
Sci Adv
Sci Adv
sciadv
advances
Science Advances
2375-2548
American Association for the Advancement of Science

adi9101
10.1126/sciadv.adi9101
Research Article
Neuroscience
SciAdv r-articles
Cellular Neuroscience
Neuroscience
Cellular Neuroscience
A highly conserved A-to-I RNA editing event within the glutamate-gated chloride channel GluClα is necessary for olfactory-based behaviors in Drosophila
GluClα editing is critical for olfaction in flies
Zak Hila Conceptualization Formal analysis Investigation Methodology Project administration Supervision Validation Visualization Writing - original draft Writing - review & editing 1 2 3
https://orcid.org/0000-0001-5316-0495
Rozenfeld Eyal Formal analysis Investigation Resources Visualization 4 5
Levi Mali Investigation Methodology Validation 1 2 3
https://orcid.org/0000-0001-9717-4143
Deng Patricia Conceptualization Resources 6
https://orcid.org/0000-0003-1124-3916
Gorelick David Formal analysis Software Validation 1 2 3
Pozeilov Hadar Investigation Methodology Software Validation 1 2 3
https://orcid.org/0000-0002-7569-8875
Israel Shai Investigation
Paas Yoav yoav.paas@biu.ac.il
Conceptualization Formal analysis Methodology Resources Visualization Writing - original draft Writing - review & editing 4 5
Paas Yoav Conceptualization Formal analysis Methodology Resources Visualization Writing - original draft Writing - review & editing 1
https://orcid.org/0000-0003-0713-1399
Li Jin Billy Conceptualization Funding acquisition Methodology Project administration Supervision 6
https://orcid.org/0000-0001-9726-1511
Parnas Moshe Conceptualization Funding acquisition Methodology Project administration Resources Supervision Validation Visualization Writing - original draft Writing - review & editing 4 5
https://orcid.org/0000-0003-1246-1827
Shohat-Ophir Galit Conceptualization Funding acquisition Methodology Project administration Resources Supervision Validation Visualization Writing - original draft Writing - review & editing 1 2 3 *
1 The Mina & Everard Goodman Faculty of Life Sciences, Bar-Ilan University, Ramat Gan 5290002, Israel.
2 Bar-Ilan University, Ramat Gan 5290002, Israel.
3 The Nanotechnology Institute, Bar-Ilan University, Ramat Gan 5290002, Israel.
4 Department of Physiology and Pharmacology, Sackler School of Medicine, Tel Aviv University, Tel Aviv 69978, Israel.
5 Sagol School of Neuroscience, Tel Aviv University, Tel Aviv 69978, Israel.
6 Department of Genetics, Stanford University, Stanford, CA 94305, USA.
* Corresponding author. Email: galit.ophir@biu.ac.il
06 9 2024
04 9 2024
10 36 eadi910129 5 2023
29 7 2024
Copyright © 2024 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works. Distributed under a Creative Commons Attribution License 4.0 (CC BY).
2024
The Authors
https://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution license, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

A-to-I RNA editing is a cellular mechanism that generates transcriptomic and proteomic diversity, which is essential for neuronal and immune functions. It involves the conversion of specific adenosines in RNA molecules to inosines, which are recognized as guanosines by cellular machinery. Despite the vast number of editing sites observed across the animal kingdom, pinpointing critical sites and understanding their in vivo functions remains challenging. Here, we study the function of an evolutionary conserved editing site in Drosophila, located in glutamate-gated chloride channel (GluClα). Our findings reveal that flies lacking editing at this site exhibit reduced olfactory responses to odors and impaired pheromone-dependent social interactions. Moreover, we demonstrate that editing of this site is crucial for the proper processing of olfactory information in projection neurons. Our results highlight the value of using evolutionary conservation as a criterion for identifying editing events with potential functional significance and paves the way for elucidating the intricate link between RNA modification, neuronal physiology, and behavior.

A conserved A-to-I RNA editing site in glutamate-gated chloride channel GluClα is necessary for olfactory behaviors in Drosophila.

http://dx.doi.org/10.13039/100006221 United States - Israel Binational Science Foundation 2019091 http://dx.doi.org/10.13039/100006221 United States - Israel Binational Science Foundation 2019026 http://dx.doi.org/10.13039/100006221 United States - Israel Binational Science Foundation 2020636 http://dx.doi.org/10.13039/501100000781 European Research Council 101085605 http://dx.doi.org/10.13039/501100003977 Israel Science Foundation 174/19 http://dx.doi.org/10.13039/501100003977 Israel Science Foundation 404/23 NIH R35 GM144100 License OptionCC BY
==== Body
pmcINTRODUCTION

Adenosine-to-inosine (A-to-I) RNA editing, catalyzed by ADAR enzymes, is a ubiquitous mechanism that generates transcriptomic and proteomic diversity in metazoans (1, 2). Most of the RNA editing events in mammals occur in noncoding parts of the transcriptome by ADAR1 and serve to prevent aberrant immune responses toward self–double-stranded RNAs (dsRNAs) (3–6). Only a small fraction of editing events occur in protein-coding sequences mainly by ADAR2, where the deamination of adenosine to inosines introduces nonsynonymous substitutions (known as recoding events) that produce different protein isoforms from a single DNA sequence (7–12). The binding of ADAR to dsRNA structures containing target sequence and an editing complementary sequence (ECS) promotes the conversion of specific adenosines within the target sequence into inosines, which are subsequently recognized by the translation machinery as guanosine (G) (11). Most recoding events occur within genes that function in neurons, accounting for neuronal deficits exhibited by ADAR2 knockout mice (13). A central example is an essential recoding event of glutamine to arginine within the M2 domain of AMPA receptor that is necessary for the proper function of the channel, the absence of which causes lethality (13–15). Another functionally important recoding event is found within the calmodulin-binding IQ domain of the voltage-activated calcium channel (Cav1.3) and functions to regulate the calcium-dependent inactivation kinetics of the channel (16), which, in turn, shapes hippocampal plasticity and memory in mice (16, 17).

Recent advances in RNA sequencing and computational analysis facilitated the identification of millions of editing sites in humans (1, 3, 18–20), tens of thousands in mice (1), and thousands in Drosophila (9, 21–23). While the function of RNA editing in Drosophila melanogaster resembles that of mammals (8, 24–32), most editing events in D. melanogaster are predicted to cause nonsynonymous protein-coding changes, making it hard to pinpoint which of their in vivo function to further study. As a consequence, research in the field was limited to studying the function of few editing events at the biochemical level (12, 33–36), leaving their physiological function at the whole organism level practically unknown. To bridge this gap, we chose to study a recoding site that shows high evolutionary conservation across several Drosophila species and is found within the extracellular domain of glutamate-gated chloride channel (GluClα) (37). The high evolutionary conservation of the particular recoding of Isoleucine at position 27 to valine is suggestive of its functional importance (37).

GluClα is an inhibitory channel in invertebrates that belongs to the Cys-loop ligand-gated family of ion channels (38), consisting of five homologous subunits that are arranged in a radial manner (Fig. 1A). Each subunit has an N-terminal extracellular hydrophilic domain containing ligand-binding residues, four transmembrane helices, and an intracellular long segment (39). Glutamate binding triggers a rapid influx of chloride ions, leading to hyperpolarization of the target cell (38, 40). Drosophila GluClα is broadly expressed in the nervous system (41, 42) and was shown in recent studies to control homeostatic modulation of presynaptic release at the neuromuscular junction (43), to regulate visual responses (44, 45) and the processing of olfactory information (40). Here, we show that recoding of Ile27 to Val is necessary for proper olfactory responses, since flies expressing endogenous GluClα in which recoding of Ile27 is prevented exhibit reduced responses to appetitive and aversive odors and impaired pheromone-based social and sexual responses. The behavioral phenotypes of GluClα-unedited flies are associated with altered activity of olfactory projection neurons (PNs) specifically in the VA1v glomeruli and can be rescued by the expression of a fully edited form of GluClα in PNs. Our results demonstrate the physiological relevance of evolutionary conserved editing events in regulating complex behavior in Drosophila and may contribute to understanding the biophysical regulation of Cys-loop receptor-channels.

Fig. 1. The recoding of Ile27 to Val is conserved across Drosophila species.

(A) Schematic representation of the Drosophila GluClα channel and the relative location of its A-to-I editing sites. (B) Relative editing levels of GluClα-editing sites in RNA extracted from various Drosophila species. n = 2 per strain. (C) Average editing levels of GluClα sites across various neuronal populations in Drosophila melanogaster brain (n = 3 for each neuronal population). (D) Schematic illustration of the approach used to generate GluClα-unedited flies. ADAR binds to a dsRNA structure formed by base pairing between an exon sequence containing the editing site and a complementary intronic sequence (ECS). Deletion of the ECS required for the editing of Ile27 was achieved via CRISPR-Cas9. (E) Pearson correlation comparing editing patterns of 1766 sites between GluClα-unedited and wt D. melanogaster flies; sites within GluClα transcripts are colored (n = 3 samples per neuronal population). (F) Relative GluClα protein levels extracted from GluClα-unedited and wt heads. Left, representative blot (α-GluClα antibodies and α-Actin as a loading control). Right, quantitation of three independent repeats, P > 0.05, t test. (G) Normalized counts of GluClα peptides analysis (MS) of GluClα protein levels in postsynaptic densities extracted from GluClα-unedited and wt flies. n = 3 P > 0.05 paired t test. ns., not significant.

RESULTS

The recoding of Ile27 to Val in GluClα is evolutionarily conserved and exhibits high editing levels

A systematic comparison of conserved RNA editing events across six different Drosophila species identified a subset of highly conserved editing sites, suggestive of their functional importance (46). One of the conserved sites is found in the inhibitory glutamate-gated chloride channel (GluClα). The transcript of GluClα has six A-to-I editing sites, five of which are classified as nonsynonymous (Fig. 1A). Comparing the editing pattern of GluClα among the six Drosophila species indicated that the recoding event of Ile27 located within the extracellular domain of the channel is conserved in D. melanogaster, yakuba, simulans, and pseudoobscura with more than 75% of the transcripts in all four species are edited (Fig. 1B and table S1). Next, we analyzed the spatial distribution of the six editing events in GluClα across the fly brain using a dataset that we previously generated, containing transcriptomes of nine different neuronal populations including the relative editing levels of thousands of editing sites (47). We found that Ile27 to Val is the most highly edited site in all tested neuronal populations, ranging from 80 to 94% editing (Fig. 1C and table S2, A and B).

The three-dimensional (3D) structure of the Drosophila GluClα channel is not known. Yet, important structural information can be derived on the basis of homology with other Cys-loop receptors. The Ile to Val editing site is situated at position 27 of the immature protein. To locate the editing site in respect to the signal peptide, we have analyzed the full-length sequence of the Drosophila GluClα subunit using SignalP 6.0 (48). SignalP 6.0 predicted, with very high probability, a signal peptide that includes amino acids 1 to 22 (fig. S1A). Hence, the edited Ile27 is probably the fifth amino acid of the mature protein. We then submitted the full-length amino acid sequence of the immature protein to a secondary-structure prediction by PSIPRED 4.0 (49) and transmembrane topology prediction (50) in the PSIPRED server (51). These predictions support the location of Ile27 C-terminally (downstream) to the signal peptide and suggest that I27 is located N-terminally to a putative α helix of the Drosophila GluClα protein (fig. S1B). Very similar secondary-structure prediction obtained with JPred4 (52) corroborated the prediction by PSIPRED 4.0 (not shown). Multiple sequence alignment between the Drosophila GluClα subunit and various subunits of anionic Cys-loop receptors [performed by Clustal Omega (53)] indicates that the predicted α helix is aligned with the first α helix (termed α1) of Cys-loop receptor subunits whose 3D structure was determined at high resolution (fig. S1C). In addition, it reveals that the editing site (I27 to V) is indeed located very close to the N terminus of the mature protein, N-terminally to the first α helix (fig. S1C). However, this position is not included in the 3D structures of the various Cys-loop receptor subunits used for the multiple alignment (fig. S1C). Noteworthy in all known 3D structures of eukaryotic Cys-loop receptors, the N-terminal α helix (α1) proceeds by a loop that connects to the first β strand (β1). This loop contains a highly conserved amino-acid cluster (fig. S1C).

The high ratio of GluClα transcripts that undergo editing at this particular site and its evolutionary conservation prompted us to further examine its functional relevance. To this end, we used the CRISPR/Cas9 system to impair editing at this site by removing the specific ECS (intronic region between positions 15,578,744 and 15,579,171) required for the binding of ADAR (Fig. 1D and fig. S2A). Comparing the editing pattern of all six sites between GluClαunedited and matched genetic control showed that the deletion of this particular ECS abolished editing of Ile27 to Val and that of a close-by synonymous site that share the same ECS (Lys26 to Lys) while leaving the editing levels of the other sites intact (Fig. 1E and table S3). The removal of the ECS did not affect protein expression, as can be seen by comparable levels of GluClα protein in protein extracts from GluClαunedited and control flies (Fig. 1F). Taking into account the close proximity of Ile27 to the signal peptide, we verified that eliminating the recoding of Ile27 to Val did not impair the proper localization of the channel to the cell membrane. To this end, we isolated synaptosomes and postsynaptic densities from wt and GluClαunedited flies (54) and compared the relative amount of GluClα protein using quantitative mass spectrometry (MS) analysis (Fig. 1G and table S4). Last, we confirmed that the elimination of editing at this site did not result in reduced life span and found that GluClαunedited flies rather have an extended life span in comparison to wt controls (fig. S2B).

Recoding of Ile27 to Val is necessary for proper olfactory responses

Given the established role of GluClα in processing olfactory information (40), we tested the behavioral responses of GluClαunedited flies to appetitive and aversive odors. Flies detect odors using olfactory receptor neurons (ORNs) located at the antennae and maxillary palps. In general, each ORN expresses a single odorant receptor gene (55–58). ORNs expressing the same receptor send their axons to the same glomerulus in the antennal lobe (AL) (59–61). In addition, the AL also contains second-order PNs, which receive input from a single glomerulus (40, 62) and are responsible for delivering odor information to higher brain regions (40, 62). The AL also contains local neurons (LNs). Most LNs are inhibitory GABAergic neurons, but about one-third of LNs are inhibitory glutamatergic neurons (40), which inhibit via GluClα both PNs and LNs (Fig. 2A) (40).

Fig. 2. Ile27 recoding is necessary for proper olfactory responses.

(A) Schematic illustration of GluClα function within the AL. PNs receive input from ORNs, and their response is dampened by glutamate that is released from LNs and acts on GluClα receptors. (B) Illustration of the two-choice olfactory trap paradigm, where flies are given a choice to enter a trap containing an odor compound (red agar) or without odor (clear agar), and the number of trapped flies is analyzed after 24 hours. Preference of GluClαunedited and wt males toward apple juice (C) and ethanol (D) containing, n = 20 ***P < 0.0001; Mann-Whitney test. (E) Graphical depiction of Y-maze assay in which flies can climb toward odors placed at the end of the arms, and their amount on each side is used to calculate preference toward apple juice or ethanol. Preference of GluClαunedited and wt males toward apple juice (F) or ethanol (G) as analyzed by the Y-maze assay. Unedited flies exhibit reduced attraction to apple juice and ethanol **P < 0.01; T test, n = 20. wt flies exhibit preference to apple juice and ethanol that differs significantly than 0 (one-sample t test; P < 0.0001), whereas GluClαunedited do not differ than 0. (H) Schematics of the multiplex system that measures the fraction of time during which single flies move toward or away from acetic acid (I) and 2-butanone (J). Mean time of wt flies away from acetic differs or 2-butanone differs from zero [one sample t test; P value (two tailed) <0.0001], while unedited is no different from 0. Unedited flies show decreased aversion to acetic acid, n = 35 ***P < 0.0001; Mann-Whitney test, and toward 2-butanone, n = 40 ***P < 0.0001; Mann-Whitney test.

Since attraction or aversion to odors requires intact motor capabilities, we first verified that GluClαunedited flies exhibit normal locomotor behaviors by analyzing various features of their motor actions using the FlyBowl tracking–based behavioral analysis system (fig. S3 and table S7). Fruit flies use odors to navigate across complex environments and locate food sources (63). Ethanol and amine volatiles emitted from rotting fruits attract Drosophila flies, who aggregate on fermenting fruits as a hub for feeding, mating, and egg laying (64–66). To examine the attraction of GluClαunedited flies to appetitive odors, we used the simple two-choice olfactory trap assay, in which one of the traps contained an appetitive odor such as apple juice or ethanol, whereas the other trap contained plan agar as control (Fig. 2B). Fifty 3-day-old flies were introduced into chambers containing two Eppendorf-based traps with either 1% agar or agar supplemented with 10% apple juice and were left uninterrupted for 24 hours before the number of flies in each trap was counted. Comparing the preference of wt flies to that of GluClαunedited flies, we found that while wt flies exhibited strong preference toward apple juice, GluClαunedited flies showed a marked reduction in the number of flies that entered the apple juice traps (Fig. 2C). The reduced preference toward appetitive odor was also apparent toward ethanol (Fig. 2D). Next, we tested the attraction toward each of these odors using a simple Y-maze assay, in which flies can choose an arm containing at its end agar or agar with 10% apple juice or ethanol (Fig. 2E). Unlike the olfactory trap assay that measures the number of flies trapped in each of the traps over 24 hours, the Y maze is based on an immediate choice between the arms of the maze. Fifty 3-day-old GluClαunedited or wt flies were inserted at the entrance of each Y maze apparatus, and the flies were given 40 s to choose between the two arms before the vials were cupped. While wt flies show a strong preference to apple juice and ethanol-containing arms, the mean preference of GluClαunedited flies is close to zero (Fig. 2, F and G), implying a random choice between the two arms. The decreased attraction of GluClαunedited flies to apple juice and ethanol in two independent assays suggests that editing of Ile27 to Val is necessary for proper olfactory responses toward appetitive odors.

Next, we analyzed the response of GluClαunedited flies to aversive odors using the multiplex system (67). This single fly assay tracks temporal approaches toward an odor or avoidance from it and hence overcome the inherent limitation of olfactory trap and Y-maze systems that measure behavioral choices of groups in which the choice of individuals may be influenced by those of other flies (Fig. 2H). We used the multiplex system to examine the responses of GluClαunedited flies to two aversive odors: acetic acid and 2-butanone. Examining the time spent avoiding each of the odors showed that while the control (wt) flies had a negative value of close to −100, in agreement with robust avoidance, the average time depicted by GluClαunedited flies was close to zero, suggesting that their movement within the chamber is random (Fig. 2, I and J). These results demonstrate that editing of Ile27 to Val in GluClα is necessary for proper chemotaxis.

Recoding of Ile27 to Val is necessary for the expression of complex olfactory guided behaviors

Given the reduced response to appetitive and aversive odors, we next extended the analysis of GluClαunedited flies to include behaviors that rely on olfactory cues, such as pheromone-based social interactions (68). A central pheromone in this respect is the male-specific pheromone cVA that is known to induce male-male aggression and promote sexual receptivity in female flies (69). Comparing aggression levels between pairs of GluClαunedited flies to those of wt flies revealed a dramatic reduction in the number of lunges exhibited by GluClαunedited flies (Fig. 3A) as well as longer duration until they exhibited first lunge (latency to first lunge, Fig. 3B). The overall reduction in aggressive behavior suggests impaired perception of cVA. These notable results prompt us to examine another cVA-dependent behavior in female flies, where the presence of cVA emitted from courting male flies promotes receptivity (69–71), measured by the time it takes from first courtship action until copulation (latency to copulation). For that, we paired GluClαunedited or wt female flies with wt male flies and assayed their receptivity. First, we verified that wt male flies exhibit similar courtship patterns toward GluClαunedited and wt females and found that the time it took male flies to initiate courtship action was similar toward the two genotypes (Fig. 3C). Next, we measured the time it took from initiation of courtship action by male flies and until the beginning of copulation. GluClαunedited females exhibited significantly longer latency to copulate, which reflects reduced receptivity (Fig. 4D). The reduced receptivity agrees with impaired perception of cVA.

Fig. 3. GluClαunedited flies exhibit impaired olfactory-based behaviors.

(A) GluClαunedited flies exhibit longer duration to first lunge during aggression assay. n = 23, ***P < 0.0001; Mann-Whitney test. (B) GluClαunedited flies exhibit a reduced ratio of lunges. n = 23, ***P < 0.0001; Mann-Whitney test. (C) wt male flies exhibit similar courtship behavior toward wt and GluClα-unedited female flies, manifested by similar latency to first courtship action. n = 50 P > 0.05 Mann-Whitney test. (D) GluClαunedited female flies exhibit reduced receptivity to courtship by wt male flies as measured by the latency to mate. Take a similar amount of time to start courting mutant and wt cs females. n = 50, **P < 0.01; Mann-Whitney test. (E) Schematic illustration of two choice ovipositional and positional preference in egg laying on substrate with acetic acid. (F) Oviposition and position indices of mated GluClαunedited and wt females. GluClα-unedited female exhibit enhanced oviposition index n = 10 (***P < 0.0001 one sample t test) and reduced positional aversion to acetic acid (**P < 0.005, t test, two-tailed), which significantly differ from zero (*P < 0.05, two-tailed t test). (G) GluClαunedited males show reduced positional aversion to substrate containing acetic acid n = 10 (test). (H) Virgin GluClαunedited female exhibit reduced positional aversion to substrate containing acetic acid n = 10 (test).

Fig. 4. Expression of an edited channel in PNs restores proper olfactory responses.

(A) Flies expressing an edited channel in PNs (blue) show increased preference toward apple juice, which is similar in its extent to wt controls (orange), compared to genetic controls (gray tones). n = 10, ***P < 0.0001; one-way analysis of variance (ANOVA). (B) Females expressing an edited channel in PNs exhibit similar receptivity to mate as wt controls, which is significantly higher than that of genetic controls. n = 20 **P < 0.001; one-way ANOVA. Males expressing an edited channel in PNs show shorter latency to first lunge (C), and an increase in the number of lunges (D) compared to the control flies. n = 12, *P < 0.005, ***P < 0.0001; one-way ANOVA.

We then explored the phenotype GluClαunedited female flies in a two-choice egg-laying paradigm, in which females exhibit attraction to food containing acetic acid as an egg-laying substrate, and at the same time show aversion to staying on acetic acid–containing food (positional avoidance) (72). The oviposition attraction to acetic acid is mediated by gustatory neurons and the positional aversion by olfactory cues (72, 73). The trade-off between oviposition preference and positional avoidance is an intriguing system to test the behavior of GluClαunedited female flies. The experimental design consists of a simple apparatus in which mated females are allowed the choice to lay eggs on regular food or on food containing 5% acetic acid, and the number of eggs laid on both sides is counted after 3 hours. During this period, the position of females is monitored every 15 min to calculate positional preference/aversion toward acetic acid–containing substrate (Fig. 3E). As expected, wt female flies show a robust preference to lay eggs on food with acetic acid (positive oviposition preference values), while they avoid staying there (negative positional preference values, Fig. 3F), demonstrating the opposing motivations between the need to lay eggs in a substrate containing acetic acid and their avoidance from its aversive odor. GluClαunedited females, on the other hand, lost the positional aversion to acetic acid and instead showed a slight preference to be on acetic acid food and accordingly showed a significantly higher preference to lay all their eggs on this substrate (Fig. 3F). Moreover, GluClαunedited males and virgin female that lack the motivation to lay eggs, and therefore do not need to stay on acetic acid as a substrate for egg laying, do not show positional aversion to acetic acid, and spend on average equal time on both substrates (Fig. 3, G and H). The results suggest that while the gustatory perception of acetic acid that is responsible for oviposition is normal in GluClαunedited females, their olfactory perception of acetic acid is impaired, pointing to the importance of Ile27 recoding in the processing of olfactory information.

The recoding of Ile27 to Val is necessary in PNs for the proper expression of olfactory-based behaviors

Considering the olfactory-related phenotypes observed in GluClαunedited flies, and the fact that these flies express an unedited form of GluClα in all the cells that normally express GluClα, we next searched for the neurons in which the recoding of Ile27 to Val has a functional relevance. Since GluClα was previously shown to function in olfactory PNs (40), we tested whether expressing a fully edited version of GluClα in olfactory PNs of GluClαunedited flies can rescue the olfactory phenotypes. To this end, we generated GluClαunedited flies harboring the GH146 driver that target ~90 of the 200 PNs in the AL (74) and a UAS transgene that allows the expression of GluClα protein in which the six editing sites are in a fully edited state (the A nucleotide is mutated to G). We assumed that introducing an excess of edited channels can restore the editing level of this particular site to normal levels. Pan neuronal expression of the fully edited channel in GluClαunedited flies resulted in 90% of the GluClα transcripts harboring a G nucleotide (edited version) at this position (fig. S4A).

We next profiled the behavior of GluClαunedited flies expressing a fully edited version of GluClα in PNs using four behavioral paradigms. First, we used the FlyBowl system to make sure that the overexpression of the edited channel did not result in any motor defects (figs. S4B and tables S6 and S7). Next, we examined their attraction to apple juice using an olfactory trap assay. While both genetic controls exhibit no preference for apple juice, the expression of edited GluClα in PNs led to a clear preference toward apple juice (similar in its extent to wt controls), suggesting that the expression edited version can rescue the impairment in olfactory response to appetitive odors (Fig. 4A). Furthermore, GluClαunedited females harboring the edited form in PNs, exhibited a significant increase in their receptivity to courting males, as the latency to copulation was dramatically shorter than the genetic controls and was similar in its extent to wt female flies (Fig. 4B). Expressing the edited form in PNs also restored aggressive display, as the experimental males exhibited shorter duration to the expression of first lunge and higher number of lunges in comparison to genetic controls (Fig. 4, C and D). Together, our results propose that the loss-of-function phenotypes of GluClαunedited flies can be partially restored by the expression of edited version in PNs, suggesting that the recoding of Ile27 to Val has a functional role in PNs to facilitate olfactory responses that are similar in their extent to wt flies.

Recoding of I27V in GluClα shapes odor responses within the antennal lobe

It was previously shown that glutamate functions as an inhibitory transmitter that shapes PNs response to olfactory stimuli (40). Glutamate that is secreted from LNs, binds to GluClα on PNs and inhibits their activity, as knockdown of GluClα in PNs results in increased responses of certain glomeruli to isoamyl acetate (IAA) and methyl salicylate (40). Thus, we first examined the responses of GluClαunedited and wt flies to glutamate. To this end, we used in vivo whole-cell patch clamp recording and measured glutamate-evoked currents in PN. GluClαunedited flies showed a strong and significant reduction in glutamate-induced currents (Fig. 5A). To strengthen the functional role of the edited form of GluClα in PNs, we next explored its effect on neuronal physiology by analyzing the odor-evoked neuronal activity of PNs within the AL. To this end, we expressed GCaMP6m in PNs in GluClαunedited and wt flies and performed two-photon functional Ca2+ imaging in response to IAA and methyl salicylate. We first validated that the GluClαunedited flies harboring either the PN driver or the GCaMP6m exhibit reduced attraction to appetitive odor compared to wt flies that harbor the same transgenes (fig. S5). Next, we compared that odor-evoked activity across all glomeruli and found similar responses to both odors in GluClαunedited and wt PNs (Fig. 5B). The overall similarity in the responses when examining all glomeruli prompted us to analyze the responses of single glomeruli across the AL including DM5, VA1v, VA2, VA3, VC1, VL2a, VL2p, and VM2. While most tested glomeruli exhibited similar responses in GluClαunedited and wt flies, the VA1v glomerulus exhibited significantly increased activity in GluClαunedited flies in response to IAA and methyl salicylate (Fig. 5C). The enhanced responses of VA1v PNs to IAA and methyl salicylate in unedited flies suggest that the recoding event of Ile27 in GluClα is necessary to dampen the activity of the PNs during the perception of these odors. We also observed a small reduction in the responses to IAA in the case of the VM2 glomerulus. These results support our behavioral finding, all together indicating that the recoding of Ile27 is required for correct function of AL neurons in regulating olfactory-based behavioral responses.

Fig. 5. GluClα-editing state affects PN odor responses.

(A) Left, inhibitory currents evoked by the application of 1 mM glutamate (arrowhead) in PNs in wt (purple) and GluClαunedited (green). Data are means (solid line) ± SEM (shaded area). Right, analysis of maximal current of traces on the left. Each dot represents measurement from a single fly. **P < 0.01, t test, n = 11 and 10 for wt and GluClαunedited, respectively. (B) ∆F/F of GCaMP6f signal in the entire AL in control GluClα (purple) and unedited GluClα (green) flies during presentation of odor pulses (isoamyl acetate, IAA, and methyl salicylate, horizontal lines). Data are means (solid line) ± SEM (shaded area). No significant difference in odor responses was observed between flies with GluClαunedited background to the control flies with wt background. Paired t test, n = 5. (C) ∆F/F of GCaMP6f signals in eight different glomeruli to IAA and methyl salicylate in GluClαunedited flies (green) compared to wt flies (purple). A significant increase in odor response was observed in VA1v glomerulus for the two tested odors. *P value < 0.05; t test, n = 6.

DISCUSSION

A-to-I RNA editing has long been suggested as cellular machinery that can provide the proteomic diversity required for the intricate function of the nervous system by shaping the spatial and temporal repertoire of expressed protein isoforms expressed in neurons (10, 75–78). Here, we dissected the physiological importance of an editing site in Drosophila and showed that a highly abundant and evolutionarily conserved recoding event that occurs within the extracellular domain of GluClα is necessary for proper olfactory responses in adult flies. First, by ablating the editing of Ile27 to Val such that the endogenous channel harbors only the unedited version (Ile27 instead of Val), we were able to examine the contribution of this particular recoding event to its function in behaving animals. Second, by using a set of behavioral assays, we found that the edited isoform is necessary for proper perception of odors, as well as for the expression of pheromone-based social behaviors. Furthermore, we mapped the spatial requirement of the edited isoform to olfactory PNs. The behavioral phenotypes observed in the unedited flies do not stem from impaired health or motor defects as they exhibit intact motor activity and even extended lifespan. The last is in agreement with findings in Caenorhabditis elegans where mutations that cause defects in sensory neurons lead to extended life span (79, 80). Moreover, the behavioral phenotypes of the unedited flies could be rescued by the expression of an edited channel in PNs, resulting in olfactory performance that is similar to that of wt flies.

Consistent with a previous study (40), our results indicate that the unedited isoform of GluClα has reduced glutamate-evoked currents and affects odor responses in VA1v glomerulus. Considering results from previous studies that found that silencing one or a few glomeruli had no effect on behavioral output (81), it is unclear how the lack of GluClα editing in PNs has such a pronounced effect on behavior. The affected glomerulus (VA1v) is sexually dimorphic (82–84) receiving input from Or47b neurons known to respond to conspecific odors. PNs responding to food odors are known to be less specific than their cognate ORNs (85, 86). In GluClαunedited flies, the VA1v glomerulus loses its specificity and responds to non-pheromone odors, suggesting that under normal conditions, glutamatergic inhibition is required to maintain VA1v specificity. It is therefore possible that the loss of odor valence observed in unedited flies results from improper activation of pheromonal pathways during the perception of non-pheromonal odor signals, leading together to contradicting consequences. Whether this is indeed the case is out of the scope of this manuscript. Furthermore, it would be intriguing to examine in future work whether GluClα has a role in controlling the activity in other pheromone related glomeruli.

The current case is not the first recoding event that is necessary for the proper function of a channel. The ablation of an editing site located within the mammalian excitatory ionotropic glutamate receptor subunit B leads to early seizure-related death in mice (87). While the ablation of Ile27 recoding in GluClα did not reduce viability or reproduction of flies under laboratory conditions, the impaired olfactory responses exhibited by unedited flies are expected to affect their fitness in natural environments, since flies heavily rely on their olfactory system to survive and reproduce.

The loss-of-function phenotype of the unedited form indicates that recoding of Ile27 to Val is necessary for the proper function of the channel, and the combination of high editing levels and evolutionary conservation are useful criteria when predicting which of the many editing events has the potential of having a function. This brings up the question of why evolution chose to maintain modulation of the sequence at the RNA level rather than inserting a mutation at the DNA level. The answer probably lies in the varying levels of its editing between different neuronal populations, suggesting that the ratio between edited and unedited forms of the channel or within subunits that construct the same channel may be spatially regulated to fine-tune its function. A relevant example in this respect is the spatial and temporal regulation of a recoding event that is found within the calmodulin-binding IQ domain of the voltage-activated calcium channel (Cav1.3) and how ablation enhances learning (16, 17). Further studies are required to test whether the editing repertoire of GluClα is regulated in response to different environmental conditions.

It is intriguing that a recoding event that is conservative in its nature (Ile to Val) and that is located close to the N terminal tip of the channel (five amino acids downstream to the putative signal peptide) has such a strong impact on function. As of this moment, there is no resolved 3D structure of Drosophila GluClα, and 3D structures of other Cys-loop receptor-channels lacked a portion of the N terminal segment that contains this particular residue, making it difficult to assess the molecular function of Ile27 recoding to Val. Still, there are notable examples of the way by which a similar recoding event (Ile400 to Val) that is located at the intracellular cavity of the potassium channel Kv1.1 strongly modifies the inactivation kinetics, where the edited form of the channel recovers from inactivation 20 times faster than the unedited form (88). The faster recovery is achieved because of reduced hydrophobic interaction between the tip of the inactivation gate and the Val residue that is found within the intracellular cavity (88). While the Ile27-containing segment of the Drosophila GluClα subunit is missing in the various 3D structures of anionic Cys-loop receptors (fig. S1), its location is clearly a few amino acids upstream to the N-terminal α helix (α1) that is typical of Cys-loop receptors (fig. S1). Helix α1 lies in close proximity to a conserved amino acid cluster that belongs to the adjacent subunit (e.g., Fig. 6, A to C; for cluster sequence, see fig. S1). This close proximity allows for van der Waals interactions (89) (Fig. 6C), and in some cases hydrogen bonding (90) [e.g., PDB 6PLR (90)], between entities of α1-helix and amino acids belonging to the aforementioned cluster in the adjacent subunit. One may therefore envision that in the Drosophila GluClα receptor the N terminus and the editing site itself might be close, at an interaction distance, to the neighboring subunit. Since inter-subunit interactions often take part in the function of Cys-loop receptors, further studies are necessary to assess how the recoding of this position affects the biophysical properties of the Drosophila GluClα receptor.

Fig. 6. Structural characteristics of GluClα receptors.

(A) Top view of the Caenorhabditis elegans GluClα receptor (GluClαcrystR; Protein Data Bank ID code 3RIF) showing five identical subunits, which are colored differently to highlight the intersubunit interfaces. Alpha-helix α1 is colored in pink in all five subunits. (B) Two adjacent subunits (colored in gray and yellow) of the GluClαcrystR are viewed from the side. The membrane region is illustrated as a gray rectangle interrupted by the transmembrane helices (OUT, extracellular; IN, intracellular). The N-terminal tip of a-helix α1 (pink), which belongs to the yellow-colored subunit, interacts with the adjacent subunit (gray spheres), as detailed in (C). The neurotransmitter glutamate (purple spheres) is bound at the intersubunit interface. (C) Space-filling models of the side chains of SER3 (belongs to Helix a1) and VAL18 (belongs to the conserved cluster shown in fig. S1) indicate van der Waals interactions between these two amino acids (3.9-Å center-to-center distance between Oγ of SER3 and Cγ2 of VAL18). Side chains of the other amino acids belonging to the conserved cluster are shown as spheres or sticks. Carbon, oxygen, nitrogen, and hydrogen atoms are colored in pink or gray, red, blue, and white, respectively. Glutamate is shown in purple spheres.

MATERIALS AND METHODS

Analyzing evolutionary conservation of GluClα editing sites across Drosophila species

The analysis is based on an existing dataset from Ramaswami et al. (91) that quantified RNA editing at 605 loci using a multiplex microfluidic polymerase chain reaction (PCR) with deep sequencing (mmPCR-seq) across 131 Drosophila strains (n = 2 per strain). The study used multiplex PCR primers to amplify 605 loci that include many known RNA editing events. PCR products of each sample were then subjected to a 15-cycle barcode PCR and pooled together. The library was sequenced using Illumina HiSeq with 101–base pair (bp) paired-end reads. Paired-end reads were combined and mapped onto the genome (dm3) using BWA samse allowing nine mismatches per read. The sequencing reads were aligned to a combination of the reference genome and 100-bp exonic sequences surrounding known splicing junctions from available gene models (obtained from the UCSC genome browser). We quantified editing levels of known D. melanogaster RNA editing sites by taking the fraction of reads containing a “G” nucleotide at that position. For editing level quantification, sites covered by ≥50 mmPCR-seq reads were used. For each strain, we excluded editing sites where the measured editing levels in the two biological replicates differed by >20% (see Fig. 1B). Custom scripts used to process data are available upon request. In this study, we compared the relative editing levels of six editing sites in GluClα.

Analysis of GluClα editing levels across the fly brain

We used an existing dataset (47) and used the following analysis pipeline: STAR (v2.4.2) (1) (--twopassMode Basic) was used to map paired-end mmPCR-seq reads and single-end RNA sequencing reads to the dm6 genome as described above. We then used the Samtools mpileup function to determine base calls from uniquely mapped reads at known and previously unidentified editing sites and calculated editing levels as number of G reads divided by the total of both A and G reads at a site. For mmPCR-seq, we required each replicate to have 100× coverage and we removed sites that were not within 20% editing between replicates, as done previously (91). Final mmPCR editing levels were determined after down sampling coverage to 200 reads for statistical analysis.

Fly lines

All D. melanogaster fly lines used in this study were kept at 25°C, ~50% humidity, light/dark of 12:12 hours, and maintained on cornmeal, yeast, molasses, and agar medium. Most fly lines were backcrossed to a Canton S background or, if mentioned, were maintained on w1118 background. wt GluClα and unedited GluClα were used for behavioral assays. The genotype of GluClα-unedited flies was validated in each experiment using PCR analysis (see below). GH146-Gal4 > UAS GCaMP6 harboring wt or unedited version of GluClα were used for the Ca2+ imaging experiments. UAS edited GluClα flies were a generous gift from L. Keegan.

Generation of GluClα-unedited flies

Ablation of Ile27 editing site was achieved by deletion of the of intronic ECS region: chr3R: 15578744-15579171 using CRISPR-Cas9 and the following gRNAs to induce double-stranded cuts to the DNA, on each side of the ECS: TCTAAACCCTAGATATACGCTGG and TATAGTATGTGACTTTGCCTGGG. In addition, we made synonymous mutations to the sgRNA target region or protospacer adjacent motif to prevent recutting by the Cas9-CRISPR system. The genotype of the GluClα-unedited flies was validated by performing simple PCR analysis (fig. S2A) using the following primers: Forward: TTGTCTCCCGCTCCACTTAC. Reverse: TTGGGCAATTTGAAAGTCGAAA.

Western blot analysis

Relative levels of GluClα was analyzed using Western blot analysis (20 heads) and anti-GluClα antibodies produced in rabbit (1:1000) and anti-actin as a loading control. Densitometric analysis was used to compare relative protein levels of GluClα unedited and wt channels. GluClα protein levels were normalized according to actin levels.

Isolation of synaptosomes and postsynaptic densities

A protocol for the isolation of synaptosomes and postsynaptic densities as described by Rajkimar et al. (54) was adapted to using fly heads as a source for protein using the changes. wt and GluClα-unedited flies were flash frozen. Nine hundred heads of each were homogenized in 1 ml of buffer A (40 strikes in dounce homogenizer). The resulting postsynaptic density fraction was dissolved in 40 μl of buffer C. Five hundred and eighty-four micrograms of protein was separated in a 10% polyacrylamide gel and stained using Comassie staining. All bands at the size of 45 to 60 KDa were sent for MS analysis.

In-gel proteolysis and MS analysis

The proteins in the gel slices were reduced with 3 mM dithiothreitol (60°C for 30 min), modified with 10 mM iodoacetamide in 100 mM ammonium bicarbonate (in the dark, room temperature for 30 min), and digested in 10% acetonitrile and 10 mM ammonium bicarbonate with modified trypsin (Promega) at a 1:10 enzyme-to-substrate ratio, overnight at 37°C. The tryptic peptides were desalted using C18 tips (Homemade stage tips), dried, and resuspended in 0.1% formic acid. The tryptic peptides were desalted using C18 tips (Top tip, Glygen), dried, and resuspended in 0.1% formic acid. The peptides were resolved by reverse-phase chromatography on 0.075 × 180-mm fused silica capillaries (J&W) packed with Reprosil reversed-phase material (Dr Maisch GmbH, Germany). The peptides were eluted with different concentration of acetonitrile with 0.1% of formic acid: a linear 60-min gradient of 6 to 34% of 80% acetonitrile followed by a 15-min gradient of 34 to 95% and 15 min at 95% of 80% acetonitrile with 0.1% formic acid in water at flow rates of 0.15 μl/min. MS was performed by Q Exactive HF mass spectrometer (Thermo Fisher Scientific) in a positive mode (m/z 300 to 1500, resolution 60,000 for MS1 and 15,000 for MS2) using repetitively full MS scan followed by higher-energy collisional dissociation (HCD; at 27 normalized collision energy) of the 18 most dominant ions (>1 charges) selected from the first MS scan. A dynamic exclusion list was enabled with exclusion duration of 20 s. The MS data were analyzed using the MaxQuant software 2.1.3.0 (92) for peak picking and identification using the Andromeda search engine, searching against the D. melanogaster proteome from the UniProt database with mass tolerance of 6 parts per million (ppm) for the precursor masses and 20 ppm for the fragment ions. Oxidation on methionine and protein N terminus acetylation were accepted as variable modifications, and carbamidomethyl on cysteine was accepted as static modifications. Minimal peptide length was set to six amino acids, and a maximum of two miscleavages was allowed. The data were quantified by label-free analysis using the same software. Peptide- and protein-level false discovery rates (FDRs) were filtered to 1% using the target-decoy strategy. Protein table were filtered to eliminate the identifications from the reverse database and common contaminants and single peptide identifications.

Life span

wt and GluClα-unedited flies (100 of each) were collected upon eclosion and placed in single food vials. The number of live flies was recorded 3 days. Flies were transferred to fresh vials once a week. Log-rank test (REF) with FDR correction was performed to compare the survival curves.

Determining relative editing levels using Sanger sequencing

GluClα-unedited flies expressing a fully edited channel in all neurons (Elav>GluClα edited) and genetic controls were flash frozen (n = 3 independent repeats). Total RNA was extracted from heads using TRIzol reagent (Ambion). One microgram of RNA was converted to cDNA using ProtoScript II reverse transcriptase, and sequences around I27V in GluClα transcripts were amplified using PCR (40 cycles) using the following primers: Forward primer: TCTTATACTTTGCCAGCCTGT. Reverse primer: TGTTCACGGAAGGTTAACTGC. PCR products were cleaned using EPPiC Fast and sent for Sanger sequencing.

Behavioral experiments

All behavioral experiments were performed about an hour after the lights on.

FlyBowl

Male flies were collected upon eclosion and aged in groups of 10 flies per vial for 4 days before test. Ten flies of each genotype were inserted in groups of 10 into Fly Bowl arenas (93), and their behavior was recorded for 30 min and analyzed using CTRAX, FixTrax (94), and JAABA (93). For kinetic features, scripts were written in MATLAB to use the JAABA code to generate the statistical features as specified in Kabra et al. (93). Quantification of complex behaviors was done using JAABA Classifiers (93) to identify specific behaviors: walk, stop, turn, approach, touch, chase, chain, song, social clustering, and grooming. Each feature of the Fly Bowl experiment was standardized according to all values calculated in our experiments for that feature to generate a z-score. Scatter plots were created using R.

Olfactory trap assay

Traps consisting of an Eppendorf tube connected to a 200-μl pipette tip were fitted to the trap to prevent flies from being able to escape the odor trap after entering it. Each trap is filled with 500 μl of a 1% agar supplemented with 10% odor apple juice (66), ethanol, or water as a control. Each setup contains two traps (with odor and control) that are glued to a 90 mm–by–15 mm petri dish. The two traps are placed in opposite directions. Fifty 3-day-old flies are introduced into each such experimental system. The flies can be attracted to one of the traps based on smell alone since they cannot make physical contact with the substrate to make the decision. To increase the motivation to choose one of the traps, the test was done in the absence of food. The experimental system is wrapped in aluminum foil to prevent the penetration of light that could disrupt their decision to enter one of the traps based solely on smell. The number of flies in each trap is counted after 24 hours to calculate preference index: preference index (PI) = (# of flies that entered the trap with the tested odor − # of flies that entered the trap that does not contain an odor source)/(total # of flies).

Immediate odor preference using Y maze

The behavioral setup is composed of two vials containing agar with and without odors that are connected to the Y-shaped adapter. A vial containing 50 3-day-old flies is placed at the entrance of the adapter, and the flies are allowed to choose one of the arms. After 40 s, the upper vials are closed and the number of flies in each is counted. Preference index is calculated as follows: PI = (# of flies in test arm − # of flies in control arm)/(total # of flies).

Single-fly assay for odor preference

Behavioral experiments were performed in a custom-built, fully automated apparatus as described in detail (67). Single flies were introduced into clear polycarbonate chambers (length, 50 mm; width, 5 mm; height, 1.3 mm) and were exposed to air containing odor flow. Odors were prepared at 10-fold dilution in mineral oil. Fresh odors were prepared daily. Air or odor streams from the two halves of the chamber converged at a central choice zone. The location of the flies was recorded and tracked. A fly’s preference was calculated as the percentage of time that it spent on one side of the chamber.

Two-choice oviposition and positional assays

The two-choice egg-laying setup contains a standard fly bottle, with the base cut off and replaced with a 60-mm petri dish lid. Half of the petri dish lid contains solid food substrate with 5% acetic acid, and the other half contains food supplemented with equal volume of water. Two-choice dishes were made by dividing a 35-mm petri dish lid with a razor blade and pouring two samples of food substrate into each half. For each test, 15 to 20 recently mated females were introduced into each two-choice apparatus allowed to sample and lay eggs for 3 hours. Oviposition preference was determined by counting the number of eggs on each half of the two-choice dish (oviposition index = # of eggs laid on acetic acid–containing food − # of eggs laid on control food/# of total eggs laid). For positional preference, the number of flies on each half of the dish was counted every 15 min along the entire duration of the experiment (3 hours) (position index = average # of flies on acetic acid half − average # of flies on control food half/total # of flies).

Courtship and mating tests

Male flies were collected upon eclosion and aged in groups of 10 flies per vial for 4 days before test. Courtship arenas were placed in behavior chambers, under controlled temperature and humidity (25°C, 70% humidity). Behavior was recorded for 1 hour from the introduction of male and female pairs using Point-Grey Flea3 cameras (1080 × 720 pixels at 30 fps). Latency to copulate was quantified for each pair as total time, starting from the first wing vibration the male exhibited and ending in successful copulation.

Aggression

Four-day-old pairs of single-housed male flies (wt, GluClα, or when indicated GH146 > UAS edited GluClα and genetic controls) were put into round aggression arenas (about 0.08 cm3 in volume). A mixture of agarose and apple juice (1% agarose, 50% apple juice) was inserted into arenas to enhance aggressive behavior. Experiments were performed at a similar time of day (lights ON +1 hour). The flies’ behavior was recorded for 30 min with Point-Grey Flea3 (1080 × 720 pixels at 60 fps). Aggressive behavior was later quantified by counting the number of lunges for each pair and latency as the time from start of experiment to first lunge for each pair.

Functional imaging

Flies used for functional imaging were raised on cornmeal agar under a 12-hour light/12-hour dark cycle at 25°C. Imaging was done as previously described (95–99). Briefly, the flies were anesthetized on ice, and then a single fly was moved to a custom-built chamber and fixed to aluminum foil using wax. Cuticle and trachea in the required area were removed, and the exposed brain was superfused with carbonated solution as described above. Odors (purest level available) were obtained from Sigma-Aldrich (Rehovot, Israel). Odor flow of 0.4 l/min (10 to 1 dilution) was combined with a carrier air stream of 0.4 l/min using mass-flow controllers (Sensirion) and software-controlled solenoid valves (The Lee Company). This resulted in a final odor dilution of 5 × 10−2 delivered to the fly. Odor flow was delivered through a 1/16-inch ultrachemical-resistant Versilon PVC tubing (Saint-Gobain, NJ, USA) placed 5 mm from the fly’s antenna. Functional imaging was performed using a two-photon laser-scanning microscope (DF-Scope installed on an Olympus BX51WI microscope). Fluorescence was excited by a Ti-Sapphire laser (Mai Tai HP DS, 100-fs pulses) centered at 910 nm, attenuated by a Pockels cell (Conoptics), and coupled to a galvo-resonant scanner. Excitation light was focused by a 20×, 1.0–numerical aperture objective (Olympus XLUMPLFLN20XW), and emitted photons were detected by GaAsP photomultiplier tubes (Hamamatsu Photonics, H10770PA-40SEL), whose currents were amplified (Hamamatsu HC-130-INV) and transferred to the imaging computer (MScan 2.3.01). All imaging experiments were acquired at 30 Hz.

Glutamate administration

Glutamate (Sigma-Aldrich, G1251) solution was prepared and diluted in external solution to the final concentration of 1 mM. A glass pipette filled with glutamate solution was placed in close proximity to the AL and was emptied using a pico injector (Harvard Apparatus, PLI-100).

Acknowledgments

We thank all members of the Shohat-Ophir laboratory for fruitful discussions and M. Noy Meshi for technical support. We thank D. Jacobson, N. Nguyen, and C. Yan for assistance with the generation of GluClα-unedited fly line. We would also like to thank the Smoler Proteomics Center (Technion, Israel) for proteomic analysis and J. I. C. Benichou for statistical consultation.

Funding: This work was supported by the U.S.-Israel Binational Science Foundation (BSF, 2019091, G.S.-O.; 2019026, M.P.; and 2020636, M.P.), the Israel Science Foundation (ISF, 174/19, G.S.-O. and 404/23, M.P.), the European Research Council (ERC, 101085605, M.P.), and the NIH (R35 GM144100, J.B.L.).

Author contributions: H.Z., G.S.-O., J.B.L., and M.P. designed research; H.Z., P.D., H.P., S.I., M.L., and E.R. performed research; D.G. and Y.P. performed computational analysis; H.Z. and E.R. analyzed data; and H.Z., M.P., Y.P., and G.S.-O. wrote the paper.

Competing interests: The authors declare that they have no competing interests.

Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.

Supplementary Materials

The PDF file includes:

Figs. S1 to S5

Legends for tables S1 to S6

Table S7

Other Supplementary Material for this manuscript includes the following:

Tables S1 to S6
==== Refs
REFERENCES AND NOTES

1 E. Eisenberg, E. Y. Levanon, A-to-I RNA editing–immune protector and transcriptome diversifier. Nat. Rev. Genet. 19 , 473–490 (2018).29692414
2 B. L. Bass, RNA editing by adenosine deaminases that act on RNA. Annu. Rev. Biochem. 71 , 817–846 (2002).12045112
3 S. H. Roth, E. Y. Levanon, E. Eisenberg, Genome-wide quantification of ADAR adenosine-to-inosine RNA editing activity. Nat. Methods 16 , 1131–1138 (2019).31636457
4 P. Reautschnig, N. Wahn, J. Wettengel, A. E. Schulz, N. Latifi, P. Vogel, T.-W. Kang, L. S. Pfeiffer, C. Zarges, U. Naumann, L. Zender, J. B. Li, T. Stafforst, CLUSTER guide RNAs enable precise and efficient RNA editing with endogenous ADAR enzymes in vivo. Nat. Biotechnol. 40 , 759–768 (2022).34980913
5 C. X. George, G. Ramaswami, J. B. Li, C. E. Samuel, Editing of cellular Self-RNAs by adenosine deaminase ADAR1 suppresses innate immune stress responses. J. Biol. Chem. 291 , 6158–6168 (2016).26817845
6 B. J. Liddicoat, R. Piskol, A. M. Chalk, G. Ramaswami, M. Higuchi, J. C. Hartner, J. B. Li, P. H. Seeburg, C. R. Walkley, RNA editing by ADAR1 prevents MDA5 sensing of endogenous dsRNA as nonself. Science 349 , 1115–1120 (2015).26275108
7 Y. Duan, S. Dou, S. Luo, H. Zhang, J. Lu, Adaptation of A-to-I RNA editing in Drosophila. PLOS Genet. 13 , e1006648 (2017).28282384
8 X. Li, I. M. Overton, R. A. Baines, L. P. Keegan, M. A. O’Connell, The ADAR RNA editing enzyme controls neuronal excitability in Drosophila melanogaster. Nucleic Acids Res. 42 , 1139–1151 (2014).24137011
9 L. Grassi, G. Leoni, A. Tramontano, RNA editing differently affects protein-coding genes in D. melanogaster and H. sapiens. Sci. Rep. 5 , 11550 (2015).26169954
10 S. Garrett, J. J. C. Rosenthal, RNA editing underlies temperature adaptation in K+ channels from polar octopuses. Science 335 , 848–851 (2012).22223739
11 M. Stapleton, J. W. Carlson, S. E. Celniker, RNA editing in Drosophila melanogaster: New targets and functional consequences. RNA 12 , 1922–1932 (2006).17018572
12 M. Y. Ryan, R. Maloney, R. Reenan, R. Horn, Characterization of five RNA editing sites in Shab potassium channels. Channels 2 , 202–209 (2008).18836299
13 M. Higuchi, S. Maas, F. N. Single, J. Hartner, A. Rozov, N. Burnashev, D. Feldmeyer, R. Sprengel, P. H. Seeburg, Point mutation in an AMPA receptor gene rescues lethality in mice deficient in the RNA-editing enzyme ADAR2. Nature 406 , 78–81 (2000).10894545
14 I. Gaisler-Salomon, E. Kravitz, Y. Feiler, M. Safran, A. Biegon, N. Amariglio, G. Rechavi, Hippocampus-specific deficiency in RNA editing of GluA2 in Alzheimer’s disease. Neurobiol. Aging 35 , 1785–1791 (2014).24679603
15 M. T. Venø, J. B. Bramsen, C. Bendixen, F. Panitz, I. E. Holm, M. Öhman, J. Kjems, Spatio-temporal regulation of ADAR editing during development in porcine neural tissues. RNA Biol. 9 , 1054–1065 (2012).22858680
16 H. Huang, B. Z. Tan, Y. Shen, J. Tao, F. Jiang, Y. Y. Sung, C. K. Ng, M. Raida, G. Köhr, M. Higuchi, H. Fatemi-Shariatpanahi, B. Harden, D. T. Yue, T. W. Soong, RNA editing of the IQ domain in Cav1.3 channels modulates their Ca2+-dependent inactivation. Neuron 73 , 304–316 (2012).22284185
17 J. Zhai, S. Navakkode, S. Q. Z. Yeow, K. Krishna-K, M. C. Liang, J. H. Koh, R. X. Wong, W. P. Yu, S. Sajikumar, H. Huang, T. W. Soong, Loss of Ca1.3 RNA editing enhances mouse hippocampal plasticity, learning, and memory. Proc. Natl. Acad. Sci. U.S.A. 119 , e2203883119 (2022).35914168
18 L. Bazak, A. Haviv, M. Barak, J. Jacob-Hirsch, P. Deng, R. Zhang, F. J. Isaacs, G. Rechavi, J. B. Li, E. Eisenberg, E. Y. Levanon, A-to-I RNA editing occurs at over a hundred million genomic sites, located in a majority of human genes. Genome Res. 24 , 365–376 (2014).24347612
19 Z. Peng, Y. Cheng, B. C.-M. Tan, L. Kang, Z. Tian, Y. Zhu, W. Zhang, Y. Liang, X. Hu, X. Tan, J. Guo, Z. Dong, Y. Liang, L. Bao, J. Wang, Comprehensive analysis of RNA-Seq data reveals extensive RNA editing in a human transcriptome. Nat. Biotechnol. 30 , 253–260 (2012).22327324
20 G. Ramaswami, W. Lin, R. Piskol, M. H. Tan, C. Davis, J. B. Li, Accurate identification of human Alu and non-Alu RNA editing sites. Nat. Methods 9 , 579–581 (2012).22484847
21 H. T. Porath, A. A. Schaffer, P. Kaniewska, S. Alon, E. Eisenberg, J. Rosenthal, E. Y. Levanon, O. Levy, A-to-I RNA editing in the earliest-diverging eumetazoan phyla. Mol. Biol. Evol. 34 , 1890–1901 (2017).28453786
22 G. S. Laurent, M. R. Tackett, S. Nechkin, D. Shtokalo, D. Antonets, Y. A. Savva, R. Maloney, P. Kapranov, C. E. Lawrence, R. A. Reenan, Genome-wide analysis of A-to-I RNA editing by single-molecule sequencing in Drosophila. Nat. Struct. Mol. Biol. 20 , 1333–1339 (2013).24077224
23 B. R. Graveley, A. N. Brooks, J. W. Carlson, M. O. Duff, J. M. Landolin, L. Yang, C. G. Artieri, M. J. van Baren, N. Boley, B. W. Booth, J. B. Brown, L. Cherbas, C. A. Davis, A. Dobin, R. Li, W. Lin, J. H. Malone, N. R. Mattiuzzo, D. Miller, D. Sturgill, B. B. Tuch, C. Zaleski, D. Zhang, M. Blanchette, S. Dudoit, B. Eads, R. E. Green, A. Hammonds, L. Jiang, P. Kapranov, L. Langton, N. Perrimon, J. E. Sandler, K. H. Wan, A. Willingham, Y. Zhang, Y. Zou, J. Andrews, P. J. Bickel, S. E. Brenner, M. R. Brent, P. Cherbas, T. R. Gingeras, R. A. Hoskins, T. C. Kaufman, B. Oliver, S. E. Celniker, The developmental transcriptome of Drosophila melanogaster. Nature 471 , 473–479 (2011).21179090
24 K. Sinigaglia, D. Wiatrek, A. Khan, D. Michalik, N. Sambrani, J. Sedmík, D. Vukić, M. A. O’Connell, L. P. Keegan, ADAR RNA editing in innate immune response phasing, in circadian clocks and in sleep. Biochim. Biophys. Acta Gene Regul. Mech. 1862 , 356–369 (2019).30391332
25 P. Deng, A. Khan, D. Jacobson, N. Sambrani, L. McGurk, X. Li, A. Jayasree, J. Hejatko, G. Shohat-Ophir, M. A. O’Connell, J. B. Li, L. P. Keegan, Adar RNA editing-dependent and -independent effects are required for brain and innate immune functions in Drosophila. Nat. Commun. 11 , 1580 (2020).32221286
26 M. J. Palladino, L. P. Keegan, M. A. O’Connell, R. A. Reenan, A-to-I pre-mRNA editing in Drosophila is primarily involved in adult nervous system function and integrity. Cell 102 , 437–449 (2000).10966106
27 Y. A. Savva, J. E. C. Jepson, A. Sahin, A. U. Sugden, J. S. Dorsky, L. Alpert, C. Lawrence, R. A. Reenan, Auto-regulatory RNA editing fine-tunes mRNA re-coding and complex behaviour in Drosophila. Nat. Commun. 3 , 790 (2012).22531175
28 J. E. C. Jepson, Y. A. Savva, C. Yokose, A. U. Sugden, A. Sahin, R. A. Reenan, Engineered alterations in RNA editing modulate complex behavior in Drosophila: Regulatory diversity of adenosine deaminase acting on RNA (ADAR) targets. J. Biol. Chem. 286 , 8325–8337 (2011).21078670
29 J. E. Jepson, R. A. Reenan, Unraveling pleiotropic functions of A-to-I RNA editing in Drosophila. Fly 4 , 154–158 (2010).20215872
30 L. P. Keegan, J. Brindle, A. Gallo, A. Leroy, R. A. Reenan, M. A. O’Connell, Tuning of RNA editing by ADAR is required in Drosophila. EMBO J. 24 , 2183–2193 (2005).15920480
31 J. E. C. Jepson, R. A. Reenan, Adenosine-to-inosine genetic recoding is required in the adult stage nervous system for coordinated behavior in Drosophila. J. Biol. Chem. 284 , 31391–31400 (2009).19759011
32 J. Ryvkin, A. Bentzur, A. Shmueli, M. Tannenbaum, O. Shallom, S. Dokarker, J. I. C. Benichou, M. Levi, G. Shohat-Ophir, Transcriptome analysis of NPFR neurons reveals a connection between Proteome diversity and social behavior. Front. Behav. Neurosci. 15 , 628662 (2021).33867948
33 L. Ingleby, R. Maloney, J. Jepson, R. Horn, R. Reenan, Regulated RNA editing and functional epistasis in Shaker potassium channels. J. Gen. Physiol. 133 , 17–27 (2009).19114634
34 A. K. Jones, S. D. Buckingham, M. Papadaki, M. Yokota, B. M. Sattelle, K. Matsuda, D. B. Sattelle, Splice-variant- and stage-specific RNA editing of the Drosophila GABA receptor modulates agonist potency. J. Neurosci. 29 , 4287–4292 (2009).19339622
35 R. O. Olson, Z. Liu, Y. Nomura, W. Song, K. Dong, Molecular and functional characterization of voltage-gated sodium channel variants from Drosophila melanogaster. Insect Biochem. Mol. Biol. 38 , 604–610 (2008).18405837
36 M. Y. Ryan, R. Maloney, J. D. Fineberg, R. A. Reenan, R. Horn, RNA editing in eag potassium channels: Biophysical consequences of editing a conserved S6 residue. Channels 6 , 443–452 (2012).23064203
37 A. L. Yablonovitch, J. Fu, K. Li, S. Mahato, L. Kang, E. Rashkovetsky, A. B. Korol, H. Tang, P. Michalak, A. C. Zelhof, E. Nevo, J. B. Li, Regulation of gene expression and RNA editing in Drosophila adapting to divergent microclimates. Nat. Commun. 8 , 1570 (2017).29146998
38 D. F. Cully, P. S. Paress, K. K. Liu, J. M. Schaeffer, J. P. Arena, Identification of a Drosophila melanogaster glutamate-gated chloride channel sensitive to the antiparasitic agent avermectin. J. Biol. Chem. 271 , 20187–20191 (1996).8702744
39 A. Etter, D. F. Cully, J. M. Schaeffer, K. K. Liu, J. P. Arena, An amino acid substitution in the pore region of a glutamate-gated chloride channel enables the coupling of ligand binding to channel gating. J. Biol. Chem. 271 , 16035–16039 (1996).8663156
40 W. W. Liu, R. I. Wilson, Glutamate is an inhibitory neurotransmitter in the Drosophila olfactory system. Proc. Natl. Acad. Sci. U.S.A. 110 , 10294–10299 (2013).23729809
41 A. J. Wolstenholme, Glutamate-gated chloride channels. J. Biol. Chem. 287 , P40232–P40238 (2012).
42 S. Kondo, T. Takahashi, N. Yamagata, Y. Imanishi, H. Katow, S. Hiramatsu, K. Lynn, A. Abe, A. Kumaraswamy, H. Tanimoto, Neurochemical organization of the Drosophila brain visualized by endogenously tagged neurotransmitter receptors. Cell Rep. 30 , 284–297.e5 (2020).31914394
43 X. Li, C. Chien, Y. Han, Z. Sun, X. Chen, D. Dickman, Autocrine inhibition by a glutamate-gated chloride channel mediates presynaptic homeostatic depression. Sci. Adv. 7 , eabj1215 (2021).34851664
44 L. N. Groschner, J. G. Malis, B. Zuidinga, A. Borst, A biophysical account of multiplication by a single neuron. Nature 603 , 119–123 (2022).35197635
45 Y. Li, P.-J. Chen, T.-Y. Lin, C.-Y. Ting, P. Muthuirulan, R. Pursley, M. Ilić, P. Pirih, M. S. Drews, K. P. Menon, K. G. Zinn, T. Pohida, A. Borst, C.-H. Lee, Neural mechanism of spatio-chromatic opponency in the Drosophila amacrine neurons. Curr. Biol. 31 , 3040–3052.e9 (2021).34033749
46 R. Zhang, P. Deng, D. Jacobson, J. B. Li, Evolutionary analysis reveals regulatory and functional landscape of coding and non-coding RNA editing. PLOS Genet. 13 , e1006563 (2017).28166241
47 A. L. Sapiro, A. Shmueli, G. L. Henry, Q. Li, T. Shalit, O. Yaron, Y. Paas, J. Billy Li, G. Shohat-Ophir, Illuminating spatial A-to-I RNA editing signatures within the Drosophila brain. Proc. Natl. Acad. Sci. U.S.A. 116 , 2318–2327 (2019).30659150
48 F. Teufel, J. J. Almagro Armenteros, A. R. Johansen, M. H. Gíslason, S. I. Pihl, K. D. Tsirigos, O. Winther, S. Brunak, G. von Heijne, H. Nielsen, SignalP 6.0 predicts all five types of signal peptides using protein language models. Nat. Biotechnol. 40 , 1023–1025 (2022).34980915
49 D. T. Jones, Protein secondary structure prediction based on position-specific scoring matrices. J. Mol. Biol. 292 , 195–202 (1999).10493868
50 T. Nugent, D. T. Jones, Transmembrane protein topology prediction using support vector machines. BMC Bioinformatics 10 , 159 (2009).19470175
51 D. W. A. Buchan, D. T. Jones, The PSIPRED protein analysis workbench: 20 years on. Nucleic Acids Res. 47 , W402–W407 (2019).31251384
52 A. Drozdetskiy, C. Cole, J. Procter, G. J. Barton, JPred4: A protein secondary structure prediction server. Nucleic Acids Res. 43 , W389–WW94 (2015).25883141
53 F. Sievers, D. G. Higgins, Clustal Omega for making accurate alignments of many protein sequences. Protein Sci. 27 , 135–145 (2018).28884485
54 S. Rajkumar, T. M. Böckers, A. Catanese, Fast and efficient synaptosome isolation and post-synaptic density enrichment from hiPSC-motor neurons by biochemical sub-cellular fractionation. STAR Protoc 4 , 102061 (2023).36853677
55 L. Buck, R. Axel, A novel multigene family may encode odorant receptors: A molecular basis for odor recognition. Cell 65 , 175–187 (1991).1840504
56 L. B. Vosshall, H. Amrein, P. S. Morozov, A. Rzhetsky, R. Axel, A spatial map of olfactory receptor expression in the Drosophila antenna. Cell 96 , 725–736 (1999).10089887
57 E. A. Hallem, M. G. Ho, J. R. Carlson, The molecular basis of odor coding in the Drosophila antenna. Cell 117 , 965–979 (2004).15210116
58 D. Task, C.-C. Lin, A. Vulpe, A. Afify, S. Ballou, M. Brbic, P. Schlegel, J. Raji, G. Jefferis, H. Li, K. Menuz, C. J. Potter, Chemoreceptor co-expression in Drosophila melanogaster olfactory neurons. eLife 11 , e72599 (2022).35442190
59 Q. Gao, B. Yuan, A. Chess, Convergent projections of Drosophila olfactory neurons to specific glomeruli in the antennal lobe. Nat. Neurosci. 3 , 780–785 (2000).10903570
60 P. Mombaerts, F. Wang, C. Dulac, S. K. Chao, A. Nemes, M. Mendelsohn, J. Edmondson, R. Axel, Visualizing an olfactory sensory map. Cell 87 , 675–686 (1996).8929536
61 L. B. Vosshall, A. M. Wong, R. Axel, An olfactory sensory map in the fly brain. Cell 102 , 147–159 (2000).10943836
62 N. K. Tanaka, K. Endo, K. Ito, Organization of antennal lobe-associated neurons in adult Drosophila melanogaster brain. J. Comp. Neurol. 520 , 4067–4130 (2012).22592945
63 A. Schneider, M. Ruppert, O. Hendrich, T. Giang, M. Ogueta, S. Hampel, M. Vollbach, A. Büschges, H. Scholz, Neuronal basis of innate olfactory attraction to ethanol in Drosophila. PLOS ONE 7 , e52007 (2012).23284851
64 A. V. Devineni, U. Heberlein, Preferential ethanol consumption in Drosophila models features of addiction. Curr. Biol. 19 , 2126–2132 (2009).20005106
65 S.-P. Wang, W.-Y. Guo, S. A. Muhammad, R.-R. Chen, L.-L. Mu, G.-Q. Li, Mating experience and food deprivation modulate odor preference and dispersal in Drosophila melanogaster males. J. Insect Sci. 14 , 131 (2014).25368075
66 R. Azanchi, K. R. Kaun, U. Heberlein, Competing dopamine neurons drive oviposition choice for ethanol in Drosophila. Proc. Natl. Acad. Sci. U.S.A. 110 , 21153–21158 (2013).24324162
67 M. Parnas, A. C. Lin, W. Huetteroth, G. Miesenböck, Odor discrimination in Drosophila: From neural population codes to behavior. Neuron 79 , 932–944 (2013).24012006
68 M. dela Paz Fernández, Y.-B. Chan, J. Y. Yew, J.-C. Billeter, K. Dreisewerd, J. D. Levine, E. A. Kravitz, Pheromonal and behavioral cues trigger male-to-female aggression in Drosophila. PLoS Biol. 8 , e1000541 (2010).21124886
69 A. Bentzur, A. Shmueli, L. Omesi, J. Ryvkin, J.-M. Knapp, M. Parnas, F. P. Davis, G. Shohat-Ophir, Odorant binding protein 69a connects social interaction to modulation of social responsiveness in Drosophila. PLOS Genet. 14 , e1007328 (2018).29630598
70 D. P. Smith, Odor and pheromone detection in Drosophila melanogaster. Pflugers Arch. 454 , 749–758 (2007).17205355
71 L. Wang, D. J. Anderson, Identification of an aggression-promoting pheromone and its receptor neurons in Drosophila. Nature 463 , 227–231 (2010).19966787
72 R. M. Joseph, A. V. Devineni, I. F. G. King, U. Heberlein, Oviposition preference for and positional avoidance of acetic acid provide a model for competing behavioral drives in Drosophila. Proc. Natl. Acad. Sci. U.S.A. 106 , 11352–11357 (2009).19541615
73 J. A. Riffell, Neuroethology: Lemon-fresh scent makes flies lay eggs. Curr. Biol. 23 , R1108–R1110 (2013).24355790
74 S.-L. Lai, T. Awasaki, K. Ito, T. Lee, Clonal analysis of Drosophila antennal lobe neurons: Diverse neuronal architectures in the lateral neuroblast lineage. Development 135 , 2883–2893 (2008).18653555
75 J. J. C. Rosenthal, The emerging role of RNA editing in plasticity. J. Exp. Biol. 218 , 1812–1821 (2015).26085659
76 G. Galarza-Muñoz, S. I. Soto-Morales, M. Holmgren, J. J. C. Rosenthal, Physiological adaptation of an Antarctic Na+/K+-ATPase to the cold. J. Exp. Biol. 214 , 2164–2174 (2011).21653810
77 J. J. C. Rosenthal, P. H. Seeburg, A-to-I RNA editing: Effects on proteins key to neural excitability. Neuron 74 , 432–439 (2012).22578495
78 S. Alon, S. C. Garrett, E. Y. Levanon, S. Olson, B. R. Graveley, J. J. C. Rosenthal, E. Eisenberg, The majority of transcripts in the squid nervous system are extensively recoded by A-to-I RNA editing. eLife 4 , e05198 (2015).25569156
79 J. Alcedo, C. Kenyon, Regulation of C. elegans longevity by specific gustatory and olfactory neurons. Neuron 41 , 45–55 (2004).14715134
80 J. Apfeld, C. Kenyon, Regulation of lifespan by sensory perception in Caenorhabditis elegans. Nature 402 , 804–809 (1999).10617200
81 L. Badel, K. Ohta, Y. Tsuchimoto, H. Kazama, Decoding of context-dependent olfactory behavior in Drosophila. Neuron 91 , 155–167 (2016).27321924
82 Y. Kondoh, K. Y. Kaneshiro, K.-I. Kimura, D. Yamamoto, Evolution of sexual dimorphism in the olfactory brain of Hawaiian Drosophila. Proc. Biol. Sci. 270 , 1005–1013 (2003).12803889
83 P. Stockinger, D. Kvitsiani, S. Rotkopf, L. Tirián, B. J. Dickson, Neural circuitry that governs Drosophila male courtship behavior. Cell 121 , 795–807 (2005).15935765
84 A. Sakurai, M. Koganezawa, K.-I. Yasunaga, K. Emoto, D. Yamamoto, Select interneuron clusters determine female sexual receptivity in Drosophila. Nat. Commun. 4 , 1825 (2013).23652013
85 M. L. Schlief, R. I. Wilson, Olfactory processing and behavior downstream from highly selective receptor neurons. Nat. Neurosci. 10 , 623–630 (2007).17417635
86 S. R. Olsen, V. Bhandawat, R. I. Wilson, Excitatory interactions between olfactory processing channels in the Drosophila antennal lobe. Neuron 54 , 89–103 (2007).17408580
87 R. Brusa, F. Zimmermann, D. S. Koh, D. Feldmeyer, P. Gass, P. H. Seeburg, R. Sprengel, Early-onset epilepsy and postnatal lethality associated with an editing-deficient GluR-B allele in mice. Science 270 , 1677–1680 (1995).7502080
88 C. Gonzalez, A. Lopez-Rodriguez, D. Srikumar, J. J. C. Rosenthal, M. Holmgren, Editing of human KV1.1 channel mRNAs disrupts binding of the N-terminus tip at the intracellular cavity. Nat. Commun. 2 , 436 (2011).21847110
89 R. E. Hibbs, E. Gouaux, Principles of activation and permeation in an anion-selective Cys-loop receptor. Nature 474 , 54–60 (2011).21572436
90 J. Yu, H. Zhu, R. Lape, T. Greiner, J. Du, W. Lü, L. Sivilotti, E. Gouaux, Mechanism of gating and partial agonist action in the glycine receptor. Cell 184 , 957–968.e21 (2021).33567265
91 G. Ramaswami, P. Deng, R. Zhang, M. Anna Carbone, T. F. C. Mackay, J. Billy Li, Genetic mapping uncovers cis-regulatory landscape of RNA editing. Nat. Commun. 6 , 8194 (2015).26373807
92 J. Cox, M. Y. Hein, C. A. Luber, I. Paron, N. Nagaraj, M. Mann, Accurate proteome-wide label-free quantification by delayed normalization and maximal peptide ratio extraction, termed MaxLFQ. Mol. Cell. Proteomics 13 , 2513–2526 (2014).24942700
93 M. Kabra, A. A. Robie, M. Rivera-Alba, S. Branson, K. Branson, JAABA: Interactive machine learning for automatic annotation of animal behavior. Nat. Methods 10 , 64–67 (2013).23202433
94 A. Bentzur, S. Ben-Shaanan, J. I. C. Benichou, E. Costi, M. Levi, A. Ilany, G. Shohat-Ophir, Early life experience shapes male behavior and social networks in Drosophila. Curr. Biol. 31 , 486–501.e3 (2021).33186552
95 J. E. Manoim, A. M. Davidson, S. Weiss, T. Hige, M. Parnas, Lateral axonal modulation is required for stimulus-specific olfactory conditioning in Drosophila. Curr. Biol. 32 , 4438–4450.e5 (2022).36130601
96 E. Rozenfeld, H. Lerner, M. Parnas, Muscarinic modulation of antennal lobe GABAergic local neurons shapes odor coding and behavior. Cell Rep. 29 , 3253–3265.e4 (2019).31801087
97 S. Israel, E. Rozenfeld, D. Weber, W. Huetteroth, M. Parnas, Olfactory stimuli and moonwalker SEZ neurons can drive backward locomotion in Drosophila. Curr. Biol. 32 , 1131–1149.e7 (2022).35139358
98 E. Rozenfeld, M. Tauber, Y. Ben-Chaim, M. Parnas, GPCR voltage dependence controls neuronal plasticity and behavior. Nat. Commun. 12 , 7252 (2021).34903750
99 E. Rozenfeld, N. Ehmann, J. E. Manoim, R. J. Kittel, M. Parnas, Homeostatic synaptic plasticity rescues neural coding reliability. Nat. Commun. 14 , 2993 (2023).37225688
