Neuronal dendrites carry out computations through compartmentalized signaling, while axons have long been considered to carry signals to their terminal fields relatively uniformly following spike initiation. However, anatomical and physiological compartmentalization of axons has been recently documented in neurons from worms through mammals (Boto et al., 2014; Cohn et al., 2015; Hendricks et al., 2012; Rowan et al., 2016). How axonal compartmentalization influences information flow across neuronal circuits and modulates behavioral outcomes is not understood. One functional role for axonal compartmentalization may be to enable localized, presynaptic plasticity to alter output from select axon compartments in a flexible, experience-dependent manner. This would vastly enhance the neuron’s flexibility and computational capabilities. A potential function of such compartmentalization would allow independent modulation of axonal segments and/or synaptic release sites by biologically salient events, such as sensory stimuli that drive learning.

The anatomical organization of the Drosophila mushroom body (MB) makes it an exemplary test bed to study how sensory information is processed during learning and rerouted to alter behavioral outcomes. The MB encodes odor in sparse representations across the intrinsic MB neurons, Kenyon cells (KCs), which are arranged in several parallel sets. They project axons in fasciculated bundles into several anatomically distinct, but spatially adjacent lobes (α/β, α′/β′, and γ) (Crittenden et al., 1998). KC axons are longitudinally subdivided into discrete tiled compartments (Aso et al., 2014a). Each compartment receives afferent neuromodulatory input from unique dopaminergic neurons (Aso et al., 2014a; Mao and Davis, 2009), and innervates unique efferent mushroom body output neurons (MBONs) (Aso et al., 2014a). Each set of dopaminergic neurons plays an individual role in learning, with some modulating aversive learning (Mao and Davis, 2009; Schroll et al., 2006; Schwaerzel et al., 2003), others modulating reward learning (Liu et al., 2012; Yamagata et al., 2015), and a third class modulating memory strength without driving behavioral valence (Boto et al., 2019). Similarly, each MBON has a unique effect on behavioral approach and avoidance, with some biasing the animal to approach, others biasing the animal to avoidance, and some having no known effect (Aso et al., 2014b; Perisse et al., 2016; Plaçais et al., 2013; Séjourné et al., 2011).

A major question in learning and memory is how presynaptic plasticity contributes to reweighting the flow of sensory signals across each of the downstream ‘approach’ or ‘avoidance’ circuits, altering action selection and memory retrieval. In naïve conditions, Drosophila dopaminergic circuits modulate cAMP in a compartmentalized fashion along the MB axons (Boto et al., 2014). This compartmentalized dopaminergic signaling can independently modulate Ca²⁺ responses in each compartment, as well as the responses of the downstream valence-coding MBONs (Cohn et al., 2015). Presynaptic plasticity within each KC compartment likely contributes to the changes in downstream MBON activity that guide learned behavioral responses (Zhang et al., 2019). However, manipulation of the ‘aversive’ protocerebral posterior lateral 1 (PPL1) dopaminergic neurons does not detectably alter Ca²⁺ signals in KCs (Boto et al., 2019; Hige et al., 2015a). Furthermore, Ca²⁺ responses in KCs are uniformly potentiated across compartments with appetitive classical conditioning protocols and unaltered in KCs following aversive protocols (Louis et al., 2018). This raises the question of how local, compartmentalized synaptic plasticity in KCs drives coherent changes across the array of downstream MBONs to modulate action selection during memory retrieval. Learning/dopamine-induced plasticity has been demonstrated in the MBONs (Berry et al., 2018; Hige et al., 2015a; Hige et al., 2015b; Owald et al., 2015), with dopamine also acting directly on them (Takemura et al., 2017) (in addition to KCs). Feedforward inhibition among MBONs that drive opposing behavioral outcomes provides a mechanism explaining how bidirectional valence coding in MBONs could be generated, with or without bidirectional presynaptic plasticity (Perisse et al., 2016). The compartmentalized, dopamine-dependent plasticity in KCs and the necessity for dopamine receptors and downstream signaling molecules in the intrinsic KCs points to a potential presynaptic contribution (Kim et al., 2007; McGuire et al., 2003; Zars et al., 2000). Thus, compartmentalized presynaptic plasticity could contribute to reweighting the flow of olfactory information to downstream circuits.

Here, we describe how learning alters the flow of information through the MB via alteration of synaptic release of the KC neurotransmitter acetylcholine (ACh) (Barnstedt et al., 2016), using a genetically encoded indicator of ACh neurotransmission. The data reveal that appetitive and aversive learning alter compartmentalized acetylcholine release in distinct spatial patterns, with differing molecular mechanisms, coherently reweighting the flow of olfactory information across the ensemble of downstream neurons that mediate action selection.

Compartmentalized plasticity in neurotransmitter release expands the potential computational capacity of learning circuits. It allows a set of odor-coding MB neurons to bifurcate their output to different downstream approach- and avoidance-driving downstream output neurons, independently modulating the synaptic connections to alter action selection based on the conditioned value of olfactory stimuli. The KCs modify the encoded value of olfactory stimuli through bidirectional plasticity in odor responses, which vary in a compartment-specific manner along the length of the axons. These changes were observed following pairing an olfactory CS with gustatory/somatosensory US (sucrose feeding or electric shock) in vivo. The CS+ and CS- drive unique patterns of plasticity in each compartment, demonstrating that olfactory stimuli are reweighted differently across compartments following learning, depending on the temporal associations of the stimuli. Different molecular mechanisms govern the potentiation of trained odor responses (Ca_V2/Cac) and maintenance of responsivity over time (IP₃R). Finally, one set of γ output neurons, γ3/γ3β′1, is important for appetitive short-term memory.

Learning-induced plasticity of ACh release in the MB was bidirectional within the compartment, depending on the valence of the US, and was coherent with the valence of the MBON downstream of the compartment. Notably, the γ2 and γ3 MB compartments, which relay information to approach-promoting MBONs (Aso et al., 2014b), exhibited plasticity that was coherent with promoting behavioral approach following appetitive conditioning and avoidance after aversive conditioning. There was an increase in the relative CS+:CS- ACh responses after appetitive conditioning, and conversely reduced CS+:CS- ACh responses following aversive conditioning. For this study, we focused on the time point 5 min following conditioning, which is consistent with behavioral short-term memory. Aversive conditioning was previously reported to decrease neurotransmitter release from KCs (Zhang and Roman, 2013; Zhang et al., 2019). Indirect evidence, via Ca²⁺ imaging in presynaptic KCs, suggested that increases in presynaptic neurotransmission could also be associated with learning. Pairing odor with stimulation of appetitive PAM dopaminergic neurons potentiates odor-evoked cytosolic Ca²⁺ transients across the KC compartments (Boto et al., 2014). Appetitive conditioning increases odor-evoked Ca²⁺ transients across KC compartments (Louis et al., 2018). Stimulation of dopaminergic circuits associated with reward learning potentiate MB γ4 connections with the respective γ4 MBON (Handler et al., 2019). We did not detect a statistically significant effect in γ4 with appetitive or aversive classical conditioning, though the CS+ and CS- trended in the same direction as the adjacent γ5 compartment following conditioning. Overall, the present data demonstrate that there are bidirectional changes in neurotransmitter (ACh) release from MB compartments following appetitive vs aversive learning and provide a window into the spatial patterns of plasticity across compartments following associative learning.

Behavioral alterations following conditioning involve changes in responses among the MBONs. As the KCs provide presynaptic olfactory input to the MBONs, it was a logical a priori assumption that presynaptic plasticity in the KCs could be altered in a compartmental manner and contribute to the changes in MBON responses after conditioning. Yet data from previous Ca²⁺ imaging experiments have not completely supported this model. Compartmentalized effects have been observed in KCs with non-associative learning protocols (Cohn et al., 2015) and within the γ4 compartment following associative learning (Handler et al., 2019). In contrast, classical conditioning produces no compartmentalized differences in odor-evoked Ca²⁺ responses. Appetitive conditioning with odor + sucrose pairing increases odor-evoked cytosolic Ca²⁺ transients in KCs across the γ lobe compartments (Louis et al., 2018). Aversive conditioning produces no net change across the compartments (Louis et al., 2018), but alters synapse-specific Ca²⁺ responses at the individual bouton level (Bilz et al., 2020). If the compartmental effects of conditioning (observed with Ca²⁺ imaging) in KCs drove a proportional change in neurotransmitter release, both the approach- and avoidance-promoting MBONs would be simultaneously potentiated. Extracellular influx of Ca²⁺ through voltage-gated calcium channels is a primary driver of neurotransmitter release; however, there are multiple sources of Ca²⁺ in the cytosol that could contribute to the GCaMP signals (Grienberger and Konnerth, 2012). A major conclusion of the present study is that learning drives compartmentalized plasticity in neurotransmitter release that is coherent with the behavioral valence of the corresponding MBON.

At least two major molecular mechanisms govern the spatial patterns of plasticity across the MB compartments: a Cac-dependent CS+ potentiation and an IP₃R-dependent maintenance of sensory responses over trials/time. This suggests that different sources of Ca²⁺ play different roles in regulating KC synaptic responses. Cac is the pore-forming subunit of the voltage-sensitive, presynaptic Ca_V2 Ca²⁺ channel in Drosophila. Ca_V2 channels regulate several forms of synaptic plasticity, including paired-pulse facilitation, homeostatic plasticity, and long-term potentiation (Frank et al., 2006; Inchauspe et al., 2004; Nanou et al., 2016). Our data suggest that these channels regulate the spatial patterns of learning-induced plasticity in the MB unidirectionally (from baseline), with Cac underlying potentiation but not depression. Ca_V2 channel activity is modulated by presynaptic calcium and G protein-coupled receptor activity (Zamponi and Currie, 2013), and channel localization in the active zone dynamically regulates synaptic strength (Gratz et al., 2019; Lübbert et al., 2019). Thus, Cac insertion into, or increased clustering within, the active zones may underlie learning-induced potentiation (e.g. in the γ1-γ2 compartments following appetitive conditioning). Conditional knockdown of Cac, which reduced Cac levels by ~29%, impaired this potentiation, likely by decreasing the number of available channels for modulation. Baseline stimulus-evoked neurotransmitter release was maintained during Cac knockdown, mediated either by the significant residual Cac expression or compensation by other intracellular Ca²⁺ channels/sources. In contrast to the Cac effect on potentiation, IP₃R was necessary to maintain normal odor responsivity when odors were presented repeatedly across multiple trials (whether those were pre/post trials in the conditioning protocol or 10× odor presentations in the adaptation protocol). This is broadly consistent with the temporal role of IP₃R in maintenance of presynaptic homeostatic potentiation at the neuromuscular junction (James et al., 2019). In addition, dopaminergic circuits associated with reward learning drive release of Ca²⁺ from the endoplasmic reticulum when activated with KCs in a backward pairing paradigm ex vivo, potentiating MB γ4 connections with the respective γ4 MBON (Handler et al., 2019). This is consistent with a role for ER calcium in positively regulating synaptic strength.

We observed potentiation and depression of ACh release across multiple MB compartments following conditioning, providing a presynaptic mechanism that potentially contributes to shaping conditioned MBON responses. Importantly, by comparing the CS+ and CS- responses to those of untrained odors, we ascribed plasticity to potentiation or depression (accounting for any non-associative olfactory adaptation) within each compartment. This is relevant for modeling efforts, where it has been unclear whether to include potentiation (along with depression) in the learning rule(s) at KC-MBON synapses (Abdelrahman et al., 2021; Bennett et al., 2021; Jiang and Litwin-Kumar, 2021; Springer and Nawrot, 2021). In addition, the experiments revealed an additional layer of spatial regulation in the γ1–γ3 compartments: a gradient of CS+ potentiation to CS- depression following appetitive conditioning. Specifically, the CS+/CS- relationship changed in a linear gradient down the γ1–γ3 compartments following appetitive conditioning. Appetitive conditioning increased CS+ responses in the γ1 compartment, while decreasing the CS- responses in the γ3 compartment. The γ2 compartment yielded a mix of these responses. These patterns of plasticity have the net effect of increasing the relative response to the CS+ odor (↑CS+:CS-). Since the MBONs postsynaptic to these compartments drive behavioral approach (Aso et al., 2014b), this would bias the animal to approach the CS+ if it encountered both odors simultaneously. Such a situation occurs at the choice point of a T-maze during retrieval in a classical conditioning assay. The CS+ and CS- produce different patterns of plasticity at different loci (e.g. γ1 vs γ3), which presumably coordinate to regulate behavior via temporal integration of the odor and US cues (Berry et al., 2018; Handler et al., 2019; Jacob and Waddell, 2020; König et al., 2018; Tanimoto et al., 2004; Tully and Quinn, 1985). The CS+ is temporally contiguous with the US, while the CS- is nonoverlapping. Therefore, the timing of CS/US pairing drives plasticity differently in each compartment. These patterns of plasticity presumably coordinate to regulate memory formation and action selection during retrieval. For instance, while the γ1–γ3 compartments exhibited ↑CS+:CS- following appetitive conditioning, the γ5 compartment exhibited plasticity in the opposite direction: decreasing the relative response to the CS+ odor (↓CS+:CS-). As the γ5 compartment is presynaptic to an avoidance-promoting MBON, this plasticity pattern would coherently contribute to biasing the animal toward CS+ approach (reducing CS+ avoidance). Thus, it would work in concert with the plasticity in γ1–γ3 to bias the animal toward behavioral approach. Overall, plasticity is regulated in each MB compartment individually by the timing of events and the valence of the US, with the changes coordinated across multiple compartments to coherently drive behavior.

Behaviorally, MBONs innervating the γ lobe variably drive behavioral approach or avoidance when stimulated (Aso et al., 2014b). Despite the approach-promoting valence of the γ2α′one and γ3/γ3β′1 MBONs, among them, only the γ3/γ3β′1 MBONs produced a loss-of-function phenotype in behavioral appetitive conditioning. This suggests that redundancy and/or different weighting across approach promoting MBONs renders the system resilient to silencing some of them. A previous study found effects of blocking the γ2α′1 MBONs, though not γ3/γ3β′1 MBONs, when blocking individual steps of memory processing (acquisition, retention, and/or retrieval) with a 1 hr appetitive memory protocol (Ichinose et al., 2021). This suggests that the different MBONs have differing roles across time, with some redundancy in appetitive processing. Blocking synaptic output of γ3/γ3β′1 MBONs reduced appetitive conditioning performance in our experiments immediately following conditioning, suggesting that these neurons play a specific role in appetitive short-term memory.

The present and previous studies suggest that alterations of MBON activity following learning are the product of both presynaptic and postsynaptic plasticity at the KC-MBON synapses, as well as feedforward inhibition (Hige et al., 2015a; Perisse et al., 2016; Pribbenow et al., 2021). Blocking synaptic output from KCs impairs the acquisition of appetitive memories (30–60 min after conditioning), suggesting a role for postsynaptic plasticity (Pribbenow et al., 2021; Schwaerzel et al., 2003). However, this does not rule out presynaptic plasticity, as blocking KC output (with R13F02) leaves signaling from reinforcing dopaminergic neurons partially intact (Cervantes-Sandoval et al., 2017), which likely shapes the presynaptic KC responses via heterosynaptic plasticity. At the circuit level, polysynaptic inhibition can convert depression from select MB compartments into potentiation in MBONs following learning; in one established example, reduction of odor-evoked responses in the GABAergic γ1pedc MBON following aversive conditioning disinhibits the downstream γ5β′2 a MBON (Owald et al., 2015; Perisse et al., 2016).

KC-MBON synapses represent one node of learning-related plasticity, which is distributed across multiple sites during learning. Short-term memory-related plasticity has been observed in multiple olfactory neurons, such as the antennal lobes (Yu et al., 2004) and KCs (Louis et al., 2018; Wang et al., 2008). In addition, connectomics studies have revealed complex connectivity within and beyond the MB, which is a multi-layered network including circuit motifs that influence the propagation of information and generation of plasticity during learning. Such connections include recurrent feedback (Aso et al., 2014a; Ichinose et al., 2015; Li et al., 2020; Otto et al., 2020; Scaplen et al., 2021; Takemura et al., 2017; Zhao et al., 2018). Some of these recurrent connections are from cholinergic MBONs that synapse within the MB, which could have contributed to the ACh signals we observed in this study. For instance, the γ2α′1 MBON is a cholinergic MBON that sends ~6% of its output back to the γ lobe (Scheffer et al., 2020). Some of the recurrent connections are formed by dopaminergic neurons, such as the PAM γ4<γ1/y2 (Aso et al., 2014a; Li et al., 2020). In addition, reciprocal connections between KCs and dopaminergic neurons in the vertical lobes are necessary for memory retrieval (Cervantes-Sandoval et al., 2017). This adds another layer of recurrent circuitry that may participate in reinforcement during associative learning (Aso et al., 2014b; Ichinose et al., 2015; Ichinose et al., 2021). Across these circuits, some neurons corelease several neurotransmitters and act on an array of postsynaptic receptors, which contribute to plasticity distributed across multiple sites (Aso et al., 2019; Barnstedt et al., 2016; Berry et al., 2012; Bielopolski et al., 2019; Cohn et al., 2015; Handler et al., 2019; Haynes et al., 2015; Keene et al., 2004; Kim et al., 2007; Liu and Davis, 2009; Pribbenow et al., 2021; Schroll et al., 2006; Séjourné et al., 2011; Silva et al., 2015; Wu et al., 2013; Zhou et al., 2019).

Overall, plasticity between KCs and MBONs may guide behavior through biasing network activation to alter action selection in a probabilistic manner. Appetitive conditioning drives compartmentalized, presynaptic plasticity in KCs that correlates with postsynaptic changes in MBONs that guide learned behaviors. Prior studies documented only depression at these synapses at short time points following conditioning (Hige et al., 2015a; Zhang and Roman, 2013; Zhang et al., 2019). Here, we observed both potentiation and depression in ACh release in the MB, suggesting that bidirectional presynaptic plasticity modulates learned behaviors. These bidirectional changes likely integrate with plasticity at downstream circuit nodes that also undergo learning-induced plasticity to produce network-level alterations in odor responses across the olfactory pathway following salient events. Thus, plasticity in ACh release from KCs functions to modulate responsivity to olfactory stimuli features across graded plasticity maps down the MB axons.

Raw data have been deposited to Dryad under https://doi.org/10.5061/dryad.dfn2z353h.

1. 1. Tomchik SM

   (2022) Dryad Digital Repository

   Data from: Associative learning drives longitudinally-graded presynaptic plasticity of neurotransmitter release along axonal compartments.

   https://doi.org/10.5061/dryad.dfn2z353h

1. 1. Abdelrahman NY
   2. Vasilaki E
   3. Lin AC

   (2021) Compensatory variability in network parameters enhances memory performance in the Drosophila mushroom body

   PNAS 118:e2102158118.

   https://doi.org/10.1073/pnas.2102158118

   • PubMed
   • Google Scholar
2. 1. Aso Y
   2. Grübel K
   3. Busch S
   4. Friedrich AB
   5. Siwanowicz I
   6. Tanimoto H

   (2009) The mushroom body of adult Drosophila characterized by GAL4 drivers

   Journal of Neurogenetics 23:156–172.

   https://doi.org/10.1080/01677060802471718

   • PubMed
   • Google Scholar
3. 1. Aso Y
   2. Hattori D
   3. Yu Y
   4. Johnston RM
   5. Iyer NA
   6. Ngo TT
   7. Dionne H
   8. Abbott L
   9. Axel R
   10. Tanimoto H
   11. Rubin GM

   (2014a) The neuronal architecture of the mushroom body provides a logic for associative learning

   eLife 3:e1.

   https://doi.org/10.7554/eLife.04577.001

   • Google Scholar
4. 1. Aso Y
   2. Sitaraman D
   3. Ichinose T
   4. Kaun KR
   5. Vogt K
   6. Belliart-Guerin G
   7. Placais PY
   8. Robie AA
   9. Yamagata N
   10. Schnaitmann C

   (2014b) Mushroom body output neurons encode valence and guide memory-based action selection

   eLife 3:e1.

   https://doi.org/10.7554/eLife.04580.001

   • Google Scholar
5. 1. Aso Y
   2. Ray RP
   3. Long X
   4. Bushey D
   5. Cichewicz K
   6. Ngo TT
   7. Sharp B
   8. Christoforou C
   9. Hu A
   10. Lemire AL
   11. Tillberg P
   12. Hirsh J
   13. Litwin-Kumar A
   14. Rubin GM

   (2019) Nitric oxide acts as a cotransmitter in a subset of dopaminergic neurons to diversify memory dynamics

   eLife 8:e49257.

   https://doi.org/10.7554/eLife.49257

   • PubMed
   • Google Scholar
6. 1. Barnstedt O
   2. Owald D
   3. Felsenberg J
   4. Brain R
   5. Moszynski JP
   6. Talbot CB
   7. Perrat PN
   8. Waddell S

   (2016) Memory-Relevant Mushroom Body Output Synapses Are Cholinergic

   Neuron 89:1237–1247.

   https://doi.org/10.1016/j.neuron.2016.02.015

   • PubMed
   • Google Scholar
7. 1. Bennett JEM
   2. Philippides A
   3. Nowotny T

   (2021) Learning with reinforcement prediction errors in a model of the Drosophila mushroom body

   Nature Communications 12:2569.

   https://doi.org/10.1038/s41467-021-22592-4

   • PubMed
   • Google Scholar
8. 1. Berry JA
   2. Cervantes-Sandoval I
   3. Nicholas EP
   4. Davis RL

   (2012) Dopamine is required for learning and forgetting in Drosophila

   Neuron 74:530–542.

   https://doi.org/10.1016/j.neuron.2012.04.007

   • PubMed
   • Google Scholar
9. 1. Berry JA
   2. Phan A
   3. Davis RL

   (2018) Dopamine Neurons Mediate Learning and Forgetting through Bidirectional Modulation of a Memory Trace

   Cell Reports 25:651–662.

   https://doi.org/10.1016/j.celrep.2018.09.051

   • PubMed
   • Google Scholar
10. 1. Bielopolski N
    2. Amin H
    3. Apostolopoulou AA
    4. Rozenfeld E
    5. Lerner H
    6. Huetteroth W
    7. Lin AC
    8. Parnas M

    (2019) Inhibitory muscarinic acetylcholine receptors enhance aversive olfactory learning in adult Drosophila

    eLife 8:e48264.

    https://doi.org/10.7554/eLife.48264

    • PubMed
    • Google Scholar
11. 1. Bilz F
    2. Geurten BRH
    3. Hancock CE
    4. Widmann A
    5. Fiala A

    (2020) Visualization of a Distributed Synaptic Memory Code in the Drosophila Brain

    Neuron 106:963–976.

    https://doi.org/10.1016/j.neuron.2020.03.010

    • PubMed
    • Google Scholar
12. 1. Boto T
    2. Louis T
    3. Jindachomthong K
    4. Jalink K
    5. Tomchik SM

    (2014) Dopaminergic Modulation of cAMP Drives Nonlinear Plasticity across the Drosophila Mushroom Body Lobes

    Current Biology 24:822–831.

    https://doi.org/10.1016/j.cub.2014.03.021

    • PubMed
    • Google Scholar
13. 1. Boto T
    2. Stahl A
    3. Zhang X
    4. Louis T
    5. Tomchik SM

    (2019) Independent Contributions of Discrete Dopaminergic Circuits to Cellular Plasticity, Memory Strength, and Valence in Drosophila

    Cell Reports 27:2014–2021.

    https://doi.org/10.1016/j.celrep.2019.04.069

    • PubMed
    • Google Scholar
14. 1. Brusich DJ
    2. Spring AM
    3. Frank CA

    (2015) A single-cross, RNA interference-based genetic tool for examining the long-term maintenance of homeostatic plasticity

    Frontiers in Cellular Neuroscience 9:107.

    https://doi.org/10.3389/fncel.2015.00107

    • PubMed
    • Google Scholar
15. 1. Cervantes-Sandoval I
    2. Phan A
    3. Chakraborty M
    4. Davis RL

    (2017) Reciprocal synapses between mushroom body and dopamine neurons form a positive feedback loop required for learning

    eLife 6:e23789.

    https://doi.org/10.7554/eLife.23789

    • PubMed
    • Google Scholar
16. 1. Cohn R
    2. Morantte I
    3. Ruta V

    (2015) Coordinated and Compartmentalized Neuromodulation Shapes Sensory Processing in Drosophila

    Cell 163:1742–1755.

    https://doi.org/10.1016/j.cell.2015.11.019

    • PubMed
    • Google Scholar
17. 1. Crittenden JR
    2. Skoulakis EM
    3. Han KA
    4. Kalderon D
    5. Davis RL

    (1998)

    Tripartite mushroom body architecture revealed by antigenic markers

    Learning & Memory (Cold Spring Harbor, N.Y.) 5:38–51.

    • PubMed
    • Google Scholar
18. 1. Dietzl G
    2. Chen D
    3. Schnorrer F
    4. Su K-C
    5. Barinova Y
    6. Fellner M
    7. Gasser B
    8. Kinsey K
    9. Oppel S
    10. Scheiblauer S
    11. Couto A
    12. Marra V
    13. Keleman K
    14. Dickson BJ

    (2007) A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila

    Nature 448:151–156.

    https://doi.org/10.1038/nature05954

    • PubMed
    • Google Scholar
19. 1. Frank CA
    2. Kennedy MJ
    3. Goold CP
    4. Marek KW
    5. Davis GW

    (2006) Mechanisms underlying the rapid induction and sustained expression of synaptic homeostasis

    Neuron 52:663–677.

    https://doi.org/10.1016/j.neuron.2006.09.029

    • PubMed
    • Google Scholar
20. 1. Gratz SJ
    2. Goel P
    3. Bruckner JJ
    4. Hernandez RX
    5. Khateeb K
    6. Macleod GT
    7. Dickman D
    8. O’Connor-Giles KM

    (2019) Endogenous Tagging Reveals Differential Regulation of Ca(2+) Channels at Single Active Zones during Presynaptic Homeostatic Potentiation and Depression

    The Journal of Neuroscience 39:2416–2429.

    https://doi.org/10.1523/JNEUROSCI.3068-18.2019

    • PubMed
    • Google Scholar
21. 1. Grienberger C
    2. Konnerth A

    (2012) Imaging calcium in neurons

    Neuron 73:862–885.

    https://doi.org/10.1016/j.neuron.2012.02.011

    • PubMed
    • Google Scholar
22. 1. Handler A
    2. Graham TGW
    3. Cohn R
    4. Morantte I
    5. Siliciano AF
    6. Zeng J
    7. Li Y
    8. Ruta V

    (2019) Distinct Dopamine Receptor Pathways Underlie the Temporal Sensitivity of Associative Learning

    Cell 178:60–75.

    https://doi.org/10.1016/j.cell.2019.05.040

    • PubMed
    • Google Scholar
23. 1. Haynes PR
    2. Christmann BL
    3. Griffith LC

    (2015) A single pair of neurons links sleep to memory consolidation in Drosophila melanogaster

    eLife 4:3868.

    https://doi.org/10.7554/eLife.03868

    • PubMed
    • Google Scholar
24. 1. Hendricks M
    2. Ha H
    3. Maffey N
    4. Zhang Y

    (2012) Compartmentalized calcium dynamics in a C. elegans interneuron encode head movement

    Nature 487:99–103.

    https://doi.org/10.1038/nature11081

    • PubMed
    • Google Scholar
25. 1. Hidalgo S
    2. Campusano JM
    3. Hodge JJL

    (2021) The Drosophila ortholog of the schizophrenia-associated CACNA1A and CACNA1B voltage-gated calcium channels regulate memory, sleep and circadian rhythms

    Neurobiology of Disease 155:105394.

    https://doi.org/10.1016/j.nbd.2021.105394

    • PubMed
    • Google Scholar
26. 1. Hige T
    2. Aso Y
    3. Modi MN
    4. Rubin GM
    5. Turner GC

    (2015a) Heterosynaptic Plasticity Underlies Aversive Olfactory Learning in Drosophila

    Neuron 88:985–998.

    https://doi.org/10.1016/j.neuron.2015.11.003

    • PubMed
    • Google Scholar
27. 1. Hige T
    2. Aso Y
    3. Rubin GM
    4. Turner GC

    (2015b) Plasticity-driven individualization of olfactory coding in mushroom body output neurons

    Nature 526:258–262.

    https://doi.org/10.1038/nature15396

    • PubMed
    • Google Scholar
28. 1. Ichinose T
    2. Aso Y
    3. Yamagata N
    4. Abe A
    5. Rubin GM
    6. Tanimoto H

    (2015) Reward signal in a recurrent circuit drives appetitive long-term memory formation

    eLife 4:e10719.

    https://doi.org/10.7554/eLife.10719

    • PubMed
    • Google Scholar
29. 1. Ichinose T
    2. Kanno M
    3. Wu H
    4. Yamagata N
    5. Sun H
    6. Abe A
    7. Tanimoto H

    (2021) Mushroom body output differentiates memory processes and distinct memory-guided behaviors

    Current Biology 31:1294–1302.

    https://doi.org/10.1016/j.cub.2020.12.032

    • PubMed
    • Google Scholar
30. 1. Inchauspe CG
    2. Martini FJ
    3. Forsythe ID
    4. Uchitel OD

    (2004) Functional compensation of P/Q by N-type channels blocks short-term plasticity at the calyx of Held presynaptic terminal

    The Journal of Neuroscience 24:10379–10383.

    https://doi.org/10.1523/JNEUROSCI.2104-04.2004

    • PubMed
    • Google Scholar
31. 1. Ishikawa T
    2. Kaneko M
    3. Shin HS
    4. Takahashi T

    (2005) Presynaptic N-type and P/Q-type Ca2+ channels mediating synaptic transmission at the calyx of Held of mice

    The Journal of Physiology 568:199–209.

    https://doi.org/10.1113/jphysiol.2005.089912

    • PubMed
    • Google Scholar
32. 1. Jacob PF
    2. Waddell S

    (2020) Spaced Training Forms Complementary Long-Term Memories of Opposite Valence in Drosophila

    Neuron 106:977–991.

    https://doi.org/10.1016/j.neuron.2020.03.013

    • PubMed
    • Google Scholar
33. 1. James TD
    2. Zwiefelhofer DJ
    3. Frank CA

    (2019) Maintenance of homeostatic plasticity at the Drosophila neuromuscular synapse requires continuous IP3-directed signaling

    eLife 8:e39643.

    https://doi.org/10.7554/eLife.39643

    • PubMed
    • Google Scholar
34. 1. Jenett A
    2. Rubin GM
    3. Ngo T-TB
    4. Shepherd D
    5. Murphy C
    6. Dionne H
    7. Pfeiffer BD
    8. Cavallaro A
    9. Hall D
    10. Jeter J
    11. Iyer N
    12. Fetter D
    13. Hausenfluck JH
    14. Peng H
    15. Trautman ET
    16. Svirskas RR
    17. Myers EW
    18. Iwinski ZR
    19. Aso Y
    20. DePasquale GM
    21. Enos A
    22. Hulamm P
    23. Lam SCB
    24. Li H-H
    25. Laverty TR
    26. Long F
    27. Qu L
    28. Murphy SD
    29. Rokicki K
    30. Safford T
    31. Shaw K
    32. Simpson JH
    33. Sowell A
    34. Tae S
    35. Yu Y
    36. Zugates CT

    (2012) A GAL4-driver line resource for Drosophila neurobiology

    Cell Reports 2:991–1001.

    https://doi.org/10.1016/j.celrep.2012.09.011

    • PubMed
    • Google Scholar
35. 1. Jiang L
    2. Litwin-Kumar A

    (2021) Models of heterogeneous dopamine signaling in an insect learning and memory center

    PLOS Computational Biology 17:e1009205.

    https://doi.org/10.1371/journal.pcbi.1009205

    • PubMed
    • Google Scholar
36. 1. Jing M
    2. Zhang P
    3. Wang G
    4. Feng J
    5. Mesik L
    6. Zeng J
    7. Jiang H
    8. Wang S
    9. Looby JC
    10. Guagliardo NA
    11. Langma LW
    12. Lu J
    13. Zuo Y
    14. Talmage DA
    15. Role LW
    16. Barrett PQ
    17. Zhang LI
    18. Luo M
    19. Song Y
    20. Zhu JJ
    21. Li Y

    (2018) A genetically encoded fluorescent acetylcholine indicator for in vitro and in vivo studies

    Nature Biotechnology 36:726–737.

    https://doi.org/10.1038/nbt.4184

    • PubMed
    • Google Scholar
37. 1. Jing M
    2. Li Y
    3. Zeng J
    4. Huang P
    5. Skirzewski M
    6. Kljakic O
    7. Peng W
    8. Qian T
    9. Tan K
    10. Zou J
    11. Trinh S
    12. Wu R
    13. Zhang S
    14. Pan S
    15. Hires SA
    16. Xu M
    17. Li H
    18. Saksida LM
    19. Prado VF
    20. Bussey TJ
    21. Prado MAM
    22. Chen L
    23. Cheng H
    24. Li Y

    (2020) An optimized acetylcholine sensor for monitoring in vivo cholinergic activity

    Nature Methods 17:1139–1146.

    https://doi.org/10.1038/s41592-020-0953-2

    • PubMed
    • Google Scholar
38. 1. Keene AC
    2. Stratmann M
    3. Keller A
    4. Perrat PN
    5. Vosshall LB
    6. Waddell S

    (2004) Diverse odor-conditioned memories require uniquely timed dorsal paired medial neuron output

    Neuron 44:521–533.

    https://doi.org/10.1016/j.neuron.2004.10.006

    • PubMed
    • Google Scholar
39. 1. Kim YC
    2. Lee HG
    3. Han KA

    (2007) D1 dopamine receptor dDA1 is required in the mushroom body neurons for aversive and appetitive learning in Drosophila

    The Journal of Neuroscience 27:7640–7647.

    https://doi.org/10.1523/JNEUROSCI.1167-07.2007

    • PubMed
    • Google Scholar
40. 1. Kitamoto T

    (2001) Conditional modification of behavior in Drosophila by targeted expression of a temperature-sensitive shibire allele in defined neurons

    Journal of Neurobiology 47:81–92.

    https://doi.org/10.1002/neu.1018

    • PubMed
    • Google Scholar
41. 1. König C
    2. Khalili A
    3. Ganesan M
    4. Nishu AP
    5. Garza AP
    6. Niewalda T
    7. Gerber B
    8. Aso Y
    9. Yarali A

    (2018) Reinforcement signaling of punishment versus relief in fruit flies

    Learning & Memory (Cold Spring Harbor, N.Y.) 25:247–257.

    https://doi.org/10.1101/lm.047308.118

    • PubMed
    • Google Scholar
42. 1. Li F
    2. Lindsey JW
    3. Marin EC
    4. Otto N
    5. Dreher M
    6. Dempsey G
    7. Stark I
    8. Bates AS
    9. Pleijzier MW
    10. Schlegel P
    11. Nern A
    12. Takemura S-Y
    13. Eckstein N
    14. Yang T
    15. Francis A
    16. Braun A
    17. Parekh R
    18. Costa M
    19. Scheffer LK
    20. Aso Y
    21. Jefferis GS
    22. Abbott LF
    23. Litwin-Kumar A
    24. Waddell S
    25. Rubin GM

    (2020) The connectome of the adult Drosophila mushroom body provides insights into function

    eLife 9:e62576.

    https://doi.org/10.7554/eLife.62576

    • PubMed
    • Google Scholar
43. 1. Liu X
    2. Davis RL

    (2009) The GABAergic anterior paired lateral neuron suppresses and is suppressed by olfactory learning

    Nature Neuroscience 12:53–59.

    https://doi.org/10.1038/nn.2235

    • PubMed
    • Google Scholar
44. 1. Liu C
    2. Plaçais PY
    3. Yamagata N
    4. Pfeiffer BD
    5. Aso Y
    6. Friedrich AB
    7. Siwanowicz I
    8. Rubin GM
    9. Preat T
    10. Tanimoto H

    (2012) A subset of dopamine neurons signals reward for odour memory in Drosophila

    Nature 488:512–516.

    https://doi.org/10.1038/nature11304

    • PubMed
    • Google Scholar
45. 1. Louis T
    2. Stahl A
    3. Boto T
    4. Tomchik SM

    (2018) Cyclic AMP-dependent plasticity underlies rapid changes in odor coding associated with reward learning

    PNAS 115:E448–E457.

    https://doi.org/10.1073/pnas.1709037115

    • PubMed
    • Google Scholar
46. 1. Lübbert M
    2. Goral RO
    3. Keine C
    4. Thomas C
    5. Guerrero-Given D
    6. Putzke T
    7. Satterfield R
    8. Kamasawa N
    9. Young SM Jr

    (2019) CaV2.1 alpha1 Subunit Expression Regulates Presynaptic CaV2.1 Abundance and Synaptic Strength at a Central Synapse

    Neuron 101:260–273.

    https://doi.org/10.1016/j.neuron.2018.11.028

    • Google Scholar
47. 1. Mao Z
    2. Davis RL

    (2009) Eight different types of dopaminergic neurons innervate the Drosophila mushroom body neuropil: anatomical and physiological heterogeneity

    Frontiers in Neural Circuits 3:5.

    https://doi.org/10.3389/neuro.04.005.2009

    • PubMed
    • Google Scholar
48. 1. McGuire SE
    2. Le PT
    3. Davis RL

    (2001) The role of Drosophila mushroom body signaling in olfactory memory

    Science (New York, N.Y.) 293:1330–1333.

    https://doi.org/10.1126/science.1062622

    • PubMed
    • Google Scholar
49. 1. McGuire SE
    2. Le PT
    3. Osborn AJ
    4. Matsumoto K
    5. Davis RL

    (2003) Spatiotemporal rescue of memory dysfunction in Drosophila

    Science (New York, N.Y.) 302:1765–1768.

    https://doi.org/10.1126/science.1089035

    • PubMed
    • Google Scholar
50. 1. Modi MN
    2. Shuai Y
    3. Turner GC

    (2020) The Drosophila Mushroom Body: From Architecture to Algorithm in a Learning Circuit

    Annual Review of Neuroscience 43:465–484.

    https://doi.org/10.1146/annurev-neuro-080317-0621333

    • PubMed
    • Google Scholar
51. 1. Müller M
    2. Davis GW

    (2012) Transsynaptic control of presynaptic Ca(2)(+) influx achieves homeostatic potentiation of neurotransmitter release

    Current Biology 22:1102–1108.

    https://doi.org/10.1016/j.cub.2012.04.018

    • PubMed
    • Google Scholar
52. 1. Nanou E
    2. Scheuer T
    3. Catterall WA

    (2016) Calcium sensor regulation of the CaV2.1 Ca2+ channel contributes to long-term potentiation and spatial learning

    PNAS 113:13209–13214.

    https://doi.org/10.1073/pnas.1616206113

    • PubMed
    • Google Scholar
53. 1. Otto N
    2. Pleijzier MW
    3. Morgan IC
    4. Edmondson-Stait AJ
    5. Heinz KJ
    6. Stark I
    7. Dempsey G
    8. Ito M
    9. Kapoor I
    10. Hsu J
    11. Schlegel PM
    12. Bates AS
    13. Feng L
    14. Costa M
    15. Ito K
    16. Bock DD
    17. Rubin GM
    18. Jefferis GSXE
    19. Waddell S

    (2020) Input Connectivity Reveals Additional Heterogeneity of Dopaminergic Reinforcement in Drosophila

    Current Biology 30:3200–3211.

    https://doi.org/10.1016/j.cub.2020.05.077

    • PubMed
    • Google Scholar
54. 1. Owald D
    2. Felsenberg J
    3. Talbot CB
    4. Das G
    5. Perisse E
    6. Huetteroth W
    7. Waddell S

    (2015) Activity of defined mushroom body output neurons underlies learned olfactory behavior in Drosophila

    Neuron 86:417–427.

    https://doi.org/10.1016/j.neuron.2015.03.025

    • PubMed
    • Google Scholar
55. 1. Perisse E
    2. Owald D
    3. Barnstedt O
    4. Talbot CB
    5. Huetteroth W
    6. Waddell S

    (2016) Aversive Learning and Appetitive Motivation Toggle Feed-Forward Inhibition in the Drosophila Mushroom Body

    Neuron 90:1086–1099.

    https://doi.org/10.1016/j.neuron.2016.04.034

    • PubMed
    • Google Scholar
56. 1. Perkins LA
    2. Holderbaum L
    3. Tao R
    4. Hu Y
    5. Sopko R
    6. McCall K
    7. Yang-Zhou D
    8. Flockhart I
    9. Binari R
    10. Shim H-S
    11. Miller A
    12. Housden A
    13. Foos M
    14. Randkelv S
    15. Kelley C
    16. Namgyal P
    17. Villalta C
    18. Liu L-P
    19. Jiang X
    20. Huan-Huan Q
    21. Wang X
    22. Fujiyama A
    23. Toyoda A
    24. Ayers K
    25. Blum A
    26. Czech B
    27. Neumuller R
    28. Yan D
    29. Cavallaro A
    30. Hibbard K
    31. Hall D
    32. Cooley L
    33. Hannon GJ
    34. Lehmann R
    35. Parks A
    36. Mohr SE
    37. Ueda R
    38. Kondo S
    39. Ni J-Q
    40. Perrimon N

    (2015) The Transgenic RNAi Project at Harvard Medical School: Resources and Validation

    Genetics 201:843–852.

    https://doi.org/10.1534/genetics.115.180208

    • PubMed
    • Google Scholar
57. 1. Pfeiffer BD
    2. Ngo T-TB
    3. Hibbard KL
    4. Murphy C
    5. Jenett A
    6. Truman JW
    7. Rubin GM

    (2010) Refinement of tools for targeted gene expression in Drosophila

    Genetics 186:735–755.

    https://doi.org/10.1534/genetics.110.119917

    • PubMed
    • Google Scholar
58. 1. Plaçais PY
    2. Trannoy S
    3. Friedrich AB
    4. Tanimoto H
    5. Preat T

    (2013) Two pairs of mushroom body efferent neurons are required for appetitive long-term memory retrieval in Drososphila

    Cell Reports 5:769–780.

    https://doi.org/10.1016/j.celrep.2013.09.032

    • PubMed
    • Google Scholar
59. 1. Pribbenow C
    2. Chen Y
    3. Heim MM
    4. Laber D
    5. Reubold S
    6. Reynolds E
    7. Balles I
    8. Grimalt RS
    9. Rauch C
    10. Rösner J
    11. Alquicira TFV
    12. Owald D

    (2021) Postsynaptic plasticity of cholinergic synapses underlies the induction and expression of appetitive memories in Drosophila

    Neuroscience 1:e450776.

    https://doi.org/10.1101/2021.07.01.450776

    • Google Scholar
60. 1. Rowan MJM
    2. DelCanto G
    3. Yu JJ
    4. Kamasawa N
    5. Christie JM

    (2016) Synapse-Level Determination of Action Potential Duration by K(+) Channel Clustering in Axons

    Neuron 91:370–383.

    https://doi.org/10.1016/j.neuron.2016.05.035

    • PubMed
    • Google Scholar
61. 1. Sayin S
    2. De Backer J-F
    3. Siju KP
    4. Wosniack ME
    5. Lewis LP
    6. Frisch L-M
    7. Gansen B
    8. Schlegel P
    9. Edmondson-Stait A
    10. Sharifi N
    11. Fisher CB
    12. Calle-Schuler SA
    13. Lauritzen JS
    14. Bock DD
    15. Costa M
    16. Jefferis GSXE
    17. Gjorgjieva J
    18. Grunwald Kadow IC

    (2019) A Neural Circuit Arbitrates between Persistence and Withdrawal in Hungry Drosophila

    Neuron 104:544–558.

    https://doi.org/10.1016/j.neuron.2019.07.028

    • PubMed
    • Google Scholar
62. 1. Scaplen KM
    2. Talay M
    3. Fisher JD
    4. Cohn R
    5. Sorkaç A
    6. Aso Y
    7. Barnea G
    8. Kaun KR

    (2021) Transsynaptic mapping of Drosophila mushroom body output neurons

    eLife 10:e63379.

    https://doi.org/10.7554/eLife.63379

    • PubMed
    • Google Scholar
63. 1. Scheffer LK
    2. Xu CS
    3. Januszewski M
    4. Lu Z
    5. Takemura S-Y
    6. Hayworth KJ
    7. Huang GB
    8. Shinomiya K
    9. Maitlin-Shepard J
    10. Berg S
    11. Clements J
    12. Hubbard PM
    13. Katz WT
    14. Umayam L
    15. Zhao T
    16. Ackerman D
    17. Blakely T
    18. Bogovic J
    19. Dolafi T
    20. Kainmueller D
    21. Kawase T
    22. Khairy KA
    23. Leavitt L
    24. Li PH
    25. Lindsey L
    26. Neubarth N
    27. Olbris DJ
    28. Otsuna H
    29. Trautman ET
    30. Ito M
    31. Bates AS
    32. Goldammer J
    33. Wolff T
    34. Svirskas R
    35. Schlegel P
    36. Neace E
    37. Knecht CJ
    38. Alvarado CX
    39. Bailey DA
    40. Ballinger S
    41. Borycz JA
    42. Canino BS
    43. Cheatham N
    44. Cook M
    45. Dreher M
    46. Duclos O
    47. Eubanks B
    48. Fairbanks K
    49. Finley S
    50. Forknall N
    51. Francis A
    52. Hopkins GP
    53. Joyce EM
    54. Kim S
    55. Kirk NA
    56. Kovalyak J
    57. Lauchie SA
    58. Lohff A
    59. Maldonado C
    60. Manley EA
    61. McLin S
    62. Mooney C
    63. Ndama M
    64. Ogundeyi O
    65. Okeoma N
    66. Ordish C
    67. Padilla N
    68. Patrick CM
    69. Paterson T
    70. Phillips EE
    71. Phillips EM
    72. Rampally N
    73. Ribeiro C
    74. Robertson MK
    75. Rymer JT
    76. Ryan SM
    77. Sammons M
    78. Scott AK
    79. Scott AL
    80. Shinomiya A
    81. Smith C
    82. Smith K
    83. Smith NL
    84. Sobeski MA
    85. Suleiman A
    86. Swift J
    87. Takemura S
    88. Talebi I
    89. Tarnogorska D
    90. Tenshaw E
    91. Tokhi T
    92. Walsh JJ
    93. Yang T
    94. Horne JA
    95. Li F
    96. Parekh R
    97. Rivlin PK
    98. Jayaraman V
    99. Costa M
    100. Jefferis GS
    101. Ito K
    102. Saalfeld S
    103. George R
    104. Meinertzhagen IA
    105. Rubin GM
    106. Hess HF
    107. Jain V
    108. Plaza SM

    (2020) A connectome and analysis of the adult Drosophila central brain

    eLife 9:e57443.

    https://doi.org/10.7554/eLife.57443

    • PubMed
    • Google Scholar
64. 1. Schroll C
    2. Riemensperger T
    3. Bucher D
    4. Ehmer J
    5. Völler T
    6. Erbguth K
    7. Gerber B
    8. Hendel T
    9. Nagel G
    10. Buchner E
    11. Fiala A

    (2006) Light-induced activation of distinct modulatory neurons triggers appetitive or aversive learning in Drosophila larvae

    Current Biology 16:1741–1747.

    https://doi.org/10.1016/j.cub.2006.07.023

    • PubMed
    • Google Scholar
65. 1. Schwaerzel M
    2. Monastirioti M
    3. Scholz H
    4. Friggi-Grelin F
    5. Birman S
    6. Heisenberg M

    (2003)

    Dopamine and octopamine differentiate between aversive and appetitive olfactory memories in Drosophila

    The Journal of Neuroscience 23:10495–10502.

    • PubMed
    • Google Scholar
66. 1. Séjourné J
    2. Plaçais P-Y
    3. Aso Y
    4. Siwanowicz I
    5. Trannoy S
    6. Thoma V
    7. Tedjakumala SR
    8. Rubin GM
    9. Tchénio P
    10. Ito K
    11. Isabel G
    12. Tanimoto H
    13. Preat T

    (2011) Mushroom body efferent neurons responsible for aversive olfactory memory retrieval in Drosophila

    Nature Neuroscience 14:903–910.

    https://doi.org/10.1038/nn.2846

    • PubMed
    • Google Scholar
67. 1. Silva B
    2. Molina-Fernández C
    3. Ugalde MB
    4. Tognarelli EI
    5. Angel C
    6. Campusano JM

    (2015) Muscarinic ACh Receptors Contribute to Aversive Olfactory Learning in Drosophila

    Neural Plasticity 2015:658918.

    https://doi.org/10.1155/2015/658918

    • PubMed
    • Google Scholar
68. 1. Springer M
    2. Nawrot MP

    (2021) A Mechanistic Model for Reward Prediction and Extinction Learning in the Fruit Fly

    ENeuro 8:ENEURO.0549-20.2021.

    https://doi.org/10.1523/ENEURO.0549-20.2021

    • PubMed
    • Google Scholar
69. 1. Takemura S-Y
    2. Aso Y
    3. Hige T
    4. Wong A
    5. Lu Z
    6. Xu CS
    7. Rivlin PK
    8. Hess H
    9. Zhao T
    10. Parag T
    11. Berg S
    12. Huang G
    13. Katz W
    14. Olbris DJ
    15. Plaza S
    16. Umayam L
    17. Aniceto R
    18. Chang L-A
    19. Lauchie S
    20. Ogundeyi O
    21. Ordish C
    22. Shinomiya A
    23. Sigmund C
    24. Takemura S
    25. Tran J
    26. Turner GC
    27. Rubin GM
    28. Scheffer LK

    (2017) A connectome of a learning and memory center in the adult Drosophila brain

    eLife 6:e26975.

    https://doi.org/10.7554/eLife.26975

    • PubMed
    • Google Scholar
70. 1. Tanaka NK
    2. Tanimoto H
    3. Ito K

    (2008) Neuronal assemblies of the Drosophila mushroom body

    The Journal of Comparative Neurology 508:711–755.

    https://doi.org/10.1002/cne.21692

    • PubMed
    • Google Scholar
71. 1. Tanimoto H
    2. Heisenberg M
    3. Gerber B

    (2004) Experimental psychology: event timing turns punishment to reward

    Nature 430:983.

    https://doi.org/10.1038/430983a

    • PubMed
    • Google Scholar
72. 1. Taufiq AM
    2. Fujii S
    3. Yamazaki Y
    4. Sasaki H
    5. Kaneko K
    6. Li J
    7. Kato H
    8. Mikoshiba K

    (2005) Involvement of IP3 receptors in LTP and LTD induction in guinea pig hippocampal CA1 neurons

    Learning & Memory (Cold Spring Harbor, N.Y.) 12:594–600.

    https://doi.org/10.1101/lm.17405

    • PubMed
    • Google Scholar
73. 1. Tomchik SM

    (2013) Dopaminergic neurons encode a distributed, asymmetric representation of temperature in Drosophila

    The Journal of Neuroscience 33:2166–76a.

    https://doi.org/10.1523/JNEUROSCI.3933-12.2013

    • PubMed
    • Google Scholar
74. 1. Tully T
    2. Quinn WG

    (1985) Classical conditioning and retention in normal and mutant Drosophila melanogaster

    Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 157:263–277.

    https://doi.org/10.1007/BF01350033

    • PubMed
    • Google Scholar
75. 1. Wang Y
    2. Mamiya A
    3. Chiang A
    4. Zhong Y

    (2008) Imaging of an early memory trace in the Drosophila mushroom body

    The Journal of Neuroscience 28:4368–4376.

    https://doi.org/10.1523/JNEUROSCI.2958-07.2008

    • PubMed
    • Google Scholar
76. 1. Wu C-L
    2. Shih M-FM
    3. Lee P-T
    4. Chiang A-S

    (2013) An octopamine-mushroom body circuit modulates the formation of anesthesia-resistant memory in Drosophila

    Current Biology 23:2346–2354.

    https://doi.org/10.1016/j.cub.2013.09.056

    • PubMed
    • Google Scholar
77. 1. Wu JK
    2. Tai CY
    3. Feng KL
    4. Chen SL
    5. Chen CC
    6. Chiang AS

    (2017) Long-term memory requires sequential protein synthesis in three subsets of mushroom body output neurons in Drosophila

    Scientific Reports 7:7112.

    https://doi.org/10.1038/s41598-017-07600-2

    • PubMed
    • Google Scholar
78. Preprint

    1. Xu CS
    2. Januszewski M
    3. Lu Z
    4. Takemura SY
    5. Hayworth K
    6. Huang G
    7. Shinomiya K
    8. Maitin-Shepard J
    9. Ackerman D
    10. Berg S

    (2020) A Connectome and Analysis of the Adult Drosophila Central Brain

    bioRxiv.

    https://doi.org/10.1101/2020.04.07.030213

    • Google Scholar
79. 1. Yamagata N
    2. Ichinose T
    3. Aso Y
    4. Plaçais PY
    5. Friedrich AB
    6. Sima RJ
    7. Preat T
    8. Rubin GM
    9. Tanimoto H

    (2015) Distinct dopamine neurons mediate reward signals for short- and long-term memories

    PNAS 112:578–583.

    https://doi.org/10.1073/pnas.1421930112

    • PubMed
    • Google Scholar
80. 1. Yamagata N.
    2. Hiroi M
    3. Kondo S
    4. Abe A
    5. Tanimoto H

    (2016) Suppression of Dopamine Neurons Mediates Reward

    PLOS Biology 14:e1002586.

    https://doi.org/10.1371/journal.pbio.1002586

    • PubMed
    • Google Scholar
81. 1. Yu D
    2. Ponomarev A
    3. Davis RL

    (2004) Altered representation of the spatial code for odors after olfactory classical conditioning; memory trace formation by synaptic recruitment

    Neuron 42:437–449.

    https://doi.org/10.1016/s0896-6273(04)00217-x

    • PubMed
    • Google Scholar
82. 1. Zamponi GW
    2. Currie KPM

    (2013) Regulation of Ca(V)2 calcium channels by G protein coupled receptors

    Biochimica et Biophysica Acta 1828:1629–1643.

    https://doi.org/10.1016/j.bbamem.2012.10.004

    • PubMed
    • Google Scholar
83. 1. Zars T
    2. Fischer M
    3. Schulz R
    4. Heisenberg M

    (2000) Localization of a short-term memory in Drosophila

    Science (New York, N.Y.) 288:672–675.

    https://doi.org/10.1126/science.288.5466.672

    • PubMed
    • Google Scholar
84. 1. Zhang S
    2. Roman G

    (2013) Presynaptic inhibition of gamma lobe neurons is required for olfactory learning in Drosophila

    Current Biology 23:2519–2527.

    https://doi.org/10.1016/j.cub.2013.10.043

    • PubMed
    • Google Scholar
85. 1. Zhang X
    2. Noyes NC
    3. Zeng J
    4. Li Y
    5. Davis RL

    (2019) Aversive Training Induces Both Presynaptic and Postsynaptic Suppression in Drosophila

    The Journal of Neuroscience 39:9164–9172.

    https://doi.org/10.1523/JNEUROSCI.1420-19.2019

    • PubMed
    • Google Scholar
86. 1. Zhao X
    2. Lenek D
    3. Dag U
    4. Dickson BJ
    5. Keleman K

    (2018) Persistent activity in a recurrent circuit underlies courtship memory in Drosophila

    eLife 7:e31425.

    https://doi.org/10.7554/eLife.31425

    • PubMed
    • Google Scholar
87. 1. Zhou M
    2. Chen N
    3. Tian J
    4. Zeng J
    5. Zhang Y
    6. Zhang X
    7. Guo J
    8. Sun J
    9. Li Y
    10. Guo A
    11. Li Y

    (2019) Suppression of GABAergic neurons through D2-like receptor secures efficient conditioning in Drosophila aversive olfactory learning

    PNAS 116:5118–5125.

    https://doi.org/10.1073/pnas.1812342116

    • PubMed
    • Google Scholar

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting the paper "Associative learning drives longitudinally-graded presynaptic plasticity of neurotransmitter release along axonal compartments" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and a Senior Editor. The reviewers have opted to remain anonymous.

Comments to the Authors:

We are sorry to say that, after consultation with the reviewers, we have decided to reject the current manuscript because of several substantial issues that were raised by the reviewers. Although we believe that these issues can be addressed experimentally and through text changes, we do not think that this substantial amount of work can be carried out within 2 months, the time eLife allows for revisions. As you will see, the reviewers were generally supportive of publication, but identified several key points such as (1) lack of behavioral experiments for the Cac knock-down experiments, (2) specific controls for some of the imaging experiments, (3) consideration of the role of dopaminergic neurons and (4) a lack of acknowledgment of the complexity of the mushroom body circuit and the literature that has addressed this previously. If the authors decide to address these issues, we are willing to consider a future manuscript, but it would be considered as a new submission.

Reviewer #1:

The paper aims to understand if and how axonal compartmentalization functions in the MB γ lobe in the context of olfactory learning. Further the authors focus on these process and find distinct molecular mechanisms underlying appetitive and aversive learning (cac and IP3). Lastly the authors focus on how this compartmentalization influences the downstream MBON function.

The paper is written well and straightforward in its approach and implications. The neuroanatomy, experimental details are clearly presented and I would rate this paper very high on a readability scale. The imaging and behavioral approaches are appropriate for the research question and address the limitations from a technical perspective. Results are presented well, but authors don't dwell on the results much before transitioning into another part of the question they seek to ask. This undermines the complexity of the MB circuits and an effort should be made to address dynamics of other lobes that underlie these behaviors. One issue is that even though the experiments work well together in supporting the results they fail to incorporate more complex dynamics of other lobes or consider the EM connectivity which might complicate the simplistic interpretations. The role of dopamine in this compartmental logic has not been directly addressed which potentially plays into this circuit.

This is impactful work as it addresses the pre-synaptic dynamics of MB with a focus on ACh release across compartments and how they differ between two forms of learning. The use of Grab Ach is a big highlight of the paper as this question could not have been asked without this tool. The authors also address the regulation of Ca²⁺ responses in the compartments that result in altered responses in valence-coding MBONs. Its also interesting to see the switch in γ2 and γ3 dynamics between appetitive and aversive learning. Lastly the match of Ca²⁺ responses in MBONs with the Ach compartmentalization highlights the behavioral relevance of axonal compartmentalization.

I like this paper as the experiments, results are clear and authors are very careful about interpreting the results. On a few occasions I feel that interpretations are over simplified and a focus on elaborating the complexities will be more helpful for the field.

Below are a few suggestions:

1) The use of Gal4 lines seems to be inconsistent between figures. The rationale for choosing a specific targeting tool should be made explicit.

2) The olfactory learning literature has many odor pairs and the one used here (ethyl butyrate and IAA) is not the most common pair. Do learning scores, neuroanatomy involved differ based on chosen odor pairs. Are learning rules generalizable across odor pairs.

3) Figure 1: Having some images from the rig set up would help in seeing how authors chose ROI's for the different compartments. The schematics are great but dont give a good sense for the real imaging resolution.

4) Figure 3: A schematic here like figure 1K would be great to align the reader. Also the n's here are half of figure 1. Is there a specific reason for that? How was power determined for these experiments.

5) Figure 4: It is not clear why γ2 is the only compartment shown here.

6) The compartmentalized pre-synaptic plasticity mapping/mirroring the MBON Ca++ response is a great result but without knocking down Ach receptor in these MBONs I am not sure a direct link between Ach release and MBON Ca²⁺ response cannot be made.

7) The above concern is somewhat addressed for γ3 as they mediate timing and MBON from γ3 are required for appetitive learning but is not addressed for other compartments and MBONs.

8) While, PAM and PPL1 neurons have been briefly discussed their role in modulating KC and MBON plasticity is not addressed experimentally which goes against what is know about this circuit. A more elaborate discussion would be beneficial. How do dopaminergic circuits associated with γ4 interact with this circuit?

Overall I find that Ach compartmentalization is well justified but the downstream circuit connectivity and plasticity seems unclear. Without addressing the role of Ach receptors it is difficult to see the causal relationship between KC Ach release and MBON activity.

Reviewer #2:

Stahl et al. investigate presynaptic Kenyon cell plasticity in the context of appetitive and aversive memories, using a genetically-encoded acetylcholine receptor-based sensor as primary read-out (instead of calcium indicators used in several previous studies). Therefore, the authors investigate one step downstream of the classically conducted calcium (or cAMP, etc.) imaging experiments – the level of neurotransmitter release. Likewise, this is one step upstream of the other widely-used readout of the postsynaptic MBON activity (or dopaminergic neurons). The authors investigate contributions of CS+ and/or CS- plasticity throughout the compartments of the γ lobe. All in all, this manuscript confirms several previous studies that investigated individual compartment plasticity (often plasticity is measured at the level of the MBONs and therefore in single compartments), and taken together with other recent publications, this study can be seen as a valuable compendium (as it addresses appetitive and aversive short-term memory protocols in the context of several compartments). As the authors address an important step between presynaptic and postsynaptic calcium transients, their work actually confirms that conclusions deduced from changed calcium transients correlate to neurotransmitter release, as expected. This study also addresses differential effects on the CS+ and the CS-, which is important, as previous studies often concentrated (partially for technical reasons) on either the CS+ specifically, or the relative changes of CS+ to CS- only. The authors further investigate the role of the active zone-localized voltage gated calcium channel cacophony in memory-related plasticity – this part needs to be strengthened by additional controls.

1. As it stands, I am not sure what to read from the Cac knock-down experiments for several reasons. First, it is entirely unclear how efficient the knock-down works. Would the amount of Cac molecules be reduced from e.g., 20 to 15 per active zone, or less, or more (the authors should quantify cacophony following knock-down)? Second, if baseline transmission is not affected (which is remarkable to a certain degree – the authors do mention potential 'residual' Cac expression in the discussion), what about e.g., high frequency odor puff protocols (again the authors mention plasticity protocols in the discussion that may help tying the observed effects to Cac). Can the authors perform additional manipulations that should phenocopy the reduction of Cac (e.g., reduction of Brp via RNAi)? Can the phenotype be modulated by titrating extracellular calcium concentrations? Would basal neurotransmission be affected when extending the time at 32{degree sign}C (the range of 4-10 days could also be quite large, depending on the turnover time for cacophony). Third, can the authors exclude that the observed effects were due to dendritic functions of Cac?

The authors need to perform additional experiments in order to manifest this important claim.

2. What is the behavioral consequence of Cac or IP3R knock-down in Kenyon cells? Is memory impaired?

3. The authors actually do not demonstrate that the acetylcholine measured is released from MB neurons (Kenyon cells) and do not exclude the possibility that the measured acetylcholine could derive from an unknown other source. The authors could address this issue by Kenyon cell-specific knock-down of VAChT or ChaT. Alternatively, acute or chronic block of KCs, while measuring sensor activation, could be an option. I am puzzled by the wording in line 80: is 'putative' meant as 'generally accepted' or 'likely'? I would strongly recommend using a less ambiguous word in this context. If not generally accepted (if so, please do explain to the reader why), this would somewhat compromise the basis of this study.

4. Figure 1 and others: Why are only the means compared statistically? The pre to post in C and E should be analysed using paired statistics – as in Figure 3E/F or S4 for example. In Figure 1: are any changes detected within flies? Figure S1 suggest that this is not the case – the change would only become apparent between conditions? Although the authors have already done a good job in illustrating their protocols, data display should be more uniform potential discrepancies should be spelt out, and claims should be adjusted accordingly (e.g., summary in K).

5. Line 189: why a trend towards decrease? Figure S4 suggests a statistically significant change.

6. Absolute conditioning protocols: Is the comparison of the 'no CS-' and 'no CS+' really suited to claim that 'no CS-' leads to decreased acetylcholine release? It appears that this would only be consistent if the 'no CS-' group was significantly different from the 'no US' group. Please clarify putative discrepancies. In this context, please also discuss the results in S5 in more detail.

7. Figure S4 shows

a. gamma1: CS+ potentiation (not seen in Figure S1?), not apparent in Cac RNAi

b. gamma2: no potentiation for the CS+ (unlike what is reported in Figure 1), yet, consistently a depression for the CS-

c. gamma3: no potentiation of CS+, yet a depression of CS-, both inconsistent with Figure 1

d. Several forms of plasticity, also to the non-trained odor in controls and Cac RNAi

Please clarify and discuss the individual findings.

8. Line 467: It is first stated that odor was presented continuously for 30s. In the next sentence the authors write that six 1-s odor pulses with a 5s inter-pulse interval were given. Please clarify. Also, the odor responses appear to typically last for about 5-6s. Obviously, this is in parts based on the kinetics of the reporter. However, would the authors not expect that continuous odor application will lead to desensitization at the level of the olfactory receptor neurons?

9. Apart from showing that postsynaptic calcium signals are altered in MBONs the authors should consider showing whether the observed effects seen with the acetylcholine indicator can also be seen when expressed in MBONs or DANs (and exclude the possibility that the potentiation/depression could be specific to KC-DAN or KC-KC synapses).

Reviewer #3:

This study shows learning-induced compartmentalized plasticity in Kenyon cells (KCs) of the mushroom body γ lobe using a genetically encoded acetylcholine sensor. The authors made several major discoveries that the plasticity is bidirectional and depends on memory valence. Interestingly, the compartments along the axon terminals of the KCs undergo spatial 'gradient' of plasticity. Knock-down of the genes that differentially regulate intracellular calcium induced distinct effects on learning-induced plasticity. These results explain, at least partly, that learning-induced plasticity at KC-MBON synapses takes place on the presynaptic side. The experiments and analysis are overall thorough, and conclusions are generally supported by the results. In the following, I list some flaws to be addressed that concern necessary controls, more careful interpretation of the statistics and results, and the choice of the cac RNAi strain.

- Odor-only control is imperfect to formally distinguish if the plasticity arises from CS+ and/or CS- changes. The effect of sugar or shock presentation per se must be measured to make this distinction. Alternatively, the authors can examine how much of the 'plasticity' is explained by unpaired presentations of CS and US.

- Conditioning effects at some compartments are on the edge of statistical significance and look inconsistent among different experimental series (e.g. Figure 1F-J vs. Figure 3F). The authors need to be cautious to interpret these statistical results. For example, knock-down of cac impaired CS- depression in y2 and y3 as well (cf. Figure 1G, H, and K), and this is contradictory to their conclusion that cac is required only for potentiation. Similarly, itpr seem to globally affect presynaptic activities.

- Some of the major findings here have been consistent with previous papers with different measurements (e.g. Bilz et al. 2020 for Figure 2). On the other hand, this study presents results not coherent with existing studies. For example, previous papers of the authors found rather homogeneous calcium increase (Boto 2014, Louis 2018), while plasticity in this study is graded along the y lobe axons and opposite in appetitive and aversive conditioning (Figure 1, 2). MBON-y3 has been reported to be dispensable (Aso 2014 and Ichinose 2021) unlike the results in Figure 6E. The authors need to provide discussion that explain the coherence/apparent inconsistencies.

- The choice of this particular cac RNAi strain is not clear, besides no control experiment for off-target effects. This is technically important, since voltage-gated calcium channels must be required more than learning-dependent potentiation. Therefore, it is blunt to conclude:

"synaptic exocytosis remained intact (L191)" upon cac knock-down;

"presynaptic potentiation, but not depression, requires the voltage-sensitive CaV2 Ca²⁺ channel cacophony across the MB compartments (L198)";

"Overall, these data demonstrate that Cac is not required for learning-induced depression of ACh release (L207)".

These conclusions need to be moderated.

Also in this context, mentioning a recent paper by Hidalgo et al. (Neurobiology of Disease, 2021) that analyzed the effect of cac-KD in KCs on aversive learning would be complementary to the physiological finding of this study.

- Compartmentalized and bidirectional plasticity in the MBONs has been repeatedly reported (e.g. Cohn, 2015; Owald, 2015). What is the intention to revisit this (Figure 5)? Also, this is essentially a detour from their main argument about the presynaptic mechanism. Consider revising the Figure or moving it to the figure supplement for readability. Similarly, it's not clear to me what Figure 6 adds to the overall conclusion.

- (Figure 6E) Did the authors examine MBON blocking in discriminative training? This needs to be clearly described. If the authors claim that the y3 compartment is important in discriminative training, they would need to show no defects in single-odor conditioning as in Figure 6A.

- Selection of ROIs for different MBONs in Figure 5 need stronger rationale. Why do the authors choose neurites for some compartments and not for others? I guess there is transformation of calcium signals in different parts of MBONs.

- Describe how the authors defined compartments in the g-KCs?

- I'm not totally sure how the results in Figure 6 support "temporal comparison (L275)". Either elaborate this conclusion more or provide an alternative interpretation.

- The quantified compartment for Figure 4A-D should be clearly stated in the text and legend, instead of simply showing y2 in 4E and let us guess.

- Is "post-conditioning odor contrast" appropriate for explaining itpr effects given the CS+ depression is seemingly exaggerated (Figure 4)?

- L256 Sentence and/or figure citation are somewhat inconsistent.

[Editors’ note: further revisions were suggested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "Associative learning drives longitudinally-graded presynaptic plasticity of neurotransmitter release along axonal compartments" for further consideration by eLife. Your revised article has been evaluated by Gary Westbrook (Senior Editor) and a Reviewing Editor. The manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:

The reviewers were overall happy with the revisions, but both raised a few points that need to be addressed before a final decision can be reached. Please pay attention in particular to the points 1. and 2. raised by reviewer 2. We believe these can be addressed by editing the text and toning down relevant statements.

Reviewer #1:

The authors have done a good job with this revised manuscript. They have addressed my previous concerns either by adding new experiments or clarifying passages in the text. Especially the role of Cac in appetitive learning and the more clear description of their analysis pipeline are convincing. Some experiments suggested, were technical not achievable with the presented techniques, and pursuing alternate methods would exceed the scope of this study.

Reviewer #2:

I've read the revised manuscript, and found it much improved. However, some of my concerns were not addressed. The authors need to clarify the following points before publication.

1. The odor-only control accounts for the effect of non-associative olfactory adaptation. But only that. The authors should be aware that other types of non-associative plasticity taking place during learning; e.g. reduced odor acuity by the previous exposure to electric shock (Preat, J Neurosci, 1998). Thus, inclusion of the US-only and/or unpaired training controls is mandatory to formally determine the absolute plasticity effect (e.g. potentiation of CS+ or depression CS- are indistinguishable; Figures 1L and 2K). Otherwise, they should moderate these claims, and focus only on the relative effects (i.e. "CS+:CS-" in the same figures)..

2. As demonstrated by the authors, Cac-KD reduces only ~29% of total Cac RNA level, suggesting that large body of Cac expression is unaffected. It is quite questionable that Cac is not required for learning-induced depression of Ach release (P10, L234) based on this partial down-regulation.

3. Somehow related to the previous point. Is it reasonable to conclude that Cac is required for learning-dependent potentiation? There is potentiation of CS- odor response in the γ 5 compartment after appetitive learning (Figure 3- sup1E)?

[Editors’ note: the authors resubmitted a revised version of the paper for consideration. What follows is the authors’ response to the first round of review.]

Reviewer #1:

[…]

I like this paper as the experiments, results are clear and authors are very careful about interpreting the results. On a few occasions I feel that interpretations are over simplified and a focus on elaborating the complexities will be more helpful for the field.

Thank you for the constructive feedback – we attempted to balance a clean focus on the results at hand with discussion of the many other impressive recent studies illuminating structural and functional aspects of the MB & associated circuitry. As suggested, in the present version, we have further elaborated on the complexities of these circuits to provide the readers with a more comprehensive picture. This is woven throughout the manuscript, particularly the discussion (e.g., pages 22‐23).

Below are a few suggestions:

1) The use of Gal4 lines seems to be inconsistent between figures. The rationale for choosing a specific targeting tool should be made explicit.

It has now been made explicit in the manuscript text (in the Methods, page 24, and main text, pages 5,9). We used two Gal4 lines that express in KCs (including γ KCs): R13F02 and 238Y. Initial ACh imaging experiments were carried out using 238Y to drive GRAB‐ACh expression; this provides maximal KC coverage, labeling 1898/~2000 KCs (Aso et al., 2009, J Neurogenet 23:156‐72). However, 238Y also labels a relatively large number of other cell types outside the MB (ibid). We often observe lethality when driving various RNAi lines with this driver, likely due to the non‐KC expression in the driver. In addition, the driver produces developmental defects when expressing various RNAi lines throughout development, reinforcing the well‐known potential for such issues when expression is not restricted to adulthood. Therefore, to mitigate these risks the molecular experiments, we switched to a more KC‐selective driver, R13F02‐Gal4, and restricted the expression of the RNAi (and GRAB‐ACh) to adulthood via Gal80^ts. The number of KCs labeled by R13F02‐Gal4 had not been documented (to our knowledge). As this is a relevant parameter for the comparison of results across drivers, we labeled and counted them, finding that the driver labels 745 ± 47 KCs; this is now included in the manuscript. Overall, the use of these two drivers/strategies provides complementary data sets, with 238Y providing insight into the plasticity across KCs and compartments, with the broadest possible KC coverage, and R13F02 providing more selective KC manipulations in the molecular experiments.

2) The olfactory learning literature has many odor pairs and the one used here (ethyl butyrate and IAA) is not the most common pair. Do learning scores, neuroanatomy involved differ based on chosen odor pairs. Are learning rules generalizable across odor pairs.

Indeed, many odor pairs have been used successfully, including various pairs and binary mixtures of 3octanol, 4‐methylcyclohexanol (Mch), benzaldehyde, 1‐octen‐3‐ol (Oct), pentyl acetate, butyl acetate, ethyl butyrate (EB), and isoamyl acetate (IA) (e.g., Barth et al., 2014 J Neurosci 34: 1819‐1837). We have previously used EB and IA as an odor pair for behavior olfactory associative conditioning (Boto et al., 2019 Cell Rep 27: 2014–2021) and for olfactory stimuli in imaging experiments (Boto et al., 2014 24: 822‐831; Louis et al., 2018 PNAS 115: E448‐E457). Performance indexes in behavioral conditioning experiments are equivalent to another odor pair (Mch vs. Oct) (Boto et al., 2019), and we see no substantive difference in odor‐evoked responses across KCs. Learning rules generalize across odors. Thus, by all available evidence, this odor pair is equally representative of odor space to any other.

3) Figure 1: Having some images from the rig set up would help in seeing how authors chose ROI's for the different compartments. The schematics are great but dont give a good sense for the real imaging resolution.

Agreed – an image is now included (Figure 1B).

4) Figure 3: A schematic here like figure 1K would be great to align the reader.

We appreciate the suggestion ‐ a schematic diagram is now included for figure 3, as well as figures 2 and 4.

Also the n's here are half of figure 1. Is there a specific reason for that? How was power determined for these experiments.

Sample sizes for aversive conditioning were based on prior published studies that detected positive effects (e.g., Louis et al., 2018). For appetitive conditioning/imaging experiments, we ran a priori power analysis to determine appropriate sample sizes using G*Power 3.1, with the following criteria: two‐tailed, α = 0.05, Power (1‐β) = 0.8, moderate effect size (α = 0.5), arriving at a total n ≥ 26 (we treated this as the minimum and ran 27).

5) Figure 4: It is not clear why γ2 is the only compartment shown here.

In figure 4, we focused on one compartment (in the main figure) to cleanly convey the conclusions in a representative compartment. Comparing Cac and IP3 RNAi, showing pre/post quantifications for CS+, CS‐, EB, and IA and box plots for one compartment introduces a relatively large number of conditions for the reader to track. Multiplying this five times over made for an unwieldy figure (we tried in earlier draft versions). For that level of detail, we direct the readers to Figure 4 – Supplement 1, which includes all compartments (stretching to 30 panels). Supplemental Figures appear side‐by‐side with a single click to toggle the view on the eLife web site upon publication, so the readers will be able to readily view all of the data. Nonetheless, to add spatial detail to the main Figure 4, we have now included a diagram of the plasticity effects across compartments (Figure 4, panels C,F,I).

6) The compartmentalized pre-synaptic plasticity mapping/mirroring the MBON Ca++ response is a great result but without knocking down Ach receptor in these MBONs I am not sure a direct link between Ach release and MBON Ca²⁺ response cannot be made.

These experiments verify that plasticity drives responses in MBONs in the same direction as the KCs. Plasticity in MBONs likely reflects both pre‐ and post‐synaptic processes (e.g., Pribbenow et al., 2021, bioRxiv doi: 10.1101/2021.07.01.450776) (we now discuss this aspect further in the manuscript). Knockdown of the postsynaptic ACh receptors in this context would remove the input to MBONs but not isolate the effects of ACh release. The best way to do so is to monitor ACh release and manipulate presynaptic processes (e.g., presynaptic Cac Cav2 channels), which is the approach we selected. We have now further enhance the rigor of these experiments by blocking synaptic release from the KCs with Shibire^ts (Figure 4 – Supplement 3), which significantly inhibited the odor‐evoked ACh signal.

7) The above concern is somewhat addressed for γ3 as they mediate timing and MBON from γ3 are required for appetitive learning but is not addressed for other compartments and MBONs.

We focused on the γ3 MBONs due to their novelty in terms of [presynaptic] CS‐ effects and the relative paucity of studies examining their roles in behavioral conditioning. In this updated manuscript, we have removed the timing experiments, focusing on the fundamental requirement for the γ3 MBONs in appetitive conditioning (γ2α’1 MBONs were also tested behaviorally).

8) While, PAM and PPL1 neurons have been briefly discussed their role in modulating KC and MBON plasticity is not addressed experimentally which goes against what is know about this circuit. A more elaborate discussion would be beneficial. How do dopaminergic circuits associated with γ4 interact with this circuit?

We have incorporated more discussion of these neurons in bidirectional plasticity in the discussion.

Overall I find that Ach compartmentalization is well justified but the downstream circuit connectivity and plasticity seems unclear. Without addressing the role of Ach receptors it is difficult to see the causal relationship between KC Ach release and MBON activity.

Our focus in this manuscript is on presynaptic plasticity in ACh release in KCs. As the reviewer notes, there is plasticity across multiple loci, including the MBONs. We added citations to additional studies to incorporate this point. In addition, we have now bolstered the experiments testing the presynaptic release mechanisms in several ways. We note that postsynaptic plasticity is being addressed in a complementary study from another group (Pribbenow et al., 2021).

Reviewer #2:

Stahl et al. investigate presynaptic Kenyon cell plasticity in the context of appetitive and aversive memories, using a genetically-encoded acetylcholine receptor-based sensor as primary read-out (instead of calcium indicators used in several previous studies). Therefore, the authors investigate one step downstream of the classically conducted calcium (or cAMP, etc.) imaging experiments – the level of neurotransmitter release. Likewise, this is one step upstream of the other widely-used readout of the postsynaptic MBON activity (or dopaminergic neurons). The authors investigate contributions of CS+ and/or CS- plasticity throughout the compartments of the γ lobe. All in all, this manuscript confirms several previous studies that investigated individual compartment plasticity (often plasticity is measured at the level of the MBONs and therefore in single compartments), and taken together with other recent publications, this study can be seen as a valuable compendium (as it addresses appetitive and aversive short-term memory protocols in the context of several compartments). As the authors address an important step between presynaptic and postsynaptic calcium transients, their work actually confirms that conclusions deduced from changed calcium transients correlate to neurotransmitter release, as expected. This study also addresses differential effects on the CS+ and the CS-, which is important, as previous studies often concentrated (partially for technical reasons) on either the CS+ specifically, or the relative changes of CS+ to CS- only. The authors further investigate the role of the active zone-localized voltage gated calcium channel cacophony in memory-related plasticity – this part needs to be strengthened by additional controls.

1. As it stands, I am not sure what to read from the Cac knock-down experiments for several reasons. First, it is entirely unclear how efficient the knock-down works. Would the amount of Cac molecules be reduced from e.g., 20 to 15 per active zone, or less, or more (the authors should quantify cacophony following knock-down)? Second, if baseline transmission is not affected (which is remarkable to a certain degree – the authors do mention potential 'residual' Cac expression in the discussion), what about e.g., high frequency odor puff protocols (again the authors mention plasticity protocols in the discussion that may help tying the observed effects to Cac).

The reviewer raises valid points about the need for more detail on the RNAi strategy and further characterization of the knockdown. The goal was to knock down Cac enough to test its role in plasticity, but not completely disrupt synaptic transmission. This is similar in approach to experiments previously carried out at the neuromuscular junction (Frank et al., 2006, Neuron 52:663‐77; Muller et al., 2012, Curr Biol 22:1102‐8). We chose an RNAi line that drives moderate knockdown (Brusich et al., 2015, Front. Cell. Neurosci 9:107) and expressed it conditionally under Gal80^ts control. These details are now discussed more clearly in the manuscript. In addition, as requested, we quantified efficiency of the knockdown with quantitative PCR, expressing the RNAi pan‐neuronally and using the same Gal80ts/induction protocol. This revealed a ~29% reduction in Cac mRNA levels. Further details about the strategy and new experiment have been added to the manuscript in the main text and Methods.

Overall, the data demonstrate that Cac is necessary for the potentiation in proximal γ lobe compartments, identifying a key mechanism of learning‐induced presynaptic plasticity in the KCs. This is consistent with the role that Cac plays at the neuromuscular junction, where Cac levels are positively correlated with synaptic potentiation (Gratz et al., 2019; J Neurosci). Furthermore, presynaptic homeostatic potentiation at the neuromuscular junction is dependent on Cac and Ca²⁺ levels (Frank et al., 2006, Neuron 52:663‐77; Muller et al., 2012, Curr Biol 22:1102‐8). Finally, Ca_V2.1 levels at the active zone are dynamically regulated and positively drive synaptic strength in mammals as well (Lubbert et al., 2018, Neuron 101:260‐273).

Can the authors perform additional manipulations that should phenocopy the reduction of Cac (e.g., reduction of Brp via RNAi)?

We have not attempted Brp knockdown in this study, as it would likely impair synaptic transmission across multiple compartments in a non‐valence‐specific manner. In addition, it would not provide direct insight into plasticity mechanisms and does not have (to our knowledge) a validated RNAi line allowing mild knockdown.

Can the phenotype be modulated by titrating extracellular calcium concentrations?

This is a conceptually interesting idea, but would not be feasible in vivo, where olfactory responses must be allowed to propagate unperturbed across multiple neurons/synapses to reach the KCs (from olfactory receptor neurons to projection neurons to KCs).

Would basal neurotransmission be affected when extending the time at 32{degree sign}C (the range of 4-10 days could also be quite large, depending on the turnover time for cacophony).

Basal neurotransmission could be affected if we were to increase the expression level of the RNAi (through various means), though doing so would be counterproductive for our experimental purpose.

Third, can the authors exclude that the observed effects were due to dendritic functions of Cac?

The axonal localization of Cac function is supported by its [axon] compartment‐specific effects on potentiation. Dendritic effects propagating into the axons would be distributed across multiple compartments (e.g., Boto et al., 2019, Cell Rep 27: 2014‐2021).

The authors need to perform additional experiments in order to manifest this important claim.

Multiple new experiments have added to the depth and rigor of the analysis. Overall, these experiments succeeded in allowing us to pinpoint Cac as a molecule that is important for regulating presynaptic plasticity and provided a molecular mechanism to manipulate/differentiate the plasticity occurring across the spatial compartments. The data demonstrated that Cac is necessary for the potentiation in proximal γ lobe compartments, identifying a key mechanism of learning‐induced presynaptic plasticity in the mushroom body. This is consistent with the role that Cac plays at the neuromuscular junction, where Cac levels are positively correlated with synaptic potentiation (Gratz et al., 2019; J Neurosci). Furthermore, presynaptic homeostatic potentiation at the neuromuscular junction is dependent on Cac and Ca²⁺ levels (Frank et al., 2006, Neuron 52:663‐77; Muller et al., 2012, Curr Biol 22:1102‐8). Finally, Ca_V2.1 levels at the active zone are also dynamically regulated and positively drive synaptic strength in mammals (Lubbert et al., 2018, Neuron 101:260‐273).

2. What is the behavioral consequence of Cac or IP3R knock-down in Kenyon cells? Is memory impaired?

To test this, we knocked down Cac and carried out appetitive conditioning, finding that short‐term memory was indeed impaired (Figure 3H).

3. The authors actually do not demonstrate that the acetylcholine measured is released from MB neurons (Kenyon cells) and do not exclude the possibility that the measured acetylcholine could derive from an unknown other source. The authors could address this issue by Kenyon cell-specific knock-down of VAChT or ChaT. Alternatively, acute or chronic block of KCs, while measuring sensor activation, could be an option.

We have now addressed this with the latter experiment that the reviewer suggested – blocking KC output while monitoring ACh release with GRAB‐ACh. The ACh signal was inhibited (Figure 4 – Supplement 3). Some signal was still observed, which likely emanates from the KCs that are not labeled by the driver and continue to respond to odors and release ACh (R13F02 labels ~746/2000 KCs). While the ACh signal imaged in the MB is clearly dominated by KCs, we do not exclude that there could be some contribution from non‐KCs, and added that caveat to the discussion.

I am puzzled by the wording in line 80: is 'putative' meant as 'generally accepted' or 'likely'? I would strongly recommend using a less ambiguous word in this context. If not generally accepted (if so, please do explain to the reader why), this would somewhat compromise the basis of this study.

Agreed – ACh is the described neurotransmitter from KCs. We were trying to convey that, without ruling out that some hypothetical future study could report cotransmission of some other neurotransmitter(s) (as has occurred in some other cell types), but the language was inaccurate on its face. This has been clarified. The sentence now reads “…synaptic release of the KC neurotransmitter…”.

4. Figure 1 and others: Why are only the means compared statistically? The pre to post in C and E should be analysed using paired statistics – as in Figure 3E/F or S4 for example. In Figure 1: are any changes detected within flies? Figure S1 suggest that this is not the case – the change would only become apparent between conditions? Although the authors have already done a good job in illustrating their protocols, data display should be more uniform potential discrepancies should be spelt out, and claims should be adjusted accordingly (e.g., summary in K).

We run all these comparisons on all data sets (Figure 1 – Supplement 1 and others). The key metric to detect potential/depression is the comparison of the Δ(post/pre) between each experimental group and its respective odor‐only control group (e.g., CS+ vs. EB). This is because olfactory conditioning generates some alteration/adaptation of the olfactory responses simply as a function of exposing the animal to the odor. The post vs. pre comparison sometimes accurately reveals such differences, particularly if they are large (e.g., in Ca²⁺ imaging experiments at the whole‐cell level, and, as the reviewer notes, in our subcellular data as well, in places). However, considering only the post vs. pre comparison would produce false negatives in some compartments that potentiate, as well as false positives in some compartments that do not. Therefore, we rely on the comparison of Δ(post/pre) between each experimental group and its respective odor‐only control as the primary metric. As noted above, we also ran the post vs. pre comparisons and report significant differences where they were found (and trends where they are important and informative).

5. Line 189: why a trend towards decrease? Figure S4 suggests a statistically significant change.

Following the rationale above, we only counted a change as significant if it differed from the odor-only control in terms of Δ(post/pre). The reviewer is correct – it is conservative here. This is often the case, as it requires that there be both a difference in Δ(post/pre) and that the difference be larger than any Δ(post/pre) in the odor‐only control.

6. Absolute conditioning protocols: Is the comparison of the 'no CS-' and 'no CS+' really suited to claim that 'no CS-' leads to decreased acetylcholine release? It appears that this would only be consistent if the 'no CS-' group was significantly different from the 'no US' group. Please clarify putative discrepancies. In this context, please also discuss the results in S5 in more detail.

Upon consideration of the reviewers’ comments and the overall theme/flow of the manuscript, we have removed this experiment from the manuscript. We intend to follow up on this result more thoroughly in a future study. For now, we focus on the conclusion that the γ3 compartment exhibits alterations in ACh release, and that its downstream MBONs are required for appetitive conditioning.

7. Figure S4 shows

a. gamma1: CS+ potentiation (not seen in Figure S1?), not apparent in Cac RNAi

b. gamma2: no potentiation for the CS+ (unlike what is reported in Figure 1), yet, consistently a depression for the CS-

c. gamma3: no potentiation of CS+, yet a depression of CS-, both inconsistent with Figure 1

d. Several forms of plasticity, also to the non-trained odor in controls and Cac RNAi

Please clarify and discuss the individual findings.

There are several differences between the genotypes in these experiments: (1) Gal4 driver (238Y vs. [sparser] R13F02), (2) inclusion of Gal80^ts, (3) temperature shifts for Gal80^ts induction. One or more of these differences may have slightly altered the plasticity effects in controls, though the overall conclusions are consistent: appetitive conditioning potentiated the relative CS+/CS‐ responses in the proximal γ compartments, while generating a trend toward decreasing responses in the distal compartments. This is now discussed in more detail in the manuscript.

8. Line 467: It is first stated that odor was presented continuously for 30s. In the next sentence the authors write that six 1-s odor pulses with a 5s inter-pulse interval were given. Please clarify. Also, the odor responses appear to typically last for about 5-6s. Obviously, this is in parts based on the kinetics of the reporter. However, would the authors not expect that continuous odor application will lead to desensitization at the level of the olfactory receptor neurons?

This was an error in the protocol description that has been corrected. For ACh imaging experiments, short odor pulses were used to avoid desensitization of the GRAB‐ACh sensor. The reviewer is correct, the responses last longer than the odor pulse, likely due to the kinetics of the reporter. Continuous odor application was used only in behavioral experiments, where this is a standard approach (calibration of the odor concentrations in the T‐maze to balance the naïve distribution accounts for any imbalance in desensitization between odors).

9. Apart from showing that postsynaptic calcium signals are altered in MBONs the authors should consider showing whether the observed effects seen with the acetylcholine indicator can also be seen when expressed in MBONs or DANs (and exclude the possibility that the potentiation/depression could be specific to KC-DAN or KC-KC synapses).

Expression levels of the GRAB‐ACh reporter are too low when expressed postsynaptically in the MBONs that we have tested so far.

Reviewer #3:

This study shows learning-induced compartmentalized plasticity in Kenyon cells (KCs) of the mushroom body γ lobe using a genetically encoded acetylcholine sensor. The authors made several major discoveries that the plasticity is bidirectional and depends on memory valence. Interestingly, the compartments along the axon terminals of the KCs undergo spatial 'gradient' of plasticity. Knock-down of the genes that differentially regulate intracellular calcium induced distinct effects on learning-induced plasticity. These results explain, at least partly, that learning-induced plasticity at KC-MBON synapses takes place on the presynaptic side. The experiments and analysis are overall thorough, and conclusions are generally supported by the results. In the following, I list some flaws to be addressed that concern necessary controls, more careful interpretation of the statistics and results, and the choice of the cac RNAi strain.

- Odor-only control is imperfect to formally distinguish if the plasticity arises from CS+ and/or CS- changes. The effect of sugar or shock presentation per se must be measured to make this distinction. Alternatively, the authors can examine how much of the 'plasticity' is explained by unpaired presentations of CS and US.

The odor‐only control accounts for the effect of non‐associative olfactory adaptation. Comparison of the CS+ and CS‐ in experimental groups reveal the associative changes in ACh release.

- Conditioning effects at some compartments are on the edge of statistical significance and look inconsistent among different experimental series (e.g. Figure 1F-J vs. Figure 3F). The authors need to be cautious to interpret these statistical results. For example, knock-down of cac impaired CS- depression in y2 and y3 as well (cf. Figure 1G, H, and K), and this is contradictory to their conclusion that cac is required only for potentiation. Similarly, itpr seem to globally affect presynaptic activities.

Cac knockdown did not affect CS‐ depression: CS‐ depression was not observed in controls for this experiment. There are several differences between the control genotypes in these experiments: (1) Gal4 driver (238Y vs. [sparser] R13F02), (2) inclusion of Gal80^ts, (3) temperature shifts for Gal80^ts induction. One or more of these differences may have slightly altered the plasticity effects in controls, though the overall conclusions are consistent: appetitive conditioning potentiated the relative CS+/CS‐ responses in the proximal γ compartments, while generating a trend toward decreasing responses in the distal compartments.

- Some of the major findings here have been consistent with previous papers with different measurements (e.g. Bilz et al. 2020 for Figure 2). On the other hand, this study presents results not coherent with existing studies. For example, previous papers of the authors found rather homogeneous calcium increase (Boto 2014, Louis 2018), while plasticity in this study is graded along the y lobe axons and opposite in appetitive and aversive conditioning (Figure 1, 2).

Indeed, this is an important finding here – Ca²⁺ in the presynaptic terminal is often, but not always, correlated with the directionality and magnitude of plasticity in ACh release across compartments. One inspiration for the present study was the independent lines of evidence suggesting presynaptic localization of plasticity associated with aversive learning (e.g., requirement for cAMP signaling molecules in KCs [e.g., Zars et al., 2000]) combined with our previous studies suggesting no net changes in odor‐evoked Ca²⁺ responses across MB γ compartments following aversive conditioning (at short‐term time points). This is discussed in the Introduction (lines 63‐67). These observations are not discrepancies, but rather reflect the different approaches used across the prior studies; the present data represent a step toward resolving these apparent discrepancies.

MBON-y3 has been reported to be dispensable (Aso 2014 and Ichinose 2021) unlike the results in Figure 6E. The authors need to provide discussion that explain the coherence/apparent inconsistencies.

Aso et al. 2014 tested 2hr appetitive memory (not STM), while silencing a subset of MBONs, which did not include the γ3 MBONs (their Figure 7B). Among the candidate GABAergic MBONs listed in the figure legend, they only tested the γ1pedc MBON in this particular assay. Ichinose et al. 2021 blocked MBONs in individual phases and tested at 1 hr, while we blocked across all phases and examined immediate memory. One of the major conclusions of their study was that multiple MBONs are involved in modulating reward memory. Given the differences in the approaches between the studies, we do not interpret the present results as necessarily inconsistent. Nonetheless, the paragraph describing the γ3 MBON behavioral data in the discussion has been edited to discuss these points.

- The choice of this particular cac RNAi strain is not clear, besides no control experiment for off-target effects. This is technically important, since voltage-gated calcium channels must be required more than learning-dependent potentiation. Therefore, it is blunt to conclude:

"synaptic exocytosis remained intact (L191)" upon cac knock-down;

"presynaptic potentiation, but not depression, requires the voltage-sensitive CaV2 Ca²⁺ channel cacophony across the MB compartments (L198)";

"Overall, these data demonstrate that Cac is not required for learning-induced depression of ACh release (L207)".

These conclusions need to be moderated.

The Cac line was chosen due to its moderate knockdown when driven without dcr‐2 (Brusich et al., 2015, Front. Cell. Neurosci 9:107). We have now included quantitative PCR demonstrating the efficacy of the RNAi line and confirming that we achieved the intended moderate knockdown effect. While we cannot rule out off‐target effects, we have confirmed that the manipulation produced the intended on‐target effect at the approximate desired level. Therefore, these statements represent conclusions that are supported by the data.

Also in this context, mentioning a recent paper by Hidalgo et al. (Neurobiology of Disease, 2021) that analyzed the effect of cac-KD in KCs on aversive learning would be complementary to the physiological finding of this study.

Agreed. We have now cited the Hidalgo et al. paper and run complementary behavioral tests using appetitive conditioning. Our results are consistent with the interpretation that Cac knockdown in KCs impacts appetitive learning, and the inclusion of these new data bolster the findings of the study.

- Compartmentalized and bidirectional plasticity in the MBONs has been repeatedly reported (e.g. Cohn, 2015; Owald, 2015). What is the intention to revisit this (Figure 5)? Also, this is essentially a detour from their main argument about the presynaptic mechanism. Consider revising the Figure or moving it to the figure supplement for readability.

The present experiments perform several functions: (1) demonstrating that plasticity is observed on the opposite side of the synapse with the same set of conditions that we used to test presynaptic plasticity, (2) testing plasticity following appetitive conditioning at the same time points we measured ACh release, (3) testing the role of the γ2α’1 and γ3 MBONs, which have not been well studied in the context of appetitive conditioning. Without these data, readers would be left to integrate conclusions across studies, assume that conditions were similar enough to compare across studies/labs, and extrapolate across multiple MBONs. While these are reasonable assumptions, they should be tested experimentally. To highlight the more novel findings, we moved the data from the γ1pedc and γ5β’2a MBONs to the supplement.

Similarly, it's not clear to me what Figure 6 adds to the overall conclusion.

In response to the reviewer’s inquiry, we reconsidered this data set and largely agree. We retained the experiment showing the requirement for the γ3 MBONs for normal appetitive short‐term conditioning (which shed light on the potential importance of plasticity in ACh release in the γ3 compartment), while eliminating the more tangential single‐odor temporal analyses.

- (Figure 6E) Did the authors examine MBON blocking in discriminative training? This needs to be clearly described. If the authors claim that the y3 compartment is important in discriminative training, they would need to show no defects in single-odor conditioning as in Figure 6A.

As above, upon consideration of the reviewers’ comments and the overall theme/flow of the manuscript, we have removed the discriminative training experiment from the manuscript. We intend to follow up on this result more thoroughly in a future study. For now, we focus on the conclusion that the γ3 compartment exhibits alterations in ACh release, and that its downstream MBONs are required for appetitive conditioning.

- Selection of ROIs for different MBONs in Figure 5 need stronger rationale. Why do the authors choose neurites for some compartments and not for others? I guess there is transformation of calcium signals in different parts of MBONs.

We selected ROIs consistently with previous studies (e.g., Berry et al., 2018; Jacob and Waddell, 2020; Zhao et al., 2018), which allowed us to reliably image the Ca²⁺ responses in these neurons. This is now noted in the Methods.

- Describe how the authors defined compartments in the g-KCs?

Based on the criteria of Aso et al., 2014 (anatomical landmarks; see also Figure 1B).

- I'm not totally sure how the results in Figure 6 support "temporal comparison (L275)". Either elaborate this conclusion more or provide an alternative interpretation.

As noted above, we removed this experiment. It deserves a more thorough treatment in a follow‐up study.

- The quantified compartment for Figure 4A-D should be clearly stated in the text and legend, instead of simply showing y2 in 4E and let us guess.

Indeed, this oversight has been corrected – the caption now indicates the compartment.

- Is "post-conditioning odor contrast" appropriate for explaining itpr effects given the CS+ depression is seemingly exaggerated (Figure 4)?

It wasn’t – the time series traces are plotted as mean values, which are strongly influenced by large values (i.e., outliers). The CS+ was significantly different in both controls and IP₃R RNAi groups (and the IP₃R RNAi p value was in fact less significant). The most important data are the ΔF/F values, not the mean time series traces. The main effect is the presence of adaptation in the CS‐ group and the odor‐only groups when knocking down IP₃R/itpr. We reworked the figure to include more data and focus on the important points (in this case, the pre and post ΔF/F values across all conditions).

Nonetheless, this raises an important related point ‐ the previous version of the manuscript left the role of IP₃R/itpr somewhat unclear. To further bolster this aspect of the study, we carried out additional experiments and clarified our interpretations. Loss of the IP₃R results in a reduction in adaptation to odors, which was noted in the initial manuscript as a reduction in odor‐evoked responses at the “post” time point across both experimental groups (CS+ and CS‐) and odor‐only controls. To explicitly test whether IP₃R knockdown increased the rate of olfactory adaptation, we tested a second protocol: presenting and imaging the odor‐evoked ACh release from KCs in response to a 1‐s odor pulse, delivered once per minute for 10 minutes. This protocol revealed increased adaptation in the IP₃R knockdown group relative to (uninduced) controls (Figure 4 K,L, Figure 4 – Supplement 2), demonstrating that IP₃R is required for maintenance of olfactory responses over time.

- L256 Sentence and/or figure citation are somewhat inconsistent.

The reviewer is correct (the figure was cited erroneously). To address other concerns, this section has now been removed, as noted above.

[Editors’ note: what follows is the authors’ response to the second round of review.]

Reviewer #2:

I've read the revised manuscript, and found it much improved. However, some of my concerns were not addressed. The authors need to clarify the following points before publication.

1. The odor-only control accounts for the effect of non-associative olfactory adaptation. But only that. The authors should be aware that other types of non-associative plasticity taking place during learning; e.g. reduced odor acuity by the previous exposure to electric shock (Preat, J Neurosci, 1998). Thus, inclusion of the US-only and/or unpaired training controls is mandatory to formally determine the absolute plasticity effect (e.g. potentiation of CS+ or depression CS- are indistinguishable; Figures 1L and 2K). Otherwise, they should moderate these claims, and focus only on the relative effects (i.e. "CS+:CS-" in the same figures)..

We have clarified and moderated the interpretation of the findings, noting that the comparison normalizes for non‐associative olfactory adaptation, as the reviewer notes. This is done in multiple places, such as on page 5, which now reads: “The Δ(post/pre) of the CS+ and CS‐ was compared to determine how each changed relative to the other, and then each was compared to its respective odor‐only control to quantify whether it was potentiated or depressed, accounting for any nonassociative olfactory adaptation (Figure 1 G‐K, Figure 1 – Supplement 2).”

2. As demonstrated by the authors, Cac-KD reduces only ~29% of total Cac RNA level, suggesting that large body of Cac expression is unaffected. It is quite questionable that Cac is not required for learning-induced depression of Ach release (P10, L234) based on this partial down-regulation.

This is a valid point; we adjusted the language accordingly. It now reads: “Overall, these data suggest that moderate knockdown of Cac does not affect learning‐induced depression of ACh release (in contrast to potentiation).”

3. Somehow related to the previous point. Is it reasonable to conclude that Cac is required for learning-dependent potentiation? There is potentiation of CS- odor response in the γ 5 compartment after appetitive learning (Figure 3- sup1E)?

There is no statistically significant difference between the CS‐ and the odor‐only control. Therefore, while we are intrigued by the quantitative increase in the CS‐ response in this group (as the reviewer is), we are unable to ascribe it to an associative effect of conditioning.

Author details

1. Aaron Stahl

   Department of Neuroscience, The Scripps Research Institute, Jupiter, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing - original draft, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-3170-1101

2. Nathaniel C Noyes

   Department of Neuroscience, The Scripps Research Institute, Jupiter, United States

   Contribution

   Data curation, Formal analysis, Investigation

   Competing interests

   No competing interests declared

3. Tamara Boto

   Department of Neuroscience, The Scripps Research Institute, Jupiter, United States

   Contribution

   Formal analysis, Investigation

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-9974-3714

4. Valentina Botero

   Department of Neuroscience, The Scripps Research Institute, Jupiter, United States

   Contribution

   Formal analysis, Investigation

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-9744-3929

5. Connor N Broyles

   Department of Neuroscience, The Scripps Research Institute, Jupiter, United States

   Contribution

   Formal analysis, Investigation

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2930-7343

6. Miao Jing

   Chinese Institute for Brain Research, Beijing, China

   Contribution

   Methodology, Resources

   Competing interests

   No competing interests declared

7. Jianzhi Zeng

   1. Peking-Tsinghua Center for Life Sciences, Peking University, Beijing, China
   2. State Key Laboratory of Membrane Biology, Peking University School of Life Sciences, Beijing, China
   3. PKU IDG/McGovern Institute for Brain Research, Beijing, China

   Contribution

   Methodology, Resources

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-5380-6281

8. Lanikea B King

   Department of Neuroscience, The Scripps Research Institute, Jupiter, United States

   Contribution

   Investigation

   Competing interests

   No competing interests declared

9. Yulong Li

   1. Chinese Institute for Brain Research, Beijing, China
   2. Peking-Tsinghua Center for Life Sciences, Peking University, Beijing, China
   3. State Key Laboratory of Membrane Biology, Peking University School of Life Sciences, Beijing, China
   4. PKU IDG/McGovern Institute for Brain Research, Beijing, China

   Contribution

   Funding acquisition, Methodology, Resources, Writing – review and editing

   Competing interests

   No competing interests declared

10. Ronald L Davis

    Department of Neuroscience, The Scripps Research Institute, Jupiter, United States

    Contribution

    Data curation, Funding acquisition, Methodology, Project administration, Resources, Supervision, Validation, Writing – review and editing

    Competing interests

    No competing interests declared

11. Seth M Tomchik

    Department of Neuroscience, The Scripps Research Institute, Jupiter, United States

    Contribution

    Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing - original draft

    For correspondence

    STomchik@scripps.edu

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0001-5686-0833

• 1,877

  Page views

• 300

  Downloads

• 8

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.