Temporally and Spatially Localized PKA Activity within Learning and Memory Circuitry Regulated by Network Feedback

Abstract Dynamic functional connectivity within brain circuits requires coordination of intercellular signaling and intracellular signal transduction. Critical roles for cAMP-dependent protein kinase A (PKA) signaling are well established in the Drosophila mushroom body (MB) learning and memory circuitry, but local PKA activity within this well-mapped neuronal network is uncharacterized. Here, we use an in vivo PKA activity sensor (PKA-SPARK) to test spatiotemporal regulatory requirements in the MB axon lobes. We find immature animals have little detectable PKA activity, whereas postcritical period adults show high field-selective activation primarily in just 3/16 defined output regions. In addition to the age-dependent PKA activity in distinct α’/β’ lobe nodes, females show sex-dependent elevation compared with males in these same restricted regions. Loss of neural cell body Fragile X mental retardation protein (FMRP) and Rugose [human Neurobeachin (NBEA)] suppresses localized PKA activity, whereas overexpression (OE) of MB lobe PKA-synergist Meng-Po (human SBK1) promotes PKA activity. Elevated Meng-Po subverts the PKA age-dependence, with elevated activity in immature animals, and spatial-restriction, with striking γ lobe activity. Testing circuit signaling requirements with temperature-sensitive shibire (human Dynamin) blockade, we find broadly expanded PKA activity within the MB lobes. Using transgenic tetanus toxin to block MB synaptic output, we find greatly heightened PKA activity in virtually all MB lobe fields, although the age-dependence is maintained. We conclude spatiotemporally restricted PKA activity signaling within this well-mapped learning/memory circuit is age-dependent and sex-dependent, driven by FMRP-Rugose pathway activation, temporally promoted by Meng-Po kinase function, and restricted by output neurotransmission providing network feedback.


Introduction
Functional connectivity and plasticity within brain circuits requires orchestration of intercellular network interactions and their downstream intracellular signaling cascades. For learning acquisition and memory consolidation, cAMP activation of protein kinase A (PKA) is widely recognized to be required for this synaptic regulation (Nguyen and Woo, 2003;Khan et al., 2021). In the Drosophila mushroom body (MB) learning/memory circuit model, both PKA and PKA regulators (e.g., Rutabaga adenylyl cyclase, Dunce phosphodiesterase) are highly expressed, and play well-characterized, pivotal roles (Nighorn et al., 1991;Crittenden et al., 1998;Lee, 2015). Within each brain hemisphere, ;2000 MB Kenyon cells (KCs) project into five distinct neuropil lobes (a/b , a'/b ', and g ) to innervate 16 defined MB output neuron (MBON) regions with en passant connectivity (Aso et al., 2014b). MB lobes receive layers of dopaminergic, GABAergic and serotonergic synaptic feedback, which provide context and regulate the KC output (Scheunemann et al., 2018;Modi et al., 2020;Yamagata et al., 2021). Investigations of a/b and g MB lobes in immobilized animals reveal low basal PKA activity, with pharmacological treatment differentially activating PKA function dependent on circuit location (Gervasi et al., 2010). However, a lack of visualization tools has limited studies of local PKA activity within the MB lobes, until the recent development of a PKA activity biosensor (PKA-SPARK), which generates bright GFP puncta on PKA phosphorylation Sears et al., 2019;Sears and Broadie, 2020). New studies of PKA-SPARK signaling within the increasingly well-defined MB circuit promise to provide critical insights into learning/memory impairments.
Fragile X mental retardation protein (FMRP) and Rugose/Neurobeachin (NBEA) are causally linked to intellectual disability and autism spectrum disorder (ASD; Wang et al., 2000;Castermans et al., 2003Castermans et al., , 2010Harris et al., 2008). FMRP drives cAMP induction and PKA activation in Fragile X syndrome (FXS) mouse/Drosophila disease models and human FXS patient cell cultures lacking FMRP (Berry-Kravis et al., 1995;Kelley et al., 2007;Sears and Broadie, 2020), and FMRP loss-offunction (LOF) impairs both learning and memory (Bolduc et al., 2008). FMRP binds rugose mRNA to positively regulate translation of this A-kinase anchor protein (AKAP) controlling MB PKA activity (Sears et al., 2019), and Rugose LOF also causes defective MB-dependent learning and memory (Volders et al., 2012;Buddika et al., 2021). Downstream of PKA, Meng-Po kinase [human SH3-binding kinase 1 (SBK1); Nara et al., 2001] is expressed within KC axons in the Drosophila MB lobes to facilitate memory formation (Lee et al., 2018). Meng-Po/ SBK1 overexpression (OE) in the MB strengthens long-term memory formation. To test MB intercellular circuit signaling requirements, the temperature-sensitive shibire ts mutant blocking Dynamin-dependent synaptic vesicle (SV) endocytosis can be used to disrupt KC neurotransmission (Koenig et al., 1983;Kitamoto, 2001). The transgenic tetanus toxin light chain (TNT) protease blocks SV fusion to more comprehensively eliminate KC neurotransmission (Sweeney et al., 1995;Haag et al., 2016). Our aim in this study was to test MB lobe PKA activity with all of the above intercellular and intracellular signaling perturbations, employing the transgenic PKA-SPARK reporter  to visualize spatiotemporal signaling regulation.
In this study, we compare immature Drosophila at 0 d posteclosion (0 dpe) with postcritical period Drosophila adults (7 dpe). Immature animals display little or no detectable PKA-SPARK activity in the MB lobes, whereas mature animals exhibit high levels of spatially-restricted signaling in a'1 and b '1 regions. In both locations, we find females have higher PKA activity. FMRP, Rugose and Meng-Po loss results in strongly reduced PKA activity in both MB lobe regions. In contrast, Meng-Po OE drives PKA activity in immature animals (0 dpe), as strongly as normal mature signaling, and also dramatically increases PKA activity in a'1, b '1, and b '1-adjacent g 3 at maturity (7 dpe). Testing circuit signaling requirements, we targeted KC synaptic output with two tools; temperature-sensitive shibire ts and TNT blockade. The shibire ts animals display both agedependent and temperature-dependent defects in PKA-SPARK activity, with expanded PKA signaling within the MB lobes at restrictive temperature. Consistently, MB-targeted TNT neurotransmission blockade dramatically expands PKA activity throughout the lobes, although the age-dependence is maintained. These findings indicate that PKA activity in the MB lobes is both age-dependent and sex-dependent in tightly restricted connectivity regions. Moreover, PKA activity is selectively induced during early adulthood in a mechanism requiring FMRP, Rugose, and Meng-Po and can be strongly potentiated by elevated Meng-Po to overcome some spatiotemporal restrictions. Finally, synaptic output suppresses PKA activity to spatially defined MB lobe regions of heightened signaling, suggesting a network feedback mechanism regulates PKA activity.

Drosophila genetics
All animals were reared and maintained on standard food in 12/12 h light/dark cycle cycling incubators at 25°C, unless otherwise noted. A combination of females and males were used for all experiments, except when females and males were compared or mutations were on the X chromosome. For mutant studies, the genetic background w 1118 (BDSC 3605) and RNAi background (P{y [1t7.7]=CaryP}attP2; BDSC 36303) were used as controls. Null dfmr1 50M (Zhang et al., 2001) and rg FDD (Volders et al., 2012) mutants were assayed in the w 1118 background. For transgenic studies, w 1118 outcrossed to the OK107-Gal4 driver line (BDSC 854) was used as the control. The OK107-Gal4 driver line at 2 dpe was used as a negative control for Western blot analyses. UAS responder lines used included: UAS-PKA-SPARK , UAS-dfmr1 (Zhang et al., 2001), UAS-hFMRP ISO7 (Sears and Broadie, 2020), UAS-DRGG-hFMRP (Coffee, 2011), UAS-rg (Volders et al., 2012), UAS-meng-po (Lee et al., 2018), UAS-meng-po RNAi (BDSC 29603;Lee et al., 2018), UAS-shibire ts (BDSC 44222), and insulated UAS-TNT (a generous gift from Brian McCabe; Brain Mind Institute, EPFL, Switzerland). For acute shibire ts experiments, animals were raised at 20°C until 7 dpe, and then either maintained at 20°C or moved to 33°C for 3 h. For longer shibire ts experiments, animals were raised at 20°C, and then either maintained at this temperature or moved to 25°C or 33°C for 7 d. Recombination and multiallele crossing schemes were performed using standard Drosophila genetic techniques.

Brain imaging
Staged brains were dissected free in PBS and fixed in 4% paraformaldehyde in PBS 1 4% sucrose for 30 min with rotation. Fixed brains were then washed 3Â in PBS, and incubated for 1.5 h in blocking buffer (PBS, 1% BSA, 0.5% goat serum, 0.2% Triton X-100). Samples were then incubated 2 h at room temperature (RT) with FITC-conjugated goat anti-GFP (1:500; Abcam, ab6662), and washed 3Â in blocking buffer. Brain preparations were mounted in Fluoromount-G between two #1 coverslips separated by die-cut Oracal 651 spacers. All imaging was Figure 1. Early life, sex-dependent PKA activity in Mushroom Body circuit regions. A, Schematic of MB lobes and defined MBON fields (dashed outlines) shown in three layers: a'/b ' (left), a/b (middle), and g lobes (right). B, C, Representative images of MB lobes with OK107-Gal4 driving UAS-PKA-SPARK at 0 dpe (B) and 7 dpe (C). The MBON fields (dashed circles) and arrows delineate the a'1 and b '1 quantified regions. D, E, PKA-SPARK::GFP puncta number in both regions, including a'1 (D) and b '1 (E). Scatter plots show all data points and mean 6 SEM. F, G, MB lobes with OK107-Gal4 driving UAS-PKA-SPARK in female (F) and male (G) at 7 dpe. H, I, Quantification of PKA-SPARK puncta in both a'1 (H) and b '1 (I). Sample size .15 fields in all conditions. Statistics show two-tailed t tests with Welch's correction (H) or Mann-Whitney tests (D, E, I). Significance: ***p , 0.001. done using a Zeiss LSM510 META with a Plan Neofluar 40Â oil-immersion (1.3 NA) objective. MB images were collected at 1-mm optical slice thickness for data analyses.

Data analyses
All image analyses were conducted blind using ImageJ. PKA-SPARK puncta were counted using the ImageJ Point Tool. All statistical analyses were conducted using GraphPad Prism (v9). Datasets passing the D'Agostino and Pearson normality test were compared with twotailed Welch's t tests (n = 2 datasets) or Brown-Forsythe and Welch ANOVA tests (n . 2 datasets). Datasets failing normality tests were compared with Mann-Whitney tests (n = 2 datasets) or with Kruskal-Wallis tests with Dunn's correction (n . 2 datasets). Graphs display all individual data points and mean 6 SEM. Significance is displayed in the figures as *p , 0.05, **p , 0.005, ***p , 0.001, and not statistically significant (p . 0.05; n.s.). Additional information is presented in the statistical table (Table 1).

Results
Age-dependent and sex-dependent PKA activity in MB lobe subfields The MB consists of two bilateral KC clusters that project into five distinct axonal lobes; a/b , a'/b ', and g (Fig.  1A). These lobes are further subdivided into discrete output neuron fields (Aso et al., 2014b). KCs can innervate multiple contiguous fields, which can have opposing behavioral outputs, raising the question of how multineuron interactions are regulated. PKA signaling required for learning and memory occurs at differential levels across the MB lobes (Skoulakis et al., 1993;Crittenden et al., 1998), suggesting local PKA activity regulation could drive developmental and circuit-level differences. To test PKA activity in defined MBON fields (Fig. 1A), we employed the Gal4/UAS system to drive expression of the UAS-PKA-separation of phases-based activity reporter of kinase (UAS-PKA-SPARK; Zhang et al., 2018) using pan-KC driver OK107-Gal4 (Connolly et al., 1996;Duffy, 2002;Aso et al., 2009). This reporter is phosphorylated specifically by PKA , and PKA-SPARK generates reversible GFP puncta that can be imaged using confocal microscopy. PKA-SPARK::GFP puncta can be fixed and immunolabeled for co-labeling in MB studies (Sears and Broadie, 2020). Brains were dissected from staged male and female animals at timed dpe at 25°C, and immediately fixed for PKA-SPARK::GFP imaging. Representative MB lobe images and quantified results are shown in Figure 1.
PKA-SPARK signaling was compared at two time points: immediately after eclosion (0 dpe; Fig. 1B) and at maturity (7 dpe; Fig. 1C) following the early-use critical period, a developmental window when sensory input experience remodels MB circuitry (Dombrovski and Condron, 2021). In the absence of signaling, diffuse GFP labels the MB lobes, allowing mapping of the MBON fields (Fig. 1A,B). At 0 dpe, there are rarely detectable PKA-SPARK::GFP puncta in the MB lobes (Fig. 1B). In contrast, mature 7 dpe animals display strong PKA activity, with many puncta localized to distinct MBON fields (Fig. 1C). The a/b lobes have inconsistent, broadly distributed puncta. Prominent signaling is tightly restricted to a'1 (left arrow) and b '1 (right arrow) fields, with lower levels in the b '2ap field (Fig. 1C). Quantification in the a'1 field by counting PKA-SPARK:: GFP puncta shows essentially no signaling at 0 dpe (0.1 6 0.04, n = 38) compared with high levels of signaling at 7 dpe (32.5 6 2.4, n = 34), a very significant increase over the short, one-week critical period (p , 0.0001; Fig. 1D). Likewise, in the b '1 field, there is essentially no detectable PKA activity at the immature 0 dpe time point (0.1 6 0.05, n = 37) compared with strong signaling at 7 dpe maturity (66.8 6 3.3, n = 38), again a highly significant elevation (p , 0.0001; Fig. 1E). b '1 displays ;2-fold more PKA activity compared with a'1. These data show a dramatic increase in PKA signaling activity between 0 and 7 dpe, tightly restricted to a few MBON fields.
At 7 dpe, the number of PKA-SPARK::GFP puncta in the MB lobes is quite variable (Fig. 1D,E). We hypothesized that this variability could be attributable in part to sex-specific differences, with sexually dimorphic levels of PKA signaling in the MB a'/b ' lobes. Comparing males and females, there is no detectable change in immature stages (0 dpe), with both sexes showing little or no PKA activity. In contrast, there is a clear sex-dependent difference in PKA-SPARK::GFP puncta number at maturity (7 dpe) with females showing higher activity (Fig. 1F) compared with males (Fig. 1G). In both sexes, prominent PKA signaling is restricted to the a'1 and b '1 fields (arrows).
In the a'1 field, females display nearly twice as many active puncta (50.1 6 3.1, n = 18) compared with the males (26.9 6 1.8, n = 16), a highly significant sex difference (p , 0.0001; Fig. 1H). Similarly in the b '1 region, females have more PKA-SPARK::GFP puncta (83.2 6 3.0, n = 18) compared with males (64.5 6 3.6, n = 16), again a very significant sex-dependent difference (p , 0.0001; Fig. 1I). Importantly, while weak PKA signaling can be detected in additional MB lobe regions in both females and males, there are no other consistent sex-specific differences outside of the striking PKA-SPARK::GFP puncta differences present in a'1 and b '1, with lower levels in the b '2ap MBON field. These data show adult-onset PKA activity in select MB lobe connectivity regions in both sexes, with females displaying comparatively more PKA signaling. To determine whether the striking age-dependent and sex-dependent differences may be because of expression level changes of the PKA-SPARK biosensor, we performed Western blot analysis of adult head lysates (Fig. 2). As above, the MB-specific OK107-Gal4 driver was used to express the PKA-SPARK::GFP reporter. Developmentally staged heads (0 and 7 dpe) were collected for both sexes. For each of the four groups (either female or male, and either 0 or 7 dpe), two heads per sample were pooled, processed and assayed in parallel for GFP levels relative to the a-tubulin loading control (Fig. 2). A single GFP-specific band was detected at ;37 kDa ( Fig. 2A), corresponding to the expected size of the GFP-containing LRRATLVD-EGFP-HOTag3 reporter construct . In all trials, the reporter levels were indistinguishable between both ages (0 and 7 dpe) and sexes (female and male; Fig. 2A). In  the quantified comparisons, PKA-SPARK::GFP levels were highly comparable between all groups, including both ages and sexes (normalized to 0 dpe female 1.0 6 0.04; 0 dpe male 1.10 6 0.03; 7 dpe female 1.09 6 0.10; 7 dpe male 0.93 6 0.08; n = 9 for each group; p . 0.05 for all comparisons; Fig. 2B). Taken together, these data show that the changes in PKA signaling are not because of reporter expression differences, but represent striking age-specific and sex-specific PKA activity levels within the MB lobes.
Localized MB lobe PKA activity driven by FMRP and Rugose We next sought to determine the molecular mechanisms for the spatiotemporal restriction of PKA activity signaling in the MB lobes. We first tested disease models causally linked to PKA misregulation and impaired MBdependent learning/memory. FMRP is involved in cAMP induction and PKA activation (Berry-Kravis et al., 1995;Sears and Broadie, 2020). In the Drosophila FXS disease model of intellectual disability, dfmr1 null mutants exhibit severely defective PKA-dependent learning/memory based on MB signaling dysfunction (Bolduc et al., 2008). The RNAbinding translational regulator FMRP binds directly to Rugose/NBEA mRNA to drive expression of this PKA anchor within KCs (Sears et al., 2019). In a second Drosophila disease model of intellectual disability, rugose (rg) null mutants also display impaired MB-dependent learning and memory (Wang et al., 2000;Volders et al., 2012). Based on these studies, we hypothesized the FMRP-Rg pathway regulates the spatiotemporal restriction of PKA activity signaling in the MB lobes ( a'1 and b '1; Fig. 1). To test this idea, we compared PKA-SPARK analyses in the genetic background control (w 1118 ) to both the dfmr1 50M (Zhang et al., 2001) and rg FDD (Volders et al., 2012) null alleles. Assays of staged animals tested differences within the defined MBON fields. Representative images and quantified data are shown in Figure 3.
At 7 dpe, w 1118 controls show striking active PKA-SPARK::GFP puncta formation primarily restricted to the MB lobe a'1 and b '1 fields (Fig. 3A, arrows). In stark contrast, dfmr1 50M nulls display a severe reduction in PKA signaling activity in both of these regions (Fig. 3A, right). Quantification of the a'1 region reveals far more puncta in the controls (43.5 6 2.8, n = 34) compared with dfmr1 50M mutants (26.1 6 3.7, n = 21), a highly significant loss of signaling (p = 0.0005; Fig. 3B). Likewise in the b '1 region, there are many more PKA-SPARK::GFP puncta in controls (75.5 6 4.2, n = 34) compared with dfmr1 50M nulls (43.9 6 7.0, n = 21), showing a significant reduction in the localized PKA activity (p = 0.0004; Fig. 3C). Similarly, in rg FDD null mutants, there is an obvious loss of signaling in both a'1 and b '1 MBON fields (Fig. 3D). Quantification of the a'1 field shows many more puncta in the controls (27.4 6 1.8, n = 24 males only) compared with rg FDD mutants (15.2 6 4.0, n = 18), a very significant reduction (p = 0.0001; Fig. 3E). Moreover, in the b '1 field, there is nearly 3-fold more signaling in the controls (58.9 6 2.6, n = 24) compared with rg FDD nulls (20.8 6 3.7, n = 18), a highly significant loss of signaling (p , 0.0001; Fig. 3F). As in controls, few PKA-SPARK::GFP puncta appear outside of these regions. These data indicate that both FMRP and Rugose are similarly required to enable PKA activity in the MB lobes, but do not alter the spatiotemporal restriction of the PKA signaling domains.  Fig. 1A). PKA-SPARK::GFP puncta number in a'1 (B) and b '1 (C) regions. D, Similar confocal imaging comparison at 7 dpe in control (left) and rg FDD null (right). E, F, Quantification of PKA-SPARK::GFP puncta number in a'1 (E) and b '1 (F) regions. Note that this comparison is done in males only owing to the X chromosome location of the rugose gene. Scatter plots show all data points and mean 6 SEM. Sample size .12 fields in all conditions. Statistics show two-tailed t tests with Welch's correction (B, C, F) or Mann-Whitney tests (E). Significance: ***p , 0.001.

Spatiotemporal PKA activity bidirectionally regulated by Meng-Po kinase
Given the tightly localized PKA activity, we next tested OE effects on spatiotemporal misregulation. Driving both Drosophila FMRP (dFMRP) and human FMRP (hFMRP) elevates PKA activity in KC somata (Sears and Broadie, 2020). We hypothesized expanded or enhanced PKA activity in the MB lobes by OE of the FMRP-Rg regulatory pathway. Using OK107-Gal4 to drive UAS-PKA-SPARK, we tested UAS-dFMRP (Zhang et al., 2001), UAS-hFMRP (Sears and Broadie, 2020), UAS-DRGG-hFMRP (Coffee, 2011), and UAS-Rg (Volders et al., 2012). DRGG-hFMRP OE localizes PKA activity to phase-separated domains in KC somata and proximal process, but the impact on lobe signaling has not previously been tested (Sears and Broadie, 2020). We also assayed OE of Meng-Po (MP) kinase, which synergizes with PKA activity in the MB (Lee et al., 2018). Meng-Po localizes to the MB lobes and UAS-MP OE dramatically strengthens MB-dependent learning and memory (Lee et al., 2018). We hypothesized enhanced spatiotemporal PKA activity in the MB lobes through increased Meng-Po function. For all of these OE comparisons, we visualized PKA-SPARK::GFP puncta throughout the MB lobes at both immature 0 dpe and mature 7 dpe time points. The MB driver OK107-Gal4 with UAS-PKA-SPARK alone was the reference control. Representative images at both time points and the quantified data for controls and all five OE conditions are shown in Figure 4.
The strong effect of Meng-Po OE suggests a requirement promoting PKA activity, consistent with its role as a PKA-synergist within the MB (Lee et al., 2018). To test this hypothesis, we next assayed MP LOF. In our hands, meng-po mutants are lethal before adult eclosion, so MB-specific meng-po RNAi was tested for changes in MB lobe PKA activity (Fig. 5). To test for spatiotemporal changes in PKA signaling, the RNAi knock-down was compared with the driver alone transgenic control at both 0 and 7 dpe. Similar to the controls at 0 dpe, there are very few or no detectable PKA-SPARK puncta observed in the meng-po RNAi animals (Fig.  5A,B, left panels). In contrast, mature 7 dpe controls exhibit strong PKA activity, again mostly localized to the a'1 and b '1 regions (Fig. 5A, right), whereas the meng-po RNAi animals exhibit no detectable PKA-SPARK puncta (Fig. 5B, right). Quantification of the a'1 field shows high puncta numbers only in the controls at 7 dpe (control 0 dpe 1.0 6 0.5, n = 10; meng-po RNAi 0 dpe 0.5 6 0.1, n = 13; control 7 dpe 31.7 6 1.8, n = 17; meng-po RNAi 7 dpe 0.3 6 0.1, n = 16; p , 0.001 between control 7 dpe and all others groups, p . 0.05 for all other comparisons; Fig. 5C). Similarly, b '1 PKA-SPARK puncta are only elevated in controls at 7 dpe (control 0 dpe 0.8 6 0.2, n = 10; meng-po RNAi 0 dpe 0.3 6 0.2, n = 13; control 7 dpe 41.7 6 2.7, n = 17; meng-po RNAi 7 dpe 0.6 6 0.2, n = 16; p , 0.001 between control 7 dpe and other groups, p . 0.05 for all other comparisons; Fig. 5D). Taken together, these data show Meng-Po promotes localized PKA activity signaling in the MB lobes.

Conditional shibire ts neurotransmission block increases PKA activity
The spatiotemporally localized PKA activity within a few MBON fields suggests local network regulation. Multiple circuit feedback interactions are proposed to provide modularity to the MB circuit, both within and between signaling compartments (Liu and Davis, 2009;Inada et al., 2017;Modi et al., 2020). We hypothesized blocking KC synaptic output could alter PKA activity signaling measured by PKA-SPARK::GFP puncta localization. To test effects of output connectivity on PKA activity, we used the temperature-sensitive shibire ts allele (Koenig et al., 1983;Kitamoto, 2001), which is frequently employed to block neurotransmission (Jacob et al., 2021;Muria et al., 2021). At restrictive temperatures (.29°C), shibire ts is defective in SV recycling, and semi-dominant over wild type (Koenig et al., 1983;Kitamoto, 2001). At reported permissive temperature (25°C) some phenotypes manifest, so we also added a 20°C condition. Using the MB-targeted OK107-Gal4 driver to express PKA-SPARK, shibire ts animals were reared at 20°C until 0 dpe, then aged for 7 d at 20°C, 25°C or 33°C.  '1 and b '1 regions (Fig. 1A) of normally heightened PKA activity. Arrowhead (F) points to the expanded PKA activity within the g 3 region. G, H, Quantification of PKA-SPARK:: GFP puncta in a'1 (G) and b '1 (H) regions. Scatter plots show all data points and mean 6 SEM. Sample size: .7 fields in every genotype and at every time point. Statistics show two-tailed t tests with Welch's correction and Mann-Whitney tests (see statistical  table; Table 1). Significance: ***p , 0.001.
Restrictive temperature shibire ts animals at 0 dpe display little to no PKA-SPARK::GFP puncta throughout the MB lobes (Fig. 6A), but have enlarged a/a' ends (arrowheads). The shibire ts animals reared to 7 dpe display variable levels of PKA activity with altered MB lobe spatial patterns depending on temperature (20-33°C). Representative images at all temperatures and quantified data are shown in Figure 6.

PKA signaling activity dramatically expanded by synaptic output block
The above shibire ts studies suggest that MB synaptic output suppresses PKA activity across MBON fields, but interpretation is complicated by temperature and pleiotropic Dynamin functions. We therefore turned to a stronger, synapse-selective neurotransmission block via targeted transgenic tetanus toxin light chain (UAS-TNT; Sweeney et al., 1995) to silence synapses (Chow et al., 2011;Haag et al., 2016;Fuenzalida-Uribe and Campusano, 2018). The TNT protease cleaves the v-SNARE Synaptobrevin to block SV exocytosis and thus presynaptic function (Link et al., 1992;Schiavo et al., 1992). Using MB-targeted OK107-Gal4 to drive UAS-TNT, PKA-SPARK animals were reared at 25°C until 0 and 7 dpe. Synaptically-silenced MBs display normal morphology (Fig. 7A,B), confirming that KC neurotransmission can be blocked without affecting MB lobe architecture. Within the defined MBON fields, PKA activity was tested by assaying PKA-SPARK::GFP puncta. TNT-blocked animals at 0 dpe continue to display very few PKA-SPARK::GFP puncta throughout the MB lobes (Fig. 7A), demonstrating that blocked KC synaptic output does not alleviate the temporal restriction on PKA signaling. In contrast, however, TNT-blocked animals at 7 dpe display greatly expanded PKA activity, with PKA-SPARK::GFP puncta appearing widely throughout the MB lobes (Fig. 7B). Representative images and quantified data from multiple MBON fields are shown in Figure 7.
In newly-eclosed animals (0 dpe), the TNT blockade of KC synaptic transmission has no effect within any MBON field (Fig. 7C-I), with very few or no active PKA-SPARK:: GFP puncta detectable in quantification trials. In sharp contrast, mature TNT-blocked animals (7 dpe) exhibit a profound increase in PKA activity signaling throughout MB lobe regions, with especially prominent puncta numbers present in both a/b and a'/b ' lobes (Fig. 7B). Interestingly, while widespread ectopic g lobe PKA activity is also frequently observed, it is not universally present in TNT-blocked animals. The a'1 and b '1 regions display numerous activated PKA-SPARK::GFP puncta, with a greater relative elevation in the a'1 region compared with the b '1 region (Fig. 7B). With the quantification of PKA-SPARK::GFP puncta in TNT-blocked animals, a'1 activity is .2-fold increased (72.4 6 3.8, n = 57; Fig. 7C) relative to controls (35.9 6 1.8, n = 68, p , 0.0001), but there is little change in the b '1 MBON field (control: 70.4 6 2.2, n = 72; UAS-TNT 75.2 6 5.3, n = 57; Fig. 7D). This striking shift shows a balance change in activity-dependent PKA signaling. Under control conditions, there is always consistently higher numbers of PKA-SPARK::GFP puncta in b '1 compared with a'1 (Fig. 1D,E). However, in TNTblocked animals these regions are no longer statistically different (p . 0.05; Fig. 7C,D). These data suggest that blocking KC neurotransmission disrupts the network signaling balance between MBON fields.

Discussion
In this study, we explore the spatiotemporal regulation of PKA activity within the MB lobes. PKA signaling initiates in early adulthood, with heightened activity in just 3/ 16 MB lobe output neuron fields (a'1, b '1, and b '2ap; Aso et al., 2014b;Modi et al., 2020). In addition to age-dependence, this regional PKA signaling displays sex-dependence, with elevation in females over males (Fig. 1). These findings were made possible with the PKA-SPARK biosensor; this fluorescent reporter uses motifs found in earlier PKA sensors, with pharmacological and genetic approaches promoting and preventing PKA activity verifying this new tool in cell culture and in vivo Sears et al., 2019;Tyurin-Kuzmin et al., 2020). While it is possible that the PKA-SPARK reporter is revealing only the strongest PKA activity , at the least a'1, b '1, and b '2ap connectivity regions have much higher PKA activity levels compared with the rest of the MB neuropil. In two disease models of intellectual disability and ASDs, from loss of either FMRP or Rugose/NBEA (Zhang et al., 2001;Volders et al., 2012), PKA activity remains spatiotemporally restricted, but is dramatically reduced (Fig. 3). There is surprisingly little change from overactivation of the FMRP!Rg control pathway (Sears et al., 2019), but PKA activity is profoundly altered by the PKA pathway Meng-Po kinase (Lee et al., 2018), with OE enhancing spatiotemporal PKA signaling (Fig. 4) and loss suppressing PKA activity (Fig. 5). Network feedback downstream of KC neurotransmission strongly suppresses PKA activity, since blocking KC synaptic output with conditional shibire ts (Koenig et al., 1983;Kitamoto, 2001) or transgenic TNT (Sweeney et al., 1995;Haag et al., 2016) induces widespread PKA activity signaling (Figs. 6, 7). Thus, localized PKA activity is highly regulated at the circuit level.
At a macro level, the a'1 and b '1 regions with the highest localized PKA activity levels ( Fig. 1) have been linked to valence (Aso et al., 2014a), i.e., whether local activation causes animals to approach or avoid a stimulus. The a'1 and b '2 fields exhibit opposing output valence (positive and negative, respectively), while the b '1 role appears less clear (Aso et al., 2014a;Modi et al., 2020). However, higher PKA signaling can be activated in the g 3 region (via Meng-Po; Fig. 4), which in combination with the b '1 region drives positive valence (Aso et al., 2014a). Moreover, dopaminergic and serotonergic biosensor signals function through the b ' region from a variety of external sense stimuli (Sun et al., 2018;Wan et al., 2021), to which the a'1/b '1 fields display high sensitivity (Inada et al., 2017), suggesting broad responsiveness. Very recent work shows high spontaneous activity in a'/b ' restricted to young animals (Leinwand and Scott, 2021), suggesting a'/b ' PKA signaling may be controlled by selective developmental activity. Moreover, inhibiting miR-92a in a/b and g , but not a'/b ', was shown to enhance memory (Guven-Ozkan et al., 2020), indicating another layer of lobe-selective circuit regulation. Note that the output neurons from these MB lobe regions are different (a'1 cholinergic, b '1 GABAergic, and b '2 glutamatergic; Aso et al., 2014b), suggesting more complex integrative circuit functions. PKA signaling activation after the early-use critical period (Fig. 1) is consistent with sensory experiencedependent regulation (Dombrovski and Condron, 2021). Importantly, MB sensory integration functions differ markedly between females and males (Park et al., 2018;Shin et al., 2020), correlating with the report here of sex-dependent PKA signaling differences in females compared with males ( Fig. 1).
Two learning and memory proteins, FMRP and Rugose, are needed for full PKA activity in the a'1 and b '1 MB lobe output neuron fields (Fig. 3). RNA-binding FMRP is a translational regulator (Darnell et al., 2011;Greenblatt and Spradling, 2018) known to facilitate PKA signaling (Berry-Kravis et al., 1995;Kelley et al., 2007), which is lost in the FXS, the commonest heritable cause of intellectual disability and ASD (Harris et al., 2008). Rugose/ NBEA is a PKA-anchor that facilitates learning and memory, and is also associated with ASDs (Wang et al., 2000;Castermans et al., 2003Castermans et al., , 2010Volders et al., 2012). Previous work has shown FMRP binds to rugose mRNA to drive KC expression (Sears et al., 2019;Buddika et al., 2021). As predicted, disruption of this FMRP!Rg regulative pathway strongly impairs PKA activity in the MB lobes (Fig. 3). In contrast, localized PKA signaling is dramatically strengthened by MB OE of the Meng-Po kinase (Lee et al., 2018), which induces early-onset PKA activity before adult sensory experience, spatially expands high PKA activity to the g 3 MBON field, and profoundly elevates PKA activity within all the normal MB lobe regions of heightened PKA signaling (Fig. 4). Consistently, Meng-Po kinase OE also greatly improves learning and memory, via PKA phosphorylation, but additionally via signaling feedback synergy (Lee et al., 2018). Moreover, meng-po RNAi causes the opposite result of reducing localized PKA-SPARK puncta (Fig. 5). Based on both loss and gain of function, we suggest Meng-Po enhances localized PKA activity, reflecting circuit level kinase regulation. Determining how Meng-Po-regulated PKA activity determines circuit excitability and regional balance will be a major subject of future research.
Two different KC synaptic output blocking methods dramatically expand PKA activity signaling in the MB lobes (Figs. 6, 7). Both conditional shibire ts and transgenic tetanus toxin tools block KC neurotransmission, but through quite different mechanisms (Kitamoto, 2001;Haag et al., 2016). At 33°C, KC-targeted shibire ts drives PKA activity expansion in the MB g lobe (Fig. 6). This change could indicate cross-compartment network interactions between the g lobe and other MB regions. Increasing g 1 PKA activity is especially interesting, as g 1 toggles inhibition of other MB regions (Perisse et al., 2016). The tetanus toxin protease blocks neurotransmission through eliminating SV exocytosis (Sweeney et al., 1995;Haag et al., 2016), and therefore provides a stronger and more selective means to silence KC synaptic output. Consistently, TNT animals show a more profound expansion of PKA activity throughout the MB lobes, albeit again affecting only spatial and not temporal patterning (Fig. 7). Neither shibire ts nor TNT blockade alters early PKA activity (Figs. 6, 7), suggesting induction of PKA signaling is determined primarily by later experience-driven activity. Localized PKA activity changes with KC output block implies active circuit balance; for example, weighing aversive versus attractive responses to sensory stimuli (Modi et al., 2020). The widespread PKA activity upregulation with KC output block leads us to hypothesize that enhancing KC neuron activity should result in elevated PKA signaling.
At the MB circuit level, multiple candidate synaptic pathways need to be explored for roles in local PKA activity regulation in different MB lobe output neuron fields. GABAergic inputs to dopaminergic neurons are one likely candidate, since GABA treatment has been shown to correct dfmr1 mutant circuit defects exacerbated by glutamate exposure (Chang et al., 2008;Yamagata et al., 2021). Moreover, GABAergic anterior paired lateral (APL) neurons broadly control MB activity through feedback to the KCs, and are most strongly activated by the a'/b ' lobes (Liu and Davis, 2009;Lin et al., 2014;Inada et al., 2017;Modi et al., 2020). Recent work shows that treatment with a dopamine transport inhibitor also ameliorates rg mutant social interaction and memory deficits (Pham et al., 2021). Another candidate is the Amnesiac neuropeptide from the serotonergic dorsal paired medial neurons, required for their normal development in the broad innervation of the MB lobes (Yu et al., 2005;Lee et al., 2011;Turrel et al., 2018). In the context of our studies, de facto depression may feedback onto KCs to promote PKA activity signaling. The impact of upstream input onto the MB lobes is an important consideration, including how this circuitry combines with spontaneous MB activity and internal lobe circuitry to determine PKA signaling. Future research directions should attempt to dissect how these different layers of neuromodulation control localized PKA activity signaling within the MB lobe circuit, and between females and males, by manipulating input-specific neuronal activity in targeted transgenic studies.
In conclusion, we report here that PKA activity signaling in the Drosophila brain MB learning and memory center is highly induced during early experiential adulthood, with selective upregulation in the a'1, b '1, and b '2ap MB lobe output neuron regions. Age-specific and sex-specific PKA signaling controlled within KCs and downstream of KC output shows that spatiotemporally restricted MB lobe PKA activity is regulated through a combination of both intracellular control and intercellular network-level mechanisms. Importantly, PKA signaling can be precociously promoted and spatial expanded though the activity of the PKA-synergist Meng-Po kinase. Moreover, KC neurotransmission inhibits localized PKA signaling within the MB circuit. Our future studies will be aimed toward generating new genetic responder tools to test KC signaling with both neurotransmission output blockade and activity promotion of upstream and downstream MB circuit components, simultaneously and independently of KCs. We also plan to test PKA activity signaling with the manipulation of specific KC partners, by altering neurotransmission signaling in combination with postsynaptic neurotransmitter receptor mutants to determine network communication cues. Taken together with the current work, these ongoing studies will continue to expand our understanding of circuit-level PKA signaling regulation in normal function and in neurologic disease model contexts.