A rapid and dynamic role for FMRP in the plasticity of adult neurons

Daniel G Gundermann; Seana Lymer; Jennifer Lennon; Justin Blau

doi:10.1038/s41467-025-66487-0

. 2025 Nov 23;16:11578. doi: 10.1038/s41467-025-66487-0

A rapid and dynamic role for FMRP in the plasticity of adult neurons

Daniel G Gundermann ^1,³, Seana Lymer ^1,⁴, Jennifer Lennon ¹, Justin Blau ^1,^2,^✉

PMCID: PMC12749007 PMID: 41276503

Abstract

Fragile X syndrome is a neurodevelopmental disorder caused by silencing Fragile X messenger ribonucleoprotein 1 (Fmr1), which encodes the FMRP RNA-binding protein. Although Fmr1 is expressed in adult neurons, it has been challenging to separate acute from chronic effects of loss of Fmr1. Here we use the precision of Drosophila genetics to show that FMRP acutely regulates neuronal plasticity in adult small ventral lateral circadian pacemaker neurons (s-LNvs), which show daily rhythms in their structure. We find that Fmr1 is required for structural plasticity in s-LNvs, and that reducing Fmr1 expression for only 4 hours prevented s-LNv projections from retracting. Conversely, briefly overexpressing Fmr1 blocks the activity-dependent expansion of s-LNv projections without major changes to the circadian clock or to activity-regulated gene expression. One FMRP target we identify in s-LNvs is sif, which encodes a Rac1 Guanine nucleotide exchange factor. Our data suggest that FMRP normally reduces sif mRNA translation at dusk to reduce Rac1 activity and regulate adult neuronal plasticity.

Subject terms: Molecular neuroscience, Synaptic plasticity, Autism spectrum disorders, Translation, Gene expression

Fragile X syndrome is a neurodevelopmental disorder caused by inactivating the RNA-binding protein FMRP. Here, the authors show that FMRP acutely regulates structural plasticity in mature adult pacemaker neurons in the Drosophila central brain.

Introduction

Cytoplasmic RNA-binding proteins (RBPs) regulate gene expression post-transcriptionally by controlling mRNA localization, stability, and translation. One advantage of regulating gene expression via RBPs is that cells can rapidly transition between different states. This is probably most clear after fertilization, when a large set of maternally deposited mRNAs is rapidly activated for translation¹. RBPs are also very important in neurons, where they help localize mRNAs to distal parts of the cell² and control translation locally in response to changes in neuronal activity³. These two functions of RBPs are key for neuronal plasticity, which underlies learning and memory³ and is disrupted in various neurological disorders⁴.

Fragile X Syndrome (FXS) exemplifies the importance of RBPs in neurons. FXS is a neurodevelopmental disorder that is the most common cause of intellectual disability in humans. FXS is caused by the expansion of CGG repeats in the 5′ UTR of the human Fragile X messenger ribonucleoprotein 1 (FMR1) gene on the X chromosome that silences FMR1 expression^5,6. FMR1 is also the most common single-gene cause of autism spectrum disorder^7–9. FMR1 encodes an RNA-binding protein known as FMRP, whose canonical function is to inhibit translation of specific mRNAs¹⁰. However, FMRP can also increase mRNA translation¹¹, regulate splicing¹², and even physically interact with ion channels to regulate neuronal excitability^13,14.

Fmr1 is expressed in many tissues, but is enriched in the brain, where its expression is developmentally regulated in mice¹⁵ and Drosophila¹⁶. For example, the highest FMRP levels in mouse somatosensory cortex coincide with the critical period of sensory-dependent plasticity, and Fmr1 mutant mice have abnormal structural and synaptic plasticity during this time^17,18. Fmr1 is also expressed in adult neurons^15,16,19, and plasticity phenotypes in the hippocampus of Fmr1 mutant mice can be rescued by reintroducing Fmr1 in adult mice²⁰, suggesting a role for FMRP in adult neurons. Indeed, FMRP is localized to dendritic spines^21,22, which are key sites of neuronal plasticity. However, it is challenging to test acute functions of FMRP in vivo in mice.

We chose to use Drosophila to explore an acute role of FMRP in adult neuronal plasticity because of the tools available to acutely alter gene expression in adult neurons. Drosophila mutant for Fmr1 show developmental abnormalities such as reduced axonal pruning of mushroom body gamma neurons during development¹⁶, which can be considered analogous to the neurodevelopmental phenotypes in mice FXS models. Drosophila Fmr1 is also important in young adult flies, where it is required to remove the PDF-Tri neurons²³. In this study, we focused on the small ventral lateral neurons (s-LNvs) to be able to study a single homogeneous cell type and thus avoid the potential confound of diverse FMRP targets in different cells. The s-LNvs are 4 of the ~75 pacemaker neurons in each brain hemisphere that control circadian rhythms in adult flies²⁴. s-LNvs have well-documented and predictable plasticity in the structure of their projections²⁵—described below.

Each s-LNv expresses a set of clock genes, including period (per) and vrille (vri)²⁶. These clock genes form an intracellular molecular clock that generates rhythms in the expression of some clock genes (e.g., per and vri), and also in the expression of clock-controlled genes that are often cell-type specific²⁷. For example, Irk1 is an s-LNv clock-controlled gene which encodes an inward rectifier potassium channel that helps control the 24 h rhythms in s-LNv excitability that peak around dawn to promote the morning bout of locomotor activity^27,28. s-LNvs also show 24 h rhythms in the structure of their projections, which are maximally expanded around dawn and retracted by dusk²⁵. This structural plasticity involves making and breaking synaptic connections with a 24 h rhythm²⁹ to likely control behavioral outputs. The predictability of these structural changes makes s-LNvs ideal to study the molecular mechanisms of plasticity.

Inducing firing in s-LNvs at dusk is sufficient to fully expand s-LNv projections within 2–3 h^30,31. Conversely, expressing a mammalian Inward rectifier K+ channel to hyperpolarize s-LNvs prevents the normal expansion of their projections at dawn³². Thus, s-LNv excitability seems to drive s-LNv structural plasticity. This idea is consistent with the requirement for Mef2—a transcription factor involved in activity-dependent transcription in many neurons³³—in s-LNv structural plasticity rhythms³⁰. Linking transcriptional changes to the s-LNv cytoskeleton, we previously showed that Puratrophin-like (Pura) encodes a Rho1 guanine nucleotide exchange factor (GEF) whose rhythmic transcription in s-LNvs peaks at dusk³¹. Rhythmic Pura expression generates rhythms in Rho1 activity that retract s-LNv projections at dusk by increasing myosin phosphorylation³¹. However, it is not yet clear what genes promote the expansion of s-LNv projections at dawn.

s-LNv projections in Drosophila Fmr1 mutant flies have expanded projections despite relatively normal molecular clock oscillations^34–36. Conversely, overexpressing Fmr1 collapses s-LNv projections³⁶, indicating that Fmr1 can regulate s-LNv plasticity bidirectionally. However, these experiments were performed before s-LNv plasticity had been described and were thus only analyzed at a single time of day. Here, we show that Fmr1 null mutants lose rhythms in the structural plasticity of s-LNv projections. We also show that only 4 h of altered Fmr1 levels in adult s-LNvs is sufficient to block the plasticity of s-LNv projections. We used RNA interference (RNAi) transgenes to show that Fmr1 is acutely required for s-LNv projections to retract at dusk. Conversely, we found that overexpressing Fmr1 in s-LNvs is sufficient to block expansion of s-LNv projections at dawn. We then sought to identify how FMRP controls s-LNv plasticity. We found that acutely altering Fmr1 expression does not dramatically affect the circadian clock or a reporter gene that reports neuronal activity in s-LNvs. Thus, Fmr1 likely acts downstream of the clock and neuronal excitability to control s-LNv plasticity. We used TRIBE-seq³⁷ to identify FMRP targets in s-LNvs. One FMRP target gene we identified is sif, which encodes a GEF for the Rac1 GTPase and which has previously been implicated in regulating synaptic growth at the Drosophila neuromuscular junction³⁸. The mammalian Sif ortholog—TIAM1—stabilizes dendrites and synapses in the developing hippocampus³⁹, and is required for structural plasticity in neurons in the spinal dorsal horn in a mouse neuropathic pain model⁴⁰. Altered Tiam1 expression is also associated with major depressive disorder⁴¹.

We found that sif and Rac1 itself are required for s-LNv projections to expand at dawn, and that transiently overexpressing sif or expressing constitutively active Rac1 is sufficient to rapidly expand s-LNv projections at dusk when they are normally retracted. Furthermore, the expanded s-LNv projections observed at dusk when Fmr1 expression is knocked down can be rescued by simultaneously reducing sif expression. Thus, we propose that FMRP normally represses sif mRNA translation at dusk to ensure that Rho1 is the dominant small GTPase and that s-LNv projections retract. s-LNv structural plasticity is thus a balance of GTPase activity: Pura—regulated transcriptionally—activates Rho1 to retract projections at dusk; while sif—regulated post-transcriptionally by Fmr1—activates Rac1 to expand s-LNv projections at dawn. Given the extensive similarities between Drosophila and mammalian FMRP, we speculate that FMRP can also rapidly control the plasticity of adult neurons in mammals. Overall, our data underscore the importance of RNA-binding proteins in acutely regulating adult neuronal plasticity.

Results

FMRP is required for s-LNv structural plasticity rhythms

Previous experiments showed that s-LNv projections are expanded in Fmr1 mutant flies compared to control flies^34–36. However, these experiments had not been performed at different times of day and thus did not test for rhythmic s-LNv structural plasticity. Control flies and flies lacking functional Fmr1 were entrained in Light:Dark (LD) cycles at 25 °C. ZT (Zeitgeber time) indicates time in a 12 h: 12 h LD cycle, with lights on from ZT0 to ZT12, and lights off from ZT12 to ZT24. Brains were dissected and fixed either 2 h after lights-on at ZT2 (“dawn”), when s-LNv projections are normally fully expanded, or 2 h after lights-off at ZT14 (“dusk”), when s-LNv projections are normally retracted²⁵.

Control and Fmr1 mutant flies also contained one copy of a Pdf-RFP transgene in which the Pigment dispersing factor (Pdf) regulatory region expresses a cytosolic Red Fluorescent Protein (RFP)²⁷. Pdf encodes a neuropeptide that is only produced by LNvs in the mature adult brain. Brains were stained using antibodies to mRFP, which fills the LNv projections for visualization. The spread of the s-LNv projections in 3D was then quantified using a previously described script³¹.

The data in Fig. 1 show that control flies have the typical expanded s-LNv projections at ZT2 and retracted projections at ZT14²⁵. However, Fmr1 null mutants have expanded s-LNv projections at both times of day. Thus, Fmr1 is required for s-LNv projections to retract and thus for rhythms in s-LNv structural plasticity.

FMRP dynamically regulates the daily structural plasticity of s-LNvs

To test if FMRP can dynamically regulate the structural plasticity of adult s-LNvs, we used an inducible system to either knock down or overexpress Fmr1 in LNvs in adult flies. We used the Pdf-Gal4 driver to restrict transgene expression to LNvs²⁴. We also used a temperature-sensitive repressor of Gal4—Gal80^TS—that is ubiquitously expressed via the tubulin enhancer (tub-Gal80^ts)⁴². This Gal4-Gal80^TS combination gives temporal control of UAS transgene expression as the temperature can be shifted from 19 °C to 30 °C to inactivate Gal80^TS and allow Gal4 to activate transcription.

We used UAS transgenes with two different short-hairpin RNAs that target Fmr1 (UAS-sh-Fmr1) to knock down Fmr1. These Fmr1-shRNA transgenes are independent and target different regions of the Fmr1 mRNA. We also used a UAS-Fmr1 transgene with an Fmr1 cDNA for overexpression. All flies also had the Pdf-RFP transgene as in Fig. 1.

Flies were raised at 19 °C, and adult flies were then entrained in Light:Dark (LD) cycles at 19 °C. The temperature in the incubator was then raised to 30 °C. One temperature shift started at ZT22 with brains dissected and fixed 4 h later at ZT2; the other temperature shift started at ZT10 with brains dissected 4 h later at ZT14. We visualized and quantified s-LNv projections as in Fig. 1.

The data in Fig. 2A show that control Pdf-Gal4, tub-Gal80^ts, Pdf-RFP flies without a UAS transgene that were kept at 19 °C and then shifted to 30 °C for 4 h have typical expanded s-LNv projections at ZT2 and retracted projections at ZT14. There was no difference between s-LNv projections from control flies and flies with expression of either of the two shFmr1 transgenes induced from ZT22 to ZT2. In contrast, s-LNv projections remained expanded when each sh-Fmr1 transgene was expressed from ZT10 to ZT14. The effects on s-LNv projections of these two shRNA transgenes are very similar to the data from Fmr1 null mutant flies in Fig. 1.

We found that inducing Fmr1 overexpression by moving flies to 30 °C for 4 h starting at ZT22 kept s-LNv projections retracted at ZT2. However, we did not observe any difference between control s-LNvs and s-LNvs in which Fmr1 was overexpressed from ZT10 to ZT14 (Fig. 2A). Thus, FMRP must be interacting with a key molecule that is required for s-LNv projections to expand at dawn rather than non-specifically shrinking s-LNv projections.

We also performed a set of control experiments to test if s-LNv structural plasticity rhythms are normal when Gal80^TS is active at the lower temperature of 19 °C. We tested 3 of the 4 genotypes used in Fig. 2A: Control flies, and flies with Pdf-Gal4, tub-Gal80^ts, and either the UAS-Fmr1 or the UAS-sh-Fmr1 (V20) transgene. We found that all 3 genotypes had normal plasticity rhythms at 19 °C with expanded s-LNv projections at ZT2, and retracted projections at ZT14 (Supplementary Fig. 1). Thus, we conclude that the phenotypes seen in Fig. 2A are rapid effects of inducing expression of the Fmr1 and Fmr1-shRNA transgenes, and that Gal80^TS blocks Gal4 activity sufficiently well at 19 °C.

The Pdf-Gal4 driver is expressed in the s-LNvs and the large LNvs (l-LNvs), with the latter more important for sleep than for circadian rhythms^43,44. To more definitively test if FMRP acts cell-autonomously in s-LNvs to regulate their structural plasticity, we used the VT027163-Gal4 driver, which expresses in s-LNvs but not in l-LNvs⁴⁵. We refer to this driver as Pura-Gal4 as it has part of the Pura regulatory region. We used Pura-Gal4 in conjunction with tub-Gal80^ts to induce Fmr1 overexpression for 4 h starting at either ZT10 or ZT22 and imaged s-LNv projections 4 h later as described above. The data in Fig. 2B show the normal plasticity rhythms in s-LNv projections from control flies with Pura-Gal4 and tub-Gal80^ts but no UAS transgene. In contrast, there was no difference in the 3D spread of the s-LNv projections between ZT2 and ZT14 in flies with Fmr1 overexpressed in s-LNvs, with s-LNv projections retracted at both timepoints.

Together, the data in Figs. 1 and 2 are consistent with previous analyses of flies with Fmr1 mutations or flies overexpressing Fmr1^34–36 that show that FMRP can function bidirectionally in s-LNv structural plasticity. However, the brief manipulations in Fig. 2 allow us to disentangle developmental effects of FMRP function in adult neurons. We conclude that FMRP can act rapidly in mature adult neurons—in this case, s-LNvs—to dynamically regulate structural plasticity.

FMRP levels in s-LNvs

The results above indicate that FMRP needs to be active at dusk for s-LNv projections to retract, but FMRP must not be too active at dawn, otherwise s-LNv projections cannot expand. We therefore wanted to measure FMRP levels to test if they change over time in s-LNvs as one way to control FMRP activity.

First, we confirmed that the fairly weak FMRP signal detected by the FMRP monoclonal antibody 6A15 in the cytoplasm of cell bodies⁴⁶ is a bona fide signal, as it is not present in Fmr1 null mutant flies (Fig. S2A). Next, we tested for any differences in FMRP immunoreactivity in s-LNv cell bodies at ZT2 and ZT14 by quantifying FMRP levels relative to levels of the Elav nuclear protein. The data in Fig. S2B show no detectable difference in FRMP levels between dawn and dusk in s-LNvs. We also detected FMRP in the projections of many neurons. However, it was not possible to determine whether this signal came from s-LNvs or adjacent projections, and therefore, we cannot exclude the possibility that FMRP levels are rhythmic in s-LNv projections.

We also tested if the UAS-sh-Fmr1 and UAS-Fmr1 transgenes affect FMRP levels. The data in Fig. S2C show that FMRP immunoreactivity in s-LNv cell bodies is decreased by expression of the UAS-sh-Fmr1 transgene for 24 h, although the effect is fairly modest. However, our conclusions from these data are limited by our inability to accurately measure endogenous FMRP levels in s-LNv projections, which could be where FMRP acts to regulate plasticity.

The data in Fig. S2D show that FMRP levels are increased ~3-fold with a 4 h induction of the UAS-Fmr1 transgene in larval LNvs, which become the s-LNvs in adult flies. We used larval LNvs as their gene expression is similar to adult s-LNvs²⁷, and the absence of l-LNvs in larvae helps visualize the beginning of their projections. Indeed, we noticed that overexpressed FMRP is clearly detectable in larval LNv projections in addition to their cell bodies (Fig. S2E). Furthermore, FMRP can be detected as speckles in larval LNvs, which is consistent with observations in mammalian neurons of FMRP forming phase-separated granules with RNA⁴⁷.

Acutely changing Fmr1 levels do not change the molecular clock

The molecular clock in s-LNvs regulates the timing of s-LNv neuronal activity, making them most active around dawn²⁸ when s-LNv projections are expanded. In addition, inducing s-LNvs to fire at dusk is sufficient to expand s-LNv projections^30,31. This leads to a model where the s-LNv molecular clock regulates rhythms in s-LNv firing, which in turn regulates s-LNv structural plasticity in an activity-dependent manner⁴⁸. Thus, FMRP could regulate s-LNv structural plasticity by altering the s-LNv molecular clock or by altering the timing of s-LNv neuronal activity. These ideas seemed plausible because constitutive long-term overexpression of Drosophila Fmr1 lengthens circadian period³⁴, and because mouse FMRP binds the mRNA of several clock genes, including mPer1 in hippocampal neurons⁴⁹. In addition, FMRP binds mRNAs that regulate neuronal excitability, and neuronal excitability is altered in Fmr1 knockout mice^10,50.

We first tested if altering Fmr1 levels for 4 h changes the phase of the molecular clock by measuring levels of the circadian transcription factor Vri. We chose Vri because its levels are rhythmic in clock neurons, its relatively short half-life makes it a more accurate marker of clock phase than longer-lived clock proteins such as Per, and because Vri protein levels mirror vri mRNA levels, which are transcriptionally controlled by the core clock⁵¹. We assayed Vri protein levels in s-LNvs by immunofluorescence and quantified the fluorescence intensity at ZT2 and ZT14 after 4 h of inducing either Fmr1 overexpression or an shFmr1 transgene.

Figure 3A shows that control flies have low Vri levels in s-LNvs at ZT2 and high levels at ZT14, as expected⁵¹. We found no significant difference in Vri levels between control s-LNvs and either experimental group. Since s-LNv structural plasticity is altered before we could detect changes in the molecular clock, the FMRP target(s) that explain its rapid effects on s-LNv plasticity is likely downstream of the molecular clock.

As a complementary way to test if altered Fmr1 levels rapidly affect the circadian clock, we measured circadian rhythms in adult locomotor activity in flies with Pdf-Gal4 and tub-Gal80^ts controlling expression of the UAS-Fmr1 or UAS-sh-Fmr1 transgenes. Flies were raised at 18 °C, and then placed into constant darkness (DD) at either 24 °C or 30 °C to measure locomotor activity rhythms. We chose 24 °C as a temperature at which Gal80TS should still repress Gal4⁴², and also because flies have weaker rhythms at 18 °C than at 24 °C in our hands, making it harder to robustly measure circadian rhythms at 18 °C.

The data in Supplementary Fig. 3 and Supplementary Table 1 show that flies from all genotypes have similar strength rhythms at 24 °C, although Pdf-Gal4; tub-Gal80^ts > Fmr1 have ~1 h longer period rhythms than the other genotypes, which suggests some leaky expression of the UAS-Fmr1 transgene at 24 °C. However, Pdf-Gal4; tub-Gal80^ts > Fmr1 flies have a dramatically longer period and weaker power at 30 °C than the other genotypes (Supplementary Fig. 3C).

Closer inspection of the actograms indicated that the very long period of Pdf-Gal4; tub-Gal80^ts > Fmr1 flies at 30 °C starts after ~5 days in DD. We compared the period and power of control and Pdf-Gal4; tub-Gal80^ts > Fmr1 flies on the first 4 full days in DD with days 6–11 (Supplementary Fig. 3D). These data confirm that the period of Pdf-Gal4; tub-Gal80^ts > Fmr1 flies changes over time: Pdf-Gal4; tub-Gal80^ts > Fmr1 flies have a 0.6 h longer period than control flies on days 1–4, but a 3.7 h longer period on days 6–11. The strength of the rhythm of Pdf-Gal4; tub-Gal80^ts > Fmr1 flies is always lower than control flies (Supplementary Fig. 3C, D). From these data, we conclude that the initial phenotype of overexpressing Fmr1 is a modest period-lengthening and a reduction in rhythm strength. This is consistent with the absence of a dramatic effect on the molecular clock after 4 h of inducing Fmr1 expression in Fig. 3A. Furthermore, weaker locomotor rhythms when s-LNv projections are retracted via Fmr1 overexpression are reminiscent of the phenotypes of flies with s-LNv projections retracted by inducing overexpression of Rho1 in adulthood³¹.

The long-period phenotype appears slowly, suggesting that the circadian clock is mainly affected only after FMRP levels increase beyond a certain threshold. Long-period rhythms were also seen when timeless-Gal4 was used to constitutively overexpress Fmr1 in all clock neurons³⁴. The main phenotype of inducing Fmr1-shRNA expression in adulthood seems to be a slight decrease in rhythm strength (Supplementary Fig. S3A–C).

Fmr1 overexpression does not dramatically change s-LNv excitability

Next, we wanted to test if transiently altering Fmr1 levels rapidly changes the excitability of s-LNvs. There is minimal electrophysiology data from s-LNvs^28,52 because their size makes recordings technically challenging. Thus, we decided to use a reporter gene whose transcription is activated by neuronal activity as a proxy for s-LNv excitability. Neuronal activity rapidly induces expression of activity-regulated genes—also known as immediate early genes—such as c-fos, Egr1, and Nr4a1³³. Expression of activity-regulated genes can thus be used to determine the recent activity of a neuron in vivo (e.g., ref. ⁵³) and has the advantage of integrating recent activity and not just activity at the instant of recording. While the Drosophila fos ortholog does not seem to be activity-regulated in insects, the orthologs of Egr1 – stripe (sr) – and Nr4a1 – Hr38 – are upregulated in response to neuronal activity in many different neurons⁵⁴.

We used a transcriptional reporter gene for Hr38, in which 4 kb of Hr38 regulatory DNA is upstream of a destabilized and nuclear-localized fluorescent Tomato reporter gene. This Hr38-Tomato reporter gene shows higher expression in larval LNvs at ZT2 than at ZT14⁵⁵, consistent with increased larval LNv excitability at dawn⁵⁶, just like adult s-LNvs²⁸. Figure 3B shows that control s-LNvs have higher Hr38-Tomato levels at ZT2 than at ZT14, and this is not significantly affected by inducing overexpression of Fmr1 for 4 h. We noticed that overexpressing FMRP increased the variability of Tomato expression levels more than in control flies. However, there was no corresponding variability in the plasticity phenotype when Fmr1 was overexpressed at dawn (Fig. 2A). Thus, the similar levels of Hr38-Tomato expression at ZT2 with and without Fmr1 overexpressed make it likely that the rapid changes in s-LNv structural plasticity caused by increasing FMRP levels are downstream of neuronal excitability.

Identifying FMRP targets in vivo in s-LNvs

We wanted to take an unbiased approach to identifying the relevant FMRP target(s) in s-LNvs that mediate the changes in s-LNv plasticity. We decided that it was important to identify targets in s-LNvs themselves since FMRP's interactions with other proteins can affect the mRNAs bound, and these proteins can differ between cells^57–59. However, there are only 8 s-LNvs in each Drosophila brain out of a total of ~200,000 neurons⁶⁰, making the task challenging. To understand how FMRP can rapidly regulate neuronal plasticity, we also wanted to identify FMRP target mRNAs that interact with FMRP in the timeframe in which we detect s-LNv plasticity phenotypes, i.e., within ~4 h of altered Fmr1 expression (Fig. 2).

Given these limitations, we decided to use TRIBE-seq³⁷. TRIBE stands for targets of RNA-binding proteins identified by editing and uses the enzyme ADAR, which changes adenosines into inosines in mRNA. Inosine is then recognized as a guanosine in vitro during RNA-sequencing library preparation. Thus, the sequence of mRNAs edited by ADAR differs from the genomic sequence. In TRIBE-seq, the ADAR catalytic domain (ADARcd) is fused to an RBP and expressed in the cells of interest. mRNA is then isolated from these cells, and RNA-sequencing is used to identify mRNAs with edited adenosines (Fig. 4A). These edited mRNAs must have been physically close to the RBP-ADARcd fusion protein, and thus likely bound by that RBP. McMahon et al.³⁷ showed that expressing an Fmr1-ADARcd transgene in a subset of Drosophila excitatory neurons led to edits in almost half of the orthologs of mammalian FMRP targets identified by cross-linking and immunoprecipitation (CLIP) from mouse brain. These FMRP-ADARcd targets included futsch, which had previously been identified genetically and biochemically as a Drosophila FMRP target⁶¹. MAP1B, the mammalian ortholog of Drosophila futsch, is one of the top FMRP targets identified by CLIP in the mouse brain¹⁰.

Fig. 4 — A Experimental procedure used for *Fmr1* TRIBE-seq in s-LNvs. Genotypes of flies used are shown on the left, followed by an example of FACS sorting of RFP⁺ events from experimental flies. RNA-seq was performed, and the HyperTRIBE pipeline⁶² used to identify FMRP targets. B Edits from the *Fmr1* TRIBE-seq experiment from experiment 1 and experiment 2 plotted against each other and normalized using the average percentage of edited reads per transcript. Transcripts were color-coded if their GO term is ‘Circadian rhythm’ (GO:0007623, red) or ‘Ion transport’ (GO:0006811, green). A select group of previously identified and validated mammalian and fly FMRP targets is labeled in cyan, and Rho-GTPase superfamily members in purple.

We used UAS transgenes that express either an FMRP-ADARcd fusion (UAS-Fmr1-ADARcd) or only the catalytic domain of a hyperactive version of the ADARcd as a control (UAS-ADARcd⁶²). Transgene expression was controlled by Pura-Gal4 to spatially limit expression to s-LNvs. Temporal control was achieved via the tub-Gal80^ts transgene, with flies raised and entrained at 19 °C to prevent UAS transgene expression. The temperature was then raised to 30 °C starting at ZT22, and the fly brain dissection began 4 h later at ZT2. To collect ~40 brains for each sample, dissections continued for 1 h, which means that s-LNvs had a range of transgene expression time of 4–5 h. These flies also contained the constitutively expressed Pdf-RFP transgene to ensure a strong fluorescent signal for Fluorescence-activated cell sorting (FACS). Pdf-RFP is expressed in both s-LNvs and l-LNvs, and thus both cell types were purified in this experiment. However, ADAR-modified mRNAs should only come from s-LNvs because Pura-Gal4 does not express in l-LNvs⁴⁵.

Adult brains were dissociated into a single cell suspension and sorted for RFP+ fluorescence. s-LNvs and particularly l-LNvs have long projections, which sometimes fragment during cell dissociation. Given the localization of RBPs like FMRP outside the cell body^21,22, and the importance of local translation in neurons³, we sorted all RFP+ events via FACS (Fig. 4A), having ensured that the collection gates would exclude any RFP- cells or fragments. We extracted RNA and then generated full-length mRNAs using the SMART-seq3 protocol⁶³, followed by RNA-seq as in ref. ³⁷. We performed two independent experiments with 39–54 brains dissected per condition. The aligned reads were run through the HyperTRIBE pipeline to identify mRNAs modified by FMRP-ADARcd using the previously described cutoffs⁶²: (i) ≥10% of the mRNA reads need to be edited in that location; (ii) ≥20 reads for that site; and (iii) no changes to G in the control dataset. The graph in Fig. 4B shows the number of edits for a transcript weighted by the average percentage of edits on that transcript. For example, an mRNA with 1 site edited in 60% of reads, and a second site edited in 40% of reads would have a score of 1 (see “Methods” section). We then plotted the weighted edits for the two independent experiments against each other.

We first checked if the relatively brief induction of Fmr1-ADAR in s-LNvs was sufficient to identify previously reported targets of Drosophila and mammalian FMRP identified by CLIP. We found several FMRP targets, including purity of essence (poe – UBR4 in mammals)¹¹, rugose (rg – Neurobeachin in mammals)⁶⁴, Calcium/calmodulin-dependent protein kinase II (CaMKII)⁶⁵, and futsch⁶¹ as FMRP targets in s-LNvs (Fig. 4B, all shown in cyan).

Previously described FMRP-mRNA targets in neurons include clock genes such as Per1⁶⁶, and genes that control neuronal excitability^50,67. We did not detect many clock genes in our dataset (red in Fig. 4B), although some clock genes, such as per and vri, are lowly expressed at ZT2 in s-LNvs. We did detect a set of genes encoding ion channels as FMRP targets in s-LNvs (green in Fig. 4B). However, these genes are unlikely to explain how rapidly increasing FMRP levels block expansion of s-LNv projections, given the lack of a major effect on the Hr38-Tomato activity-dependent transcriptional reporter gene in 4 h (Fig. 3B). We therefore examined the list of FMRP targets for alternative ways to regulate s-LNv plasticity.

We noticed two Rho-GTPase activity regulators in the top 50 mRNAs edited by FMR1-ADAR in s-LNvs: still life (sif) and CG43102 (purple in Fig. 4B). Supplementary Data 1 gives a full list of FMRP targets in s-LNvs. The presence of sif and CG43102 is interesting because expression of the Rho1 GEF Pura peaks at dusk in s-LNvs to drive rhythmic Rho1 activity and retract s-LNv projections³¹. Pura itself was not detected in our FMRP TRIBE-seq experiments, although it is not highly expressed at dawn. The importance of GEFs in s-LNv plasticity was further supported by a study showing that overexpressing the GEF trio keeps s-LNvs in a dusk-like retracted state even at dawn⁶⁸. sif was also one of the top 20 FMRP targets identified in Drosophila cholinergic neurons³⁷, and CG43102 was also identified by FMRP TRIBE-seq in cholinergic neurons³⁷, although with many fewer edits than sif. Given the importance of Rho GTPases in s-LNv plasticity, we decided to focus on sif and CG43102.

We also sequenced the genomes of flies that contained the genotypes that differed between the two lines used for TRIBE-seq and used the GATK HaplotypeCaller to identify SNPs that could be detected as edits, and used the Broad Institute GATK Best Practices recommendations for filtering SNPs. We removed all the A to G SNPs, as well as the reverse complementary SNPs that were identified by this analysis, from the HyperTRIBE results. Supplementary Fig. 4 shows that sif is still in the top targets even after this stringent cutoff, although CG43102 is now further down the list.

We also used an independent method to test if FMRP binds sif mRNA in vivo. For this, we crossed elav-GeneSwitch flies (elav-GS⁶⁹) to flies with either a UAS-GFP transgene as a control or to UAS-Fmr1-GFP flies. elav-GS is inactive until flies are placed on food containing mifepristone, which activates the GS protein to transcribe a UAS transgene⁶⁹. We isolated fly heads 48 h after placing flies on mifepristone, and then immunoprecipitated either GFP or FMRP-GFP and any associated mRNA. Importantly, we used the same anti-GFP nanobody for both control and experimental flies. RNA was then reverse transcribed, and we compared the fraction of immunoprecipitated RNA to input RNA. The data in Supplementary Fig. 5 show that sif mRNA was immunoprecipitated by the FMRP-GFP fusion protein but not by GFP alone. This was not a non-specific interaction of all mRNAs with FMRP-GFP, as Beadex (Bx) mRNA was not immunoprecipitated by either GFP or FMRP-GFP. We chose Bx because it is expressed in LNvs⁷⁰, but it was not identified as an FMRP target in our experiments. As elav-GS is expressed in the vast majority of differentiated neurons⁶⁹, this experiment lacks the cell-type specificity of TRIBE-seq. However, it validates the FMRP-sif mRNA interaction detected by TRIBE-seq in s-LNvs (Fig. 4) and in cholinergic neurons³⁷.

sif and Rac1 regulate the expansion of s-LNv projections at dawn

We next tested whether sif and CG43102 are involved in the daily structural plasticity of s-LNvs. We used the same transient expression strategy as in Fig. 2, using UAS-shRNA transgenes to reduce expression and either a UAS-sif transgene to overexpress sif or a P-element insertion line (EY01540) that contains UAS binding sites inserted just upstream of CG43102. We found no changes in the structure of s-LNv projections with either strategy for altering CG43102 expression (Supplementary Fig. 6). However, we cannot formally exclude a role for CG43102 in s-LNv plasticity, as we did not check the efficacy of overexpression or knockdown in the lines we used.

In contrast, we found strong s-LNv plasticity phenotypes from altering sif expression. First, we found that overexpressing sif for 4 h from ZT10 to ZT14 leaves s-LNv projections in an expanded state at ZT14, when they are normally retracted. sif overexpression from ZT22 to ZT2 had no effect, and s-LNv projections remained fully expanded, indicating that the sif-mediated expansion of s-LNv projections at ZT14 is a time-specific expansion of s-LNv projections (Fig. 5A).

Fig. 5 — A Left: Representative images of s-LNv projections from flies with *Pdf-Gal4* and *tubulin-Gal80*^ts crossed to w flies (Control: *Pdf, tub-Gal80*^ts > +), to *UAS-sif* flies (Overexpression: *Pdf, tub-Gal80*^ts > *UAS-sif*), or to *sif* short-hairpin RNA flies (Knockdown: *Pdf, tub-Gal80*^ts > *UAS-sh-sif)*. Transgenes were induced for 4 h as in Fig. 2. Flies contained a *Pdf-RFP* transgene and projections visualized using an RFP antibody. Right: Quantification of the 3D spread of s-LNv projections. A one-way ANOVA followed by a Tukey post-hoc test was used to assess significance. s-LNv projections in control flies (gray) at ZT2 differ from ZT14 (p < 0.001, denoted by ***), and from flies expressing *sif-shRNA* (yellow) in LNvs at ZT2 (p = 0.045, denoted as *p < 0.05). s-LNv projections in control flies at ZT14 differ from flies overexpressing *sif* (orange) in LNvs at ZT14 (p = 0.028, denoted as *p < 0.05). Sample sizes pooled from 4 independent experiments: Controls – 16 hemispheres at ZT2, 11 at ZT14; *sif* overexpression – 11 hemispheres at ZT2, 11 at ZT14; *sif-shRNA* – 10 hemispheres at ZT2, 12 at ZT14. B Top: Representative images of s-LNv projections from flies with *Pdf-Gal4* and *tubulin-Gal80*^ts crossed to *UAS-myrRFP* flies (Control: *Pdf, tub-Gal80*^ts > *UAS-myrRFP*), or flies with a constitutively active *UAS-Rac1* transgene (Active Rac1: *Pdf, tub-Gal80*^ts > *UAS-Rac1*^CA). Transgenes were induced at ZT12, and brains dissected 2 h later at ZT14. s-LNv projections visualized using PDF antisera. Bottom: Quantification of 3D spread of s-LNv projections with a two-sided t-test to test for differences between genotypes. Error bars show SEM. s-LNv projections in control flies at ZT14 (gray bar) differ from flies expressing *Rac1*^CA in LNvs at ZT14 (orange, p = 0.016, denoted as *p < 0.05). Sample sizes: 10 control and 8 experimental hemispheres pooled from 2 independent experiments. C Top: Representative images of s-LNv projections from flies with *Pdf-Gal4* and *tubulin-Gal80*^ts crossed to *UAS-myrRFP* flies (Control: *Pdf, tub-Gal80*^ts > *UAS-myrRFP*), or to flies with a *Rac1* short-hairpin RNA (Knockdown: *Pdf, tub-Gal80*^ts > *UAS-sh-Rac1)*. Transgenes were induced at ZT12 and dissected 14 h later at ZT2. s-LNv projections visualized as in (B). Bottom: Quantification of 3D spread of s-LNv projections with a two-sided t-test to test for differences between genotypes. Error bars show SEM. s-LNv projections in control flies at ZT2 (gray) differ from flies expressing *Rac1-shRNA* in LNvs at ZT2 (pink, p = 0.015, denoted as *p < 0.05). Sample sizes: 6 control and 7 experimental hemispheres pooled from 2 independent experiments.

In the complementary experiment, we found that inducing expression of sif-shRNA from ZT22 to ZT2 led to retracted s-LNv projections at dawn, in contrast to the expanded s-LNv projections in control flies. However, expressing sif-shRNA from ZT10 to ZT14 had no effect, and s-LNv projections stayed retracted. Therefore, we conclude that sif is normally required for s-LNv projections to expand at dawn. Taking these data together with sif being sufficient to expand s-LNv projections when overexpressed at dusk (Fig. 5A), we conclude that controlling Sif protein levels in s-LNvs is very important for their rhythmic structural plasticity.

sif encodes a Rac1 GEF which localizes to the larval neuromuscular junction and to the synaptic terminal in adult photoreceptor cells³⁸. TIAM1 (T cell lymphoma invasion and metastasis 1) is the mammalian Sif ortholog and activates Rac1 after NMDA receptor stimulation to reorganize the actin cytoskeleton in mammalian hippocampal neurons⁷¹. Interestingly, sif was one of only two genes that showed rhythmic alternative splicing in the three different classes of clock neurons assayed (including LN_vs, DN1s⁷²), supporting the idea that regulation of sif is important for clock neuron function.

Since sif encodes a Rac1 GEF, we wanted to test whether Rac1 itself can regulate the plasticity of s-LNv projections. We first used a UAS transgene expressing a constitutively active version of Rac1 (UAS-Rac1^CA) that acts on downstream targets independently of GEF activity⁷³. We used Pdf-Gal4 and tub-Gal80^ts to induce expression of UAS-Rac1^CA at dusk, and quantified the 3D spread of s-LNv projections using antibodies to PDF, as the flies did not contain the Pdf-RFP transgene. We found that inducing expression of Rac1^CA for just 2 h from ZT12 to ZT14 increased the 3D spread of the s-LNvs when control s-LNv projections are normally retracted (Fig. 5B). Conversely, we found that s-LNv projections remained retracted when an RNAi targeting Rac1 was expressed for 14 h starting at ZT12 and ending at ZT2, again using PDF to visualize s-LNv projections (Fig. 5C). This UAS-Rac1-RNAi transgene was previously shown to reduce Rac1 expression in fly heads when crossed to elav-Gal4⁷⁴. Together, these data show that Rac1 is normally involved in expanding s-LNv projections at dawn, and that uncoupling the regulation of Rac1 from a GEF is sufficient to rapidly expand s-LNv projections when they are normally retracted at dusk.

A genetic interaction between sif and Fmr1

We have shown that sif mRNA is an FMRP target in s-LNvs (Fig. 4B), that sif is required at dawn for s-LNv projections to expand, and that abnormally high sif mRNA levels at dusk keep s-LNv projections expanded (Fig. 5A). We therefore hypothesized that FMRP normally represses sif mRNA translation at dusk to help retract s-LNv projections.

We used genetics to test this idea by asking if the abnormally expanded s-LNv projections at dusk caused by Fmr1-shRNA can be rescued by co-expressing sif-shRNA. Our logic was as follows: If FMRP normally represses sif translation at dusk, then FMRP knockdown at dusk increases sif mRNA translation, leading to expanded s-LNv projections. Thus, co-expressing sif-shRNA alongside Fmr1-shRNA might return s-LNv projections to their normally retracted ZT14 phenotype if the FMRP-sif mRNA interaction is important for s-LNv structural plasticity.

We again induced expression of the transgenes for 4 h using Pdf-Gal4 and tub-Gal80^ts, with flies also containing the Pdf-RFP transgene. The results in Fig. 6 confirm that expressing Fmr1-shRNA alone from ZT10 to ZT14 prevented s-LNv projections from retracting, as in Fig. 2A. However, we found that co-expressing Fmr1-shRNA with sif-shRNA gave retracted s-LNv projections at ZT14 that were not significantly different from control s-LNvs at ZT14. Expressing sif-shRNA alone did not significantly affect the 3D spread of s-LNv projections (Fig. 5A).

Fig. 6 — Left: Representative confocal images of s-LNv projections from flies with *Pdf-Gal4* and *tubulin-Gal80*^ts crossed to w flies (Control: *Pdf, tub-Gal80*^ts > +), flies with an shRNA transgene against *Fmr1* only (*Fmr1* knockdown: *Pdf, tub-Gal80*^ts > *UAS-sh-Fmr1*), flies with shRNA transgenes against both *Fmr1* and *sif* (*Fmr1* + *sif* knockdown: *Pdf, tub-Gal80*^ts > *UAS-sh-Fmr1* + *UAS-sh-sif)* or flies with shRNA transgenes against both *Fmr1* and *CG43102* (*Fmr1* + *CG43102* knockdown: *Pdf, tub-Gal80*^ts > *UAS-sh-Fmr1* + *UAS-sh-CG43102*). Transgenes were induced by moving flies to 30 °C at ZT10 and dissecting 4 h later at ZT14. All flies also contained a *Pdf-RFP* transgene, and projections were visualized using an antibody to RFP. Right: Quantification of the 3D spread of s-LNv projections as in Fig. 1. A one-way ANOVA followed by a Tukey post-hoc test was used to assess significance. s-LNv projections in control flies at ZT14 (gray bar) differ from flies expressing *Fmr1-shRNA* in LNvs at ZT14 (pale blue bar, p = 0.029, denoted by *p < 0.05), which in turn differ from s-LNv projections in flies expressing *Fmr1-shRNA* and *sif-shRNA* at ZT14 (yellow stripes on pale blue bar, p = 0.048, denoted by *p < 0.05). s-LNv projections at ZT14 in flies expressing *Fmr1-shRNA* and *sif-shRNA* at ZT14 differ from flies expressing *Fmr1-shRNA* and *CG43102-shRNA* (green stripes on pale blue bar, p = 0.16, denoted by *p < 0.05). Sample sizes pooled from 6 independent experiments: Controls – 7 hemispheres; *Fmr1-shRNA* – 23 hemispheres; *Fmr1-shRNA* + *sif-shRNA* – 13 hemispheres; *Fmr1-shRNA* + *CG43102-shRNA* – 7 hemispheres.

To test if expressing a second shRNA transgene diminishes the efficacy of the UAS-Fmr1-shRNA transgene, we used the UAS-CG43102-shRNA transgene as a control. We found that co-expressing Fmr1-shRNA and CG43102-shRNA gave expanded projections that were not significantly different from expressing Fmr1-shRNA alone. Thus, the rescue of the Fmr1-shRNA phenotype by sif-shRNA is not a trivial consequence of overwhelming either the RNAi pathway or the Gal4/UAS system with two UAS-shRNA transgenes. Instead, we conclude that reducing sif expression specifically overrides the defect caused by reduced Fmr1 expression. This result is consistent with FMRP normally repressing sif translation at dusk to allow s-LNvs to retract.

Discussion

It has been challenging to address the roles of FMRP in mature neurons, given the effects of Fmr1 null mutants on neuronal development. We used Drosophila genetics to test the acute role of FMRP in the plasticity of adult neurons and found that FMRP rapidly and bidirectionally regulates the plasticity of adult s-LNv circadian pacemaker neurons: Too much FMRP blocks the activity-dependent expansion of s-LNv projections at dawn, and too little FMRP prevents their retraction at dusk. Working in this single cell type, we identified a set of mRNAs targeted by FMRP, and we showed that one target—a Rac1 GEF encoded by sif—explains at least some of the s-LNv cellular phenotypes caused by altering Fmr1 expression.

Our study adds to the roles of RBPs in Drosophila circadian pacemaker neurons, where RBPs can regulate circadian period length. One mechanism for this is the regulation of per mRNA translation by the RBPs Atx-2 and Tyf^75–77. Another example is the RBP Lark, which targets double-time (dbt) RNA in clock neurons⁷⁸. dbt encodes the kinase that controls PER protein stability and thus circadian period length⁷⁹.

Identifying sif as an FMRP target in s-LNvs allowed us to establish that Sif/Rac1 is normally involved in expanding s-LNv projections at dawn. Our data support the model that FMRP reduces translation of sif mRNA at dusk to regulate s-LNv structural plasticity (Fig. 7), although we were unable to directly test this model by measuring Sif protein levels in s-LNvs. FMRP decreasing Sif levels at dusk would reduce Rac1 activity at the same time as rising Pura mRNA levels increase Rho1 activity³¹. It is relatively common for cells to decrease one process while simultaneously increasing the opposing process. This can help turn the output of two individual graded signals into a switch – for example, in the interplay between transcriptional activators and repressors that bind the same or overlapping DNA-sequences^80,81. An analogous process with small GTPases could help ensure that s-LNv projections are either expanding or retracting, but not trying to do both simultaneously. Indeed, s-LNv plasticity phenotypes tend to leave projections either constitutively expanded or constitutively retracted, and rarely stuck in between states (e.g., Figs. 2, 5, and 6)^30,31. In addition, crosstalk between Rac1 and Rho1 GTPases themselves has been observed⁸², and this could also help ensure switch-like behavior between expansion and retraction in s-LNv projections. We also note that the pre-synaptic protein Bruchpilot (Brp) associates with Sif and RhoGAP100F⁸³. Thus, Sif could also be involved in the daily rhythms observed in the number of active zones in s-LNvs projections^29,31. In addition, it is intriguing that Brp binds RhoGAP100F, which could inactivate Rho1 and could thereby help the switch from Rho1 to Rac1 activity, although this remains to be tested. Finally, it should also be noted that FMRP itself indirectly regulates the Rac1-Wave regulatory complex via its interactions with CYFIP1 in mice^84,85.

Fig. 7 — Top: *sif* mRNA (orange) is translated around dawn, which allows Sif protein (orange) to activate Rac1 (dark orange), which in turn expands s-LNv projections. Bottom: Increased FMRP activity (blue) at dusk reduces *sif* mRNA translation, thereby reducing activation of Rac1 and helping to retract s-LNv projections.

For FMRP to dynamically regulate Sif protein levels, FMRP activity must change over 24 h so that FMRP is more active at dusk than at dawn. In theory, FMRP activity could be regulated at any level from transcription to post-transcriptional modifications. We could not detect any changes in FMRP levels in s-LNvs between dawn and dusk (Supplementary Fig. 2)—although we had to focus on s-LNv cell bodies due to challenges in detecting endogenous FMRP in neuronal projections. Mammalian FMRP is dephosphorylated in response to neuronal activity, which disaggregates FMRP-mRNA granules, thereby increasing translation of mRNAs that were previously bound by FMRP⁴⁷. Such a mechanism could help explain how experimentally increasing neuronal activity at dusk expands s-LNv projections in only 2 h³⁰. However, phosphorylation of FMRP is unlikely to fully explain how neuronal activity regulates s-LNv plasticity: Overexpressing FMRP at dawn blocks expansion of s-LNv projections even though s-LNvs seem to continue to fire normally, as shown by expression of the activity-dependent Hr38 transcriptional reporter gene (Fig. 3B). We speculate that FMRP activity is normally regulated by activity-mediated dephosphorylation in s-LNvs, but that inducing FMRP overexpression can override this. We also note that translation of Fmr1 mRNA can be regulated by neuronal activity²¹. Understanding the regulation of FMRP activity in s-LNvs will ultimately require developing tools to label FMRP in a specific cell type to clearly visualize FMRP throughout the entirety of a neuron—as has been done for other neuronal proteins such as Brp⁸⁶.

TRIBE-seq also identified sif mRNA as an FMRP target in cholinergic and GABAergic neurons in Drosophila³⁷, raising the possibility that the mechanism described here is a general mechanism for controlling neuronal plasticity. Indeed, Tiam1 was identified as an FMRP target in mouse brains using CLIP¹⁰. This idea is further supported by the genetic interaction of Fmr1 and Rac1 in the development of dendrites in the Drosophila peripheral nervous system⁸⁷.

The post-translational regulation of Sif/Rac1 activity described here contrasts with the transcriptional regulation of Pura/Rho1 activity that drives retraction of s-LNv projections at dusk³¹. However, the requirement of Mef2 in expanding s-LNv projections³⁰ suggests that activity-dependent transcriptional regulation is also important at dawn. Thus, s-LNv structural plasticity is similar to the plasticity associated with learning and memory by using both transcription factors and RNA-binding proteins to regulate gene expression^3,88, making s-LNvs an interesting model to study the molecular mechanisms of neuronal plasticity. In conclusion, our data reveal a rapid and direct role for FMRP in regulating neuronal plasticity in adult neurons, and underscore the importance of RNA-binding proteins in plasticity.

Methods

Fly rearing and transgenes

For transgene induction experiments, flies were raised at 19 °C and then entrained as adults for at least 4 days in 12:12 light-dark cycles at 19 °C. To induce transgene expression, the incubator temperature was changed to 30 °C for 2, 4, or 14 h before dissections. Transgenic flies used were as follows, with Bloomington Stock IDs in parentheses:

Gal4 and Gal80 lines

Pdf-Gal4 (Bloomington Stock# 80939)²⁴; Pura-Gal4 (VT027163-Gal4), described in ref. ⁴⁵; tub-Gal80^ts (7018)⁴²; and elav-GS (43642)⁶⁹.

Reporter genes

Pdf-RFP²⁷; Hr38-Tomato⁵⁵

UAS transgenes

UAS-Fmr1 (6931)⁶¹; UAS-ADARcd (UAS-ADAR-E488Q from ref. ⁸⁹); UAS-Fmr1-ADAR³⁷; UAS-sif (9127)⁹⁰; UAS-myrRFP⁸⁶; UAS-Rac1^CA (6291)⁷³; UAS-GFP (90912)⁹¹ and UAS-Fmr1-GFP (92833)⁹². We used the EY01540 EPgy2 insertion line inserted just upstream of CG43102 for the CG43102 overexpression experiment in Supplementary Fig. 6 (15515).

UAS-shRNA lines

UAS-sh-Fmr1 (V20) (34944); UAS-sh-Fmr1 (V22) (35200); UAS-sh-sif (61934); UAS-sh-CG43102 (62371). We also used UAS-sh-Rac1 (34910), which was previously shown to reduce Rac1 expression in fly heads when crossed to elav-Gal4⁷⁴.

Immunofluorescence and imaging

We used a mix of female and male flies for all imaging experiments, with immunofluorescence performed similarly to ref. ³¹. Briefly, dissected brains were fixed in 4% formaldehyde for 20 min at room temperature before permeabilizing with 1% Triton X-100 in PBS and blocking with 10% heat-inactivated goat serum in PBS with 0.5% Triton X-100 (PBST). Samples were incubated with primary antibodies in the same PBST buffer with serum at 4 °C for two nights in all experiments except for Fig. 4B, which was for one night. After washing in PBST, samples were incubated with the appropriate secondary antibodies in PBST + 10% serum for 1 h at room temperature before further washing in PBST. Finally, samples were allowed to sink in 50% glycerol and then mounted on a glass slide and coverslip with SlowFade™ Gold Antifade Mountant (Thermo Fisher). Slides were stored at 4 °C in the dark until imaging on a Leica SP8 confocal microscope for all images except those in Fig. 4B, which were taken on a Leica SP5. Brains were imaged on the SP8 with the 63× lens at 1.94× zoom and a resolution of 1024 × 1024. For imaging s-LNv projections, a hemisphere would be picked (right or left) and all of the brains would be imaged from that hemisphere to reduce bias in picking the projection to image, unless there was a problem with that hemisphere (e.g., dust on the slide at that position). Antibodies used were Takara Living Colors dsRed polyclonal rabbit antibody (catalog number 632496) at 1:500 to recognize RFP, mouse anti-PDF C7 at 1:50⁹³ (provided by the Developmental Studies Hybridoma Bank), guinea Pig anti-Vri at 1:5000 (generously provided by Paul Hardin), and 6A15 monoclonal mouse anti-FMRP at 1:50 (Santa Cruz Biotechnology, catalog number sc-57005).

Quantification of the 3D spread of s-LNv projections

s-LNv projections were quantified using a custom MATLAB script as described previously³¹. Briefly, the confocal stacks were saved as separate JPEG files using ImageJ. The script was then run on those images, and a polygon was drawn around the dorsal projections, starting where the projections turn. The script generates the variation along the x, y, and z axes, and these values are multiplied together to determine the spread in 3 dimensions (3D spread) of the projections. The 3D spread data were then loaded into RStudio, and each condition was tested for normality with the Shapiro–Wilk method. This normality test was only significant for Fig. 2A for which a Kruskal–Wallis test followed by a Wilcoxon test was used between the conditions of interest, and the p-value was adjusted using the ‘p.adjust’ function in R with the Benjamini–Hochberg method with 7 as the number of comparisons. For all other figures, a one-way ANOVA was performed, followed by a Tukey HSD post-hoc test to test for significant differences between the conditions.

Quantification of immunofluorescence

ImageJ was used to quantify the intensity in s-LNv cell nuclei. The intensity value was obtained by drawing a polygon around the desired area and then using the ‘Measure’ function in ImageJ. An area outside the cell bodies was recorded as background, and the average of the s-LNv intensity across 4 cells was divided by the background to obtain a single datapoint.

FMRP TRIBE-seq in s-LNvs

Control flies had the genotype: Pdf-RFP / UAS-ADARcd; Pura-Gal4, tub-Gal80^ts / Sb or TM6B. Experimental flies had the genotype: Pdf-RFP / UAS-Fmr1-ADARcd; Pura-Gal4, tub-Gal80^ts / +. Flies were raised at 19 °C and then transgene expression was induced at ZT22 by shifting to 30 °C. Brains were dissected in Schneider’s media (Sigma-Aldrich) for 1 h from ZT2 to ZT3, and then dissociation was started by incubating a 0.5% Trypsin solution in PBS for 30 min at 25 °C in a shaking dry bath. Brains were then washed with ice-cold Schneider’s media and then with ice-cold D-PBS before pipetting up and down 50 times for mechanical disassociation. Tissue clumps were removed by passing the sample through a 20 µm filter. RFP+ events were sorted using a FACSAria II (BD Biosciences). A control sample of brains from w¹¹¹⁸ flies with no RFP transgene was used to set a gate to exclude non-fluorescent events. These were dissected and dissociated in parallel to Pdf-RFP brains, and a gate was set so it captured RFP+ events with a threshold so that none or only very rare events would be above the threshold for w¹¹¹⁸ brains. All samples were kept on ice while they waited to be sorted. In experiment 1, we sorted 61 and 76 RFP+ events per 100,000 events for the UAS-ADARcd control and the UAS-Fmr1-ADAR experimental group, respectively. In experiment 2, we sorted 26 and 41 RFP+ events per 100,000 events for the UAS-ADARcd control and the UAS-Fmr1-ADAR experimental group, respectively. RFP+ fragments were sorted into 300μL of Dynabeads™ mRNA Purification Kit lysis buffer. Samples were frozen at −80 °C until processing.

A Dynabeads™ mRNA purification kit was used to extract mRNA. SMART-seq3⁶³ was performed for bulk sorted events using 15 amplification cycles. Libraries were prepared with the Illumina XT DNA from 1 ng of amplified cDNA with 6 cycles of amplification when adding indices. The sizes of the library were checked using a Tapestation (Agilent Technologies) to ensure a peak centered around 300–800 nt. Libraries were sequenced on the Illumina NextSeq500 platform with paired-end reads at 150 base pairs in length. (2 × 150)

Analysis of TRIBE-seq data

Trimmomatic was used to trim sequencing data with the following parameters:

-phred33 HEADCROP:6 LEADING:25 TRAILING:25 AVGQUAL:25 MINLEN:19.

Sequences were aligned to the Drosophila genome version dm6 using STAR and the following parameters:

--outFilterMismatchNoverLmax 0.07 --outFileNamePrefix aligned_ADARcd_exp2 --outFilterMatchNmin 16 --outFilterMultimapNmax 1.

Samtools was then used to filter alignments that had a higher than 10 phred score with Samtools view and the -q 10 parameter. Duplicates were removed using Picard. The HyperTRIBE protocol⁶² was then followed to obtain the edited sites of the Fmr1-ADARcd condition using ADARcd RNA-seq as the control. The final file of the HyperTRIBE pipeline produced by the script ‘summarize_results.pl’ was imported into RStudio for analysis.

For Supplementary Fig. 4, genomic DNA was extracted from UAS-ADARcd; Sb / TM6b and UAS-Fmr1-ADAR flies using standard procedures. Libraries were prepared using the NEBNext Ultra II FS DNA Library Prep Kit for Illumina, using 30 min of fragmentation in the first step. Libraries were sequenced with MiSeq 2 × 75 − 150 Cycle v3. Sequences were aligned to the Drosophila genome version dm6 using BWA-MEM and SNPs identified using GATK HaplotypeCaller with the parameters recommended by the Broad Institute: QD < 2.0, FS > 60.0, MQ < 40.0, SOR > 4.0, MQRankSum < −12.5, and ReadPosRankSum < −8.0. Finally, the .vcf files were filtered to find SNPs that change from an A to G or the reverse complement T to C in the ADARcd to the Fmr1-ADARcd condition. Finally, these SNPs were converted into a .bed file and subtracted from the .bedgraph files obtained from the HyperTRIBE pipeline using bedtools.

Immunoprecipitation from fly heads

Flies that contained the transgenes elav-GeneSwitch (elav-GS)⁶⁹ and either UAS-Fmr1-GFP⁹² or UAS-GFP⁹¹ were raised on normal food until adulthood, and then placed onto food containing mifepristone at a concentration of 0.5 μg/μl for 48 h. We followed the protocol of ref. ⁹⁴, but we scaled the reagents used down to 1/20th and only followed the cytoplasmic part of the protocol. Briefly, a mix of 40 male and female flies was collected for each condition. Flies were anesthetized with CO₂, placed into 1.7 mL tubes, and frozen at −70 °C. Fly heads were separated from bodies using a small sieve pre-cooled to −70 °C. The frozen heads were ground into a powder with a cooled plastic homogenizer, and lysis buffer was added to the ground heads and homogenized again until the solution became dark in color. The lysate was centrifuged at 100 × g for 5 min at 4 °C to eliminate cuticle and debris, and then at 16,000 × g for 20 min at 4 °C to obtain the cytoplasmic fraction. 100 μl of the 400 μl total was saved as the input fraction.

25 μl of ChromoTek GFP-Trap magnetic agarose bead slurry (Proteintech) was equilibrated with lysis buffer by washing 3 times. Lysates were then incubated with beads for 1 h at 4 °C in a nutator. Beads were washed 3 times with lysis buffer, and then incubated with 3 μL of 20 μg/μl Proteinase K in 200 μl lysis buffer for 30 min at 55 °C with shaking. Finally, 1 mL of Trizol was added to each sample and stored at −70 °C until further processing.

RNA was extracted following the Trizol reagent protocol. The pellet was resuspended in 22 μl of DEPC-treated water. 9 μl was used for reverse transcription using the SuperScript™ III Reverse Transcriptase (Thermo Fisher) according to the manufacturer’s protocol. qPCR was performed on a CFX Opus 96 Real-Time PCR System with 0.5 μl of 1:5 diluted cDNA and Applied Biosystems™ Power SYBR™ Green PCR Master Mix for 40 cycles with an annealing temperature of 50 °C. We performed three technical replicates per condition, and the whole experiment was repeated twice.

We used the following primers for qPCR in Supplementary Fig. 5:

sif forward primer: 5′-ATCCCAAGCGCGACTACATC-3′

sif reverse primer: 5′-TTCCAGTTTTTCCGCGATGC-3′

Bx forward primer: 5′-AGCAACAGTAAGAGCCTTGT-3′

Bx reverse primer: 5′-TGCTGTTGGTTGGAGGAAAG-3′

Supplementary information

Supplementary information^{(7.2MB, pdf)}

41467_2025_66487_MOESM2_ESM.docx^{(14.4KB, docx)}

Description of additional supplementary files

Supplementary dataset 1^{(13.5KB, xlsx)}

Transparent Peer Review file^{(175.3KB, pdf)}

Source data

Source data^{(33.8KB, xlsx)}

Acknowledgements

We are very grateful to Eugenia Olesnicky for suggesting that we use TRIBE-Seq and to Bassem Hassan for first encouraging us to study FMRP many years ago. We are also very grateful to Michael Rosbash, Kate Abruzzi, and Aoife MacMahon for TRIBE flies, advice, and for sharing unpublished data. We also thank Tom Jongens for generously providing Fmr1 null mutants, fly food on which these mutants survive, and for helpful discussions. We also thank Eric Lai for advice throughout the course of this project, David Owald and Christine Vogel for helpful discussions, and Eric Klann for comments on the manuscript. We thank the Drosophila Transgenic RNAi Project at Harvard Medical School (NIGMS R01-GM084947) and the Bloomington Drosophila Stock Center for flies, and Paul Hardin and the Developmental Studies Hybridoma Bank for antibodies. We also thank Simon Kidd, Sofia Luminari, and Angelina Xu for help with dissections. This work was supported by the NYU IT High Performance Computing resources, services, and staff expertise. We acknowledge the Zegar Family Foundation for their generous support and the NYU Center for Genomics and Systems Biology Genomics Core. This investigation was conducted in facilities constructed with support from Research Facilities Improvement Grant Number C06 RR-15518-01 from the NIH National Center for Research Resources. DG and SL were partly supported by the NYU’s Graduate School of Arts & Science MacCracken Program. This work was supported by NIH grant GM136363 to J.B. and the NYUAD Center for Genomics & Systems Biology (J.B.).

Author contributions

D.G.G., S.L., and J.B. conceived and designed the project. D.G.G., S.L., J.L., and J.B. conducted experiments and analyzed data. D.G.G. and J.B. wrote the manuscript. All co-authors contributed to the final edited manuscript.

Peer review

Peer review information

Nature Communications thanks Thomas Jongens and the other anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

Source data are provided with this paper. TRIBE-seq data is available via GEO accession number GSE241810 at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE241810 Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-025-66487-0.

References

1.Winata, C. L. & Korzh, V. The translational regulation of maternal mRNAs in time and space. FEBS Lett.592, 3007–3023 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Abouward, R. & Schiavo, G. Walking the line: mechanisms underlying directional mRNA transport and localisation in neurons and beyond. Cell Mol. Life Sci.78, 2665–2681 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Ule, J. & Darnell, R. B. RNA binding proteins and the regulation of neuronal synaptic plasticity. Curr. Opin. Neurobiol.16, 102–110 (2006). [DOI] [PubMed] [Google Scholar]
4.Thelen, M. P. & Kye, M. J. The role of RNA binding proteins for local mrna translation: Implications in neurological disorders. Front. Mol. Biosci.6, 161 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Pieretti, M. et al. Absence of expression of the FMR-1 gene in Fragile X syndrome. Cell66, 817–822 (1991). [DOI] [PubMed] [Google Scholar]
6.Verkerk, A. J. et al. Identification of a gene (FMR-1) containing a CGG repeat coincident with a breakpoint cluster region exhibiting length variation in Fragile X syndrome. Cell65, 905–914 (1991). [DOI] [PubMed] [Google Scholar]
7.Nelson, D. L., Orr, H. T. & Warren, S. T. The unstable repeats-three evolving faces of neurological disease. Neuron77, 825–843 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Wang, L. W., Berry-Kravis, E. & Hagerman, R. J. Fragile X: leading the way for targeted treatments in autism. Neurotherapeutics7, 264–274 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Schaefer, G. B. & Mendelsohn, N. J. Genetics evaluation for the etiologic diagnosis of autism spectrum disorders. Genet. Med.10, 4–12 (2008). [DOI] [PubMed] [Google Scholar]
10.Darnell, J. C. et al. FMRP stalls ribosomal translocation on mRNAs linked to synaptic function and autism. Cell146, 247–261 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Greenblatt, E. J. & Spradling, A. C. Fragile X mental retardation 1 gene enhances the translation of large autism-related proteins. Science361, 709–712 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Zhou, L. T. et al. A novel role of Fragile X mental retardation protein in pre-mRNA alternative splicing through RNA-binding protein 14. Neuroscience349, 64–75 (2017). [DOI] [PubMed] [Google Scholar]
13.Brown, M. R. et al. Fragile X mental retardation protein controls gating of the sodium-activated otassium channel Slack. Nat. Neurosci.13, 819–821 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Deng, P. Y. et al. FMRP regulates neurotransmitter release and synaptic information transmission by modulating action potential duration via BK channels. Neuron77, 696–711 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Gholizadeh, S., Halder, S. K. & Hampson, D. R. Expression of Fragile X mental retardation protein in neurons and glia of the developing and adult mouse brain. Brain Res.1596, 22–30 (2015). [DOI] [PubMed] [Google Scholar]
16.Tessier, C. R. & Broadie, K. Drosophila Fragile X mental retardation protein developmentally regulates activity-dependent axon pruning. Development135, 1547–1557 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Bureau, I., Shepherd, G. M. & Svoboda, K. Circuit and plasticity defects in the developing somatosensory cortex of FMR1 knock-out mice. J. Neurosci.28, 5178–5188 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Harlow, E. G. et al. Critical period plasticity is disrupted in the barrel cortex of FMR1 knockout mice. Neuron65, 385–398 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Tamanini, F. et al. Differential expression of FMR1, FXR1 and FXR2 proteins in human brain and testis. Hum. Mol. Genet.6, 1315–1322 (1997). [DOI] [PubMed] [Google Scholar]
20.Zeier, Z. et al. Fragile X mental retardation protein replacement restores hippocampal synaptic function in a mouse model of Fragile X syndrome. Gene Ther.16, 1122–1129 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Weiler, I. J. et al. Fragile X mental retardation protein is translated near synapses in response to neurotransmitter activation. Proc. Natl. Acad. Sci. USA94, 5395–5400 (1997). 21. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Ferrari, F. et al. The Fragile X mental retardation protein-RNP granules show an mGluR-dependent localization in the post-synaptic spines. Mol. Cell Neurosci.34, 343–354 (2007). [DOI] [PubMed] [Google Scholar]
23.Gatto, C. L. & Broadie, K. Fragile X mental retardation protein is required for programmed cell death and clearance of developmentally-transient peptidergic neurons. Dev. Biol.356, 291–307 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Renn, S. C., Park, J. H., Rosbash, M., Hall, J. C. & Taghert, P. H. A pdf neuropeptide gene mutation and ablation of PDF neurons each cause severe abnormalities of behavioral circadian rhythms in Drosophila. Cell99, 791–802 (1999). [DOI] [PubMed] [Google Scholar]
25.Fernandez, M. P., Berni, J. & Ceriani, M. F. Circadian remodeling of neuronal circuits involved in rhythmic behavior. PLoS Biol.6, e69 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Allada, R. & Chung, B. Y. Circadian organization of behavior and physiology in Drosophila. Annu. Rev. Physiol.72, 605–624 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Ruben, M., Drapeau, M. D., Mizrak, D. & Blau, J. A mechanism for circadian control of pacemaker neuron excitability. J. Biol. Rhythms27, 353–364 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Cao, G. & Nitabach, M. N. Circadian control of membrane excitability in Drosophila melanogaster lateral ventral clock neurons. J. Neurosci.28, 6493–6501 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Gorostiza, E. A., Depetris-Chauvin, A., Frenkel, L., Pirez, N. & Ceriani, M. F. Circadian pacemaker neurons change synaptic contacts across the day. Curr. Biol.24, 2161–2167 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Sivachenko, A., Li, Y., Abruzzi, K. C. & Rosbash, M. The transcription factor Mef2 links the Drosophila core clock to Fas2, neuronal morphology, and circadian behavior. Neuron79, 281–292 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Petsakou, A., Sapsis, T. P. & Blau, J. Circadian rhythms in Rho1 activity regulate neuronal plasticity and network hierarchy. Cell162, 823–835 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Depetris-Chauvin, A. et al. Adult-specific electrical silencing of pacemaker neurons uncouples molecular clock from circadian outputs. Curr. Biol.21, 1783–1793 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Flavell, S. W. & Greenberg, M. E. Signaling mechanisms linking neuronal activity to gene expression and plasticity of the nervous system. Annu. Rev. Neurosci.31, 563–590 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Dockendorff, T. C. et al. Drosophila lacking dfmr1 activity show defects in circadian output and fail to maintain courtship interest. Neuron34, 973–984 (2002). [DOI] [PubMed] [Google Scholar]
35.Sekine, T. et al. Circadian phenotypes of Drosophila Fragile X mutants in alternative genetic backgrounds. Zool. Sci.25, 561–571 (2008). [DOI] [PubMed] [Google Scholar]
36.Gatto, C. L. & Broadie, K. Temporal requirements of the Fragile X mental retardation protein in modulating circadian clock circuit synaptic architecture. Front. Neural Circuits3, 8 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.McMahon, A. C. et al. TRIBE: Hijacking an RNA-editing enzyme to identify cell-specific targets of RNA-binding proteins. Cell165, 742–753 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Sone, M. et al. Still life, a protein in synaptic terminals of Drosophila homologous to GDP-GTP exchangers. Science275, 543–547 (1997). [DOI] [PubMed] [Google Scholar]
39.Cheng, J. et al. The Rac-GEF Tiam1 promotes dendrite and synapse stabilization of dentate granule cells and restricts hippocampal-dependent memory functions. J. Neurosci.41, 1191–1206 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Li, L. et al. Tiam1 coordinates synaptic structural and functional plasticity underpinning the pathophysiology of neuropathic pain. Neuron111, 2038–2050.e2036 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Aston, C., Jiang, L. & Sokolov, B. P. Transcriptional profiling reveals evidence for signaling and oligodendroglial abnormalities in the temporal cortex from patients with major depressive disorder. Mol. Psychiatry10, 309–322 (2005). [DOI] [PubMed] [Google Scholar]
42.McGuire, S. E., Le, P. T., Osborn, A. J., Matsumoto, K. & Davis, R. L. Spatiotemporal rescue of memory dysfunction in Drosophila. Science302, 1765–1768 (2003). [DOI] [PubMed] [Google Scholar]
43.Parisky, K. M. et al. PDF cells are a GABA-responsive wake-promoting component of the Drosophila sleep circuit. Neuron60, 672–682 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Sheeba, V. et al. Large ventral lateral neurons modulate arousal and sleep in Drosophila. Curr. Biol.18, 1537–1545 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Sekiguchi, M., Inoue, K., Yang, T., Luo, D. G. & Yoshii, T. A catalog of GAL4 drivers for labeling and manipulating circadian clock neurons in Drosophila melanogaster. J. Biol. Rhythms35, 207–213 (2020). [DOI] [PubMed] [Google Scholar]
46.Morales, J. et al. Drosophila Fragile X protein, DFXR, regulates neuronal morphology and function in the brain. Neuron34, 961–972 (2002). [DOI] [PubMed] [Google Scholar]
47.Tsang, B. et al. Phosphoregulated FMRP phase separation models activity-dependent translation through bidirectional control of mRNA granule formation. Proc. Natl Acad. Sci. USA116, 4218–4227 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Lymer, S., Patel, K., Lennon, J. & Blau, J. Circadian clock neurons use activity-regulated gene expression to control structural plasticity. Preprint at bioRxiv10.1101/2024.05.25.595887 (2024).
49.Sawicka, K. et al. FMRP has a cell-type-specific role in CA1 pyramidal neurons to regulate autism-related transcripts and circadian memory. eLife10.7554/eLife.46919 (2019). [DOI] [PMC free article] [PubMed]
50.Contractor, A., Klyachko, V. A. & Portera-Cailliau, C. Altered neuronal and circuit excitability in Fragile X Syndrome. Neuron87, 699–715 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Cyran, S. A. et al. vrille, Pdp1, and dClock form a second feedback loop in the Drosophila circadian clock. Cell112, 329–341 (2003). [DOI] [PubMed] [Google Scholar]
52.Fernandez-Chiappe, F. et al. High-frequency neuronal bursting is essential for circadian and sleep behaviors in Drosophila. J. Neurosci.41, 689–710 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Lin, D. et al. Functional identification of an aggression locus in the mouse hypothalamus. Nature470, 221–226 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Chen, X., Rahman, R., Guo, F. & Rosbash, M. Genome-wide identification of neuronal activity-regulated genes in Drosophila. eLife 5, e19942 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Zhu, Z. et al. A gene expression program activated by neuronal inactivity. Preprint at bioRxiv10.1101/636878 (2024).
56.Collins, B., Kane, E. A., Reeves, D. C., Akabas, M. H. & Blau, J. Balance of activity between LN_vs and glutamatergic dorsal clock neurons promotes robust circadian rhythms in Drosophila. Neuron74, 706–718 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Worpenberg, L. et al. Ythdf is a N6-methyladenosine reader that modulates Fmr1 target mRNA selection and restricts axonal growth in Drosophila. EMBO J.40, e104975 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Bardoni, B., Schenck, A. & Mandel, J. L. A novel RNA-binding nuclear protein that interacts with the Fragile X mental retardation (FMR1) protein. Hum. Mol. Genet.8, 2557–2566 (1999). [DOI] [PubMed] [Google Scholar]
59.Napoli, I. et al. The Fragile X syndrome protein represses activity-dependent translation through CYFIP1, a new 4E-BP. Cell134, 1042–1054 (2008). [DOI] [PubMed] [Google Scholar]
60.Raji, J. I. & Potter, C. J. The number of neurons in Drosophila and mosquito brains. PLoS ONE16, e0250381 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Zhang, Y. Q. et al. Drosophila Fragile X-related gene regulates the MAP1B homolog Futsch to control synaptic structure and function. Cell107, 591–603 (2001). [DOI] [PubMed] [Google Scholar]
62.Rahman, R., Xu, W., Jin, H. & Rosbash, M. Identification of RNA-binding protein targets with HyperTRIBE. Nat. Protoc.13, 1829–1849 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Hagemann-Jensen, M. et al. Single-cell RNA counting at allele and isoform resolution using Smart-seq3. Nat. Biotechnol.38, 708–714 (2020). [DOI] [PubMed] [Google Scholar]
64.Sears, J. C., Choi, W. J. & Broadie, K. Fragile X mental retardation protein positively regulates PKA anchor Rugose and PKA activity to control actin assembly in learning/memory circuitry. Neurobiol. Dis.127, 53–64 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
65.Sudhakaran, I. P. et al. FMRP and Ataxin-2 function together in long-term olfactory habituation and neuronal translational control. Proc. Natl. Acad. Sci. USA111, E99–E108 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
66.Tang, X. et al. FMRP binds Per1 mRNA and downregulates its protein expression in mice. Mol. Brain16, 33 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
67.Gibson, J. R., Bartley, A. F., Hays, S. A. & Huber, K. M. Imbalance of neocortical excitation and inhibition and altered UP states reflect network hyperexcitability in the mouse model of Fragile X syndrome. J. Neurophysiol.100, 2615–2626 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
68.Polcownuk, S., Yoshii, T. & Ceriani, M. F. Decapentaplegic acutely defines the connectivity of central pacemaker neurons in Drosophila. J. Neurosci.41, 8338–8350 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
69.Nicholson, L. et al. Spatial and temporal control of gene expression in Drosophila using the inducible GeneSwitch GAL4 system. I. Screen for larval nervous system drivers. Genetics178, 215–234 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
70.Tsai, L. T., Bainton, R. J., Blau, J. & Heberlein, U. Lmo mutants reveal a novel role for circadian pacemaker neurons in cocaine-induced behaviors. PLoS Biol.2, e408 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
71.Tolias, K. F. et al. The Rac1-GEF Tiam1 couples the NMDA receptor to the activity-dependent development of dendritic arbors and spines. Neuron45, 525–538 (2005). [DOI] [PubMed] [Google Scholar]
72.Wang, Q., Abruzzi, K. C., Rosbash, M. & Rio, D. C. Striking circadian neuron diversity and cycling of Drosophila alternative splicing. eLife10.7554/eLife.35618 (2018). [DOI] [PMC free article] [PubMed]
73.Luo, L., Liao, Y. J., Jan, L. Y. & Jan, Y. N. Distinct morphogenetic functions of similar small GTPases: Drosophila Drac1 is involved in axonal outgrowth and myoblast fusion. Genes Dev.8, 1787–1802 (1994). [DOI] [PubMed] [Google Scholar]
74.Kikuchi, M. et al. Disruption of a RAC1-centred network is associated with Alzheimer’s disease pathology and causes age-dependent neurodegeneration. Hum. Mol. Genet.29, 817–833 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
75.Lim, C. et al. The novel gene twenty-four defines a critical translational step in the Drosophila clock. Nature470, 399–403 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
76.Lim, C. & Allada, R. ATAXIN-2 activates PERIOD translation to sustain circadian rhythms in Drosophila. Science340, 875–879 (2013). [DOI] [PubMed] [Google Scholar]
77.Zhang, Y., Ling, J., Yuan, C., Dubruille, R. & Emery, P. A role for Drosophila ATX2 in activation of PER translation and circadian behavior. Science340, 879–882 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
78.Huang, Y., McNeil, G. P. & Jackson, F. R. Translational regulation of the DOUBLETIME/CKIdelta/epsilon kinase by LARK contributes to circadian period modulation. PLoS Genet.10, e1004536 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
79.Price, J. L. et al. double-time is a novel Drosophila clock gene that regulates PERIOD protein accumulation. Cell94, 83–95 (1998). [DOI] [PubMed] [Google Scholar]
80.Stanojevic, D., Small, S. & Levine, M. Regulation of a segmentation stripe by overlapping activators and repressors in the Drosophila embryo. Science254, 1385–1387 (1991). [DOI] [PubMed] [Google Scholar]
81.Rossi, F. M., Kringstein, A. M., Spicher, A., Guicherit, O. M. & Blau, H. M. Transcriptional control: rheostat converted to on/off switch. Mol. cell6, 723–728 (2000). [DOI] [PubMed] [Google Scholar]
82.Guilluy, C., Garcia-Mata, R. & Burridge, K. Rho protein crosstalk: another social network?. Trends Cell Biol.21, 718–726 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
83.Owald, D. et al. A Syd-1 homologue regulates pre- and postsynaptic maturation in Drosophila. J. Cell Biol.188, 565–579 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
84.De Rubeis, S. et al. CYFIP1 coordinates mRNA translation and cytoskeleton remodeling to ensure proper dendritic spine formation. Neuron79, 1169–1182 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
85.Santini, E. et al. Reducing eIF4E-eIF4G interactions restores the balance between protein synthesis and actin dynamics in Fragile X syndrome model mice. Sci. Signal10.1126/scisignal.aan0665 (2017). [DOI] [PMC free article] [PubMed]
86.Chen, Y. et al. Cell-type-specific labeling of synapses in vivo through synaptic tagging with recombination. Neuron81, 280–293 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
87.Lee, A. et al. Control of dendritic development by the Drosophila Fragile X-related gene involves the small GTPase Rac1. Development130, 5543–5552 (2003). [DOI] [PubMed] [Google Scholar]
88.Kandel, E. R. The molecular biology of memory storage: a dialogue between genes and synapses. Science294, 1030–1038 (2001). [DOI] [PubMed] [Google Scholar]
89.Xu, W., Rahman, R. & Rosbash, M. Mechanistic implications of enhanced editing by a HyperTRIBE RNA-binding protein. RNA24, 173–182 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
90.Sone, M. et al. Synaptic development is controlled in the periactive zones of Drosophila synapses. Development127, 4157–4168 (2000). [DOI] [PubMed] [Google Scholar]
91.Shearin, H. K., Macdonald, I. S., Spector, L. P. & Stowers, R. S. Hexameric GFP and mCherry reporters for the Drosophila GAL4, Q, and LexA transcription systems. Genetics196, 951–960 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
92.Buddika, K., Xu, J., Ariyapala, I. S. & Sokol, N. S. I-KCKT allows dissection-free RNA profiling of adult Drosophila intestinal progenitor cells. Development10.1242/dev.196568 (2021). [DOI] [PMC free article] [PubMed]
93.Cyran, S. A. et al. The Double-time protein kinase regulates the subcellular localization of the Drosophila clock protein Period. J. Neurosci.25, 5430–5437 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
94.Heim, M., Blot, L. & Besse, F. An RNA-immunoprecipitation protocol to identify RNAs associated with RNA-binding proteins in cytoplasmic and nuclear Drosophila head fractions. STAR Protoc.3, 101415 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary information^{(7.2MB, pdf)}

41467_2025_66487_MOESM2_ESM.docx^{(14.4KB, docx)}

Description of additional supplementary files

Supplementary dataset 1^{(13.5KB, xlsx)}

Transparent Peer Review file^{(175.3KB, pdf)}

Source data^{(33.8KB, xlsx)}

Data Availability Statement

[CR1] 1.Winata, C. L. & Korzh, V. The translational regulation of maternal mRNAs in time and space. FEBS Lett.592, 3007–3023 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR2] 2.Abouward, R. & Schiavo, G. Walking the line: mechanisms underlying directional mRNA transport and localisation in neurons and beyond. Cell Mol. Life Sci.78, 2665–2681 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR3] 3.Ule, J. & Darnell, R. B. RNA binding proteins and the regulation of neuronal synaptic plasticity. Curr. Opin. Neurobiol.16, 102–110 (2006). [DOI] [PubMed] [Google Scholar]

[CR4] 4.Thelen, M. P. & Kye, M. J. The role of RNA binding proteins for local mrna translation: Implications in neurological disorders. Front. Mol. Biosci.6, 161 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR5] 5.Pieretti, M. et al. Absence of expression of the FMR-1 gene in Fragile X syndrome. Cell66, 817–822 (1991). [DOI] [PubMed] [Google Scholar]

[CR6] 6.Verkerk, A. J. et al. Identification of a gene (FMR-1) containing a CGG repeat coincident with a breakpoint cluster region exhibiting length variation in Fragile X syndrome. Cell65, 905–914 (1991). [DOI] [PubMed] [Google Scholar]

[CR7] 7.Nelson, D. L., Orr, H. T. & Warren, S. T. The unstable repeats-three evolving faces of neurological disease. Neuron77, 825–843 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR8] 8.Wang, L. W., Berry-Kravis, E. & Hagerman, R. J. Fragile X: leading the way for targeted treatments in autism. Neurotherapeutics7, 264–274 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR9] 9.Schaefer, G. B. & Mendelsohn, N. J. Genetics evaluation for the etiologic diagnosis of autism spectrum disorders. Genet. Med.10, 4–12 (2008). [DOI] [PubMed] [Google Scholar]

[CR10] 10.Darnell, J. C. et al. FMRP stalls ribosomal translocation on mRNAs linked to synaptic function and autism. Cell146, 247–261 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR11] 11.Greenblatt, E. J. & Spradling, A. C. Fragile X mental retardation 1 gene enhances the translation of large autism-related proteins. Science361, 709–712 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR12] 12.Zhou, L. T. et al. A novel role of Fragile X mental retardation protein in pre-mRNA alternative splicing through RNA-binding protein 14. Neuroscience349, 64–75 (2017). [DOI] [PubMed] [Google Scholar]

[CR13] 13.Brown, M. R. et al. Fragile X mental retardation protein controls gating of the sodium-activated otassium channel Slack. Nat. Neurosci.13, 819–821 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR14] 14.Deng, P. Y. et al. FMRP regulates neurotransmitter release and synaptic information transmission by modulating action potential duration via BK channels. Neuron77, 696–711 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR15] 15.Gholizadeh, S., Halder, S. K. & Hampson, D. R. Expression of Fragile X mental retardation protein in neurons and glia of the developing and adult mouse brain. Brain Res.1596, 22–30 (2015). [DOI] [PubMed] [Google Scholar]

[CR16] 16.Tessier, C. R. & Broadie, K. Drosophila Fragile X mental retardation protein developmentally regulates activity-dependent axon pruning. Development135, 1547–1557 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR17] 17.Bureau, I., Shepherd, G. M. & Svoboda, K. Circuit and plasticity defects in the developing somatosensory cortex of FMR1 knock-out mice. J. Neurosci.28, 5178–5188 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR18] 18.Harlow, E. G. et al. Critical period plasticity is disrupted in the barrel cortex of FMR1 knockout mice. Neuron65, 385–398 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR19] 19.Tamanini, F. et al. Differential expression of FMR1, FXR1 and FXR2 proteins in human brain and testis. Hum. Mol. Genet.6, 1315–1322 (1997). [DOI] [PubMed] [Google Scholar]

[CR20] 20.Zeier, Z. et al. Fragile X mental retardation protein replacement restores hippocampal synaptic function in a mouse model of Fragile X syndrome. Gene Ther.16, 1122–1129 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR21] 21.Weiler, I. J. et al. Fragile X mental retardation protein is translated near synapses in response to neurotransmitter activation. Proc. Natl. Acad. Sci. USA94, 5395–5400 (1997). 21. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR22] 22.Ferrari, F. et al. The Fragile X mental retardation protein-RNP granules show an mGluR-dependent localization in the post-synaptic spines. Mol. Cell Neurosci.34, 343–354 (2007). [DOI] [PubMed] [Google Scholar]

[CR23] 23.Gatto, C. L. & Broadie, K. Fragile X mental retardation protein is required for programmed cell death and clearance of developmentally-transient peptidergic neurons. Dev. Biol.356, 291–307 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR24] 24.Renn, S. C., Park, J. H., Rosbash, M., Hall, J. C. & Taghert, P. H. A pdf neuropeptide gene mutation and ablation of PDF neurons each cause severe abnormalities of behavioral circadian rhythms in Drosophila. Cell99, 791–802 (1999). [DOI] [PubMed] [Google Scholar]

[CR25] 25.Fernandez, M. P., Berni, J. & Ceriani, M. F. Circadian remodeling of neuronal circuits involved in rhythmic behavior. PLoS Biol.6, e69 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR26] 26.Allada, R. & Chung, B. Y. Circadian organization of behavior and physiology in Drosophila. Annu. Rev. Physiol.72, 605–624 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR27] 27.Ruben, M., Drapeau, M. D., Mizrak, D. & Blau, J. A mechanism for circadian control of pacemaker neuron excitability. J. Biol. Rhythms27, 353–364 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR28] 28.Cao, G. & Nitabach, M. N. Circadian control of membrane excitability in Drosophila melanogaster lateral ventral clock neurons. J. Neurosci.28, 6493–6501 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR29] 29.Gorostiza, E. A., Depetris-Chauvin, A., Frenkel, L., Pirez, N. & Ceriani, M. F. Circadian pacemaker neurons change synaptic contacts across the day. Curr. Biol.24, 2161–2167 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR30] 30.Sivachenko, A., Li, Y., Abruzzi, K. C. & Rosbash, M. The transcription factor Mef2 links the Drosophila core clock to Fas2, neuronal morphology, and circadian behavior. Neuron79, 281–292 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR31] 31.Petsakou, A., Sapsis, T. P. & Blau, J. Circadian rhythms in Rho1 activity regulate neuronal plasticity and network hierarchy. Cell162, 823–835 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR32] 32.Depetris-Chauvin, A. et al. Adult-specific electrical silencing of pacemaker neurons uncouples molecular clock from circadian outputs. Curr. Biol.21, 1783–1793 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR33] 33.Flavell, S. W. & Greenberg, M. E. Signaling mechanisms linking neuronal activity to gene expression and plasticity of the nervous system. Annu. Rev. Neurosci.31, 563–590 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR34] 34.Dockendorff, T. C. et al. Drosophila lacking dfmr1 activity show defects in circadian output and fail to maintain courtship interest. Neuron34, 973–984 (2002). [DOI] [PubMed] [Google Scholar]

[CR35] 35.Sekine, T. et al. Circadian phenotypes of Drosophila Fragile X mutants in alternative genetic backgrounds. Zool. Sci.25, 561–571 (2008). [DOI] [PubMed] [Google Scholar]

[CR36] 36.Gatto, C. L. & Broadie, K. Temporal requirements of the Fragile X mental retardation protein in modulating circadian clock circuit synaptic architecture. Front. Neural Circuits3, 8 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR37] 37.McMahon, A. C. et al. TRIBE: Hijacking an RNA-editing enzyme to identify cell-specific targets of RNA-binding proteins. Cell165, 742–753 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR38] 38.Sone, M. et al. Still life, a protein in synaptic terminals of Drosophila homologous to GDP-GTP exchangers. Science275, 543–547 (1997). [DOI] [PubMed] [Google Scholar]

[CR39] 39.Cheng, J. et al. The Rac-GEF Tiam1 promotes dendrite and synapse stabilization of dentate granule cells and restricts hippocampal-dependent memory functions. J. Neurosci.41, 1191–1206 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR40] 40.Li, L. et al. Tiam1 coordinates synaptic structural and functional plasticity underpinning the pathophysiology of neuropathic pain. Neuron111, 2038–2050.e2036 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR41] 41.Aston, C., Jiang, L. & Sokolov, B. P. Transcriptional profiling reveals evidence for signaling and oligodendroglial abnormalities in the temporal cortex from patients with major depressive disorder. Mol. Psychiatry10, 309–322 (2005). [DOI] [PubMed] [Google Scholar]

[CR42] 42.McGuire, S. E., Le, P. T., Osborn, A. J., Matsumoto, K. & Davis, R. L. Spatiotemporal rescue of memory dysfunction in Drosophila. Science302, 1765–1768 (2003). [DOI] [PubMed] [Google Scholar]

[CR43] 43.Parisky, K. M. et al. PDF cells are a GABA-responsive wake-promoting component of the Drosophila sleep circuit. Neuron60, 672–682 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR44] 44.Sheeba, V. et al. Large ventral lateral neurons modulate arousal and sleep in Drosophila. Curr. Biol.18, 1537–1545 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR45] 45.Sekiguchi, M., Inoue, K., Yang, T., Luo, D. G. & Yoshii, T. A catalog of GAL4 drivers for labeling and manipulating circadian clock neurons in Drosophila melanogaster. J. Biol. Rhythms35, 207–213 (2020). [DOI] [PubMed] [Google Scholar]

[CR46] 46.Morales, J. et al. Drosophila Fragile X protein, DFXR, regulates neuronal morphology and function in the brain. Neuron34, 961–972 (2002). [DOI] [PubMed] [Google Scholar]

[CR47] 47.Tsang, B. et al. Phosphoregulated FMRP phase separation models activity-dependent translation through bidirectional control of mRNA granule formation. Proc. Natl Acad. Sci. USA116, 4218–4227 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR48] 48.Lymer, S., Patel, K., Lennon, J. & Blau, J. Circadian clock neurons use activity-regulated gene expression to control structural plasticity. Preprint at bioRxiv10.1101/2024.05.25.595887 (2024).

[CR49] 49.Sawicka, K. et al. FMRP has a cell-type-specific role in CA1 pyramidal neurons to regulate autism-related transcripts and circadian memory. eLife10.7554/eLife.46919 (2019). [DOI] [PMC free article] [PubMed]

[CR50] 50.Contractor, A., Klyachko, V. A. & Portera-Cailliau, C. Altered neuronal and circuit excitability in Fragile X Syndrome. Neuron87, 699–715 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR51] 51.Cyran, S. A. et al. vrille, Pdp1, and dClock form a second feedback loop in the Drosophila circadian clock. Cell112, 329–341 (2003). [DOI] [PubMed] [Google Scholar]

[CR52] 52.Fernandez-Chiappe, F. et al. High-frequency neuronal bursting is essential for circadian and sleep behaviors in Drosophila. J. Neurosci.41, 689–710 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR53] 53.Lin, D. et al. Functional identification of an aggression locus in the mouse hypothalamus. Nature470, 221–226 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR54] 54.Chen, X., Rahman, R., Guo, F. & Rosbash, M. Genome-wide identification of neuronal activity-regulated genes in Drosophila. eLife 5, e19942 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR55] 55.Zhu, Z. et al. A gene expression program activated by neuronal inactivity. Preprint at bioRxiv10.1101/636878 (2024).

[CR56] 56.Collins, B., Kane, E. A., Reeves, D. C., Akabas, M. H. & Blau, J. Balance of activity between LN_vs and glutamatergic dorsal clock neurons promotes robust circadian rhythms in Drosophila. Neuron74, 706–718 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR57] 57.Worpenberg, L. et al. Ythdf is a N6-methyladenosine reader that modulates Fmr1 target mRNA selection and restricts axonal growth in Drosophila. EMBO J.40, e104975 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR58] 58.Bardoni, B., Schenck, A. & Mandel, J. L. A novel RNA-binding nuclear protein that interacts with the Fragile X mental retardation (FMR1) protein. Hum. Mol. Genet.8, 2557–2566 (1999). [DOI] [PubMed] [Google Scholar]

[CR59] 59.Napoli, I. et al. The Fragile X syndrome protein represses activity-dependent translation through CYFIP1, a new 4E-BP. Cell134, 1042–1054 (2008). [DOI] [PubMed] [Google Scholar]

[CR60] 60.Raji, J. I. & Potter, C. J. The number of neurons in Drosophila and mosquito brains. PLoS ONE16, e0250381 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR61] 61.Zhang, Y. Q. et al. Drosophila Fragile X-related gene regulates the MAP1B homolog Futsch to control synaptic structure and function. Cell107, 591–603 (2001). [DOI] [PubMed] [Google Scholar]

[CR62] 62.Rahman, R., Xu, W., Jin, H. & Rosbash, M. Identification of RNA-binding protein targets with HyperTRIBE. Nat. Protoc.13, 1829–1849 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR63] 63.Hagemann-Jensen, M. et al. Single-cell RNA counting at allele and isoform resolution using Smart-seq3. Nat. Biotechnol.38, 708–714 (2020). [DOI] [PubMed] [Google Scholar]

[CR64] 64.Sears, J. C., Choi, W. J. & Broadie, K. Fragile X mental retardation protein positively regulates PKA anchor Rugose and PKA activity to control actin assembly in learning/memory circuitry. Neurobiol. Dis.127, 53–64 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR65] 65.Sudhakaran, I. P. et al. FMRP and Ataxin-2 function together in long-term olfactory habituation and neuronal translational control. Proc. Natl. Acad. Sci. USA111, E99–E108 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR66] 66.Tang, X. et al. FMRP binds Per1 mRNA and downregulates its protein expression in mice. Mol. Brain16, 33 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR67] 67.Gibson, J. R., Bartley, A. F., Hays, S. A. & Huber, K. M. Imbalance of neocortical excitation and inhibition and altered UP states reflect network hyperexcitability in the mouse model of Fragile X syndrome. J. Neurophysiol.100, 2615–2626 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR68] 68.Polcownuk, S., Yoshii, T. & Ceriani, M. F. Decapentaplegic acutely defines the connectivity of central pacemaker neurons in Drosophila. J. Neurosci.41, 8338–8350 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR69] 69.Nicholson, L. et al. Spatial and temporal control of gene expression in Drosophila using the inducible GeneSwitch GAL4 system. I. Screen for larval nervous system drivers. Genetics178, 215–234 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR70] 70.Tsai, L. T., Bainton, R. J., Blau, J. & Heberlein, U. Lmo mutants reveal a novel role for circadian pacemaker neurons in cocaine-induced behaviors. PLoS Biol.2, e408 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR71] 71.Tolias, K. F. et al. The Rac1-GEF Tiam1 couples the NMDA receptor to the activity-dependent development of dendritic arbors and spines. Neuron45, 525–538 (2005). [DOI] [PubMed] [Google Scholar]

[CR72] 72.Wang, Q., Abruzzi, K. C., Rosbash, M. & Rio, D. C. Striking circadian neuron diversity and cycling of Drosophila alternative splicing. eLife10.7554/eLife.35618 (2018). [DOI] [PMC free article] [PubMed]

[CR73] 73.Luo, L., Liao, Y. J., Jan, L. Y. & Jan, Y. N. Distinct morphogenetic functions of similar small GTPases: Drosophila Drac1 is involved in axonal outgrowth and myoblast fusion. Genes Dev.8, 1787–1802 (1994). [DOI] [PubMed] [Google Scholar]

[CR74] 74.Kikuchi, M. et al. Disruption of a RAC1-centred network is associated with Alzheimer’s disease pathology and causes age-dependent neurodegeneration. Hum. Mol. Genet.29, 817–833 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR75] 75.Lim, C. et al. The novel gene twenty-four defines a critical translational step in the Drosophila clock. Nature470, 399–403 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR76] 76.Lim, C. & Allada, R. ATAXIN-2 activates PERIOD translation to sustain circadian rhythms in Drosophila. Science340, 875–879 (2013). [DOI] [PubMed] [Google Scholar]

[CR77] 77.Zhang, Y., Ling, J., Yuan, C., Dubruille, R. & Emery, P. A role for Drosophila ATX2 in activation of PER translation and circadian behavior. Science340, 879–882 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR78] 78.Huang, Y., McNeil, G. P. & Jackson, F. R. Translational regulation of the DOUBLETIME/CKIdelta/epsilon kinase by LARK contributes to circadian period modulation. PLoS Genet.10, e1004536 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR79] 79.Price, J. L. et al. double-time is a novel Drosophila clock gene that regulates PERIOD protein accumulation. Cell94, 83–95 (1998). [DOI] [PubMed] [Google Scholar]

[CR80] 80.Stanojevic, D., Small, S. & Levine, M. Regulation of a segmentation stripe by overlapping activators and repressors in the Drosophila embryo. Science254, 1385–1387 (1991). [DOI] [PubMed] [Google Scholar]

[CR81] 81.Rossi, F. M., Kringstein, A. M., Spicher, A., Guicherit, O. M. & Blau, H. M. Transcriptional control: rheostat converted to on/off switch. Mol. cell6, 723–728 (2000). [DOI] [PubMed] [Google Scholar]

[CR82] 82.Guilluy, C., Garcia-Mata, R. & Burridge, K. Rho protein crosstalk: another social network?. Trends Cell Biol.21, 718–726 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR83] 83.Owald, D. et al. A Syd-1 homologue regulates pre- and postsynaptic maturation in Drosophila. J. Cell Biol.188, 565–579 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR84] 84.De Rubeis, S. et al. CYFIP1 coordinates mRNA translation and cytoskeleton remodeling to ensure proper dendritic spine formation. Neuron79, 1169–1182 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR85] 85.Santini, E. et al. Reducing eIF4E-eIF4G interactions restores the balance between protein synthesis and actin dynamics in Fragile X syndrome model mice. Sci. Signal10.1126/scisignal.aan0665 (2017). [DOI] [PMC free article] [PubMed]

[CR86] 86.Chen, Y. et al. Cell-type-specific labeling of synapses in vivo through synaptic tagging with recombination. Neuron81, 280–293 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR87] 87.Lee, A. et al. Control of dendritic development by the Drosophila Fragile X-related gene involves the small GTPase Rac1. Development130, 5543–5552 (2003). [DOI] [PubMed] [Google Scholar]

[CR88] 88.Kandel, E. R. The molecular biology of memory storage: a dialogue between genes and synapses. Science294, 1030–1038 (2001). [DOI] [PubMed] [Google Scholar]

[CR89] 89.Xu, W., Rahman, R. & Rosbash, M. Mechanistic implications of enhanced editing by a HyperTRIBE RNA-binding protein. RNA24, 173–182 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR90] 90.Sone, M. et al. Synaptic development is controlled in the periactive zones of Drosophila synapses. Development127, 4157–4168 (2000). [DOI] [PubMed] [Google Scholar]

[CR91] 91.Shearin, H. K., Macdonald, I. S., Spector, L. P. & Stowers, R. S. Hexameric GFP and mCherry reporters for the Drosophila GAL4, Q, and LexA transcription systems. Genetics196, 951–960 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR92] 92.Buddika, K., Xu, J., Ariyapala, I. S. & Sokol, N. S. I-KCKT allows dissection-free RNA profiling of adult Drosophila intestinal progenitor cells. Development10.1242/dev.196568 (2021). [DOI] [PMC free article] [PubMed]

[CR93] 93.Cyran, S. A. et al. The Double-time protein kinase regulates the subcellular localization of the Drosophila clock protein Period. J. Neurosci.25, 5430–5437 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR94] 94.Heim, M., Blot, L. & Besse, F. An RNA-immunoprecipitation protocol to identify RNAs associated with RNA-binding proteins in cytoplasmic and nuclear Drosophila head fractions. STAR Protoc.3, 101415 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

A rapid and dynamic role for FMRP in the plasticity of adult neurons

Daniel G Gundermann

Seana Lymer

Jennifer Lennon

Justin Blau

Abstract

Introduction

Results

FMRP is required for s-LNv structural plasticity rhythms

Fig. 1. FMRP is required for the structural plasticity of s-LNvs.

FMRP dynamically regulates the daily structural plasticity of s-LNvs

Fig. 2. FMRP dynamically regulates the structural plasticity of s-LNvs.

FMRP levels in s-LNvs

Acutely changing Fmr1 levels do not change the molecular clock

Fig. 3. Transiently altering Fmr1 levels does not dramatically affect the s-LNv molecular clock or expression of an activity-dependent reporter gene.

Fmr1 overexpression does not dramatically change s-LNv excitability

Identifying FMRP targets in vivo in s-LNvs

Fig. 4. Fmr1 TRIBE-seq in s-LNvs.

sif and Rac1 regulate the expansion of s-LNv projections at dawn

Fig. 5. sif and Rac1 regulate the expansion of s-LNv projections at dawn.

A genetic interaction between sif and Fmr1

Fig. 6. Knocking down sif blocks the effect of Fmr1-shRNA on s-LNv projections.

Discussion

Fig. 7. Model to explain how FMRP rapidly controls s-LNv structural plasticity.

Methods

Fly rearing and transgenes

Gal4 and Gal80 lines

Reporter genes

UAS transgenes

UAS-shRNA lines

Immunofluorescence and imaging

Quantification of the 3D spread of s-LNv projections

Quantification of immunofluorescence

FMRP TRIBE-seq in s-LNvs

Analysis of TRIBE-seq data

Immunoprecipitation from fly heads

Supplementary information

Source data

Acknowledgements

Author contributions

Peer review

Peer review information

Data availability

Competing interests

Footnotes

Supplementary information

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases